diff --git a/marked/F/T-REC-F.1-199803-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.1-199803-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a7da83ecb82b51a4cef7694858eef2c35f1986ff --- /dev/null +++ b/marked/F/T-REC-F.1-199803-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d77a478c5b78d330a2a6abbd32e7d0ac053d8b9c1ffe8a96476214afa6ddf7ae +size 8196 diff --git a/marked/F/T-REC-F.1-199803-I_PDF-E/74448f9178da618d823e5a5dadc56fb5_img.jpg b/marked/F/T-REC-F.1-199803-I_PDF-E/74448f9178da618d823e5a5dadc56fb5_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9f688007af3c3c9e01941ef6a48de141ddb8a967 --- /dev/null +++ b/marked/F/T-REC-F.1-199803-I_PDF-E/74448f9178da618d823e5a5dadc56fb5_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2ecfef9c05ae310a2088d379cb916a42d0a0faae8231fae4e8263151f6960f64 +size 106474 diff --git a/marked/F/T-REC-F.10-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.10-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..64672153e39c851d9e16fdbc702e526c7ba7e8a7 --- /dev/null +++ b/marked/F/T-REC-F.10-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:454a9958ffe168868cb7d38a0eb24418dafe31a7a4245c992089b2316ac37d3e +size 7392 diff --git a/marked/F/T-REC-F.10-198811-I_PDF-E/raw.md b/marked/F/T-REC-F.10-198811-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..8080d0607ea0fe76b7306cc5ed96d39f17b1c565 --- /dev/null +++ b/marked/F/T-REC-F.10-198811-I_PDF-E/raw.md @@ -0,0 +1,63 @@ + + +![ITU logo](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +The logo of the International Telecommunication Union (ITU) features a globe with a lightning bolt superimposed on it, and the letters 'ITU' written across the center. + +ITU logo + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +**F.10** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +# **OPERATIONS AND QUALITY OF SERVICE TELEGRAPH SERVICES** --- + +## **CHARACTER ERROR RATE OBJECTIVE FOR TELEGRAPH COMMUNICATION USING 5 - UNIT START-STOP EQUIPMENT** + +**ITU-T Recommendation F.10** + +(Extract from the *Blue Book*) + +--- + +## NOTES + +1 ITU-T Recommendation F.10 was published in Fascicle II.4 of the *Blue Book*. This file is an extract from the *Blue Book*. While the presentation and layout of the text might be slightly different from the *Blue Book* version, the contents of the file are identical to the *Blue Book* version and copyright conditions remain unchanged (see below). + +2 In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +## Recommendation F.10 + +## CHARACTER ERROR RATE OBJECTIVE FOR TELEGRAPH COMMUNICATION USING 5-UNIT START-STOP EQUIPMENT + +The CCITT, + +*considering* + +- (a) that it would be useful to have a common standard for assessing the quality of telegraph communications, +- (b) that the character error rate [1] of a telegraph communication would be suitable for this purpose; +- (c) that a defined error rate objective should be established; +- (d) that all forms of current wideband transmission techniques have intrinsic characteristics that make some errors economically unavoidable; +- (e) that occasional bursts of errors occur (errors that are concentrated into a relatively short time interval, i.e. seconds), + +*unanimously declares* + +- (1) that the quality of service must be the same for telegraph communication in the public telegram service, the telex service and the leased circuit service; +- (2) that the objective, irrespective of transmission media and intervening equipment, should be an error rate not greater than 3 in 100 000 alphabetic telegraph signals transmitted; +- (3) that the error rate objective should have not less than a 95% probability; +- (4) that, in the determination of error rate, measurements should be made for relatively long time periods, i.e. for at least several hours (see Note 1); +- (5) that the effects of operator error rates (e.g. in the public telegram service) and of local ends and their terminations (e.g. teleprinters) should be excluded when determining the error rate. + +*Note 1* – The minimum time interval should include the busy hour and be of at least 12 hours duration. + +*Note 2* – The absolute value for the error rate objective requires further study. + +## Reference + +- [1] CCITT Definition: *Character error rate*, Vol. I, Fascicle I.3 (Terms and Definitions). \ No newline at end of file diff --git a/marked/F/T-REC-F.100-198811-I_PDF-E/raw.md b/marked/F/T-REC-F.100-198811-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..0f376cf79c2662f488054cf78c840308d72f51c6 --- /dev/null +++ b/marked/F/T-REC-F.100-198811-I_PDF-E/raw.md @@ -0,0 +1,83 @@ + + +![ITU logo](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +The logo of the International Telecommunication Union (ITU) features a globe with a lightning bolt superimposed on it, and the letters 'ITU' in a bold, sans-serif font. + +ITU logo + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +**F.100** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**OPERATIONS AND QUALITY OF SERVICE +TELEGRAPH SERVICES** + +--- + +**SCHEDULED RADIOCOMMUNICATION +SERVICES** + +**ITU-T Recommendation F.100** + +(Extract from the *Blue Book*) + +--- + +# NOTES + +1 ITU-T Recommendation F.100 was published in Fascicle II.4 of the *Blue Book*. This file is an extract from the *Blue Book*. While the presentation and layout of the text might be slightly different from the *Blue Book* version, the contents of the file are identical to the *Blue Book* version and copyright conditions remain unchanged (see below). + +2 In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +# **SCHEDULED RADIOCOMMUNICATION SERVICES** + +The CCITT, + +*unanimously declares* + +that the following rules should be adopted for the scheduled radiocommunication service. + +## **1 General** + +1.1 These rules are to be observed in the scheduled radiocommunication service in which radiocommunications are transmitted to one or more destinations. + +1.2 Only those senders and addressees who satisfy the provisions and conditions agreed between the Administrations concerned shall be allowed to participate in this service. + +## **2 Conditions of acceptance** + +2.1 Transmissions over the scheduled radiocommunication service must consist only of information and news relating to politics, commerce, etc., and must contain no communication of a private nature and no messages on behalf of third parties. They may include, however, brief indications as to how they should be passed on and to whom, provided that the time required to transmit them does not exceed 5 per cent of the total time taken to transmit the information or news, or, where appropriate, 5 per cent of the number of words of which the information or news consists. + +2.2 The sender shall communicate to the Administration of the country of emission the addresses of any intended recipient or recipients. + +2.3 The radiocommunications may be expressed either in plain language or in secret language, as decided by the Administrations of the countries of emission and of reception. In the absence of special arrangements between the Administrations concerned, the only languages authorized for plain language shall be French, one of the languages designated by the country of origin or one of the languages of one of the countries of reception. The Administrations of the countries of emission and of reception shall reserve to themselves the right to require the deposit of the codes used. + +2.4 The radio transmissions shall bear, as the address, an arbitrary word placed immediately before the text. + +## **3 Conditions of transmission** + +3.1 The Administration of the country of emission shall communicate to the other Administrations the address of any person residing in their territory for whom the radiocommunications are intended. It shall notify, in addition, in respect of each addressee, the date fixed for the first reception, the name of the emitting station and the address of the sender. The Administrations shall notify one another of any changes which occur in the number and the addresses of the senders and recipients. + +3.2 Where the services are operated by recognized private operating agencies, Administrations may authorize such agencies to communicate the notifications required under the preceding paragraph. + +3.3 Each Administration shall take, as far as is practicable, suitable measures to ensure that only the stations authorized for this special service of communication make use of the radiocommunications in question and then only of those intended for them. The provisions of the Convention relating to the secrecy of telecommunications shall apply to these radiocommunications. + +3.4 These radiocommunications shall be transmitted at fixed times. + +## **4 Conditions of reception** + +4.1 The Administration of the country of reception shall decide whether reception shall be permitted in its country. It may authorize the direct reception of such communications by the addressees designated by the sender or may itself make reception equipment available to the addressees to this end. It shall notify the Administration of the country of emission of the conditions under which reception takes place. + +## **5 Charging** + +5.1 The charge to be collected from the sender shall be fixed by the Administration of the country of emission. + +5.2 Apart from any charges levied for the establishment and working of private receiving stations or for the lease of receiver equipment, the addressees of these radiocommunications may be subjected by the Administration of their country to the payment of a receiver charge, the amount and method of assessment of which shall be decided on by that Administration. + +5.3 The charges for these radiocommunications shall not enter into the international accounts. \ No newline at end of file diff --git a/marked/F/T-REC-F.106-198811-I_PDF-E/raw.md b/marked/F/T-REC-F.106-198811-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..7bbe51353e56906cca7010f714ea8694a398182a --- /dev/null +++ b/marked/F/T-REC-F.106-198811-I_PDF-E/raw.md @@ -0,0 +1,69 @@ + + +![ITU logo](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +The logo of the International Telecommunication Union (ITU) features the letters 'ITU' in a bold, sans-serif font, superimposed on a stylized globe with intersecting lines. + +ITU logo + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.106** + +# **OPERATIONS AND QUALITY OF SERVICE TELEGRAPH SERVICES** --- + +## **OPERATIONAL PROVISIONS FOR PRIVATE PHOTOTELEGRAPH CALLS** + +**ITU-T Recommendation F.106** + +(Extract from the *Blue Book*) + +--- + +### NOTES + +1 ITU-T Recommendation F.106 (new numbering of Rec. F.80 *bis*) was published in Fascicle II.4 of the *Blue Book*. This file is an extract from the *Blue Book*. While the presentation and layout of the text might be slightly different from the *Blue Book* version, the contents of the file are identical to the *Blue Book* version and copyright conditions remain unchanged (see below). + +2 In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +### **OPERATIONAL PROVISIONS FOR PRIVATE PHOTOTELEGRAPH CALLS1)** + +#### **1 Field of application** + +1.1 These provisions apply to calls between private stations or between (outgoing) private stations and (incoming) public stations. + +1.2 The rules governing phototelegrams exchanged between private stations and public stations are embodied in Recommendations F.80 and D.80. + +#### **2 General** + +2.1 Private phototelegraph stations may be authorized by the Administration concerned to communicate with one another and to exchange phototelegrams with public stations. + +2.2 Administrations undertake to set up connections for such transmissions or to make suitable leased circuits available to private stations at their request. + +#### **3 Conditions of acceptance** + +3.1 Private stations may communicate with all (public and private) phototelegraph stations connected to the international phototelegraph network. + +3.2 Calls between private stations set up on the international phototelegraph network are allowed without any time limit. However, when telephone traffic is subjected to restrictions, the exchange of phototelegraph calls between private stations may be delayed or limited by agreement between the terminal centres concerned. + +3.3 Connections with a public station may be made available to a private station only during the business hours of the public phototelegraph office. However, the public station may not close until it has accepted all the phototelegrams that the private station has announced it wishes to hand in. + +3.4 The conditions under which booked phototelegraph calls are made available are specified in Recommendation F.82. + +#### **4 Charging** + +4.1 Charges for phototelegraph calls in continental and in intercontinental relations exchanged between private stations or from a private station to a public station are governed by Recommendation D.83. + +#### **5 Special services** + +5.1 The special urgent service shall be allowed in relations where it exists for telephone traffic. + +--- + +1) See also Recommendation D.81. \ No newline at end of file diff --git a/marked/F/T-REC-F.107-198811-I_PDF-E/raw.md b/marked/F/T-REC-F.107-198811-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..0428afaaabab64f9ebd60c2be1fb67769c0ec6f3 --- /dev/null +++ b/marked/F/T-REC-F.107-198811-I_PDF-E/raw.md @@ -0,0 +1,157 @@ + + +![ITU logo](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +The logo of the International Telecommunication Union (ITU) features the letters 'ITU' in a bold, sans-serif font, superimposed on a stylized globe with latitude and longitude lines. A lightning bolt symbol is positioned to the right of the globe. + +ITU logo + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.107** + +# **OPERATIONS AND QUALITY OF SERVICE TELEGRAPH SERVICES** --- + +## **RULES FOR PHOTOTELEGRAPH CALLS ESTABLISHED OVER CIRCUITS NORMALLY USED FOR TELEPHONE TRAFFIC** + +**ITU-T Recommendation F.107** + +(Extract from the *Blue Book*) + +--- + +## NOTES + +1 ITU-T Recommendation F.107 (new numbering of Rec. F.82) was published in Fascicle II.4 of the *Blue Book*. This file is an extract from the *Blue Book*. While the presentation and layout of the text might be slightly different from the *Blue Book* version, the contents of the file are identical to the *Blue Book* version and copyright conditions remain unchanged (see below). + +2 In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +## **RULES FOR PHOTOTELEGRAPH CALLS ESTABLISHED OVER CIRCUITS NORMALLY USED FOR TELEPHONE TRAFFIC** + +The CCITT, + +*considering* + +(a) that, in international phototelegraph communications, the occupation time of international telephone circuits often greatly exceeds the duration of the actual phototelegraph call; + +(b) that this drawback results *in part* from the inadequacy of existing rules on the setting up, supervision and clearing of phototelegraph calls over circuits normally used for telephone traffic, even if these circuits have been designated in advance as capable of carrying phototelegraph calls; + +(c) that phototelegraph communications between public stations on the one hand and public and private stations on the other, require close collaboration between the telegraph and telephone services of the various Administrations; + +(d) that, on the other hand, phototelegraph communications between private stations do not concern the telegraph services, although it is desirable for all phototelegraph calls between public stations, between public and private stations, and between private stations to be established in the same way, + +*unanimously declares* + +that the following rules should be applied to the provision of international phototelegraph connections. + +### **1 Field of application** + +1.1 The rules below define the operating procedure to be followed in the international phototelegraph service when phototelegraph calls are set up on circuits normally used for telephone traffic. They do not apply to phototelegraph transmissions on circuits in permanent use for that purpose or on leased circuits. + +1.2 These rules define the procedure applicable to the setting up, supervision and clearing of international phototelegraph calls: + +- between public stations; +- between a public and a private station; +- between private stations. + +### **2 General provisions** + +2.1 In relations where telephone circuits are used for both the phototelegraph service and the telephone service, the Administrations concerned shall assign by mutual agreement a certain number of circuits for phototelegraph transmissions, taking into account the usual requirements of both phototelegraphy and the telephone service. + +2.2 These circuits shall be specially marked at terminal exchanges and repeater stations with a view to the protection of the phototelegraph transmissions. + +2.3 The telephone circuits used for international phototelegraph transmissions shall, as far as practicable, be four-wire circuits. For phototelegraph transmission, they shall normally be disconnected from the switching equipment used for telephone calls. Interconnection of circuits for setting up phototelegraph calls should be four-wire-four-wire, as far as possible, both on the international and the national side. + +2.4 Administrations shall designate in each international phototelegraph terminal centre an authority responsible for the international phototelegraph calls. This authority is in a position to carry out, or cause to be carried out, all the operations necessary for the establishment of international phototelegraph calls. This authority shall henceforth be called the *international phototelegraph position (IPP)*. + +2.5 Administrations are recommended to centralize, as far as possible, in one place all the technical, operational and charging procedures necessary in an international centre when telephone circuits are used for phototelegraph calls. + +2.6 A booking for a phototelegraph call emanating from a public or private phototelegraph station is routed to (or arrives directly at) the IPP of the origin country responsible for setting up the international phototelegraph call that has been booked. This IPP then becomes the control IPP for establishing the call. + +### **3 Setting up, supervision and clearing of international phototelegraph calls** + +3.1 If the telephone service on the international circuit needed for setting up a phototelegraph call is operated with advance preparation, requests for phototelegraph calls rank in the order in which they are accepted among requests for telephone calls of the same category. + +3.2 In this case the control IPP advises the telephone office responsible for these circuits that a phototelegraph transmission is to take place. The control IPP agrees with the telephone service on the probable time at which the phototelegraph transmission will be taking place. + +3.3 The IPPs shall proceed as follows when setting up an international call. + +3.3.1 The control IPP transmits the following information as quickly as possible to the destination IPP: + +- designation of the transmitting station; +- designation of the destination station; +- probable time at which the phototelegraph call will take place; +- where necessary, indication of the subscriber responsible for the charges; and, in addition: + - a) for calls requested by public stations: + - category of phototelegram to be transmitted; + - date and time of handing-in of the phototelegram. + - b) for calls requested by private stations: + - category of call requested; + - date and time of the request. + +3.3.2 The destination IPP shall take the necessary steps to advise immediately the destination phototelegraph station by telephone that a phototelegraph transmission is about to take place. + +3.3.3 If the called phototelegraph station is in a position to receive the phototelegraph call at the time fixed, the destination IPP informs the control IPP. At the said time, the two IPPs take the necessary steps, in agreement with the telephone service, to establish the call. Care must be taken to avoid interrupting telephone calls in progress. + +3.3.4 If the called phototelegraph station is not in a position to receive the call at the time fixed, the destination IPP fixes the time when the transmission is to take place, taking into account the information received from the receiving phototelegraph station. It then communicates the time fixed to the control IPP, which informs the calling station. + +3.3.5 The control IPP then takes the necessary measures, in agreement with the telephone service, to establish the phototelegraph call between the stations concerned at the agreed time. + +3.4 If the telephone service involved is demand service, the outgoing IPP shall take an available circuit for the phototelegraph call, after ensuring that telephone calls in progress are not interrupted; it shall use this circuit to call the incoming IPP. + +3.4.1 To establish a phototelegraph call, it shall transmit the data mentioned under § 3.3.1 above to the incoming IPP, except for the probable time of the phototelegraph call. + +3.4.2 The incoming IPP shall take the necessary steps to advise immediately the called phototelegraph station by telephone that a phototelegraph transmission is about to take place. + +3.4.3 If the called phototelegraph station is in a position to receive the phototelegraph call immediately, the two IPPs shall straight away establish the necessary connection. + +3.4.4 If the called phototelegraph station is not in a position to receive the call immediately, the destination IPP fixes the time when the transmission is to take place, taking into account the information received from the receiving phototelegraph station. It then communicates the time fixed to the control IPP, which informs the calling station. The two immediately clear the international telephone circuit. + +3.4.5 At the time agreed upon, the outgoing IPP shall take the necessary steps to establish the phototelegraph call. + +3.5 The control IPP shall note the time when the phototelegraph call starts. The beginning of the call is the moment when the connection with the called station is offered to the caller. When the international circuit is extended towards a national PP at the caller's end, the latter shall determine the beginning of the call and indicate it to the IPP in its country when the call is cleared (see § 3.7 below). + +- 3.6 The control IPP supervises the transmission in progress: +- a) on the transmitting (go) path by means of a device enabling it to check, without risk of interference, that transmission is taking place; + - b) on the return path by means of a device enabling it to listen to service conversation from the phototelegraph receiving station. + +Intervention in the circuits should be avoided after the call has been established, unless such intervention has been requested by one of the IPPs or one of the phototelegraph stations connected. + +3.7 After consulting the receiving phototelegraph station, the calling phototelegraph station announces the end of the call either direct to its IPP, or, in the case of extension of an international circuit, to the national PP on which it depends. + +3.7.1 The latter must inform its IPP as quickly as possible, giving the time at which it received notice of the end of the call. The control IPP notes the end-of-transmission time and immediately communicates the notice announcing the end to the incoming IPP and, if necessary, to the transit IPP. + +3.7.2 The outgoing and incoming IPPs and any transit IPPs then take the necessary measures to restore the international circuit to the telephone service without delay. + +3.7.3 It is recommended that the called station should likewise announce the end of transmission so that the called station may be cleared more quickly. + +3.8 Unless the Administrations concerned decide to the contrary, the terminal IPPs do not come to an agreement on the chargeable duration, since this is determined by the control IPP. + +### **4 Special procedures for phototelegraph stations** + +4.1 For each phototelegram to be transmitted, the outgoing public station shall prepare a narrow tape comprising the preamble and address (and, if necessary, the signature and special service indications), unless these indications have been written on the phototelegram by the sender. This tape is transmitted with the phototelegram. + +4.2 As soon as the call is established, the interconnected phototelegraph stations proceed to adjust the equipment and to transmit, in accordance with the instructions of the receiving station, adopting the following order of operations: + +- a) if necessary, agreement on the index of cooperation and speed of transmission, then synchronization adjustment by means of the synchronization frequency; +- b) adjustment of the white level; +- c) adjustment of the black level; +- d) phasing; +- e) start; +- f) transmission. + +4.3 If the phototelegram is passed to a public station by a private station, the public station shall ask the private station, if necessary, for information regarding establishment of the preamble and conditions of delivery to the addressee. + +### **5 Faulty transmissions** + +5.1 In the case of fault conditions, the control IPP shall immediately make arrangements to clear the fault or make another circuit available. + +5.2 When, after completion of the call, it is seen that the transmission was faulty, the receiving phototelegraph station shall inform its IPP. If it so desires, the receiving phototelegraph station can make a new booking with its IPP for a phototelegraph call, in the manner defined in § 2.6, with the sending station. + +5.3 If the phototelegraph station that receives the faulty picture and books a new call is a private station, its attention should be drawn to the fact that both calls will be chargeable if the faults in the picture are not due to the telephone or telegraph services. \ No newline at end of file diff --git a/marked/F/T-REC-F.108-198811-I_PDF-E/raw.md b/marked/F/T-REC-F.108-198811-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..fd58c934c6f938a8b2b83d7975a1a2681bbfa100 --- /dev/null +++ b/marked/F/T-REC-F.108-198811-I_PDF-E/raw.md @@ -0,0 +1,139 @@ + + +![ITU logo](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +The logo of the International Telecommunication Union (ITU) features the letters 'ITU' in a bold, sans-serif font, superimposed on a stylized globe with intersecting lines. + +ITU logo + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +**F.108** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +# **OPERATIONS AND QUALITY OF SERVICE TELEGRAPH SERVICES** --- + +## **OPERATING RULES FOR INTERNATIONAL PHOTOTELEGRAPH CALLS TO MULTIPLE DESTINATIONS** + +**ITU-T Recommendation F.108** + +(Extract from the *Blue Book*) + +--- + +### NOTES + +- 1 ITU-T Recommendation F.108 (new numbering of Rec. F.85) was published in Fascicle II.4 of the *Blue Book*. This file is an extract from the *Blue Book*. While the presentation and layout of the text might be slightly different from the *Blue Book* version, the contents of the file are identical to the *Blue Book* version and copyright conditions remain unchanged (see below). +- 2 In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +## **OPERATING RULES FOR INTERNATIONAL PHOTOTELEGRAPH CALLS TO MULTIPLE DESTINATIONS1)** + +The CCITT, + +*considering* + +(a) that it seems advisable to provide for rules to which the Administrations may refer in the case where they decide to allow calls enabling several phototelegraph stations in different countries to receive a transmission simultaneously; + +(b) that for multiple calls of this kind *international dissemination* (i.e., a distribution of the transmission to different countries) is necessary and possibly *national dissemination* in the incoming countries (i.e., to the various receiving stations belonging to the same national network); + +(c) that the participants in a multiple call may be both public stations and private stations (primarily press agencies); + +(d) that press agencies are anxious to transmit pictures to their customers (newspaper offices) directly – without retransmission; + +(e) that in the case of transmission by series, the agencies also wish to add or disconnect certain customers between two successive transmissions; + +(f) that operation over the international part of the collective connection should not be help up by modifications in an incoming country; + +*considering further* + +(g) that dissemination equipment can be set up either in the offices of Administrations or on the premises of private enterprises; + +(h) that press agencies operate private phototelegraph networks for their own requirements; + +(i) that private enterprises should be allowed under certain conditions to use their own equipment and networks to effect an additional dissemination service to their customers; + +*and recognizing* + +(j) that satisfactory transmission of phototelegrams to multiple destinations can be obtained only if all the countries concerned employ a uniform mode of operation; + +*unanimously declares the view* + +that multiple calls may be allowed in the international service to enable several phototelegraph stations in different countries to receive a transmission from a transmitting station simultaneously. + +The rules below define the procedure to be followed for multiple destination calls. The conditions for ordinary connections that are part of the collective link are governed by Recommendations F.82 and D.83. + +--- + +1) See also Recommendation D.85. + +#### 1 Conditions of acceptance + +- 1.1 A multiple call may be requested for *primary dissemination* to various destination countries (international dissemination) and for *secondary dissemination* in the incoming countries (national dissemination). Simultaneously a national dissemination in the origin country can be combined with the international dissemination. +- 1.2 Private stations of a receiving country participating in a multiple call (generally press agencies) are allowed to effect simultaneously with reception *further dissemination* to other private stations (generally newspaper offices). +- 1.3 *Further dissemination* may be made either over a private network or over circuits of the public network made available by the Administration. In the latter case the further dissemination is allowed only within the country where the main receiving station is situated. +- 1.4 The primary dissemination (including, where necessary, dissemination in the origin country) and the secondary dissemination must be effected by the Administration concerned. +- 1.5 Any further dissemination over a private network shall in any case by the agency to which the network belongs. When circuits on the public network are used for dissemination to subordinate stations, the Administration concerned shall decide who should effect this further dissemination. For technical and operational reasons (see § 3.4 below), it is preferable that every further dissemination should be carried out by the agency using its own distribution equipment, provided that the equipment has been inspected and approved by the Administration. +- 1.6 In any case, the installation at the main station shall be such that the operator using it can so control operations as to ensure that the subordinate stations cannot enter into communication with the transmitting station. + +#### 2 Setting up and constitution of a multiple call connection + +- 2.1 To obtain a multiple call, the transmitting station shall contact the international phototelegraph position (IPP) of its country and indicate, separately for each country, which phototelegraph stations are to receive the simultaneous transmission. Requests should be made as soon as possible, and in any case not later than two hours before the transmission is to take place, to enable the Administrations concerned to make arrangements to set up the call. +- 2.2 The IPP of the origin country (see Figure 1/F.85, country A) shall inform the IPPs of the destination countries (countries B, C and D) giving the names of the called participants and the time at which the transmission is to take place (see § 5.2 below). So as not to surcharge the outgoing IPP, the incoming IPPs shall, at the same time, be asked to set up an international connection from their own end at the scheduled time. +- 2.3 The incoming IPPs shall arrange, where necessary, to make facilities available for national dissemination to participants in the multiple call (see § 5.3 below). If only one participant is designated in an incoming country (C), the international circuit shall be switched directly to the receiving station in question. +- 2.4 An intermediate connection with the international dissemination shall be requested in a transit country for destination countries to which it is not possible to connect directly for the dissemination from the origin country. +- 2.5 The outgoing IPP is the control IPP for the multiple call (i.e., as far as the main receiving stations). If an intermediate connection with the international dissemination service has been set up, the transit IPP acts as the sub-control IPP for the part of the multiple call farther down the circuit. +- 2.6 The control IPP notes the time when the multiple call begins and ends and the time and duration of any interruption or irregularity that may occur during transmission (for the purpose of calculating a refund). The start of the call is the moment at which the multiple connection has been placed at the disposal of the caller. The end of the call is the moment at which it is released by the transmitting station. +- 2.7 On clearing the multiple call, the control IPP must notify the incoming IPPs concerned of the time at which the call began and ended (to ensure conformity with the national dissemination charges). + +![Diagram of phototelegraph transmissions to multiple destinations across four countries (A, B, C, D). It shows various stations including Transmitter, PTT, Public, Ag 1, Ag 2, Ag 3, Ag 4, Control station, and Sub-control station. A legend at the bottom defines symbols for primary, secondary, and further dissemination, as well as station types and network types.](f9a14fbfecbd7d059226cc93677d721b_img.jpg) + +The diagram illustrates a network of phototelegraph transmissions across four countries: Country A (Country of origin), Country B, Country C, and Country D. In Country A, a Transmitter connects to a PTT, which then connects to a Public network. From the Public network, the signal is disseminated to Ag 1 (press agency) and Ag 2. Ag 1 connects to a Control station, which in turn connects to a Sub-control station. Ag 2 connects to Ag 3 in Country C. Ag 3 connects to Ag 4 in Country D. Ag 4 connects to a Two-wire network, which then connects to a private network. The diagram uses various symbols to represent different types of connections and stations. A legend at the bottom defines these symbols: + +- Primary dissemination (solid square) +- Secondary dissemination (open square) +- Further dissemination (square with 'pp') +- Chief receiving station (solid square) +- Subordinate station (open square) +- Public control station (square with 'P') +- Press agency (Ag) +- Private network (two parallel lines) + +Diagram of phototelegraph transmissions to multiple destinations across four countries (A, B, C, D). It shows various stations including Transmitter, PTT, Public, Ag 1, Ag 2, Ag 3, Ag 4, Control station, and Sub-control station. A legend at the bottom defines symbols for primary, secondary, and further dissemination, as well as station types and network types. + +FIGURE 1/F.108 +Phototelegraph transmissions to multiple destinations + +#### 3 Extension of a multiple call + +3.1 When a private dissemination is added to the multiple call, the Administration shall merely make the required circuits available to the user. The individual calls shall then be set up successively to the calling station and shall be charged from the moment they are made available. + +3.2 The routing to public network circuits of a phototelegraph transmission received on a leased circuit (see Ag 1 of country D) is not allowed. + +3.3 Each press agency is the control station for the connections with its customers. In secondary private dissemination (see Ag 1 in country D) the second connecting station becomes the sub-control station for the part farther down the circuit. + +3.4 Should an Administration ensure itself (see § 1.5 above) the further dissemination on the public network (country D), two separate distribution panels (I and II, country D) will be necessary to connect the incoming circuit separately to the main station (Ag 4) and to its group of customers. To prevent subordinate stations from coming in on the international part of the collective call, the interconnection between I and II must be made by a one-way channel. The private enterprise (Ag 4) must direct the operations for which it is responsible in the two sections of the further dissemination facilities. Since the Administration is responsible for making its dissemination facilities available at the right time, and in view of the operational difficulties, in particular resulting from any change required in this part of the further dissemination, this mode of operation is not recommended. + +#### **4 Speech circuit** + +4.1 The speech circuit is a leased telephone circuit that provides a direct connection between the site of the transmitting equipment and the control IPP. This type of connection expedites the procedures preparatory to the call and enables rapid action to be taken to overcome any difficulties experienced during transmission. It also allows timely notice to be given of the end of the multiple call and, further, it is a suitable means for determining exactly the chargeable duration of the call. + +4.2 The speech circuit may, however, be replaced by a telephone connection set up from the transmitting station over the general telephone network. + +#### **5 General provisions** + +5.1 The setting up of a multiple call may involve unforeseen delays in practice, particularly when intermediate connection centres are used in the international dissemination sector. For these reasons, it is impossible for Administrations to give any guarantee that a multiple call will be made available at a specific time. + +5.2 It is for the control IPP to estimate the time required to make the requested call available. It must inform all IPPs (incoming and transit, if need be) of the time at which transmission is to take place. + +5.3 The incoming IPPs must do their utmost to respect the scheduled time for setting up the international connection to the control (or sub-control) IPP. To avoid any delay in commencing the simultaneous transmission because of arrangements to be made in an incoming country, the national extension (dissemination or simple prolongation) should in all cases be ready before the international circuit is made available. + +5.4 If a called station is unable to accept the call at the scheduled time, the incoming IPP shall so inform the control IPP. It will be for the station requesting the multiple call to decide whether transmission should be held up until the station is ready to receive, or whether the station should be brought in later or whether it should be excluded from the call. In any case, the call charge begins from the time the caller is informed of the position. + +5.5 When it is not possible to set up an international or national connection required for the multiple call within six minutes after the appointed time, the control IPP shall draw the caller's attention to the difficulties that have arisen. Whatever course the caller decides to take, the part of the call that has taken place will be charged for. + +5.6 When a request is made to extend to other relations a call (single or multiple) that has already begun, it must be regarded as a new call request. This call will be made available to the caller as soon as it is set up, and will be charged for separately from that time. The caller must indicate the time at which it should be combined with the original call. \ No newline at end of file diff --git a/marked/F/T-REC-F.11-199110-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.11-199110-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9ff35fb80886d11ede195ab5a46c22ce14710d49 --- /dev/null +++ b/marked/F/T-REC-F.11-199110-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:30287180fe1dd821f9408c4cc58b365a422436acfaa9134dcb98f9f1fe8a50fe +size 7439 diff --git a/marked/F/T-REC-F.11-199110-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg b/marked/F/T-REC-F.11-199110-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a9c77affeca83cca6b4ab6698782ec655fcb989f --- /dev/null +++ b/marked/F/T-REC-F.11-199110-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9d7b028b7cee87f9ac0699f12018e3622d0345d0ac5cef0de878d4deaf50e11c +size 5724 diff --git a/marked/F/T-REC-F.11-199110-I_PDF-E/raw.md b/marked/F/T-REC-F.11-199110-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..e7c81e3dbde73929f00dc738324e6552f2696d6f --- /dev/null +++ b/marked/F/T-REC-F.11-199110-I_PDF-E/raw.md @@ -0,0 +1,73 @@ + + +![ITU logo: A globe with a lightning bolt and the letters ITU.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with a lightning bolt and the letters ITU. + +INTERNATIONAL TELECOMMUNICATION UNION + +**CCITT** + +**F.11** + +THE INTERNATIONAL +TELEGRAPH AND TELEPHONE +CONSULTATIVE COMMITTEE + +**TELEGRAPH AND MOBILE SERVICES +OPERATIONS AND QUALITY OF SERVICE** + +--- + +**CONTINUED AVAILABILITY +OF TRADITIONAL SERVICES** + +**Recommendation F.11** + +--- + +![CCITT logo: A globe with the letters CCITT.](6ed175c791b5e156d9c98a8dbcc3318c_img.jpg) + +CCITT logo: A globe with the letters CCITT. + +Geneva, 1991 + +# FOREWORD + +The CCITT (the International Telegraph and Telephone Consultative Committee) is a permanent organ of the International Telecommunication Union (ITU). CCITT is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The Plenary Assembly of CCITT which meets every four years, establishes the topics for study and approves Recommendations prepared by its Study Groups. The approval of Recommendations by the members of CCITT between Plenary Assemblies is covered by the procedure laid down in CCITT Resolution No. 2 (Melbourne, 1988). + +Recommendation F.11 was prepared by Study Group I and was approved under the Resolution No. 2 procedure on the 11th of October 1991. + +# --- CCITT NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication Administration and a recognized private operating agency. + +## **Recommendation F.11** + +## **CONTINUED AVAILABILITY OF TRADITIONAL SERVICES** + +## **1 Background** + +1.1 At the World Administrative Telegraph and Telephone Conference (Melbourne, 1988) considerable attention was given to ensuring a wide availability of telecommunication services with effective interconnection and interoperability. This implied a continuing availability of basic or traditional services in many international relations, despite the evolution of new or more sophisticated services. Article 4 of the International Telecommunication Regulation (Melbourne, 1988) expects Administrations to provide “a wide range of international telecommunication services” including “at least a form of telecommunication which is reasonably accessible to the public, including those who may not be subscribers to a specific telecommunication service”. + +1.2 Resolution No. 6 of the same World Administrative Telegraph and Telephone Conference (Melbourne, 1988) called for provisions to be made “to allow, through available communication infrastructures, continued availability of traditional services so as to enable effective communications on a world-wide basis”. In particular this concerned services in “certain rural areas and developing countries” that “may need to rely on existing widely available services for a relatively long period of time”. + +## **2 Scope** + +2.1 Pursuant to the International Telecommunication Regulations (Melbourne, 1988), this Recommendation lays down general principles on the continued availability of traditional services to facilitate global interconnection and interoperability of telecommunication facilities, pending the establishment of new services and to ensure the efficiency, usefulness and availability to the public of international telecommunication services. + +2.2 It addresses the coordination needed between Administrations to achieve that continued availability. Members should take any necessary steps to ensure that coordination. + +2.3 Within the framework of this Recommendation, “traditional” services include the international public telegram, telex and telephone services. + +## **3 Services principles** + +3.1 Pursuant to Article 4 (4.3) of the International Telecommunication Regulations, and subject to national coordination arrangements, Administrations should provide and maintain at least a minimum quality of service corresponding to the relevant CCITT Recommendations for the various telecommunication services that they offer to the public. + +3.2 Administrations may take action to develop, to improve or to rationalize this range of services, including an integration of elements of various services (interworking or accessing arrangements) or the introduction of new services. Provisions to ensure adequate interconnection and interoperability may be incorporated into existing Recommendations or covered by new Recommendations. In particular the aspects of quality of service (as perceived by the user) and tariff principles need to be covered in the appropriate Recommendations. + +3.3 When the Administration (or Administrations) in one country suspends an international service definitively, either in all or certain relations, it shall make arrangements to ensure to the greatest extent practicable that the service may effectively continue to operate in some acceptable form for other Administrations, pending the establishment of the relevant new or replacement services in their countries. In this context, the continuing service is the end result perceived by the user. The actual telecommunication infrastructure used to furnish that end service may vary. For example in some instances and for agreed classes of traffic the telemessage and/or bureaufax services could be considered to provide an adequate replacement for an international public telegram service. + +3.4 In principle, such replacement services or interworking arrangements should not unduly burden Administrations that may reasonably be considered to rely on the continued availability of the relevant traditional service. Consequently, the necessary changes in the technical facilities and in the operational procedures (including accounting) should be arranged to have minimum impact on Administrations needing to retain the traditional service. \ No newline at end of file diff --git a/marked/F/T-REC-F.110-199607-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.110-199607-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..bb91db91fb24cacf9135b70f6b9c8cc1b19350e2 --- /dev/null +++ b/marked/F/T-REC-F.110-199607-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d9b60eca3bb8a9408f91cab114d9faf702a2d3ab136c3bfee43f65e41f3cf4dd +size 8252 diff --git a/marked/F/T-REC-F.111-199103-I_PDF-E/raw.md b/marked/F/T-REC-F.111-199103-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..78098aa7410edd4816795eab41eab95df029bc5d --- /dev/null +++ b/marked/F/T-REC-F.111-199103-I_PDF-E/raw.md @@ -0,0 +1,89 @@ + + +![ITU logo: A globe with a lightning bolt and the letters ITU.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with a lightning bolt and the letters ITU. + +INTERNATIONAL TELECOMMUNICATION UNION + +**CCITT** + +**F.111** + +THE INTERNATIONAL +TELEGRAPH AND TELEPHONE +CONSULTATIVE COMMITTEE + +**TELEGRAPH AND MOBILE SERVICE +OPERATIONS AND QUALITY OF SERVICE** + +--- + +**PRINCIPLES OF SERVICE FOR +MOBILE SYSTEMS** + +**Recommendation F.111** + +--- + +![CCITT logo: A globe with the letters CCITT.](6ed175c791b5e156d9c98a8dbcc3318c_img.jpg) + +CCITT logo: A globe with the letters CCITT. + +Geneva, 1991 + +# FOREWORD + +The CCITT (the International Telegraph and Telephone Consultative Committee) is a permanent organ of the International Telecommunication Union (ITU). CCITT is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The Plenary Assembly of CCITT which meets every four years, establishes the topics for study and approves Recommendations prepared by its Study Groups. The approval of Recommendations by the members of CCITT between Plenary Assemblies is covered by the procedure laid down in CCITT Resolution No. 2 (Melbourne, 1988). + +Recommendation F.111 was prepared by Study Group I and was approved under the Resolution No. 2 procedure on the 11 of March 1991. + +## --- CCITT NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication Administration and a recognized private operating agency. + +# **PRINCIPLES OF SERVICE FOR MOBILE SYSTEMS** + +## **1 Scope** + +1.1 This Recommendation defines the principles applicable to international public correspondence service provided by mobile systems connected to the international telecommunication (fixed) networks. This Recommendation applies to aeronautical, land and maritime mobile-satellite and terrestrial systems. + +## **2 Introduction** + +2.1 There is an increased availability of mobile telecommunication systems on an international, regional and national basis. International mobile satellite systems offer telephone, data and telex services to aeronautical, land and maritime mobile users. Regional and national cellular systems, both analogue and digital, are in operation or at an advanced planning stage. Users of these mobile systems expect to be connected to the international telecommunications networks and be able to obtain international services available to fixed network customers. In order to ensure that both fixed and mobile user needs are met, principles of service for both satellite and terrestrial mobile systems when interworking with the international telecommunication networks are presented in this Recommendation. + +## **3 Services** + +3.1 The services offered by mobile systems (e.g. voice, data, telex) when interworking with the international telecommunication networks should, as far as practicable, be similar to those offered on fixed networks. + +## **4 Operation** + +4.1 When interworking between the mobile system and the fixed network the operational procedures for mobile system users should, wherever possible, be the same as for fixed networks. The same tones, announcements, service codes and signals, etc., used in the fixed network should be recognized and where appropriate returned by the mobile system. + +4.2 Any numbering plan for a national mobile system normally is part of the relevant national numbering plan of the country in which it is resident. + +4.3 For international mobile systems the numbering plan should either be derived from the national numbering plan of the participating countries or follow the international numbering plan applicable for the services to be provided. + +4.4 There should be mechanisms in place to cater for the needs of any call recording, billing and international accounting functions that might be required. + +## **5 Service between mobile systems** + +5.1 There should be no restrictions to establishing international calls between mobile terminals in different mobile systems via the relevant fixed network(s). For example, a user of a land mobile cellular terminal should be able to make an international telephone call to a mobile satellite terminal using the established international routing methods and procedures. + +## **6 Quality of service** + +6.1 The quality of service (e.g. bit error rate for data) of a mobile system should be such that the service experienced by the mobile user should be, where practical, similar to that of a user of a fixed network service. + +## **7 Service limitations** + +7.1 It is recognized that because of various reasons that mobile systems may not provide all the same services to their users as provided to users of the fixed telecommunication networks. Any limitation on service imposed by the mobile system on its users should, as far as practicable, not adversely affect the quality of service provided to fixed network users when the systems are interconnected. + +## **8 Special mobile services** + +8.1 It is recognized that mobile systems often provide services other than the conveyance of public correspondence. Mobile services may involve special services, e.g. priority and/or preemption treatment for safety of life messages, navigation, etc. + +8.2 Facilities for such special services are not necessarily available within the fixed networks. Where necessary Administrations should make any arrangements for the provision of these services (e.g. dedicated circuits), which is then a national matter outside the scope of CCITT Recommendations. + +8.3 CCITT Recommendations cover the effects of such special services on calls to/from the fixed networks. \ No newline at end of file diff --git a/marked/F/T-REC-F.112-198811-I_PDF-E/raw.md b/marked/F/T-REC-F.112-198811-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..97551cd72a891d2c15871725c781b2e3167f5763 --- /dev/null +++ b/marked/F/T-REC-F.112-198811-I_PDF-E/raw.md @@ -0,0 +1,57 @@ + + +![ITU logo](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +The logo of the International Telecommunication Union (ITU) features the letters 'ITU' in a bold, sans-serif font, superimposed on a stylized globe with intersecting lines. + +ITU logo + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +**F.112** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**OPERATIONS AND QUALITY OF SERVICE** + +**MOBILE SERVICE** + +--- + +**QUALITY OBJECTIVES FOR 50-BAUD +START - STOP TELEGRAPH TRANSMISSION +IN THE MARITIME MOBILE-SATELLITE +SERVICE** + +**ITU-T Recommendation F.112** + +(Extract from the *Blue Book*) + +--- + +# NOTES + +1 ITU-T Recommendation F.112 was published in Fascicle II.4 of the *Blue Book*. This file is an extract from the *Blue Book*. While the presentation and layout of the text might be slightly different from the *Blue Book* version, the contents of the file are identical to the *Blue Book* version and copyright conditions remain unchanged (see below). + +2 In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +## **Recommendation F.112** + +# **QUALITY OBJECTIVES FOR 50-BAUD START-STOP TELEGRAPH TRANSMISSION IN THE MARITIME MOBILE-SATELLITE SERVICE** + +The CCITT, + +*considering* + +- (a) that proper interworking of this telegraph transmission with the international telegraph services must be ensured; +- (b) CCITT Recommendation F.10 concerning character error rate for telegraph communications, + +*unanimously recommends* + +that, for the coast-earth station-to-mobile terminal and mobile terminal-to-coast-earth station links, sufficient margin should be included to overcome adverse propagation conditions. The objective should be that propagation conditions should not contribute any character errors for at least 95% of all calls with mobile terminals within the satellite service area. With the exception of blockage effects, propagation conditions should not contribute more than 8 errors in 100 000 characters with a 99% confidence level for mobile terminals at the edge of the service area. + +*Note* - This Recommendation corresponds to CCIR Recommendation 552. \ No newline at end of file diff --git a/marked/F/T-REC-F.113-199208-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.113-199208-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..094ad08ab3e4c1ce3aeef6280180677fb3dd1b65 --- /dev/null +++ b/marked/F/T-REC-F.113-199208-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e26379c3cb691f71e32b0eb8696c67adccc82d9f31a5becb97e336c35f665527 +size 8250 diff --git a/marked/F/T-REC-F.113-199208-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.113-199208-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..46b30147748f5f8a9d094c09a17555e4ad040af8 --- /dev/null +++ b/marked/F/T-REC-F.113-199208-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0500e6c8dd9deaa913c87391400693f6c0d505530a6dbb6c1894c3263bb2e15c +size 6886 diff --git a/marked/F/T-REC-F.113-199208-I_PDF-E/raw.md b/marked/F/T-REC-F.113-199208-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..b564c940eff9700497689ae94ca273d60b08d016 --- /dev/null +++ b/marked/F/T-REC-F.113-199208-I_PDF-E/raw.md @@ -0,0 +1,170 @@ + + +![ITU logo: A globe with a lightning bolt and the letters ITU.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with a lightning bolt and the letters ITU. + +INTERNATIONAL TELECOMMUNICATION UNION + +**CCITT** + +THE INTERNATIONAL +TELEGRAPH AND TELEPHONE +CONSULTATIVE COMMITTEE + +**F.113** + +(08/92) + +**TELEGRAPH AND MOBILE SERVICES +OPERATIONS AND QUALITY OF SERVICE** + +--- + +**SERVICE PROVISIONS FOR +AERONAUTICAL PASSENGER +COMMUNICATIONS SUPPORTED +BY MOBILE-SATELLITE SYSTEMS** + +**Recommendation F.113** + +--- + +![CCITT logo: A globe with the letters CCITT.](84a1d09fb489061482111515543b60dc_img.jpg) + +CCITT logo: A globe with the letters CCITT. + +Geneva, 1992 + +# FOREWORD + +The CCITT (the International Telegraph and Telephone Consultative Committee) is a permanent organ of the International Telecommunication Union (ITU). CCITT is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The Plenary Assembly of CCITT which meets every four years, establishes the topics for study and approves Recommendations prepared by its Study Groups. The approval of Recommendations by the members of CCITT between Plenary Assemblies is covered by the procedure laid down in CCITT Resolution No. 2 (Melbourne, 1988). + +Recommendation F.113 was prepared by Study Group I and was approved under the Resolution No. 2 procedure on the 4th of August 1992. + +## --- CCITT NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized private operating agency. + +# **SERVICE PROVISIONS FOR AERONAUTICAL PASSENGER COMMUNICATIONS SUPPORTED BY MOBILE-SATELLITE SYSTEMS** + +(1992) + +## **1 Introduction** + +### **1.1 Scope** + +1.1.1 This Recommendation makes provision for the operational and Quality of Service requirements for both air-to-ground and ground-to-air aeronautical passenger communications mobile-satellite services. This Recommendation follows the principles laid down in Recommendation F.111. + +1.1.2 Operation in the AMS(R)S1) band must meet the standards established by the International Civil Aviation Organization (ICAO). This includes the technical specifications, preemptive and priority functions, as defined in ICAO Annex 10, Aeronautical Telecommunications. + +1.1.3 This Recommendation only covers aeronautical passenger communications. Other service and operational provisions are not covered by this Recommendation. + +1.1.4 Aeronautical passengers communications (APC), a form of communications, unrelated to safety and regularity of flight or the aircraft operating agency, which is at the disposal of the passenger and crew of an aircraft in flight status (see Note) on a controlled basis subject to flight safety requirements. Aeronautical passenger communications is one form of public correspondence. + +*Note* – An aircraft may be considered to be in flight status, while on the ground waiting for departure, while taxiing for take-off, while flying, while taxiing following landing, and while waiting for passengers to leave the aircraft. + +## **2 System description** + +2.1 An aeronautical mobile-satellite system provides two-way voice and data communication with aircraft operating within the field-of-view of a satellite. + +An aeronautical mobile-satellite system comprises of three basic elements: + +- *Aircraft earth stations (AES)* – Transmitting and receiving stations located on the aircraft capable of communicating with the satellite and interfacing with terminals within the aircraft. +- *Aeronautical ground earth stations (GES)* – Fixed terrestrial transmitting and receiving stations communicating with the satellite and interfacing with terrestrial public and private telecommunication networks. For the purpose of this Recommendation the acronym GES will be used to denote an aeronautical ground earth station to avoid confusion over identical acronyms, although it is recognized that *Radio Regulation 77* defines this as an aeronautical earth station (see Note). + +*Note* – Fixed terrestrial transmitting and receiving stations jointly providing a combination of maritime, land and aeronautical mobile services are commonly referred to as land earth stations (LES) in the INMARSAT system. + +- *Satellite* – Providing communication between AESs and GESs. + +--- + +1) AMS(R)S Aeronautical mobile-satellite (route) service. + +2.2 The system may support: + +- i) aeronautical passenger communications (APC); +- ii) aeronautical administrative communications (AAC); +- iii) aeronautical operational control (AOC) communications; and +- iv) air traffic services (ATS) communications. + +2.3 Terrestrial networks that may be interconnected with aeronautical mobile satellite systems are: + +- public telephone, telex, packet switched and circuit switched data networks; +- dedicated private networks for aeronautical telecommunication services. + +## 3 Service description + +3.1 Aeronautical passenger communications support the following services: + +- telephone; +- telex; +- packet switched data transmission; +- circuit switched data transmission; +- facsimile. + +Administrations may provide one or more of these services. + +## 4 Air-to-ground calls + +4.1 *General* + +4.1.1 Fully automatic service operation should be employed, where practicable, for all services. + +4.1.2 The AES and the GES provide preemptive and priority functions for the control of air-to-ground calls, in accordance with the International Civil Aviation Organization (ICAO) standards and recommended practices and the ITU Radio Regulations. + +4.1.3 Provision shall be made for the collection of information relevant to billing and international accounting functions. + +4.1.4 When a credit card system is employed, the relevant details of the credit card must also be recorded for validity check purposes. + +4.1.5 Procedures for call establishment should be in accordance with CCITT Recommendations, e.g. Recommendations E.216, F.126 and X.350 are applicable in respect to the INMARSAT-aeronautical system. + +4.2 *Telephone service* + +4.2.1 The service requirements and interworking procedures with the telephone service as defined in Recommendation E.105, are for further study. + +4.3 *Telex service* + +4.3.1 The service requirements and interworking procedures with the telex service as defined in Recommendation F.60, are for further study. + +### 4.4 *Data transmission services* + +#### 4.4.1 *Packet switched data transmission services* + +The service attributes, facilities, numbering and selection procedures are for further study. + +#### 4.4.2 *Circuit switched data transmission services* + +The service attributes, facilities, numbering and selection procedures are for further study. + +## **5 Ground-to-air calls** + +### 5.1 *General* + +5.1.1 The unique aircraft number identification is standardized by ICAO. A unique AES mobile number based upon this aircraft identification number should be used regardless of the types of service provided, on the radio link between the ground and aircraft. + +5.1.2 The AES and GES provide preemptive and priority functions for the control of ground-to-air calls, in accordance with ICAO standards and recommended practices and the ITU Radio Regulations. + +5.1.3 Numbering plans and call selection procedures should be in accordance with CCITT Recommendations, e.g. Recommendations E.215, F.125 and X.121, applicable to the INMARSAT-aeronautical system. + +5.1.4 Non-automatic operation for ground-to-air communication is for further study. + +### 5.2 *Telephone service* + +5.2.1 The service requirements are for further study. + +### 5.3 *Telex service* + +5.3.1 The service requirements are for further study. + +### 5.4 *Data transmission services* + +#### 5.4.1 *Packet switched data transmission services* + +5.4.1.1 The service requirements are for further study. + +#### 5.4.2 *Circuit switched data transmission services* + +5.4.2.1 The service requirements are for further study. \ No newline at end of file diff --git a/marked/F/T-REC-F.115-199502-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.115-199502-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7e86be364c2e70dc0688295c12b1f0dfde871be1 --- /dev/null +++ b/marked/F/T-REC-F.115-199502-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9f5a9bf4a6fdd6b2b1432e4a58154012788a8fe0b6a49989a97624d32d6d507c +size 8198 diff --git a/marked/F/T-REC-F.115-199502-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg b/marked/F/T-REC-F.115-199502-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ac26e12812fc76e23c23261ef3a3333b797575ba --- /dev/null +++ b/marked/F/T-REC-F.115-199502-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d2988a65a770697ef26b060f2f8e745a770a842a2b03bcf511ae935f4fac6ae8 +size 92548 diff --git a/marked/F/T-REC-F.115-199502-I_PDF-E/raw.md b/marked/F/T-REC-F.115-199502-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..2a0c941f64c7fa60747d18038844286ce5cf8c70 --- /dev/null +++ b/marked/F/T-REC-F.115-199502-I_PDF-E/raw.md @@ -0,0 +1,388 @@ + + +![ITU logo](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +The logo of the International Telecommunication Union (ITU) features the letters 'ITU' in a bold, sans-serif font, superimposed on a stylized globe with intersecting lines. + +ITU logo + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +**F.115** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +(02/95) + +**OPERATIONS AND QUALITY OF SERVICE +MOBILE SERVICE** + +--- + +**SERVICE OBJECTIVES AND PRINCIPLES +FOR FUTURE PUBLIC LAND MOBILE +TELECOMMUNICATION SYSTEMS** + +**ITU-T Recommendation F.115** + +(Previously "CCITT Recommendation") + +--- + +# FOREWORD + +The ITU-T (Telecommunication Standardization Sector) is a permanent organ of the International Telecommunication Union (ITU). The ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Conference (WTSC), which meets every four years, establishes the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics. + +The approval of Recommendations by the Members of the ITU-T is covered by the procedure laid down in WTSC Resolution No. 1 (Helsinki, March 1-12, 1993). + +ITU-T Recommendation F.115 was prepared by ITU-T Study Group 1 (1993-1996) and was approved under the WTSC Resolution No. 1 procedure on the 21st February 1995. + +# --- NOTE + +In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +# CONTENTS + +| | Page | +|----------------------------------------------------------------|-------------| +| 1 Introduction ..... | 1 | +| 2 Scope ..... | 1 | +| 3 References ..... | 1 | +| 3.1 CCITT and ITU-T Recommendations ..... | 1 | +| 3.2 ITU-R Recommendations ..... | 2 | +| 4 Definitions ..... | 2 | +| 5 Service principles ..... | 3 | +| 6 Overall service objectives ..... | 4 | +| 7 Service requirements ..... | 4 | +| 7.1 General service requirements ..... | 4 | +| 7.2 General access requirements ..... | 4 | +| 7.3 Numbering considerations ..... | 5 | +| 8 Services supported ..... | 5 | +| 8.1 Introduction ..... | 5 | +| 8.2 Common FPLMTS services ..... | 5 | +| 8.3 FPLMTS specific services ..... | 5 | +| 9 Quality of Service ..... | 5 | +| 9.1 General ..... | 6 | +| 9.2 Speech quality ..... | 6 | +| 9.3 Traffic engineering aspects ..... | 7 | +| Appendix I – Recommendations relevant to common services ..... | 7 | + +# **SUMMARY** + +The introduction of third generation digital mobile systems known as FPLMTS, which are scheduled to start service around the year 2000 subject to market considerations, will result in mobile communication becoming available to mass markets for the purpose of global communication. This Recommendation defines the service objectives and principles for FPLMTS which will enable the development of FPLMTS into the future. These objectives and principles are produced to aid the intercommunication of services provided over both FPLMTS and the fixed telecommunication network and to provide guidance for further development of FPLMTS. + +# **SERVICE OBJECTIVES AND PRINCIPLES FOR FUTURE PUBLIC LAND MOBILE TELECOMMUNICATION SYSTEMS** + +*(Geneva, 1994)* + +# **1 Introduction** + +**1.1** Future Public Land Mobile Telecommunication Systems (FPLMTS) will provide access, by means of one or more radio links, to a range of telecommunication services which are specific to mobile users and to telecommunication services supported by the fixed telecommunication network. FPLMTS are third generation mobile systems which may be terrestrial or satellite based and are scheduled to start service around the year 2000 subject to market considerations. They may be utilized as stand alone mobile systems or as part of the fixed network. The frequency bands 1885-2025 and 2110-2200 MHz have been identified by Radio Regulation WARC-92 (Footnote 746A) for use on a worldwide basis for implementation of FPLMTS. The high level description of FPLMTS and the framework of services are covered in ITU-R Recommendations M.687-1 and M.816 respectively. + +**1.2** Key features of FPLMTS are expected to be: + +- applicability to land mobile, aeronautical mobile, and maritime mobile environments; +- incorporation of a variety of systems; +- high degree of commonality of design; +- compatibility of services within FPLMTS and with the fixed network; +- quality of service comparable to that of the fixed network; +- security of communications comparable to that of the fixed network; +- worldwide intercommunication and interoperability; +- use of a lightweight, pocket-sized terminal. + +**1.3** FPLMTS services may be divided into those which are specific to FPLMTS and those which are common to the fixed network. + +# **2 Scope** + +The purpose of this Recommendation is to identify and define the service objectives and principles for the Future Public Land Mobile Telecommunication Systems (FPLMTS) and to provide guidance for the further development of FPLMTS. This Recommendation is part of a family of Recommendations dealing with FPLMTS which are defined in clause 3. + +# **3 References** + +FPLMTS are defined by a set of related (FPLMTS and non-FPLMTS specific) Recommendations. Recommendations of general interest are listed in this clause while Recommendations of a more specific nature are referenced with the applicable text. + +## **3.1 CCITT and ITU-T Recommendations** + +- CCITT Recommendation E.164 (1991), *Numbering plan for the ISDN era*. +- ITU-T Recommendation E.750 (1993), *Introduction to E.750-Series of Recommendations on traffic engineering aspects of mobile networks*. +- ITU-T Recommendation E.751 (1993), *Reference connections for traffic engineering of land mobile networks*. + +- ITU-T Recommendation E.770 (1993), *Land mobile and fixed network interconnection traffic grade of service concept*. +- ITU-T Recommendation E.771 (1993), *Network grade of service parameters and target values for circuit-switched land mobile services*. +- CCITT Recommendation F.111 (1991), *Principles of service for mobile systems*. +- ITU-T Recommendation F.850 (1993), *Principles of Universal Personal Telecommunication (UPT)*. +- ITU-T Recommendation F.851 (1995), *Universal Personal Telecommunication (UPT) – Service description (service set one)*. +- ITU-T Recommendation G.114 (1993), *One-way transmission time*. +- ITU-T Recommendation G.174 (1994), *Transmission performance objectives for terrestrial digital wireless systems using portable terminals to access the PSTN*. +- ITU-T Recommendation I.114 (1993), *Vocabulary of terms for Universal Personal Telecommunication*. + +## 3.2 ITU-R Recommendations + +Recommendation ITU-R M.687-1 (Future Public Land Mobile Telecommunication Systems) is the core document for the definition of FPLMTS. + +- ITU-R Recommendation M.687-1, *Future Public Land Mobile Telecommunication Systems (FPLMTS)*. +- ITU-R Recommendation M.816, *Framework for services supported on Future Public Land Mobile Telecommunication Systems (FPLMTS)*. +- ITU-R Recommendation M.817, *FPLMTS network architectures*. +- ITU-R Recommendation M.818, *Satellite operation within FPLMTS*. +- ITU-R Recommendation M.819-1, *FPLMTS for developing countries*. +- ITU-R Recommendation M.[Doc. 8/154(+ Corr.1)], *Security principles for FPLMTS*. +- ITU-R Recommendation M.[Doc. 8/155], *Speech and voiceband data performance requirements for FPLMTS*. +- Recommendation FPLMTS.TMLG, *Vocabulary of terms for FPLMTS*. +- Recommendation FPLMTS.RSEL, *Procedures for selection of radio transmission technologies for FPLMTS*. +- Recommendation FPLMTS.NMGM, *Framework of FPLMTS network management*. +- ITU-R Recommendation M.1034, *Requirements for the radio interface(s) for FPLMTS*. +- ITU-R Recommendation M.1035, *Framework for the radio interface(s) and radio subsystem functionality for FPLMTS*. +- ITU-R Recommendation M.1036, *Spectrum considerations for implementation of Future Public Land Mobile Telecommunication Systems (FPLMTS) in the bands 1885-2025 MHz and 2110-2200 MHz*. + +NOTE – ITU-R Recommendation numbers will be checked and updated prior to the printing of this Recommendation. + +# 4 Definitions + +For the purposes of this Recommendation, the following specifications apply: + +**4.1 Future Public Land Mobile Telecommunication Systems (FPLMTS):** FPLMTS are third generation mobile systems, which are scheduled to start service around the year 2000 subject to market considerations. They may be utilized as stand alone mobile systems or as part of the fixed network. They are systems which will provide telecommunication services to mobile or stationary users by means of one or more radio links. The frequency bands 1885-2025 and 2110-2200 MHz have been identified by Radio Regulation WARC-92 (Footnote 746A) for use on a + +worldwide basis for implementation of FPLMTS. This mobility will be unrestricted in terms of location within the radio coverage area. FPLMTS will extend the telecommunication services of the fixed network to these users over wide geographic areas, subject to constraints imposed by spectrum allocation and radio propagation. In addition, FPLMTS will support a range of services particular to mobile radio systems (see Recommendation ITU-R M.687-1). Provision of services may be limited by the capabilities of the involved terminals and network. + +**4.2 terminal mobility:** The ability of the terminal to access telecommunication services from different locations and while in motion, and the capability of the network to identify and locate that terminal (see Recommendation I.114). + +**4.3 FPLMTS user mobility:** FPLMTS user mobility is a feature which enables a FPLMTS user to transfer his/her identity between FPLMTS mobile terminals. + +**4.4 Universal Personal Telecommunication (UPT):** UPT enables access to telecommunication services while allowing personal mobility. It enables each UPT user to participate in a user-defined set of subscribed services and to initiate and receive calls on the basis of a personal, network-transparent, UPT number across multiple networks at any terminal, fixed or mobile, irrespective of geographic location, limited only by terminal and network capabilities and restrictions imposed by the network operator (see Recommendation F.850). + +**4.5 privacy:** The right of individuals to control or influence what information related to them may be collected or stored and by whom and to whom that information may be disclosed. + +**4.6 mobile terminal:** For the purposes of this Recommendation this term includes both the radiocommunication equipment and specific telecommunication service equipment, e.g. telephone (handset) and fax machine. + +# **5 Service principles** + +**5.1** FPLMTS are intended to provide users with services which will be available globally. The provision of these services may involve cooperation between a number of operating entities (e.g. service providers, mobile network operators and fixed network operators). + +**5.2** At any geographical location there may be more than one network operator or service provider. + +**5.3** As an aid to users, it is desirable that any service offered should have a similarity of access interaction and operational procedures on a global basis. + +**5.4** FPLMTS will integrate a terrestrial component and a satellite-based component (see Recommendations ITU-R M.687-1 and M.818). Satellite based services may however have some properties different from those provided terrestrially. + +**5.5** To satisfy the need for call detail recording between different countries and service providers, there may be a requirement for FPLMTS to be able to include, in call records, the geographical location of the calling and called parties. In order to maintain the privacy of the FPLMTS user as the called party, the geographic location of the FPLMTS user should not appear on any bill or statement submitted to any party other than the FPLMTS subscriber or appropriate service provider. See 7.1.5. + +**5.6** Interworking of FPLMTS with the fixed network should be in accordance with Recommendation F.111, Principles of service for mobile systems. + +**5.7** Terminal mobility which enables a user to be in motion during all aspects of a call give rise to specific service requirements whether the service being used is a FPLMTS specific service or one of the services of the fixed networks. It will require authentication and validation procedures to verify the right of the user to access the service. The recording of call details may capture an indication of user location, network access point and called party location, in addition to other billing, charging and accounting information. See 5.5 concerning FPLMTS user privacy. + +# **6 Overall service objectives** + +- 6.1** To provide FPLMTS users with continuous availability of their subscribed telecommunication services regardless of their roaming situation, geographic location, or network access, on a global basis but subject to the terms and conditions under which the visited network offers service to roaming users. +- 6.2** To provide to FPLMTS users, where feasible, services which are identical, or similar, to those offered by the fixed telecommunication networks. (Refer to 7.1/F.111.) +- 6.3** To facilitate introduction of new customised services, in particular for multimedia applications, which allow the use of modular reusable service components in accordance with ITU-T Recommendations. +- 6.4** To provide telecommunications of high quality that is comparable to that of the fixed network. +- 6.5** To accommodate a variety of mobile terminals ranging from those which are small enough to be carried to those which are mounted in vehicles or otherwise used in conjunction with a mobile network. +- 6.6** To provide commonality of service presentation and operation (the means by which the service and the user interact) among FPLMTS. +- 6.7** To provide for forward compatibility with developing telecommunication services. +- 6.8** To support a range of telecommunication services in various configurations such as point-to-point, multipoint and point-to-multipoint. +- 6.9** To provide for backward compatibility with services provided by existing mobile systems to the extent practicable. + +# **7 Service requirements** + +## **7.1 General service requirements** + +- 7.1.1** Provide for service availability such that the users of one service provider may request service in real time from another service provider without the need for prior arrangements or agreements on both an international and national basis, where national regulations and implementations permit. +- 7.1.2** Provide for the secure and timely communication of billing, charging and accounting information. +- 7.1.3** Provide for the secure authentication and validation of mobile terminals and users. +- 7.1.4** Provide for appropriate levels of security for telecommunication services to ensure that the user perceives the security of the communication as being comparable to that of the fixed network. +- 7.1.5** Provide for optional location privacy at the request of the user. +- 7.1.6** Provide for the secure and accurate creation of appropriate data records to facilitate charging, billing, accounting and network management functions in accordance with the relevant D- and M-Series Recommendations. +- 7.1.7** Provide a capability for UPT support (i.e. personal mobility aspects, see Recommendation F.851). +- 7.1.8** Provide for FPLMTS terminal mobility and FPLMTS user mobility both within and among FPLMTS. + +## **7.2 General access requirements** + +- 7.2.1** Features and facilities of FPLMTS should, to the extent feasible, enable mobile users of one service provider to access and utilize the system of a service provider in another country, on an automatic basis when visiting that country. Such international, automatic roaming would be subject to national regulation, international agreement and any limitations imposed by service providers. + +**7.2.2** In a similar manner to international operation, access to FPLMTS should be available in the maritime and aeronautical environments to the extent permitted by national and international regulatory authorities, the characteristics of the system used and any limitations imposed by service providers. + +**7.2.3** Access to FPLMTS satellite services should be available to the extent permitted by national and international regulatory authorities, the characteristics of the system used and any limitations imposed by service providers. + +**7.2.4** The FPLMTS support of access to UPT will have the objective of maintaining UPT's common presentation functionality to users. + +## **7.3 Numbering considerations** + +**7.3.1** A FPLMTS user requires a number formatted in accordance with Recommendation E.164 by which he/she can be addressed by callers. + +**7.3.2** FPLMTS users may participate in closed user groups utilizing non-E.164 numbers as closed user group identities. + +# **8 Services supported** + +## **8.1 Introduction** + +The services to be supported by FPLMTS reflect the needs of users to communicate in a mobile environment. + +Each FPLMTS provider will determine what services will be offered to its users. However, it is recommended that the following be offered: + +- i) voice communication with other users of fixed or mobile terminals (telephone service) connected to private or public networks; and +- ii) data communications with other fixed or mobile terminals. + +It should be noted that this does not prevent other services being offered. + +Figure 1 illustrates the service relationships. + +## **8.2 Common FPLMTS services** + +The services which are applicable to the common area shown in Figure 1 are those of the fixed network. The definitions, service and operational provisions relating to these services may be found in the relevant E-, F- and I-Series Recommendations as listed in Appendix I. This list is not intended to be exclusive. + +## **8.3 FPLMTS specific services** + +The following services are applicable to the FPLMTS specific area as shown in the diagram above. This list is not intended to be exclusive. + +- Location services. +- Paging service. +- Short message services. +- Traffic and navigation services. + +# **9 Quality of Service** + +The Quality of Service offered by the FPLMTS should be comparable to that of the PSTN/ISDN and public data networks as far as practicable. [See ITU-R M.(Doc. 8/155), Recommendation FPLMTS.PRQ and Recommendation G.174.] + +NOTE – ITU-R Recommendation numbers will be checked and updated prior to the printing of this Recommendation. + +![Conceptual model of network infrastructures and their services relationship diagram](a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg) + +The diagram illustrates the conceptual model of network infrastructures and their services relationship. At the top, three boxes represent service types: 'Fixed network specific services', 'Common services', and 'FPLMTS specific services'. A horizontal dashed line separates these from the infrastructure layer below. To the right of this line, the text 'ITU service Recommendations' is above 'ITU technical Recommendations'. Below the dashed line, two large overlapping ovals represent 'Fixed network infrastructure' and 'FPLMTS infrastructure'. Lines connect the service boxes to these infrastructure ovals. Below the infrastructure ovals, various devices are shown: 'Fixed links (Physical or radio)' (represented by a telephone icon), a 'PBX' (Private Branch Exchange) connected to multiple telephones and a computer, and 'Mobile Radio links' (represented by a zigzag line) connecting to a telephone and a computer. The text 'T0103790-94/d01' is located in the bottom right corner of the diagram area. + +Conceptual model of network infrastructures and their services relationship diagram + +FIGURE 1/F.115 +**Conceptual model of network infrastructures + and their services relationship** + +## 9.1 General + +Consideration will need to be given to maintaining Quality of Service during handover between cells and to any implications in relation to cell size, frequency requirements, protocols, etc. + +To maintain the Quality of Service for data services, additional error control facilities for the radiocommunication path may be required (e.g. for facsimile machines). + +## 9.2 Speech quality + +The quality of speech services in FPLMTS should be comparable to that of the PSTN/ISDN. Speech services are sensitive to delay and this needs to be taken into consideration (see Recommendation G.114). + +The quality of speech supported within FPLMTS should be nominally that recommended in the P-Series Recommendations. However, it is acknowledged that in some applications where low bit rate coding is employed, some minimal degradation of speech quality may occur. + +## 9.3 Traffic engineering aspects + +As per the E.700-Series Recommendations on traffic engineering of mobile networks. + +# Appendix I + +## Recommendations relevant to common services + +(This appendix does not form an integral part of this Recommendation) + +As noted in 8.2, this list is not intended to be exclusive. + +- F.110 *Operational provisions for the maritime mobile service.* +- F.111 *Principles of service for mobile systems.* +- F.113 *Service provision for aeronautical passenger communications supported by mobile satellite systems.* +- F.122 *Operational procedures for the maritime satellite data transmission service.* +- F.160 *General operational provisions for the international public facsimile services.* +- F.180 *General operational provisions for the international public facsimile service between subscribers stations.* +- F.182 *Operational provisions for the international public facsimile service between subscriber stations with groups 2 and 3 facsimile machines.* +- F.184 *Operational provisions for the international public facsimile service between subscriber stations with group 4 facsimile machines.* +- F.200 *Teletex service.* +- F.300 *Videotex service.* +- F.310 *Broadband Videotex services.* +- F.600 *Service and operational principles for public data transmission services.* +- F.701 *Teleconference service.* +- F.710 *General principles for audiographic conference services.* +- F.711 *Audiographic conference teleservice for ISDN.* +- F.720 *Videotelephony services general.* +- F.721 *Videotelephony service for ISDN.* +- F.722 *Broadband videotelephony services.* +- F.730 *Videoconference service general.* +- F.732 *Broadband videoconference services.* +- F.740 *Audiovisual interactive services.* +- F.761 *Service-oriented requirements for telewriting applications.* +- F.811 *Broadband connection-oriented bearer services.* +- F.812 *Broadband connectionless data bearer service.* +- F.813 *Virtual path service for reserved and permanent communications.* +- F.821 *Broadband TV distribution services.* +- F.822 *Broadband HDTV distribution services.* +- F.850 *Principles of UPT.* +- F.851 *Universal Personal Telecommunications (UPT) service description (service set one).* + +- F.400 *Message handling system and service overview.* +- F.401 *Message handling services: Naming and addressing for public message handling services.* +- F.410 *Message handling services: The public message transfer service.* +- F.415 *Message handling services: Intercommunication with public physical delivery services.* +- F.420 *Message handling services: The public interpersonal messaging service.* +- F.435 *Message handling: EDI messaging service.* +- F.440 *Message handling services: The voice messaging service.* +- F.450 *Asynchronous group communication.* +- F.500 *International public directory services.* +- F.16 *Global Virtual Network Service.* +- I.241.8 *Teleaction stage one service description.* +- E.105 *International telephone service.* +- E.116 *International telecommunications charge card service.* +- E.140 *Operator-assisted telephone service.* +- E.152 *International freephone service.* +- E.330 *User control of ISDN-supported services.* + +### **ISDN supplementary services** + +- I.230 *Definition of bearer service categories.* +- I.231 *Circuit-mode bearer service categories.* +- I.232 *Packet mode bearer service categories.* +- I.240 *Teleservices supported by ISDN.* +- I.241.7 *Telephony 7 kHz teleservice.* +- I.250 *Definition of supplementary services.* +- I.251 *NUMBER IDENTIFICATION SUPPLEMENTARY SERVICES.* +- I.251.1 *Direct-Dialling-In.* +- I.251.2 *Multiple Subscriber Number.* +- I.251.3 *Calling Line Identification Presentation.* +- I.251.4 *Calling Line Identification Restriction.* +- I.251.5 *Connected Line Identification Presentation.* +- I.251.6 *Connected Line Identification Restriction.* +- I.251.7 *Malicious Call Identification.* +- I.251.8 *Sub-addressing.* +- I.252 *CALL OFFERING SUPPLEMENTARY SERVICES.* +- I.252.1 *Call Transfer (normal).* +- I.252.2 *Call Forwarding Busy.* +- I.252.3 *Call Forwarding No Reply.* +- I.252.4 *Call Forwarding Unconditional.* +- I.252.5 *Call Deflection.* +- I.252.6 *Line Hunting.* +- I.252.7 *Call Transfer (explicit).* +- I.252.8 *Call Transfer (single step).* +- I.253 *CALL COMPLETION SUPPLEMENTARY SERVICES.* +- I.253.1 *Call Waiting.* +- I.253.2 *Call Hold.* + +- I.253.3 *Completion of calls to busy subscribers.* +- I.253.4 *Completion of calls to no reply.* +- I.254 *MULTIPARTY SUPPLEMENTARY SERVICES.* +- I.254.1 *Conference Calling.* +- I.254.2 *Three-Party Service.* +- I.254.3 *Preset Conference Calling Service.* +- I.254.4 *Conference Call, Booked add-on.* +- I.254.5 *Conference Call, Meet me.* +- I.255 *COMMUNITY OF INTEREST SUPPLEMENTARY SERVICES.* +- I.255.1 *Closed User Group.* +- I.255.2 *Support of Private Numbering Plans.* +- I.255.3 *Multi-level precedence and pre-emption service.* +- I.255.4 *Priority service.* +- I.255.5 *Outgoing call barring.* +- I.256 *CHARGING SUPPLEMENTARY SERVICES.* +- I.256.1 *Credit Card Calling.* +- I.256.2 a) *Advice of Charge: Charging information at call set-up time.* +- I.256.2 b) *Advice of Charge: Charging information during the call.* +- I.256.2 c) *Advice of Charge: Charging information at the end of the call.* +- I.256.3 *Reverse charging.* +- I.256.4 *ISDN freephone service.* +- I.257 *ADDITIONAL INFORMATION TRANSFER.* +- I.257.1 *User-to-user signalling.* + +NOTE – This list will be checked and updated, if required, prior to the printing of this Recommendation. \ No newline at end of file diff --git a/marked/F/T-REC-F.116-200003-I_PDF-E/raw.md b/marked/F/T-REC-F.116-200003-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..0c885f04ac7db29e91f852ed0b72841d483be751 --- /dev/null +++ b/marked/F/T-REC-F.116-200003-I_PDF-E/raw.md @@ -0,0 +1,504 @@ + + +![ITU logo: A globe with a lightning bolt and the letters ITU.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with a lightning bolt and the letters ITU. + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.116** + +(03/00) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Mobile service – Mobile services and multdestination +satellite services + +--- + +**Service features and operational provisions in +IMT-2000** + +ITU-T Recommendation F.116 + +(Previously CCITT Recommendation) + +--- + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| AUDIOVISUAL SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to ITU-T List of Recommendations.* + +## **SERVICE FEATURES AND OPERATIONAL PROVISIONS IN IMT-2000** + +## **Summary** + +IMT-2000 systems are third generation mobile systems, which may be terrestrial, or satellite-based and may be characterized as networks in terms of service provision. This Recommendation describes service features and operational provisions that apply to IMT-2000 systems. It defines a set of service features and capabilities to standardize service presentation and to facilitate global mobility for users across network boundaries. Some of these services features and capabilities are essential whilst others are optional. + +## **Source** + +ITU-T Recommendation F.116 was prepared by ITU-T Study Group 2 (1997-2000) and was approved under the WTSC Resolution No. 1 procedure on 13 March 2000. + +## FOREWORD + +ITU (International Telecommunication Union) is the United Nations Specialized Agency in the field of telecommunications. The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of the ITU. The ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Conference (WTSC), which meets every four years, establishes the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics. + +The approval of Recommendations by the Members of the ITU-T is covered by the procedure laid down in WTSC Resolution No. 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +## INTELLECTUAL PROPERTY RIGHTS + +The ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. The ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, the ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementors are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database. + +© ITU 2000 + +All rights reserved. No part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and microfilm, without permission in writing from the ITU. + +## CONTENTS + +###### Page + +| | | | +|--------|-------------------------------------------------|---| +| 1 | Introduction..... | 1 | +| 2 | Scope..... | 1 | +| 3 | References..... | 1 | +| 4 | Terminology..... | 1 | +| 5 | Mobility concept..... | 2 | +| 5.1 | Terminal mobility ..... | 2 | +| 5.2 | Standard terminal equipment portability ..... | 2 | +| 5.3 | IMT-2000 user mobility..... | 2 | +| 6 | Support of flexible service provision..... | 2 | +| 7 | Universal bearer capability handling ..... | 3 | +| 7.1 | Data service types ..... | 3 | +| 7.2 | Quality of service categories..... | 4 | +| 7.3 | Example combinations..... | 4 | +| 8 | Support of multimedia services in IMT-2000..... | 4 | +| 9 | IMT-2000 service capabilities ..... | 5 | +| 9.1 | Privacy and fraud prevention ..... | 5 | +| 9.2 | Emergency call support service ..... | 5 | +| 9.3 | Man-Machine Interface (MMI) functions..... | 6 | +| 9.4 | User registration..... | 6 | +| 9.5 | Multiple profiles/subscriptions ..... | 6 | +| 9.6 | IMT-2000 service profile management..... | 6 | +| 9.6.1 | IMT-2000 profile interrogation ..... | 6 | +| 9.6.2 | IMT-2000 profile modification..... | 6 | +| 9.6.3 | IMT-2000 profile portability ..... | 7 | +| 9.7 | Roaming capability ..... | 7 | +| 9.8 | Addressing capabilities ..... | 7 | +| 9.9 | Over-the-Air services..... | 7 | +| 9.9.1 | Over-the-Air service provisioning..... | 7 | +| 9.9.2 | Over-the-Air profile administration..... | 7 | +| 9.10 | Priority access ..... | 7 | +| 9.11 | Services based on location..... | 7 | +| 9.11.1 | Position determination..... | 7 | +| 9.11.2 | Emergency calls..... | 7 | +| 9.11.3 | Location registration..... | 8 | + +| | Page | +|-----------------------------------------------------|-------------| +| 9.12 Identification services ..... | 8 | +| 9.13 Teleservice support capability ..... | 8 | +| 9.14 Supplementary service support capability ..... | 8 | +| 9.15 Voice Recognition support capability..... | 9 | +| 9.16 Preferred language ..... | 9 | + +# **Introduction** + +IMT-2000 are third generation mobile systems, which may be terrestrial or satellite-based. These systems may be characterized as networks in terms of service provision. They may be utilized as stand-alone mobile systems or as part of the fixed network. + + + +## **SERVICE FEATURES AND OPERATIONAL PROVISIONS IN IMT-2000** + +*(Geneva, 2000)* + +# **1 Introduction** + +IMT-2000 are third generation mobile systems, which may be terrestrial or satellite-based. These systems may be characterized as networks in terms of service provision. They may be utilized as stand-alone mobile systems or as part of the fixed network. + +# **2 Scope** + +This Recommendation describes service features and operational provisions that apply to IMT-2000 systems. It defines a set of service features and capabilities, some of which are essential and some optional. The intent is to standardize service presentation and to facilitate global mobility for users across network boundaries. + +# **3 References** + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; all users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. + +- ITU-T Recommendation F.111 (1991), *Principles of service for mobile systems*. +- ITU-T Recommendation F.115 (1995), *Service objectives and principles for future public land mobile telecommunication systems*. +- ITU-T Recommendation F.700 (2000), *Framework Recommendation for audiovisual/multimedia services*. +- ITU-T Recommendation Q.1701 (1999), *Framework for IMT-2000 networks*. +- Recommendation ITU-R M.816-1 (1997), *Framework for services supported on International Mobile Telecommunications 2000 (IMT-2000)*. +- Recommendation ITU-R M.1078 (1993), *Security principles for International Mobile Telecommunications 2000 (IMT-2000)*. +- Recommendation ITU-R M.1223 (1997), *Evaluation of security mechanisms for IMT-2000*. + +# **4 Terminology** + +This Recommendation defines the following terms: + +**4.1 mobile terminal:** For the purpose of this Recommendation, this term includes both the radiocommunication equipment, specific service equipment (e.g. telephone (handset), fax machine) and user identity module (UIM). Several types of terminal equipment (e.g. ISDN TE) may be connected to a mobile terminal. + +**4.2 User Identity Module (UIM):** A standard device or functionality providing secure procedures in support of user or terminal registration, authentication, and privacy for wireless access to IMT-2000. It may also contain application data to be used to facilitate telecommunication services and other services (e.g. UPT, banking). + +**4.3 virtual home environment:** Virtual home environment (VHE) is a capability whereby a user is offered the same service experience in a visited network as in his home network. + +The degree to which the VHE matches the actual home environment may, for example, be subject to the degree of cooperation between the visited and home network, their relative technical capabilities and the compatibility of the user terminal. + +# **5 Mobility concept** + +### **5.1 Terminal mobility** + +The ability of the terminal to access telecommunication services from different locations while in motion and the capability of the network to identify and locate that terminal. + +NOTE – The term "motion" includes the stationary state. This mobility aspect does not preclude the concept of fixed terminals participating in IMT-2000. + +## **5.2 Standard terminal equipment portability** + +The ability to connect terminal equipment to IMT-2000 mobile terminals. + +### **5.3 IMT-2000 user mobility** + +IMT-2000 user mobility is a feature which enables an IMT-2000 user to transfer his/her identity between IMT-2000 mobile terminals. + +The IMT-2000 user mobility affords the IMT-2000 user, with the application of a user identity module (e.g. "smart card"), to make an IMT-2000 mobile terminal his/her own mobile terminal. In this manner, the IMT-2000 user can access telecommunication services specified in their service profile with any IMT-2000 mobile terminal. + +# **6 Support of flexible service provision** + +In IMT-2000, services should be offered in a flexible way so that the preferences of each user can be fulfilled but without unduly constraining interoperability of, and roaming between, networks. For example, the user might be offered a choice of language in which to receive network prompts and announcements. The concept of Virtual Home Environment (VHE) has been proposed as a means of facilitating such an approach. + +When VHE is applied, depending on appropriate business arrangements and agreements, the following options exist: + +- the service profile of a user and the MMI (Man Machine Interface) which he is accustomed to are recorded, so that they can be transferred electronically from the home network to the visited network; +- the service profile of a user and/or the MMI which he is accustomed to are partially stored in the UIM and the remaining information stored in the appropriate databases in the home network; +- the service profile of a user and the MMI which he is accustomed to are stored completely in the UIM. + +For users who roam from their home network, each visited network can provide the same service-related functions in the same way as the home network, provided sufficient information has been supplied by the home network. + +# 7 Universal bearer capability handling + +IMT-2000 systems should support services based on digital transmission in an efficient, economical and user-friendly way. The service requester should have maximum freedom in communicating the characteristics of the requested service to the network. These characteristics might include a combination of service types and their quality of service. The network should try to fulfill the requirement or propose a reasonable alternative to the requester. + +The universal bearer services should allow for the emulation of, and interworking with, existing second generation services, e.g. 3.1 kHz audio UDI, Multicast, Group Broadcast, Voice Broadcast, Groupcall. + +The following service types may be provided using ITU defined protocols or using IETF (Internet Engineering Task Force) defined protocols (IP). In the specific case of IP, both static and dynamic IP-based addressing capabilities may be used to provide associated mobility support for services. + +Multiple concurrent session controls may also be provided as a service provider option to avoid routing ambiguities. + +### 7.1 Data service types + +The data service types supported by IMT-2000 are classified as follows: + +#### *Communication configurations* + +- PTP: Point-to-point service. +- PTM-B: Point-to-multipoint broadcast service. +- PTM-M: Point-to-multipoint multicast service. +- MTM: Multipoint-to-multipoint (e.g. needed for multimedia services with different sources of content). + +#### *Area selection* + +- geographical area selection; +- address domain selection; +- multicast group selection; +- static and dynamic addressing in IP. + +#### *Communication modes for network services* + +- CLNS: Connectionless network service; +- CONS: Connection oriented network service; +- via ITU protocols or via IETF protocols. + +#### *Timing relationship* + +- timing relationship between source and destination required; +- timing relationship between source and destination not required. + +#### *Bit rate* + +- constant bit rate; +- variable bit rate; +- asymmetry. + +## **7.2 Quality of service categories** + +The quality of service that a user can request is described by categories such as: + +- Best effort + +or: + +- QoS guaranteed with parameters such as: + - bit rate; + - throughput; + - delay characteristics; + - Maximum Bit Error Rate. + +### **7.3 Example combinations** + +Depending on the attributes, Timing Relationship, Bit rate and Communication mode, different bearer classes can be distinguished, which can occur within different communication configurations and area selections. They are further qualified by the set of QoS parameters. + +# **8 Support of multimedia services in IMT-2000** + +Multimedia services combine two or more media into a single integrated service. From the user's point of view, a multimedia telecommunication service is the combination of IMT-2000 telecommunication capabilities required to support a particular multimedia application. From the IMT-2000 provider's point of view, a multimedia telecommunication service is a combination or set of combinations of two or more media components (e.g. audio, video, graphics, etc.). + +Examples of IMT-2000 multimedia services which could be supported by one or several multimedia communication tasks are: + +- mobility services (specifically related to the mobility of the user, like location service); +- conference services (providing bidirectional and synchronized real-time transfer of voice and possibly moving pictures); +- conversation services (allowing bidirectional dialogue communication with real-time end-to-end information transfer); +- distribution services (providing continuous flow of information from a central source); +- retrieval services (allowing to retrieve information from one source at a time); +- collection services (allowing to retrieve information from several sources in parallel); +- message services (offering user-to-user communications with store-and-forward capabilities). + +# **9 IMT-2000 service capabilities** + +### **9.1 Privacy and fraud prevention** + +Security mechanisms for the protection of the IMT-2000 user, service provider and network operator should provide for the following features in order to guarantee a maximum of privacy and fraud prevention, that are at least as good as in second generation wireless systems. + +The special security issues that arise from user/terminal mobility and inter-PLMN roaming shall be considered: + +- confidentiality of user messages; +- privacy of user/subscriber related data; +- privacy of billing data; +- authentication of user to UIM, UIM to the network, network to UIM, service provider to network operator, visited network to home network, etc.; +- event reporting and event limitation. + +The features above should be fulfilled in a way that is transparent to the user or is at least user-friendly. The impact on the quality of service in normal operation should be kept to a minimum. + +Service-related security features shall operate when roaming. + +The home network is informed if the visited network cannot provide the security functions deemed essential by the home network operator, who shall then be enabled to take the appropriate action and assure the integrity of the service. + +Some security features could be an integral part of the services and should be defined by the service provider. + +For further details on security features and security mechanisms for IMT-2000, refer to Recommendations ITU-R M.1078 and M.1223. + +### **9.2 Emergency call support service** + +Any IMT-2000 mobile terminal should be allowed to make a call attempt to an emergency call centre. + +The emergency call centre should be able to infer that the call was initiated by a mobile terminal (e.g. based on trunk group). The current location of the mobile terminal, if available, is to be supplied to the emergency call centre. Upon answering the call, the emergency call centre shall be able to communicate with the caller over a normal voice connection. After establishment of the call with the emergency call centre, the caller may allow a party or parties previously put on "hold" to join the conversation. Release occurs when either the caller or the emergency call centre disconnects. The emergency call centre should be supplied with sufficient information to enable it to re-establish the call to the caller after the emergency call is released. + +The emergency call concept is applicable to voice and Telecommunications Device for the Deaf/Text Telephony (TDD/TTY). + +National and/or service provider options include: + +- a) the concept of access for emergency call attempts to be given preferential or priority access to the scarce radio resources compared with normal call attempts; +- b) the facility to permit call attempts to an emergency call centre to be subjected to a reduced level of authentication and validation. For example, a call may be made without a UIM in the terminal. + +### **9.3 Man-Machine Interface (MMI) functions** + +MMI plays a role of the user's terminal access, network access, call setup, and service control. It should be possible for the user to customize certain aspects of the MMI. The presentation of a familiar MMI to the user should form a central part of the Virtual Home Environment (VHE). The service provider may be able to offer tailored MMI solutions using VHE. The MMI should meet the requirements to support the VHE to allow seamless service provision even across different types of networks and access terminals. This Recommendation is not intended to provide detailed MMI specifications, although it is desirable for the user to be offered a friendly MMI, integrating functions provided by the terminal manufacturer with those requested by the VHE. + +### **9.4 User registration** + +User registration is a feature by which a user associates himself and his service profile with the terminal for the purpose of accessing telecommunication services (e.g. receiving incoming and originate outgoing calls) and by which a network becomes aware of the existence and location of a terminal and its associated user(s). + +In order for an IMT-2000 user to be registered on an IMT-2000 terminal, a UIM associated with the user has to be physically present in the terminal. + +User authentication should be executed at each user registration. A user may explicitly deregister. + +IMT-2000 networks should provide a capability to support UPT and therefore UPT user registration. + +### **9.5 Multiple profiles/subscriptions** + +It shall be possible for a user to obtain certain services from one service provider and some other services from another service provider using a single UIM. Where multiple subscriptions on a single UIM are in use the following principles shall apply: + +- if there are several service providers subscriptions on a single UIM, then different numbers are to be used for each service provider, i.e. shared numbering shall not be used; +- for outgoing calls, the user shall be able to select which service provider to use on a per call basis or on a subscription basis; +- it shall be possible to have multiple subscriptions active simultaneously; +- there is a requirement for a terminal supported by one network operator to be able to have users from different service providers associated with that terminal. Such a capability infers that the service provider of each associated user will have an agreement with the service provider supporting that terminal. + +### **9.6 IMT-2000 service profile management** + +#### **9.6.1 IMT-2000 profile interrogation** + +This feature enables a user to interrogate the current status of their service profile (e.g. for location information, availability of services, charging information, etc.). + +Some data included in the profile is restricted from view of the user. User authentication/validation security procedures should be executed at each profile interrogation. + +#### **9.6.2 IMT-2000 profile modification** + +This feature enables a user to modify their current service profile (e.g. for availability of services, changing of password, etc.). + +Some data included in the profile is restricted from the view of the user. User authentication/validation security procedures should be executed at each profile interrogation. + +#### **9.6.3 IMT-2000 profile portability** + +The concept of VHE in IMT-2000 requires that the service profile of a user is used when roaming outside the home network. Technical or business constraints or other reasons may limit its full application. + +### **9.7 Roaming capability** + +IMT-2000 networks may provide VHE support to facilitate roaming outside the home network in an environment of flexible service provision. + +### **9.8 Addressing capabilities** + +IMT-2000 should be able to support the traditional addressing mechanisms that exist in existing networks, e.g. E.164 telephone number, X.121 address on packet data networks or X.400 or IP address or name for electronic mail. IMT-2000 should also support a more integrated approach, where multimedia calls can be set up without having to use all these mechanisms together. + +Therefore, there is a general service requirement that the signalling capability of IMT-2000 systems should support the use of numeric, alpha and alphanumeric naming schemes. + +### **9.9 Over-the-Air services** + +#### **9.9.1 Over-the-Air service provisioning** + +This is a feature which allows a subscriber to IMT-2000 to negotiate an initial subscription with a service provider and subsequently amend that subscription using automatic procedures over the air. (The term "automatic procedures" is intended to convey the fact that no human intervention is required on the part of the service provider.) + +#### **9.9.2 Over-the-Air profile administration** + +This is a feature, which enables a service provider to automatically update the user profile data as required. For example, updating the system selection preference list for use when the user is roaming. + +### **9.10 Priority access** + +As a service provider option and subject to national requirements it should be possible to create different classes of service to enable users to be offered Priority Access and Assignment. This feature functions when no radio channels are available. Users are placed in a queuing system until radio channels are available and then assigned channels on a first come first served basis subject to their class of service priority. Whilst in the queue awaiting channel assignment, a user is considered to be in the busy condition. + +### **9.11 Services based on location** + +#### **9.11.1 Position determination** + +Location allows an IMT-2000 user's position to be determined. Position determination may be initiated by the user, the network, or an external party (e.g. an Emergency Centre). Position determination may be subject to various restrictions based on capability, security, service profiles, etc. + +#### **9.11.2 Emergency calls** + +For the specific case of emergency calls, this capability is needed to identify the location of an IMT-2000 terminal from which an emergency call is being placed to within the limits specified by the appropriate national authority. + +NOTE – In the United States of America, for example, this is to within a radius of 125 metres in 67 per cent of all cases or better. + +The implementation of location services by satellite-based systems is not covered by this Recommendation. + +#### **9.11.3 Location registration** + +Services based on location registration will offer subscribers the opportunity to tailor their set of active services according to their current serving location. For example, there may be a specific set of active services for the residential location and a separate set of active services for the business location. + +### **9.12 Identification services** + +For reasons of public safety and service, it should be possible to make available within the network and in some cases to an end user, IMT-2000 user parameters such as user identity, network address and location information. + +For example, for emergency calls, the emergency call-handling centre may need all of these IMT-2000 user parameters to be able to: + +- identify a calling party; +- re-establish a terminated call to a calling party; +- pass calling party location information, to an emergency service. + +These IMT-2000 user parameters should be generated at the calling/originating end of the network, and be available for delivery as required. + +Public safety and service needs may also require the same information for the called IMT-2000 party, i.e. the called IMT-2000 user identity, the IMT-2000 network address of the call termination, and the location information for the call recipient. These IMT-2000 user parameters should be generated at the called/terminating end of the network and be available for delivery as required. + +### **9.13 Teleservice support capability** + +New teleservices as well as those that are supported in existing telecommunication systems should be supported by IMT-2000 (e.g. messaging, speech, facsimile, paging). + +### **9.14 Supplementary service support capability** + +New and existing supplementary services should be supported by IMT-2000. Examples of such supplementary services are as follows: + +- a) Identification Supplementary Services (e.g. Abbreviated Dialling, Calling Line Identification Presentation/Restriction, Connected Line Identification Presentation/Restriction, Malicious Call Identification, Universal Access Number, Multiple Subscriber Number, Calling Name Identification Presentation/Restriction, Calling Number Identification Presentation/Restriction). +- b) Call Offering Supplementary Services (e.g. Call Distribution, Call Deflection, Call Forwarding Unconditional/Busy/No Reply/Not Reachable/default, Call Transfer, Destination Call Routing, Follow-Me Diversion, Mobile Access Hunting, Universal Access Number, Flexible Alerting). +- c) Call Completion Supplementary Services (e.g. Call Hold, Call Waiting, Completion of Calls to Busy Subscriber). +- d) Multiparty Supplementary Services (e.g. Conference Calling, 3-way calling). + +- e) Community of Interest Supplementary Services (e.g. Closed User Group, Virtual Private Network). +- f) Charging Supplementary Services (Charge Card Calling, Advice of Charge, Credit Card Calling). +- g) Additional Information Transfer Supplementary Services (User-to-user signalling, Message Waiting Notification). +- h) Call Restriction Supplementary Services (e.g. Selective barring or screening of Incoming/outgoing Calls, Password and PIN restrictions). + +### **9.15 Voice Recognition support capability** + +Voice Recognition will enable a user to control features and supplementary services using spoken commands: (e.g. call origination, activation and de-activation of features, registration operations, user identification, validation and authentication, creation of short text messages and access to the short message service). + +### **9.16 Preferred language** + +This feature enables the subscriber to specify the language for network services. Network services, which could be offered in the subscriber's preferred language, include: + +- recorded announcements; +- directory assistance; +- operator services; +- emergency services; +- "Help" lines; +- Message Waiting Notification; +- CLIP number not available; +- CLIP number restricted. + + + + + +# ITU-T RECOMMENDATIONS SERIES + +| | | +|-----------------|--------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of the ITU-T | +| Series B | Means of expression: definitions, symbols, classification | +| Series C | General telecommunication statistics | +| Series D | General tariff principles | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Construction, installation and protection of cables and other elements of outside plant | +| Series M | TMN and network maintenance: international transmission systems, telephone circuits, telegraphy, facsimile and leased circuits | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks and open system communications | +| Series Y | Global information infrastructure | +| Series Z | Languages and general software aspects for telecommunication systems | + +\*18042\* \ No newline at end of file diff --git a/marked/F/T-REC-F.122-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.122-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..64672153e39c851d9e16fdbc702e526c7ba7e8a7 --- /dev/null +++ b/marked/F/T-REC-F.122-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:454a9958ffe168868cb7d38a0eb24418dafe31a7a4245c992089b2316ac37d3e +size 7392 diff --git a/marked/F/T-REC-F.122-198811-I_PDF-E/b28af4985cdef1e519e3aaf26561dcb3_img.jpg b/marked/F/T-REC-F.122-198811-I_PDF-E/b28af4985cdef1e519e3aaf26561dcb3_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..caa3dd56b8948c8a3aac4813ad99fc01ab38c6f7 --- /dev/null +++ b/marked/F/T-REC-F.122-198811-I_PDF-E/b28af4985cdef1e519e3aaf26561dcb3_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cb14fcdcf5de047135a5b11685679772c84691507da7441d9b72058230e0a29a +size 21246 diff --git a/marked/F/T-REC-F.127-199610-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.127-199610-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3a58f68885c03bc2fb95bb409d68f7f55947475f --- /dev/null +++ b/marked/F/T-REC-F.127-199610-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:97adcbdebb3326a1af9817c0266d6c1cef088a8f97dc377b1e5fd01e721d14f7 +size 8269 diff --git a/marked/F/T-REC-F.127-199610-I_PDF-E/raw.md b/marked/F/T-REC-F.127-199610-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..70a3407eb81d49374a96ecc7e84e9b4850f27f76 --- /dev/null +++ b/marked/F/T-REC-F.127-199610-I_PDF-E/raw.md @@ -0,0 +1,285 @@ + + +![ITU logo: A globe with a lightning bolt and the letters ITU.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with a lightning bolt and the letters ITU. + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.127** + +(10/96) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Mobile service – Mobile services and multdestination +satellite services + +--- + +**Operational procedures for interworking +between the international telex service and the +service offered by the INMARSAT-C system** + +ITU-T Recommendation F.127 + +(Previously CCITT Recommendation) + +--- + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1–F.109 | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | F.110–F.159 | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | F.160–F.399 | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | F.550–F.599 | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| AUDIOVISUAL SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to ITU-T List of Recommendations.* + +## FOREWORD + +The ITU-T (Telecommunication Standardization Sector) is a permanent organ of the International Telecommunication Union (ITU). The ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Conference (WTSC), which meets every four years, establishes the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics. + +The approval of Recommendations by the Members of the ITU-T is covered by the procedure laid down in WTSC Resolution No. 1 (Helsinki, March 1-12, 1993). + +ITU-T Recommendation F.127 was revised by ITU-T Study Group 1 (1993-1996) and was approved by the WTSC (Geneva, 9-18 October 1996). + +--- + +## NOTE + +In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +© ITU 1997 + +All rights reserved. No part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and microfilm, without permission in writing from the ITU. + +## CONTENTS + +| | Page | +|--------------------------------------------------------------------------------|-------------| +| 1 Definitions..... | 1 | +| 2 Scope..... | 1 | +| 3 Introduction..... | 2 | +| 4 Principles of operation ..... | 2 | +| 5 Operational procedures ..... | 2 | +| 5.1 International telex service to INMARSAT-C system..... | 2 | +| 5.2 Calls from an MES to subscribers of the international telex service ..... | 4 | +| 5.3 MES-to-MES calls utilizing the international telex service..... | 4 | +| Annex A – Alphabetical list of abbreviations used in this Recommendation ..... | 4 | + +# OPERATIONAL PROCEDURES FOR INTERWORKING BETWEEN THE INTERNATIONAL TELEX SERVICE AND THE SERVICE OFFERED BY THE INMARSAT-C SYSTEM + +*(revised 1992 and 1996)* + +The ITU-T, + +*considering* + +- a) that INMARSAT has introduced various mobile services based on the INMARSAT-A, INMARSAT-B and INMARSAT-C systems; +- b) that Recommendation F.120 specifies ship station identification for the maritime mobile-satellite service; +- c) that Recommendation F.125 specifies the numbering plan for the maritime mobile-satellite service of INMARSAT; +- d) that the provision of interworking with the international telex service is a standard requirement within the INMARSAT-C system, + +*unanimously recommends* + +that operational procedures of interworking between the international telex service and the service provided by the INMARSAT-C system be in accordance with this Recommendation. + +## 1 Definitions + +This Recommendation defines the following terms: + +**1.1 ship earth station** is defined in Article 1, section 4.16 of the *Radio Regulations*, ITU Geneva, 1982. For the purpose of this Recommendation, the terms “ship earth station” and “mobile earth station” are interchangeable. + +**1.2 coast earth station** is defined in Article 1, section 4.14 of the *Radio Regulations*, ITU Geneva, 1982. For the purpose of this Recommendation, the terms “coast earth station” and “land earth station” are interchangeable. + +**1.3 The mobile satellite store-and-forward unit** is the functional interface between the mobile satellite message transmission system and the international telex service. + +## 2 Scope + +**2.1** The purpose of this Recommendation is: + +- a) to standardize procedures for the subscribers of the international telex service calling Mobile Earth Stations (MESs) in the INMARSAT-C system; +- b) to standardize procedures for an MES calling subscribers of the international telex service; +- c) to standardize procedures for MES-to-MES calls that transit the international telex network. + +## 3 Introduction + +3.1 The characteristics of the mobile satellite circuit provided by the INMARSAT-C system are such that only store-and-forward operation is supported when interworking with the international telex service. + +3.2 A general description of the INMARSAT-C system and the services it may support is given in Supplement No. 3 to the F-Series Recommendations (*Blue Book*, Fascicle II.4, Melbourne, 1988). + +3.3 In the interests of ensuring that the optimum use is made of the international telex service, Administrations may wish to advise their customers of any differences in messaging procedures between the service offered by the INMARSAT-C system and normal telex procedures as offered by other INMARSAT standard systems. + +3.4 The provisions of this Recommendation should be applied to future implementation and, as far as practicable, to existing implementations of Mobile Satellite Store-and-Forward Unit (MSSFU). + +## 4 Principles of operation + +4.1 Communication between subscribers of the international telex service and an MES is on a store-and-forward basis. Thus, conversational mode interworking between terminals is not provided. + +4.2 Interworking between the international telex service and the INMARSAT-C system may employ either one-stage or two-stage selection procedures for the delivery of telex messages to an MES. Administrations may provide either or both modes of operation. + +These services may be extended across international boundaries on a bilateral basis. Where no such bilateral agreement exists, the Administration operating the MSSFU may clear the call and return the service signal (**NA**). + +4.3 In mobile to fixed direction, subscribers to the mobile-satellite service provided by the INMARSAT-C system may send single messages to the subscribers of the international telex service and to the appropriate applications. The messages are forwarded by the MSSFU via the international telex service. + +4.4 For calls from MES-to-MES, the procedures for handling calls between different satellite regions is not the subject for international standardization. Such calls which utilize the international telex service should follow normal telex procedures. + +## 5 Operational procedures + +### 5.1 International telex service to INMARSAT-C system + +#### 5.1.1 One-stage selection by the telex customer + +5.1.1.1 A customer of the international telex service may place a call to a mobile earth station using normal telex procedures. + +5.1.1.2 The originating telex subscriber, when calling an INMARSAT-C MES, should input the following selection information, plus any necessary prefix and end-of-selection signal: + +58S4X1.....X8 + +where 58S is the telex destination code in accordance with Recommendation F.69 (S = 1, 2, 3 or 4), 4 is the T-digit in accordance with Recommendation F.125, and X1.....X8 are part of the MES identity in accordance with Recommendation F.125. + +5.1.1.3 On receipt of this address, the MSSFU should check that the addressed MES is logged into the satellite region and should, if the result is positive, return the call connect signal in accordance with the relevant U-Series Recommendations. If the check fails, the call attempt should be cleared by the MSSFU with the appropriate service signal in accordance with Recommendation F.131. + +5.1.1.4 The MSSFU should return the MES answerback, formatted in accordance with Recommendation F.74. + +5.1.1.5 The MSSFU should transmit a **WRU** signal following the MES answerback in order to capture the answerback of the originating telex subscriber. + +**5.1.1.5 bis** Where the call to the MSSFU is being handled by a telex SFU operating in accordance with Recommendations F.72 and U.81 on behalf of a customer of the international telex service, the telex SFU identification should be forwarded to the MSSFU immediately following receipt of the called MES answerback. This should comprise the code expression **CI**, the characters **SFU** and the TNIC of the parent network of the telex SFU. This will then allow the procedures which subsequently follow to be under the full control of the telex SFU with no prompt signals returned by the MSSFU. + +Where the call to the MSSFU is originated by a subscriber of the international telex service operating in interactive or manual mode, the procedures of 5.1.1.6 to 5.1.1.8 should be applied. + +**5.1.1.6** The calling telex address shall be determined from the received telex answerback in accordance with the rules described in Recommendation U.74. + +Where the calling telex address cannot be determined, the MSSFU should wait 3 seconds from the end of the received answerback to enable the originating telex customer to voluntarily input his telex address, in accordance with Recommendation U.208. At the end of this period, the MSSFU shall return the prompt signal **ADD**. If the calling telex address is not received within 15 seconds of the **ADD** prompt, another **ADD** prompt signal shall be returned. If another 15 seconds elapse without the receipt of the calling telex address, the connection shall be cleared by the MSSFU. + +Optionally, the call may be accepted by the MSSFU, and the **GA** prompt returned, where, in the event of non-delivery of an accepted message, alternative arrangements are in place to provide a non-delivery notification message to the originating telex subscriber, for example, spill out to an assistance position. + +It should be noted that some MSSFU do not support the **ADD** procedure. In this case, the MSSFU should return the code expression **REJ** and clear the call. + +**5.1.1.7** Input of the message may now be commenced by the originating telex customer. + +**5.1.1.8** The preferred method of clearing to be applied by the originating telex customer is by the use of the End-of-Transmission (EOT) signal, four times combination 26 in figures case (++++) which need not be passed to the intended recipient. This method enables the provision of a message reference number by the MSSFU which could be useful to the message originator should a Non-Delivery Notification (NDN) be subsequently received. However, it should be recognized that the originating telex customer could clear the connection in one of the following ways: + +- a) The customer may close his message with the End-of-Message (EOM) signal **NNNN** which is followed by a period of idle condition of 30 seconds. This should be interpreted by the MSSFU as being equivalent to an EOT signal and return the input transaction accepted for delivery (ITD) sequence as specified in Recommendation U.208. +- b) The customer may close his message with the end-of-message signal **NNNN** followed immediately by a clear, having optionally exchanged answerbacks, in which case the MSSFU should proceed as in c). +- c) The call may be cleared without EOT or EOM, in which case all text received should be forwarded to the MES in the normal way, optionally including a statement which would advise the recipient that the message originator did not clear the connection using the EOT procedure and that consequently the message may be incomplete. +- d) To facilitate clearing by Telex Automatic Emitting Device (TAED) which may interpret a clear initiated by the MSSFU as an abnormal condition, the **ITD** sequence should be followed by a period of idle condition to enable clearing by TAEDs in accordance with Recommendation S.20. + +**5.1.1.9** Following clearing of the connection across the international telex network, delivery of the message to the address MES should commence at the earliest opportunity. + +**5.1.1.10** The message should not be held in the MSSFU for more than 24 hours. + +**5.1.1.11** In the event of non-delivery of the message to the MES, it is the responsibility of the MSSFU to return an NDN directly to the originating telex customer in accordance with Recommendation U.208. + +**5.1.1.12** If message delivery has not commenced at the expiry of the 24-hour period, an NDN should be provided to the originating telex customer. However, any delivery procedure which has already been commenced by the MSSFU that would exceed the 24-hour period, should be completed. + +#### **5.1.2 Two-stage selection** + +**5.1.2.1** Subscribers of the international telex service should use normal telex call establishment procedures to access the MSSFU, which is allocated a national number for this purpose, the general service principles and access procedures and protocols being in accordance with Recommendations F.72 and U.80 for two-stage access to a store-and-forward unit. + +**5.1.2.2** The MSSFU should accept both single and multi-addressed calls, the address line containing an optional attention information field in accordance with Recommendation F.72 to indicate the expected answerback of the recipient, attention information, class of delivery indicator, and a request for positive delivery notification. The list of addresses, either single or multiple, should be closed by the End-of-Address (EOA) signal **BT**. + +As a minimum, the address line should contain the selection information in accordance with 5.1.1.2. + +**5.1.2.3** If the calling address cannot be determined from the calling subscriber's answerback for the purpose of delivering a non-delivery advice, the call should be cleared. + +The call may be accepted where, in the event of non-delivery of a message, alternative arrangements for delivery are provided. The alternative arrangements, for example, may be the provision of an operator position. + +**5.1.2.4** The action to be taken when abnormal conditions are encountered during message input shall be in accordance with Recommendations F.72 and U.80. + +### **5.2 Calls from an MES to subscribers of the international telex service** + +**5.2.1** Subscribers to the Mobile Satellite service provided by the INMARSAT-C system may send messages to subscribers of the international telex service. + +**5.2.2** The messages should be forwarded by the MSSFU to the addressed telex subscriber via the international telex service. The answerback sent to the called telex subscriber should be that of the MES. + +**5.2.3** Upon delivery of the message to the destination telex subscriber, the MSSFU, if requested by the originating MES, should provide a Positive Delivery Notification (PDN) to the MES. In the event of non-delivery of the message to the destination telex subscriber, it is an INMARSAT requirement of the MSSFU to provide a Non-Delivery Notification (NDN) to the MES. The manner in which these notifications are provided to the MES is not the subject for international standardization. + +### **5.3 MES-to-MES calls utilizing the international telex service** + +**5.3.1** Subscribers to the Mobile Satellite service provided by the INMARSAT-C system may send messages to another MES. + +**5.3.2** Where the call is to an MES in another satellite region and the call is carried by the international telex service, the procedures to be followed should be in accordance with 5.2. + +The procedures to be adopted where the destination MSSFU does not support a one-stage selection method on calls from the international telex service are for further study. + +**5.3.3** The procedures for calls between MESs in the same satellite region or in satellite regions served by the same Land Earth Station (LES), are not the subject for international standardization. + +## **Annex A** + +### **Alphabetical list of abbreviations used in this Recommendation** + +EOA End-of-Address + +EOM End-of-Message + +EOT End-of-Transmission + +ITD Input Transaction accepted for Delivery +LES Land Earth Station +MES Mobile Earth Station +MSSFU Mobile Satellite Store-and-Forward Unit +NDN Non-Delivery Notification +PDN Positive Delivery Notification +TAED Telex Automatic Emitting Device + +# ITU-T RECOMMENDATIONS SERIES + +- Series A Organization of the work of the ITU-T +- Series B Means of expression +- Series C General telecommunication statistics +- Series D General tariff principles +- Series E Telephone network and ISDN +- Series F Non-telephone telecommunication services** +- Series G Transmission systems and media +- Series H Transmission of non-telephone signals +- Series I Integrated services digital network +- Series J Transmission of sound-programme and television signals +- Series K Protection against interference +- Series L Construction, installation and protection of cables and other elements of outside plant +- Series M Maintenance: international transmission systems, telephone circuits, telegraphy, facsimile and leased circuits +- Series N Maintenance: international sound-programme and television transmission circuits +- Series O Specifications of measuring equipment +- Series P Telephone transmission quality +- Series Q Switching and signalling +- Series R Telegraph transmission +- Series S Telegraph services terminal equipment +- Series T Terminal equipments and protocols for telematic services +- Series U Telegraph switching +- Series V Data communication over the telephone network +- Series X Data networks and open system communication +- Series Z Programming languages \ No newline at end of file diff --git a/marked/F/T-REC-F.130-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.130-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..64672153e39c851d9e16fdbc702e526c7ba7e8a7 --- /dev/null +++ b/marked/F/T-REC-F.130-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:454a9958ffe168868cb7d38a0eb24418dafe31a7a4245c992089b2316ac37d3e +size 7392 diff --git a/marked/F/T-REC-F.131-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.131-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..64672153e39c851d9e16fdbc702e526c7ba7e8a7 --- /dev/null +++ b/marked/F/T-REC-F.131-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:454a9958ffe168868cb7d38a0eb24418dafe31a7a4245c992089b2316ac37d3e +size 7392 diff --git a/marked/F/T-REC-F.141-199406-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.141-199406-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..8744879774a3227e993500501850f3664ed73265 --- /dev/null +++ b/marked/F/T-REC-F.141-199406-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4ec56b8096c2499ef1c111246890c60c550c68fe68608a9be12a368fc54fc83e +size 8269 diff --git a/marked/F/T-REC-F.141-199406-I_PDF-E/f6d72d7c790e7f585532140f3971639a_img.jpg b/marked/F/T-REC-F.141-199406-I_PDF-E/f6d72d7c790e7f585532140f3971639a_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..83907ee59af9365a3ec62b7aac6a73fbbe9acba1 --- /dev/null +++ b/marked/F/T-REC-F.141-199406-I_PDF-E/f6d72d7c790e7f585532140f3971639a_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bcd513957a14c84b8a16870661326f57c3bce3098525243d9eb394ecef157744 +size 107796 diff --git a/marked/F/T-REC-F.141-199406-I_PDF-E/raw.md b/marked/F/T-REC-F.141-199406-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..edc390100223fc34d34dac9298c5975b8a896909 --- /dev/null +++ b/marked/F/T-REC-F.141-199406-I_PDF-E/raw.md @@ -0,0 +1,154 @@ + + +![ITU logo: A globe with a lightning bolt and the letters ITU.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with a lightning bolt and the letters ITU. + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +**F.141** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +(06/94) + +**OPERATIONS AND QUALITY OF SERVICE +MOBILE SERVICE** + +--- + +**INTERNATIONAL TWO-WAY MULTIPOINT +TELECOMMUNICATION SERVICE +VIA SATELLITE** + +**ITU-T Recommendation F.141** + +(Previously "CCITT Recommendation") + +--- + +# FOREWORD + +The ITU-T (Telecommunication Standardization Sector) is a permanent organ of the International Telecommunication Union (ITU). The ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Conference (WTSC), which meets every four years, establishes the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics. + +The approval of Recommendations by the Members of the ITU-T is covered by the procedure laid down in WTSC Resolution No. 1 (Helsinki, March 1-12, 1993). + +ITU-T Recommendation F.141 was prepared by ITU-T Study Group 1 (1993-1996) and was approved under the WTSC Resolution No. 1 procedure on the 1st of June 1994. + +--- + +# NOTE + +In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +# CONTENTS + +| | Page | +|----------------------------------------------------|-------------| +| 1 Scope ..... | 1 | +| 1.1 Definition of service ..... | 1 | +| 2 Service description ..... | 1 | +| 2.1 Functional elements of service provision ..... | 1 | +| 2.2 Service provision ..... | 1 | +| 2.3 Types of service ..... | 1 | +| 2.4 Area of service ..... | 1 | +| 2.5 Service configuration ..... | 2 | +| 3 Quality of Service ..... | 3 | +| 4 Service availability ..... | 3 | +| 5 Class of space segment capacity ..... | 3 | + +# **SUMMARY** + +This Recommendation provides service and operational guidelines for the provision of international two-way multipoint telecommunication service via satellite. + +International two-way multipoint telecommunication service via satellite is defined as a service provided to customers by Recognized Operating Agencies (ROAs) for the two-way transmission of information via satellite, between a central facility and a multiplicity of points where transmit/receive earth stations are located. + +# INTERNATIONAL TWO-WAY MULTIPOINT TELECOMMUNICATION SERVICE VIA SATELLITE + +*(Geneva, 1994)* + +# 1 Scope + +This Recommendation provides operational guidelines and a service description including quality of service requirements for international two-way multipoint telecommunication service via satellite. + +## 1.1 Definition of service + +**international two-way multipoint telecommunication service via satellite** is defined as a service provided to customers by Recognized Operating Agencies (ROAs) for the two-way transmission of information via satellite, between a central facility and a multiplicity of points where transmit/receive earth stations are located. + +# 2 Service description + +## 2.1 Functional elements of service provision + +An international two-way multipoint telecommunication service via satellite generally includes the following nine elements (see Figure 1): + +- 1) the customer central facility equipment; +- 2) the links between the customer central facility equipment and the service provider command and control management centre; +- 3) the service provider command and control management centre; +- 4) the link between the master earth station and the service provider command and control management centre; +- 5) the master earth station (hub); +- 6) the space segment capacity; +- 7) the remote earth station antennas; +- 8) the link between transmit/receive remote earth station antenna and user terminal equipment; +- 9) the user terminal equipment. + +## 2.2 Service provision + +The service may be provided on either a full-time (24-hour) basis, a scheduled part-time basis or an occasional basis. Where the hub and one or more of the remote earth stations are in different countries, the provision of service is governed by the appropriate national and international regulations and subject to agreement between the ROAs. The ROA in each country should be made aware of their scope of the service coverage and the traffic routing capabilities. + +## 2.3 Types of service + +The service may be digital or analogue, provided in the form of inbound (remote to hub) and outbound (hub to remote) channels at rates depending on the application and within the capacity available. The data rates, bandwidths, and frequencies of the outbound channels and those of the inbound channels may be different. + +## 2.4 Area of service + +The service may be provided on a national and/or international basis in accordance with relevant national regulatory requirements. + +## 2.5 Service configuration + +2.5.1 As illustrated in Figure 1, there are generally nine functional elements in the provision of an international two-way multipoint telecommunication service via satellite. + +![Diagram of international two-way multipoint telecommunication service via satellite showing functional elements 1 through 9.](f6d72d7c790e7f585532140f3971639a_img.jpg) + +The diagram illustrates the functional elements of an international two-way multipoint telecommunication service via satellite. It is divided into three main sections: Customer Central Facility (1), Service Provider Command and Control Management Centre (3), and Remote E.S. (7, 8, 9) for three different customers (A, B, and C). The central part of the diagram shows the 'Space segment' containing a satellite (6) and a 'Master E.S.' (5). The 'Service Provider Command and Control Management Centre' contains a 'Protocol Conversion Multiplexing Addressing Network Control Master ES TE' (4). The 'Customer Central Facility' (1) contains three 'Customer' units (2). Arrows indicate the flow of communication: from the Customer Central Facility to the Service Provider Command and Control Management Centre, from the Service Provider Command and Control Management Centre to the Master E.S., from the Master E.S. to the Space segment, and from the Space segment to each of the Remote E.S. units. Each Remote E.S. unit contains three 'User TE' units (8) connected to a 'Remote E.S.' (7). The diagram is labeled 'T0103400-94/d01' in the bottom right corner. + +Diagram of international two-way multipoint telecommunication service via satellite showing functional elements 1 through 9. + +FIGURE 1/F.141 +International two-way multipoint telecommunication service via satellite + +2.5.2 Owing to the need for flexibility, the service may be adapted to meet a variety of requirements taking into account the regulations governing the operation of the space segment providers and ROAs involved. + +2.5.3 Subject to national regulatory arrangements, interconnection may be offered with the public switched network, a customer's private network and/or by other arrangements designed to meet specific customer requirements. + +2.5.4 The two-way operation is in a star configuration enabling communications between several remote locations and the hub, and communications among remote locations through the hub. The two-way operation can be achieved under proven multiple access procedures, such as those defined in G-, M-, and N-Series Recommendations. + +2.5.5 The master hub, consisting of a master earth station and a service provider command and control management centre, may be dedicated to a single customer or may be shared by a number of customers; it may be located either on a customer's site or may be remotely connected. + +2.5.6 The conditions for use of the master earth station (5), the remote transmit/receive earth stations (7), the service provider command and control management centre (3) and the links (2, 4) are a national matter. + +2.5.7 The conditions for the use of the space segment capacity (6) are defined by the satellite system space segment providers. + +2.5.8 The remote transmit/receive earth stations (7) would normally be located at the user premises near the user equipment; but the user equipment may also be remotely connected by a link (8) to the transmit/receive earth stations. + +# 3 Quality of Service + +The efficiency of operation and therefore the quality of service provided to the customers are linked to the relationship of all parties which contribute to the provision of the service, i.e. the technical equipment and the entities in charge of the operation. + +# 4 Service availability + +The service availability, as defined in CCITT Recommendation X.140, is the ratio of aggregate time, during which satisfactory or tolerable service is or could be provided, to the total observation period. Service availability depends on the class of space segment capacity, the earth station configuration and interference effects, and the bit error rate required. It is not possible to specify a service availability requirement for all two-way multipoint telecommunication services implementations via satellite. The availability for each implementation will have to be calculated on an individual basis considering all the points mentioned above. + +# 5 Class of space segment capacity + +Services offered may take account of classes of space segment capacity available from the space segment provider. The following classes of space segment capacity may be utilized to provide services: + +- a) *Non-pre-emptible* – A service which may not be interrupted or terminated for the provision of a service to another customer. There are two types of non-pre-emptible services: + - 1) *protected* – A service for which restoration is guaranteed; and + - 2) *unprotected* – A service for which restoration is not guaranteed and which may only be restored subject to availability of an alternate facility. +- b) *Pre-emptible* – A service which may be interrupted to provide a service of higher priority. \ No newline at end of file diff --git a/marked/F/T-REC-F.150-199110-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.150-199110-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..724021fa63095476b9a4e6ffeffc60e0f8dda051 --- /dev/null +++ b/marked/F/T-REC-F.150-199110-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:caf1204fd6f0ed83eedb51efd449e55093c1923c435d9bd406e22465634428d9 +size 7389 diff --git a/marked/F/T-REC-F.150-199110-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg b/marked/F/T-REC-F.150-199110-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3c8c794dc2514e550c73f9e4c9f36ef6d99162ed --- /dev/null +++ b/marked/F/T-REC-F.150-199110-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a947bef025fe582c4f37bd6855fb69d50974d34da5a256ef6657c738e982217d +size 5966 diff --git a/marked/F/T-REC-F.150-199110-I_PDF-E/raw.md b/marked/F/T-REC-F.150-199110-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..122c9c97b552653e0b7b820c3bcd4d2bac45b9d6 --- /dev/null +++ b/marked/F/T-REC-F.150-199110-I_PDF-E/raw.md @@ -0,0 +1,198 @@ + + +![ITU logo: A globe with the letters ITU and a lightning bolt.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with the letters ITU and a lightning bolt. + +INTERNATIONAL TELECOMMUNICATION UNION + +**CCITT** + +**F.150** + +THE INTERNATIONAL +TELEGRAPH AND TELEPHONE +CONSULTATIVE COMMITTEE + +**TELEGRAPH AND MOBILE SERVICES +OPERATIONS AND QUALITY OF SERVICE** + +--- + +**SERVICE AND OPERATIONAL +PROVISIONS FOR THE +INTEX SERVICE** + +**Recommendation F.150** + +--- + +![CCITT logo: A globe with the letters CCITT.](6ed175c791b5e156d9c98a8dbcc3318c_img.jpg) + +CCITT logo: A globe with the letters CCITT. + +Geneva, 1991 + +# FOREWORD + +The CCITT (the International Telegraph and Telephone Consultative Committee) is a permanent organ of the International Telecommunication Union (ITU). CCITT is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The Plenary Assembly of CCITT which meets every four years, establishes the topics for study and approves Recommendations prepared by its Study Groups. The approval of Recommendations by the members of CCITT between Plenary Assemblies is covered by the procedure laid down in CCITT Resolution No. 2 (Melbourne, 1988). + +Recommendation F.150 was prepared by Study Group I and was approved under the Resolution No. 2 procedure on the 11th of October 1991. + +## --- CCITT NOTES + +- 1) In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication Administration and a recognized private operating agency. +- 2) A list of abbreviations used in this Recommendation can be found in Annex A. + +## **Recommendation F.150** + +# **SERVICE AND OPERATIONAL PROVISIONS FOR THE INTEX1) SERVICE** + +## **1 Introduction** + +1.1 With the continuing development of the infrastructures of the networks which currently support, amongst others, the international telex service, it is feasible that some of these networks could support other services operating at speeds in excess of 50 bauds and using character sets other than those of International Telegraph Alphabet No. 2 (ITA2). There may be benefits to both customer and Administrations if a new text service could be introduced that would utilize the same network infrastructure as the international telex service. + +## **2 Scope** + +2.1 This Recommendation sets out the service and operational principles for the introduction of a new text service, initially operating asynchronously at 300 bit/s and using International Alphabet No. 5 (IA5). + +## **3 Service outline** + +3.1 This service will be real-time circuit switched and provide asynchronous operation with a basic transmission rate of 300 bit/s. The use of other transmission rates at a future date is for further study. + +3.2 The service will be half-duplex in its normal mode (see § 4.5) but will allow for full duplex operation in the transparent mode (see § 4.8). + +3.3 The Intex service will use the destination codes detailed in Recommendation F.69. + +3.4 It is recommended that the national numbering scheme for the Intex service be similar to and compatible with that used for the international telex service. + +## **4 Service description** + +4.1 Call set up procedures should be in accordance with the relevant Recommendations. + +4.2 To allow maximum flexibility of use of the service and to ensure the compatibility of terminals interworking at the higher speed, four separate operating modes may be provided. These are normal mode, telex interworking mode, extended mode and transparent mode. As a minimum, normal and telex interworking mode should be provided, with the choice of operating in three modes, other than telex interworking, being a customer option on a call-by-call basis, or within a call. + +4.3 At the beginning of any incoming or outgoing call an Intex terminal shall always default to the normal mode. + +4.4 Character sets, structure and procedures for generating mode changes should generally be in accordance with the relevant CCITT Recommendation. + +--- + +1) The use of the name Intex is conditional on its being freely available in Member countries and hence the name of the service offered may vary from country to country. + +### 4.5 *Normal mode* + +Normal mode is the mode of the basic Intex call for the exchange of text. There is no need for any prearrangement or pre-knowledge with regard to the capabilities of the called terminal over and above that of normal mode functions. The character set in this mode shall be derived from the International Alphabet No. 5 in accordance with Recommendation S.34. + +### 4.6 *Telex interworking mode* + +Telex interworking mode is intended for interworking with the international telex service (see § 7). This mode may be invoked at the start of a call where it is known that the called party is a terminal participating in the international telex service or it will be entered into automatically where the call is found to terminate on such a terminal and the service identifier **TLX** will be indicated at the Intex terminal. + +When the call is established in manual mode, the Intex operator should be informed if a previously-prepared message contains non-ITA2 characters. The procedures to be applied for automatic call set-up in the case where the Intex operator is not aware of the characteristics of the called party terminal are for further study. + +### 4.7 *Extended mode* + +Extended mode allows for the exchange of messages using the full character set of International Alphabet No. 5 in accordance with Recommendation T.50. This mode shall only be used when it has been established that it is supported by both terminals. + +### 4.8 *Transparent mode* + +In the transparent mode the character set to be used is defined by the customers involved in the call, subject only to the restriction on character structure within the network used. This mode shall only be used when it has been established that it is supported by both terminals. + +## 5 Answerback composition + +5.1 The Intex service shall employ the principle of the exchange of answerbacks, resident in the terminals and under the control of the service provider. The composition of the answerbacks for the Intex service must facilitate interworking, especially with the international telex service where many terminals employ answerback checking mechanisms based on format and length as well as content. + +5.2 Intex answerbacks shall include: + +- a) the customer's national number; +- b) the terminal identity letter or letters (if required); +- c) an abbreviated name designating the customer; +- d) the identification code, equivalent to the telex network identification code (TNIC) as specified in Recommendation F.69, preceded by a space character. + +5.3 The Intex answerback shall comprise a sequence of 20 IA5 characters, formatted in accordance with Recommendation S.35, as follows: + +- a) for terminals without identification letters: + - character US (unit separator); + - carriage return; + +- line feed; + - numeric characters representing the national number of the customer; + - character RS (record separator); + - space; + - alpha characters indicating the name of the customer; + - space; + - one or two characters representing the network identification code; + - character RS; +- b) for terminals with identification letters: +- character US; + - carriage return; + - line feed; + - numeric characters representing the national number of the customer; + - character RS; + - one or two alpha characters identifying the terminal; + - space; + - alpha characters indicating the name of the customer; + - space; + - one or two characters representing the network identification code; + - character RS; +- c) for terminals without identification letters and whose answerback code does not include letters indicating the name of the customer: +- character US; + - carriage return; + - line feed; + - numeric characters representing the national number of the customer; + - character RS; + - space; + - one or two characters representing the network identification code; + - character RS; +- d) for terminals with identification letters and whose answerback code does not include letters indicating the name of the customer: +- character US; + - carriage return; + - line feed; + - numeric characters representing the national number of the customer; + - character RS; + - one or two alpha characters identifying the terminal; + - space; + +- one or two characters representing the network identification code; +- character RS; +- e) for terminals providing mailbox or similar store and retrieval facilities: + - character US; + - carriage return; + - line feed; + - numeric characters representing the national number of the customer; + - = (“equals” sign); + - character RS; + - two alpha characters identifying the mailbox; + - space; + - one or two characters representing the network identification code; + - character RS. + +If the signals in the answerback code do not fill all the 20 places available, the unused places shall be filled by the necessary number of characters RS inserted before the network identification code. + +5.4 The alpha fields of Intex answerbacks shall comprise either upper- or lower case latin letters, but not mixed. + +5.5 Intex answerbacks, when subjected to the code conversion rules applied by networks, should result in answerback formats compatible with the requirements of the service with which Intex is interworking. + +## 6 Operations + +6.1 International interconnection of Intex services should be agreed bilaterally. + +6.2 International signalling and information transfer procedures should be in accordance with the relevant CCITT Recommendations. + +## 7 Interworking with the international telex service + +7.1 Interworking with the international telex service should be in accordance with Recommendation F.82. + +## ANNEX A (to Recommendation F.150) + +### Alphabetical list of abbreviations used in this Recommendation + +| | | +|------|----------------------------------------| +| IA5 | International Alphabet No. 5 | +| ITA2 | International Telegraph Alphabet No. 2 | +| RS | Record separator | +| TNIC | Telex network identification code | +| US | Unit separator | \ No newline at end of file diff --git a/marked/F/T-REC-F.16-199502-I_PDF-E/raw.md b/marked/F/T-REC-F.16-199502-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..7d207a84e211ad9a57866f215a2f92a34133f108 --- /dev/null +++ b/marked/F/T-REC-F.16-199502-I_PDF-E/raw.md @@ -0,0 +1,855 @@ + + +![ITU logo](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +The logo of the International Telecommunication Union (ITU) features a globe with a lightning bolt superimposed on it, and the letters 'ITU' written across the center. + +ITU logo + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.16** + +(02/95) + +**OPERATIONS AND QUALITY OF SERVICE +TELEGRAPH SERVICES** + +--- + +**GLOBAL VIRTUAL NETWORK SERVICE** + +**ITU-T Recommendation F.16** + +(Previously "CCITT Recommendation") + +--- + +# **FOREWORD** + +The ITU-T (Telecommunication Standardization Sector) is a permanent organ of the International Telecommunication Union (ITU). The ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Conference (WTSC), which meets every four years, establishes the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics. + +The approval of Recommendations by the Members of the ITU-T is covered by the procedure laid down in WTSC Resolution No. 1 (Helsinki, March 1-12, 1993). + +ITU-T Recommendation F.16 was prepared by ITU-T Study Group 1 (1993-1996) and was approved under the WTSC Resolution No. 1 procedure on the 21st of February 1995. + +# --- **NOTE** + +In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +# CONTENTS + +Page + +| | | | +|-----|--------------------------------------------------------------------|----| +| 1 | Definition ..... | 1 | +| 1.1 | Scope ..... | 1 | +| 1.2 | References..... | 1 | +| 1.3 | Definitions ..... | 1 | +| 1.4 | Abbreviations..... | 4 | +| 2 | Description ..... | 4 | +| 2.1 | General description..... | 4 | +| 2.2 | GVNS access ..... | 4 | +| 2.3 | GVNS calling..... | 6 | +| 2.4 | GVNS features description..... | 6 | +| 2.5 | GVNS operation, administration, maintenance and provisioning ..... | 7 | +| 2.6 | Applicability ..... | 8 | +| 3 | Procedures..... | 8 | +| 3.1 | Provision/withdrawal ..... | 8 | +| 3.2 | Normal procedures..... | 8 | +| 3.3 | Exceptional procedures..... | 8 | +| 3.4 | Alternate procedures ..... | 8 | +| 4 | Network capabilities for charging ..... | 9 | +| 5 | Interworking requirements ..... | 9 | +| 5.1 | Interworking with non-ISDN networks ..... | 9 | +| 5.2 | Interworking with the private networks (non-ISDN and ISDN) ..... | 9 | +| 6 | Interactions with supplementary services..... | 9 | +| 6.1 | Number related supplementary services ..... | 9 | +| 6.2 | Community of interest supplementary services ..... | 10 | +| 6.3 | Charging supplementary services ..... | 10 | +| 6.4 | Other supplementary services ..... | 10 | +| 7 | SDL diagrams..... | 10 | + +# **SUMMARY** + +The Global Virtual Network Service (GVNS) is a multi-network international service which provides private network functions to users at geographically dispersed international locations while minimizing the need for dedicated network resources. It may be offered to customers over the PSTN and/or ISDN. + +# **INTRODUCTION** + +This Recommendation focuses on how interworking service providers can provide the seamless Global Virtual Network Service (GVNS). It presents the service description of GVNS that supports global virtual networking. The GVNS is described here from the user's point of view and is characterized by a prose service definition description in accordance with Recommendation I.130. + +# GLOBAL VIRTUAL NETWORK SERVICE + +(Geneva, 1994) + +# 1 Definition + +The **Global Virtual Network Service (GVNS)** is a multi-network international service which provides private network functions to users at geographically dispersed international locations while minimizing the need for dedicated network resources. It may be offered to customers over the PSTN and/or ISDN. + +The Global Virtual Network Service is a feature-rich communications service. It provides the functions typically associated with the private networks (see 1.3.10), but utilizing the public switched network(s). The GVNS Customer's network configuration is defined per the customer's direction using customer-specific service information resident in multiple networks. The network configurations may be administered by the GVNS Customer directly, the GVNS Participating Service Provider(s) and/or the GVNS Coordinator(s) (see 1.3.11 and 1.3.12). + +The GVNS provides the customers with global services as a result of internetworking among the GVNS Participating Service Providers in various countries. GVNS may accommodate this interconnection both via ISDN and non-ISDN facilities. + +## 1.1 Scope + +This Recommendation contains the service description of GVNS for the PSTN and ISDN. This Recommendation defines the service from a user's perspective. This Recommendation does not specify the requirements where the service is provided to the user by the private networks; however, it addresses the interworking requirements with private networks. + +This Recommendation is the Stage 1 service description for the ISDN and is applicable to the ISDN Stages 2 and 3 Recommendations for GVNS. The terms Stages 1, 2 and 3 are defined in CCITT Recommendation I.130. + +## 1.2 References + +The following Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision: all users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. + +- CCITT Recommendation I.130 (1988), *Method for the characterization of telecommunication services supported by an ISDN and network capabilities of an ISDN*. +- ITU-T Recommendation I.112 (1993), *Vocabulary of terms for ISDNs*. +- ITU-T Recommendation I.210 (1993), *Principles of telecommunication services supported by an ISDN and the means to describe them*. +- ITU-T Recommendation Z.100 (1993), *CCITT specification and description language (SDL)*. + +## 1.3 Definitions + +**1.3.1 GVNS customer:** A **GVNS Customer (e.g. a person or an organization)** is the entity that purchases the service (e.g. the telecommunication manager of a corporation). The GVNS Customer functions as an interface between the GVNS Service Provider(s) and the GVNS Users. The GVNS Customer can define aspects of the service; for example, the on-net locations, numbering plan, and/or the calling privileges of the GVNS Users. + +**1.3.2 GVNS user:** A **GVNS User** is the end-user of the service. Typically, GVNS Users are not allowed to change their calling privileges. + +**1.3.3 on-net locations:** These are stations, user-network interfaces and authorized remote access locations that are logically defined by the GVNS Customer to be part of its virtual network during service provisioning or remote access (see 2.2.2.3). + +GVNS Customers may define a physical entity (station or user network interface) as multiple on-net locations during service provisioning. As a result, one physical entity may be registered to one or more GVNS User Groups. + +The relationship between the logical virtual networks and the physical networks is depicted in Figure 1. + +**1.3.4 off-net locations:** These are stations and user-network interfaces that are not defined by the GVNS Customer to be part of the GVNS. In some cases, a customer may define off-net locations as part of its customer-defined numbering plan. Such locations are termed as virtual on-net locations. + +**1.3.5 virtual on-net locations:** These are off-net locations included in the customer-defined numbering plan. The numbers that are defined in the numbering plan to reach to these off-net locations may be subject to GVNS-specific treatments (e.g. Call Screening). + +**1.3.6 GVNS calls:** GVNS Calls are calls from an on-net location. It may terminate to an on-net, off-net, or virtual on-net location. GVNS Calls include the following two types of calls: + +### – *On-net Calls:* + +An On-net Call is a call from an on-net location to another on-net location of the same GVNS User Group. + +### – *Off-net Calls:* + +An Off-net Call is a call from an on-net location to an off-net location. These calls can be specified with a customer-defined number (to a virtual on-net location) or as a public number. + +**1.3.7 GVNS user group and subgroup:** A GVNS user group is made up of a group of on-net locations defined by the GVNS Customer (including authorized remote access locations, see 2.2.2.3). + +A GVNS Customer may optionally define GVNS subgroups when they are offered by the GVNS Participating Service Provider. A GVNS subgroup is made up of a subset of the locations defined by the GVNS Customer as on-net locations (including authorized remote access locations). An on-net location belongs to one and only one GVNS subgroup. Subgroup identification is required to support subnetworking (see 2.4.8). The relationship between the user group and subgroups is depicted in Figure 1. + +**1.3.8 GVNS user group and subgroup identifier:** A GVNS user group is assigned a unique GVNS user group identifier to identify a group among supporting networks. The subgroup identifier uniquely identifies a group of users within a customer's GVNS user group, and has meaning only within that user group. + +**1.3.9 numbering plan:** A GVNS Customer has a numbering plan which associates addresses with locations in the GVNS (on-net addresses). These numbers may be public numbers or customer-defined (private). The GVNS Customers may also define private numbers to identify off-net addresses. When a GVNS Customer uses customer-defined numbers, the capabilities may include, but are not limited to those of the Support of Private Numbering Plans supplementary service. These capabilities are defined by each GVNS Participating Service Provider. + +**1.3.10 private network functions:** Private network functions include those provided by dedicated private networks and dedicated business groups. The private network functionality can be customer based (e.g. PBX), network based (e.g. business group) or a combination of both. + +**1.3.11 GVNS participating service provider:** The GVNS Participating Service Provider is a ROA that provides customer access and egress to and from the GVNS. The GVNS Participating Service Provider optionally offers the features defined in 2.4 on a national and/or international basis. The interworking of such services between ROAs in multiple countries creates the GVNS offering. + +![Diagram showing four virtual networks: User group W, User group X, User group Y, and User group Z. User group W contains Subgroup W1 (A1.A5.B1, B2.C2.C3, C4) and Subgroup W2 (A2.B4.B5, B6.C1). User group X contains Subgroup X1 (C5.C6.C7.A8) and Subgroup X2 (A4.B3.B7, A3.A6.A7). User group Y contains A8.B8.C8.C9. User group Z contains A9.B9.B10.](fc46871d72c65d3381d9201646d23439_img.jpg) + +Virtual (logical) networks + +Diagram showing four virtual networks: User group W, User group X, User group Y, and User group Z. User group W contains Subgroup W1 (A1.A5.B1, B2.C2.C3, C4) and Subgroup W2 (A2.B4.B5, B6.C1). User group X contains Subgroup X1 (C5.C6.C7.A8) and Subgroup X2 (A4.B3.B7, A3.A6.A7). User group Y contains A8.B8.C8.C9. User group Z contains A9.B9.B10. + +![Diagram showing three physical networks: Network A, Network B, and Network C. Network A has connections A1, A2, A3, A4, A5, A6, A7, A8, A9. Network B has connections B1, B2, B3, B4, B5, B6, B7, B8, B9, B10. Network C has connections C1, C2, C3, C4, C5, C6, C7, C8, C9. Connection A8 is shared between Network A and Network C.](1956f44611abd5c3c41049836aa78ad8_img.jpg) + +Physical networks (GVNS Participating Networks) + +Diagram showing three physical networks: Network A, Network B, and Network C. Network A has connections A1, A2, A3, A4, A5, A6, A7, A8, A9. Network B has connections B1, B2, B3, B4, B5, B6, B7, B8, B9, B10. Network C has connections C1, C2, C3, C4, C5, C6, C7, C8, C9. Connection A8 is shared between Network A and Network C. + +T0102660-92/d01 + +User Group W +Subgroup W1 +Subgroup W2 + +User Group X +Subgroup X1 +Subgroup X2 + +User Group Y +No Subgroups + +User Group Z +No Subgroups + +GVNS participating service provider for W = A, B, C + GVNS participating service provider for X = A, B, C + GVNS participating service provider for Y = C, A, B + GVNS participating service provider for Z = B, A + A1 ... B9 = physical connections to Network A + B1 ... B10 = physical connections to Network B + C1 ... C9 = physical connections to Network C + A8 = one physical connection belonging to two user groups (X and Y) + +FIGURE 1/F.16 +**GVNS user groups and subgroups** + +**1.3.12 GVNS coordinator:** One or more GVNS Coordinators may be elected by the GVNS Customer to provide and arrange with other ROAs (GVNS Participating Service Providers) the provision of GVNS. + +The GVNS Coordinator(s) will arrange customers ordering, provisioning, and billing procedures currently undertaken by each GVNS Participating Service Provider for the establishment and ongoing provision of GVNS. + +## 1.4 Abbreviations + +| | | +|-------|---------------------------------------------------------| +| CLI | Calling Line Identification | +| GVNS | Global Virtual Network Service | +| ISDN | Integrated Services Digital Network | +| PBX | Private Branch Exchange | +| PIN | Personal Identification Number | +| OAM&P | Operation, Administration, Maintenance and Provisioning | +| PSTN | Public Switched Telephone Network | +| ROA | Recognized Operating Agency | +| SDL | Specification and Description Language | + +# 2 Description + +## 2.1 General description + +The GVNS allows service providers to offer subscribing customers a service with the features and functionality similar to that of a private network while minimizing the need for dedicated network resources. However, dedicated network resources may be used to access GVNS or in conjunction with GVNS. + +A GVNS Customer may optionally elect one or more GVNS Coordinators to provide one or more of the following functions: + +- a) Single End Service Ordering; +- b) Single End Fault Reporting; +- c) Maintenance Coordination; +- d) Configuration Management; +- e) Network Management Reporting Tools. + +A GVNS Customer is assigned a unique identifier which identifies the global GVNS User Group (see 1.3.7). GVNS User Group identification provides a means by which various networks can recognize the GVNS Customer. + +A GVNS Customer may choose to group users into one or more subgroups (see 1.3.7) in order to allow subgroup dependent service interactions and operations. + +GVNS users can be located in one or more networks, and may interwork with private network resources. + +A variety of routing privileges may be assigned to GVNS calls. GVNS users can make calls to other users within their GVNS and to non-GVNS users. GVNS calls can also be made from non-GVNS locations. At the customer's request, appropriate verification of the calling party may be required. + +## 2.2 GVNS access + +Dependent upon GVNS Participating Service Provider capabilities, the customer can choose from various user access procedures and physical access arrangements. + +### **2.2.1 User access procedures** + +Based upon the GVNS Participating Service Provider capabilities, the customer can choose either direct or indirect access to the GVNS. + +#### **2.2.1.1 Direct access** + +This type of access allows the user to reach the GVNS without providing any code(s). This is only possible from on-net locations if the network recognizes the GVNS location by performing Calling Line Identification (CLI) screening or using another automatic mechanism (e.g. physical network connection). + +An example of User Access Procedures: + +- a) the GVNS user lifts the handset and waits for network response; +- b) the network sends an acknowledgement (e.g. dial tone); +- c) depending upon customer requirements, the user may be required to provide an authorization code and the network sends an acknowledgement to the user; +- d) the user is now connected to the GVNS and can use the service features available. + +If the GVNS location from which the user accesses its GVNS is a PBX extension, the procedure could change in a way that is PBX dependent (e.g. dialling a prefix). + +#### **2.2.1.2 Indirect access** + +This type of access allows the GVNS user to reach its GVNS by providing a service access code and/or authorization code. This type of access is available both from on-net and off-net locations. + +##### **2.2.1.2.1 Indirect access from on-net locations** + +When indirect access is used from an on-net location, the user is only required to provide a service access code. The network recognizes the GVNS location by CLI screening or using another automatic mechanism (e.g. physical network connection). + +An example of User Access Procedures: + +- a) the GVNS user lifts the handset and waits for a network response; +- b) the network sends a network acknowledgement; +- c) the user provides a service access code; +- d) the network sends another network acknowledgement; +- e) depending upon customer requirement, the user may be required to provide an authorization code and the network sends an acknowledgement to the user; +- f) the user is now connected to the GVNS and can use the service features available. + +##### **2.2.1.2.2 Indirect access from off-net locations** + +When indirect access is used from an off-net location (see Remote Access 2.2.2.3), the network cannot identify the user via CLI or another automated mechanism. Therefore, the user is required to provide both a Remote Access Number and an authorization code. The user privileges and network authentication are based upon the authorization code. + +An example of User Access Procedures: + +- a) the GVNS user lifts the handset and waits for a network response; +- b) the network sends a network acknowledgement; +- c) the Remote Access Number (e.g. a public number) is provided by the user; +- d) an indication is provided to guide the caller to provide an authorization code; +- e) the authorization code is provided. + +Depending upon the GVNS Participating Network Provider's implementation, the authorization code may be a single set of digits or split (e.g. separate network prompts for authorization code and PIN); + +- f) the network informs the user by means of tone(s), announcement(s), and/or other mechanisms that indicate access to the GVNS has been successful. + +### **2.2.2 Physical access arrangements** + +Dependent upon GVNS Participating Service Provider capability, the customer can choose from various Dedicated, Switched and Remote Access arrangements to receive the features and functions of the GVNS. + +#### **2.2.2.1 Dedicated access** + +Dedicated access is a physical connection from the user to a network that supports the relevant GVNS for only one GVNS Customer. If the GVNS for only one GVNS Customer uses the same dedicated access for more than one service, the user may need to dial a special code (e.g. prefix or access code) to identify GVNS calls. + +#### **2.2.2.2 Switched access** + +Switched access is a switched connection from the user to a GVNS via one or more networks on which: + +- a) the particular GVNS is not supported; and +- b) the circuits to the GVNS Participating Service Provider carry the GVNS traffic of various GVNS User Groups and possibly non-GVNS traffic. + +The user may need to dial a special code (e.g. prefix or access code) to identify the desired GVNS Participating Service Provider. The availability of Calling Line Identification (network provided or network verified) is necessary to provide authorized access to the GVNS. + +#### **2.2.2.3 Remote access** + +Remote access can be used to make GVNS calls from locations not identified as part of the GVNS at service provisioning. The caller makes a call to a Remote Access Number (e.g. public number or service code) and, optionally, provides an authorization code. Either the Remote Access Number or the authorization code provides authorized access to the GVNS. Once a non-GVNS location receives authorized access from a GVNS participating service provider, it becomes an on-net location (authorized remote access location, see 1.3.3). + +## **2.3 GVNS calling** + +All calls made from an on-net location may terminate on an on-net, off-net, or virtual on-net location. If terminating on an on-net location, it could be done via dedicated or a switched access. The user may provide a customer-defined or a public number to reach on-net, off-net or virtual on-net locations. + +## **2.4 GVNS features description** + +A GVNS Participating Service Provider may support one or more of the following optional features. The GVNS Participating Service Providers will interwork their systems and procedures to provide GVNS features for customers in multiple countries. + +### **2.4.1 Call screening** + +The GVNS Customer can define call screening to determine what type of calls are allowed from on-net or remote access locations. This may include, but is not limited to, restrictions such as, allow private number dialled calls only, or allow all private number and public number calls in a particular country only. + +Call screening is also a mechanism which may be used by the GVNS Customer to restrict GVNS users access and/or calling privileges. + +### **2.4.2 Customer-defined numbering** + +GVNS allows the customer to define a private numbering plan for some or all of their dialling needs. The number of digits sent and received between the customer's equipment and the GVNS service point will be at the customer's discretion, within a given range, specified by each GVNS Participating Service Provider. The GVNS Customers should have the capability to define an on-net location with a customer-defined private number or a public number. + +### **2.4.3 Private network interface** + +The GVNS Customers should be able to interconnect with their private network(s) using GVNS. The GVNS must be able to communicate to and from the customer's private network(s) (see 5.2). + +### **2.4.4 Standard announcements** + +General and network announcements are to be provided when appropriate for GVNS calls. A verbal announcement may be played when a certain predefined condition occurs (e.g. "All circuits are busy" at the general level; "You are not authorized to dial this number" at the network level). + +### **2.4.5 Customized announcements** + +A GVNS Customer can select a particular announcement to the caller given the dialled digits or any of several intercept conditions (e.g. "Welcome to XYZ company's private network"). The customer sets up the routing to the announcements. The customized announcements capability may have the flexibility to support multiple messages in different languages and countries. + +### **2.4.6 Authorization code** + +Authorization codes may be used by the GVNS Customer to identify whether the GVNS User has the authorization to place the call. Authorization codes may also be used to assign varying levels of functionality. As a GVNS Service Provider option, the calling party could place additional calls without requiring re-entry of the authorization code. + +### **2.4.7 Accounting code** + +The accounting code may be used by the user to identify an individual or group with which a GVNS call should be associated. + +### **2.4.8 Subnetworking (subgrouping)** + +A customer of the GVNS service may choose to group its users into subnetworks with their own access and routing restrictions. Optionally, each subnetwork may have its own numbering plan. The subnetworking provides the flexibility to the GVNS Customer by allowing its users to associate subnetworks with functional organizations, e.g. manufacturing, marketing, or with geographical locations. + +### **2.4.9 Direct termination overflow** + +If the dedicated access for the GVNS call termination is busy, the incoming call can overflow to another dedicated access, switched access, or to an announcement. + +## **2.5 GVNS operation, administration, maintenance and provisioning** + +The GVNS coordinator(s), if elected (see 1.3.12) will coordinate the Operation, Maintenance, Administration and Provisioning (OAM&P) aspects of the GVNS with the service providers involved in providing the service in multiple countries. The supporting networks should be able to identify the GVNS coordinator(s), i.e. the country and the network in order to provide some of the OAM&P functionalities, e.g. billing. Some examples of such OAM&P activities are listed below1). + +### **2.5.1 Provisioning** + +This function provides means of order processing, provisioning and installing the GVNS service for the customer. + +### **2.5.2 Configuration management** + +The customer should be able to configure its GVNS according to its needs from domestic and international locations2). The configuration management involves GVNS dialling plan changes, activation/deactivation of authorization codes, routing changes, etc. + +--- + +1) The concept of Telecommunication Management Network (TMN) is in the M-Series Recommendations. The TMN embodies these OAM&P aspects. TMN would facilitate the customer being able to access one system for configuration management. + +2) The GVNS coordinator can also provide the configuration management capability on the customer's behalf. + +### **2.5.3 Performance monitoring** + +In order to properly monitor the performance of its GVNS, the customer should be able to obtain various traffic statistics, e.g. trunk group usage, originating/terminating traffic from/to a particular location. The GVNS coordinator or the customer may create reports based on these statistics. These reports could provide input for decisions to reconfigure and/or add additional resources to the GVNS. + +## **2.6 Applicability** + +This service is applicable to the PSTN and circuit mode bearer services in the ISDN. + +# **3 Procedures** + +Specific procedures for the provision and withdrawal of access methods (see 2.2) and features (see 2.4) could vary by GVNS Participating Service Providers and, therefore, are not defined below. + +## **3.1 Provision/withdrawal** + +The GVNS is provided to the customer by subscription. Provision and coordination of GVNS may be required in multiple networks depending on the location of the GVNS Users. + +## **3.2 Normal procedures** + +### **3.2.1 Activation/deactivation/registration** + +Activation, deactivation, and registration of GVNS may be performed by the GVNS Participating Service Providers and/or Coordinator(s) by arrangements with the customer. If the customer elects a GVNS Coordinator, the Coordinator may arrange activation, deactivation, and registration among the GVNS Participating Service Providers. The GVNS Participating Service Providers and/or Coordinator(s) and the customer can change the information relating to GVNS, e.g. on-net and off-net locations by following appropriate procedures. + +### **3.2.2 Invocation and operation** + +Depending upon the access arrangement (see 2.2.2), a GVNS User may be required to dial an access code, Remote Access Number, or Authorization Code which identifies the call as being a GVNS call. Optionally, the user only dials a customer-defined or public number. The network then, handles the call according to the customer-specific record stored in the network, (e.g. routes to the proper destination), and creates a record for billing, international accounting, and various reporting purposes. Call screening may be performed before the call is routed to ensure calling security and authorization. + +## **3.3 Exceptional procedures** + +If the call cannot be routed to the proper destination due to, e.g. unavailability of network resources, unauthorized access, an appropriate indication shall be returned to the calling user. + +### **3.3.1 Activation/deactivation/registration** + +None identified. + +### **3.3.2 Invocation and operation** + +None identified. + +## **3.4 Alternate procedures** + +None identified. + +# **4 Network capabilities for charging** + +The GVNS Participating Service Provider may choose to offer a call charge for GVNS which is different from other network charges as part of the GVNS service. The exact nature of the call charges is the subject of the D-Series Recommendations. In any case, the GVNS Participating Service Provider should record the relevant details in order to be able to charge the customer accurately for the service. + +# **5 Interworking requirements** + +## **5.1 Interworking with non-ISDN networks** + +If during interworking it is not possible to sustain the association of the call with the specific GVNS information, the call may still be allowed to proceed if the network has enough information to bill the call properly and the customer has chosen this treatment. + +## **5.2 Interworking with the private networks (non-ISDN and ISDN)** + +The GVNS Customers should be able to interconnect with their private network(s) using GVNS. The GVNS must be able to communicate to and from the customer's private network(s). + +The GVNS service provides the same features and functions to users directly connected to the public networks as to users connected via private networks. Locations connected to GVNS via private network(s) are not required to send or to receive additional information for GVNS calls over and beyond those required to initiate or to complete basic calls via PSTN or ISDN. + +Certain features and functions provided within the private networks may not be implemented on GVNS. + +# **6 Interactions with supplementary services** + +Interaction with the ISDN supplementary services defined in the I.250-Series Recommendations is described in this clause. The GVNS interactions for some of these supplementary services are defined below. Interactions with the remaining supplementary services are for further study. + +A description of the interaction with PSTN supplementary services is not possible as there are no descriptions that have been standardized on an international basis. + +A list of possible supplementary telephone services is contained in Supplement No. 1 of the CCITT *Blue Book*, Fascicle II.2. + +## **6.1 Number related supplementary services** + +### **6.1.1 Calling Line Identification Presentation** + +No customer-defined number will be provided to an off-net location. + +If a GVNS call is made to an off-net location (including to a virtual on-net location), the calling line identity is the public number. If no public number is available, the calling line identity will not be provided to the called user. If the call is forwarded to an on-net location, the customer-defined number is provided unless the customer specifies a public number. + +If a GVNS call is made to an on-net location, the calling line identity is the customer-defined number; however, if a public number is provided by the caller, the calling line identity is the public number. If the call is forwarded to another on-net location, the same calling line identity treatment is provided. + +### **6.1.2 Calling Line Identification Restriction** + +No impact. + +### **6.1.3 Connected Line Identification Presentation** + +No customer-defined number will be provided to an off-net location. + +If a GVNS call is made to a public number, the connected line identity is the public number. If the call is forwarded to another public number, that public number will be the connected line identity. If the call is forwarded to an on-net or virtual on-net location, the connected line identity is the customer-defined number of that location; however, if a public number is provided by the connecting user, the connected line identity is the public number. + +If a GVNS call is made to an on-net location or virtual on-net location, the connected line identity is the customer-defined number of that location; however, if a public number is provided by the connecting user, the connected line identity is the public number. If the call is forwarded to another on-net location, the connected line identity is the customer-defined number of that location; if the call is forwarded to a public number, the connected line identity is the public number of that location. + +### **6.1.4 Connected Line Identification Restriction** + +No impact. + +### **6.1.5 Sub-addressing** + +No impact. Subaddressing can be employed as an addition to the GVNS defined number. + +## **6.2 Community of interest supplementary services** + +### **6.2.1 Support of Private Numbering Plans** + +It is possible for the GVNS Customer to assign private numbers for all of their calling needs. Additional aspects are for further study. + +## **6.3 Charging supplementary services** + +### **6.3.1 Telecommunications Charge Card Service** + +It may be possible to make GVNS calls using a charge card. Additional aspects are for further study. + +## **6.4 Other supplementary services** + +### **6.4.1 User-to-User Signalling** + +No impact. + +# **7 SDL diagrams** + +Figures 2 and 3 specify the dynamic description of GVNS Stage 1 using graphic means. + +Process GVNS User + +![SDL diagram for user's view of GVNS. The process starts with a note about the calling party being a GVNS user who sends a service request. It then enters a state, sends a GVNS Request to Network (linked to section 3.2.2), and waits for a response. The response can be a Dialled Number Request, Authorization Code Request, Account Code Request, or Interim Calling User Call Set-up Success. The first three lead to corresponding network responses and then back to a waiting state. The fourth leads to the GVNS Active state.](bd0b93e7a46ede276d0a3b79ac487bd9_img.jpg) + +``` + +stateDiagram-v2 + [*] --> State1 + State1 --> GVNS_Request_to_Network : GVNS Request to Network [ See 3.2.2 ] + GVNS_Request_to_Network --> Waiting1 : Waiting for GVNS Response + Waiting1 --> Dialled_Number_Request_From_Network : Dialled Number Request From Network + Waiting1 --> Authorization_Code_Request_from_Network : Authorization Code Request from Network + Waiting1 --> Account_Code_Request_from_Network : Account Code Request from Network + Waiting1 --> Interim_Calling_User_Call_Set-up_Success : Interim Calling User Call Set-up Success + Dialled_Number_Request_From_Network --> Dialled_Number_to_Network : Dialled Number to Network + Authorization_Code_Request_from_Network --> Authorization_Code_to_Network : Authorization Code to Network + Account_Code_Request_from_Network --> Account_Code_to_Network : Account Code to Network + Dialled_Number_to_Network --> Waiting2 : Waiting for GVNS Response + Authorization_Code_to_Network --> Waiting2 + Account_Code_to_Network --> Waiting2 + Interim_Calling_User_Call_Set-up_Success --> GVNS_Active : GVNS Active + note left of State1 : /* Calling party is a GVNS User + The user knows the type of access being used and sends the appropriate GVNS service request. + */ + note right of GVNS_Request_to_Network : T0104310-95/d02 + +``` + +SDL diagram for user's view of GVNS. The process starts with a note about the calling party being a GVNS user who sends a service request. It then enters a state, sends a GVNS Request to Network (linked to section 3.2.2), and waits for a response. The response can be a Dialled Number Request, Authorization Code Request, Account Code Request, or Interim Calling User Call Set-up Success. The first three lead to corresponding network responses and then back to a waiting state. The fourth leads to the GVNS Active state. + +FIGURE 2/F.16 (sheet 1 of 2) +**SDL diagram for user's view of GVNS** + +Process GVNS User + +![SDL diagram for user's view of GVNS. The diagram shows two state transition paths. The top path starts with 'Waiting for GVNS Response' (state), followed by 'Interim Calling User Call Set-up Failure' (event). It then reaches a decision 'Follow-on Status' (X). If 'Prompt not received' (Z), it goes to 'Follow-on Request to Network' (output), then 'Waiting for Follow-on Response' (state), leading to 'Follow-on Offer' (output) or 'Follow-on Limit Exceeded' (output). 'Follow-on Offer' leads to state Y. If 'Follow-on not available', the path ends. If 'Prompt received' (Y), it reaches 'User Action?' (decision). 'Reject' leads to 'Disconnect Request to Network' (output) and ends. 'Continue' leads to 'Dialled Number to Network' (output), then 'Waiting for GVNS Response' (state). The bottom path starts with 'GVNS Active' (state), leading to three parallel outputs: 'Calling User Disconnect' (output) leading to 'Disconnect Request to Network' (output) and ending; 'Calling User Follow-on' (output) leading to state Z; and 'Calling User Disconnect' (output) leading to state X.](5cab96b2d23174c25919840ecd50aa48_img.jpg) + +``` + +stateDiagram-v2 + [*] --> S1: Waiting for GVNS Response + S1 --> E1: Interim Calling User Call Set-up Failure + E1 --> D1: Follow-on Status + D1 --> O1: Follow-on Request to Network: Prompt not received (Z) + O1 --> S2: Waiting for Follow-on Response + S2 --> O2: Follow-on Offer + S2 --> O3: Follow-on Limit Exceeded + O2 --> S3: Y + O3 --> [*] + D1 --> D2: Prompt received: Follow-on not available + D2 --> [*] + D1 --> D3: User Action?: Prompt received (Y) + D3 --> O4: Dialled Number to Network: Continue + O4 --> S4: Waiting for GVNS Response + D3 --> O5: Disconnect Request to Network: Reject + O5 --> [*] + + S5: GVNS Active --> O6: Calling User Disconnect + S5 --> O7: Calling User Follow-on + S5 --> O8: Calling User Disconnect + O6 --> O9: Disconnect Request to Network + O9 --> [*] + O7 --> S6: Z + O8 --> S7: X + +``` + +SDL diagram for user's view of GVNS. The diagram shows two state transition paths. The top path starts with 'Waiting for GVNS Response' (state), followed by 'Interim Calling User Call Set-up Failure' (event). It then reaches a decision 'Follow-on Status' (X). If 'Prompt not received' (Z), it goes to 'Follow-on Request to Network' (output), then 'Waiting for Follow-on Response' (state), leading to 'Follow-on Offer' (output) or 'Follow-on Limit Exceeded' (output). 'Follow-on Offer' leads to state Y. If 'Follow-on not available', the path ends. If 'Prompt received' (Y), it reaches 'User Action?' (decision). 'Reject' leads to 'Disconnect Request to Network' (output) and ends. 'Continue' leads to 'Dialled Number to Network' (output), then 'Waiting for GVNS Response' (state). The bottom path starts with 'GVNS Active' (state), leading to three parallel outputs: 'Calling User Disconnect' (output) leading to 'Disconnect Request to Network' (output) and ending; 'Calling User Follow-on' (output) leading to state Z; and 'Calling User Disconnect' (output) leading to state X. + +T0104320-95/d03 + +FIGURE 2/F.16 (sheet 2 of 2) +**SDL diagram for user's view of GVNS** + +# PROCESS GVNS\_Service\_Provider + +![SDL diagram for GVNS service provider's view of GVNS. The diagram shows a state transition from 'Idle' to 'Calling User GVNS Access Request'. From there, it branches based on 'Access arrangement' ('Dedicated', 'Switched', 'Remote'). The 'Dedicated' and 'Switched' paths lead to 'Response To Calling User' and then 'Wait for Dialled Number'. The 'Remote' path leads to a decision 'Collect dialled number first?'. If 'Yes', it goes to 'Collect Dialled Number' and then 'Success?'. If 'No', it goes to 'Collect Authorization Code' and then 'Success?'. Both 'Success?' decisions lead to 'Terminate Call' if 'No', or to a connector 'G' if 'Yes'. A note on the left indicates that for remote access, the user has dialed an appropriate code. Another note indicates that the B party number may have been included in the access request. Four boxes at the top right list actions: Terminate Call, Collect Dialled Number, Collect Authorization Code, and Collect Account Code.](c85ded401105f62f2d6ff26b3b5eb4af_img.jpg) + +``` + +stateDiagram-v2 + [*] --> Idle + Idle --> CallingUserGVNSAccessRequest: Calling User GVNS Access Request + CallingUserGVNSAccessRequest --> AccessArrangement: Access arrangement? + AccessArrangement --> Dedicated: 'Dedicated' + AccessArrangement --> Switched: 'Switched' + AccessArrangement --> Remote: 'Remote' + Dedicated --> ResponseToCallingUser: Response To Calling User + Switched --> ResponseToCallingUser + Remote --> CollectDialledNumberFirst: 'Collect dialled number first?' + CollectDialledNumberFirst --> Yes1: 'Yes' + CollectDialledNumberFirst --> No1: 'No' + Yes1 --> CollectDialledNumber: Collect Dialled Number + CollectDialledNumber --> Success1: 'Success?' + Success1 --> Yes2: 'Yes' + Success1 --> No2: 'No' + No2 --> CollectAuthorizationCode: Collect Authorization Code + CollectAuthorizationCode --> Success2: 'Success?' + Success2 --> Yes3: 'Yes' + Success2 --> No3: 'No' + No3 --> TerminateCall: Terminate Call + Yes2 --> G: G + Yes3 --> G + TerminateCall --> [*] + Note left of CallingUserGVNSAccessRequest: /* High level description of network actions provided by GVNS service provide */ + Note right of AccessArrangement: For remote access, user has dialled appropriate code. For direct or switched access, user may have dialled access code to identify GVNS service or GVNS service provider. Call has been routed to correct service provider. + Note left of ResponseToCallingUser: If required. B party number may have been included in access request + Note right of TerminateCall: Terminate Call /* Q.71 actions */ + Note right of CollectDialledNumber: Collect Dialled Number + Note right of CollectAuthorizationCode: Collect Authorization Code + Note right of CollectAccountCode: Collect Account Code + +``` + +SDL diagram for GVNS service provider's view of GVNS. The diagram shows a state transition from 'Idle' to 'Calling User GVNS Access Request'. From there, it branches based on 'Access arrangement' ('Dedicated', 'Switched', 'Remote'). The 'Dedicated' and 'Switched' paths lead to 'Response To Calling User' and then 'Wait for Dialled Number'. The 'Remote' path leads to a decision 'Collect dialled number first?'. If 'Yes', it goes to 'Collect Dialled Number' and then 'Success?'. If 'No', it goes to 'Collect Authorization Code' and then 'Success?'. Both 'Success?' decisions lead to 'Terminate Call' if 'No', or to a connector 'G' if 'Yes'. A note on the left indicates that for remote access, the user has dialed an appropriate code. Another note indicates that the B party number may have been included in the access request. Four boxes at the top right list actions: Terminate Call, Collect Dialled Number, Collect Authorization Code, and Collect Account Code. + +T0103490-94/d04 + +FIGURE 3/F.16 (sheet 1 of 7) + SDL diagram for GVNS service provider's view of GVNS + +![SDL diagram for GVNS service provider's view of GVNS. The process starts with 'Wait for Dialled Number', leading to 'Calling Party Disconnect' and 'Dialled Number /* from Calling user */'. 'Calling Party Disconnect' leads to 'Terminate Call' and then an exit. 'Dialled Number' leads to 'Identify GVNS User Group', which then leads to a decision 'More Information?'. 'More Information?' has 'Yes' and 'No' paths. 'Yes' leads to 'Account code first?'. 'No' leads to connector D. 'Account code first?' has 'Yes' and 'No' paths. 'Yes' leads to connector B and 'Collect Account code'. 'No' leads to connector A and 'Collect Authorization Code'. Both 'Collect Account code' and 'Collect Authorization Code' lead to 'Success?' decisions. 'Success?' has 'Yes' and 'No' paths. 'Yes' leads to 'Authorization code also and not yet collected?' or 'Account code also and not yet collected?'. 'No' leads to 'Terminate Call'. 'Authorization code also and not yet collected?' has 'Yes' and 'No' paths. 'Yes' leads to connector A. 'No' leads to an exit. 'Account code also and not yet collected?' has 'Yes' and 'No' paths. 'Yes' leads to connector B. 'No' leads to an exit. Both exits lead to a decision 'Dialled number collected?'. 'Dialled number collected?' has 'Yes' and 'No' paths. 'Yes' leads to connector D. 'No' leads to connector C. Connector G also leads to the 'Dialled number collected?' decision.](76b0cd79baaedd942af4dc42f2e764b8_img.jpg) + +PROCESS GVNS\_Service\_Provider + +``` + +stateDiagram-v2 + [*] --> Wait + state "Wait for Dialled Number" as Wait + Wait --> CPD + state "Calling Party Disconnect" as CPD + CPD --> TC1 + state "Terminate Call" as TC1 + TC1 --> [*] + Wait --> DN + state "Dialled Number /* from Calling user */" as DN + DN --> IGUG + state "Identify GVNS User Group" as IGUG + IGUG --> MI + state "More Information?" as MI + MI --> D1((D)) : No + MI --> ACF + state "Account code first?" as ACF + ACF --> B1((B)) : Yes + ACF --> A1((A)) : No + ACF --> CAC + state "Collect Account code" as CAC + ACF --> CACode + state "Collect Authorization Code" as CACode + CAC --> S1 + state "Success?" as S1 + S1 --> [*] : No + S1 --> ACNYC + state "Authorization code also and not yet collected?" as ACNYC + ACNYC --> A2((A)) : Yes + ACNYC --> [*] : No + CACode --> S2 + state "Success?" as S2 + S2 --> [*] : No + S2 --> ACCNYC + state "Account code also and not yet collected?" as ACCNYC + ACCNYC --> B2((B)) : Yes + ACCNYC --> [*] : No + [*] --> DNC + state "Dialled number collected?" as DNC + DNC --> C((C)) : No + DNC --> D2((D)) : Yes + G((G)) --> DNC + +``` + +SDL diagram for GVNS service provider's view of GVNS. The process starts with 'Wait for Dialled Number', leading to 'Calling Party Disconnect' and 'Dialled Number /\* from Calling user \*/'. 'Calling Party Disconnect' leads to 'Terminate Call' and then an exit. 'Dialled Number' leads to 'Identify GVNS User Group', which then leads to a decision 'More Information?'. 'More Information?' has 'Yes' and 'No' paths. 'Yes' leads to 'Account code first?'. 'No' leads to connector D. 'Account code first?' has 'Yes' and 'No' paths. 'Yes' leads to connector B and 'Collect Account code'. 'No' leads to connector A and 'Collect Authorization Code'. Both 'Collect Account code' and 'Collect Authorization Code' lead to 'Success?' decisions. 'Success?' has 'Yes' and 'No' paths. 'Yes' leads to 'Authorization code also and not yet collected?' or 'Account code also and not yet collected?'. 'No' leads to 'Terminate Call'. 'Authorization code also and not yet collected?' has 'Yes' and 'No' paths. 'Yes' leads to connector A. 'No' leads to an exit. 'Account code also and not yet collected?' has 'Yes' and 'No' paths. 'Yes' leads to connector B. 'No' leads to an exit. Both exits lead to a decision 'Dialled number collected?'. 'Dialled number collected?' has 'Yes' and 'No' paths. 'Yes' leads to connector D. 'No' leads to connector C. Connector G also leads to the 'Dialled number collected?' decision. + +T0103500-94/d05 + +FIGURE 3/F.16 (sheet 2 of 7) + SDL diagram for GVNS service provider's view of GVNS + +PROCESS GVNS\_Service\_Provider + +![SDL diagram for GVNS service provider's view of GVNS. The process starts at connector 'C', proceeds to 'Collect Dialled Number', then checks 'Success?'. If 'No', it goes to connector 'D'. If 'Yes', it checks 'GVNS User Group Identified?'. If 'No', it goes to 'Identify GVNS User Group' then checks 'Success?'. If 'Yes', it goes to 'Call Screening' then checks 'Passed?'. If 'No', it goes to connector 'D'. If 'Yes', it checks 'Remote Access?'. If 'Yes', it checks 'Account Code Needed?'. If 'Yes', it goes to 'Collect Account Code' then checks 'Success?'. If 'No', it goes to 'Terminate Call'. If 'Yes', it goes to connector 'H'. If 'No', it goes to connector 'D'. If 'No' to 'Remote Access?', it goes to connector 'H'. From 'H', it goes to 'Set up GVNS call' then checks 'Success?'. If 'No', it goes to connector 'E'. If 'Yes', it goes to 'GVNS Active'. From 'E', it checks 'Follow on available?'. If 'No', it goes to connector 'F'. If 'Yes', it goes to 'Follow on Offer to the Calling User' then 'Wait for user action'. From 'F', it goes to 'Play Announcement To the Calling User' then 'Terminate Call'.](7f25db95ce3916c0e09803b861a2f7bc_img.jpg) + +``` + +stateDiagram-v2 + [*] --> C + C --> CollectDialledNumber[Collect Dialled Number] + CollectDialledNumber --> Success1{Success?} + Success1 -- No --> D + Success1 -- Yes --> GVNSUserGroupIdentified{GVNS User Group Identified?} + GVNSUserGroupIdentified -- No --> IdentifyGVNSUserGroup[Identify GVNS User Group] + IdentifyGVNSUserGroup --> Success2{Success?} + Success2 -- No --> D + Success2 -- Yes --> CallScreening[Call Screening] + CallScreening --> Passed{Passed?} + Passed -- No --> D + Passed -- Yes --> RemoteAccess{Remote Access?} + RemoteAccess -- Yes --> AccountCodeNeeded{Account Code Needed?} + AccountCodeNeeded -- Yes --> CollectAccountCode[Collect Account Code] + CollectAccountCode --> Success3{Success?} + Success3 -- No --> TerminateCall1[Terminate Call] + Success3 -- Yes --> H + AccountCodeNeeded -- No --> H + RemoteAccess -- No --> H + H --> SetupGVNSCall[Set up GVNS call] + SetupGVNSCall --> Success4{Success?} + Success4 -- No --> E + Success4 -- Yes --> GVNSActive([GVNS Active]) + E --> FollowOnAvailable{Follow on available?} + FollowOnAvailable -- No --> F + FollowOnAvailable -- Yes --> FollowOnOffer[Follow on Offer to the Calling User] + FollowOnOffer --> WaitForUserAction([Wait for user action]) + F --> PlayAnnouncement[Play Announcement To the Calling User] + PlayAnnouncement --> TerminateCall2[Terminate Call] + +``` + +Reasons can include: + - not subscribed; + - not provided by N/W; + - allowed number of calls exceeded. + +SDL diagram for GVNS service provider's view of GVNS. The process starts at connector 'C', proceeds to 'Collect Dialled Number', then checks 'Success?'. If 'No', it goes to connector 'D'. If 'Yes', it checks 'GVNS User Group Identified?'. If 'No', it goes to 'Identify GVNS User Group' then checks 'Success?'. If 'Yes', it goes to 'Call Screening' then checks 'Passed?'. If 'No', it goes to connector 'D'. If 'Yes', it checks 'Remote Access?'. If 'Yes', it checks 'Account Code Needed?'. If 'Yes', it goes to 'Collect Account Code' then checks 'Success?'. If 'No', it goes to 'Terminate Call'. If 'Yes', it goes to connector 'H'. If 'No', it goes to connector 'D'. If 'No' to 'Remote Access?', it goes to connector 'H'. From 'H', it goes to 'Set up GVNS call' then checks 'Success?'. If 'No', it goes to connector 'E'. If 'Yes', it goes to 'GVNS Active'. From 'E', it checks 'Follow on available?'. If 'No', it goes to connector 'F'. If 'Yes', it goes to 'Follow on Offer to the Calling User' then 'Wait for user action'. From 'F', it goes to 'Play Announcement To the Calling User' then 'Terminate Call'. + +T0103510-94/d06 + +FIGURE 3/F.16 (sheet 3 of 7) + SDL diagram for GVNS service provider's view of GVNS + +Process GVNS\_Service\_Provider + +![State transition diagram for GVNS Active state](83852ec55d4802521a727926336bedab_img.jpg) + +This state transition diagram starts with an initial state labeled "GVNS Active" in a double oval. Three arrows lead to different input/output blocks: "Calling User Disconnect", "Calling User Follow on Request", and "Called User Disconnect". The "Calling User Disconnect" block leads to a "Terminate Call" process block (double rectangle), which then leads to an exit symbol (a circle with an X). The "Calling User Follow on Request" block leads to a decision diamond labeled "Follow on available?". From this diamond, a "Yes" path leads to a circle labeled "C", and a "No" path leads to a circle labeled "F". The "Called User Disconnect" block leads to a circle labeled "E". + +State transition diagram for GVNS Active state + +![State transition diagram for Wait for user action state](5445597cceefaca1ac89e710fe339325_img.jpg) + +This state transition diagram starts with an initial state labeled "Wait for user action" in a double oval. Two arrows lead to different input/output blocks: "Calling User Disconnect" and "Calling User Follow on Request". The "Calling User Disconnect" block leads to a "Terminate Call" process block (double rectangle), which then leads to an exit symbol (a circle with an X). The "Calling User Follow on Request" block leads to a circle labeled "C". Below the diagram is the text "T0103810-94/d07". + +State transition diagram for Wait for user action state + +FIGURE 3/F.16 (sheet 4 of 7) +SDL diagram for GVNS service provider's view of GVNS + +## PROCEDURE Collect Authorization Code + +![SDL diagram for GVNS service provider's view of GVNS. The process starts with an initial state, followed by 'Request Authorization Code from Calling User', then 'GVNS Waiting for Calling User Response'. From there, three paths emerge: 'Timer expiry' leading to 'Retry allowed?' (Yes loops back, No leads to 'Play Announcement To User'), 'Authorization Code /* from Calling user */' leading to 'Code valid?' (Yes leads to end, No loops back), and 'Calling User Disconnect' leading to end. Reference T0103530-94/d08.](812e188283162af0b54fb3e30ffee51b_img.jpg) + +``` + +stateDiagram-v2 + [*] --> Request + state "Request Authorization Code from Calling User" as Request + Request --> Waiting + state "GVNS Waiting for Calling User Response" as Waiting + Waiting --> TimerExpiry + Waiting --> AuthCode + Waiting --> Disconnect + state "Timer expiry" as TimerExpiry + TimerExpiry --> RetryAllowed{Retry allowed?} + RetryAllowed -- Yes --> Request + RetryAllowed -- No --> PlayAnnouncement + state "Play Announcement To User" as PlayAnnouncement + PlayAnnouncement --> [*] + state "Authorization Code /* from Calling user */" as AuthCode + AuthCode --> CodeValid{Code valid?} + CodeValid -- Yes --> [*] + CodeValid -- No --> Request + state "Calling User Disconnect" as Disconnect + Disconnect --> [*] + note right of Disconnect: T0103530-94/d08 + +``` + +SDL diagram for GVNS service provider's view of GVNS. The process starts with an initial state, followed by 'Request Authorization Code from Calling User', then 'GVNS Waiting for Calling User Response'. From there, three paths emerge: 'Timer expiry' leading to 'Retry allowed?' (Yes loops back, No leads to 'Play Announcement To User'), 'Authorization Code /\* from Calling user \*/' leading to 'Code valid?' (Yes leads to end, No loops back), and 'Calling User Disconnect' leading to end. Reference T0103530-94/d08. + +FIGURE 3/F.16 (sheet 5 of 7) +SDL diagram for GVNS service provider's view of GVNS + +## PROCEDURE Collect Account Code + +![SDL diagram for GVNS service provider's view of GVNS. The process starts with an initial state, followed by 'Request Account Code from Calling User', then 'GVNS Waiting for Calling User Response'. It then branches into three events: 'Timer expiry', 'Account Code /* from Calling User */', and 'Calling User Disconnect'. 'Timer expiry' leads to a decision 'Retry allowed?'. If 'Yes', it loops back to the request. If 'No', it leads to 'Code valid?'. If 'Code valid?' is 'Yes', it ends. If 'Code valid?' is 'No', it leads to 'Play Announcement To User' and then ends. 'Calling User Disconnect' also ends the process. A label T0103540-94/d09 is present.](08a978a124d3ed6cf1a3d0cfd89418d0_img.jpg) + +``` + +stateDiagram-v2 + [*] --> Request + state "Request Account Code from Calling User" as Request + Request --> Waiting + state "GVNS Waiting for Calling User Response" as Waiting + Waiting --> TimerExpiry + Waiting --> AccountCode + Waiting --> Disconnect + state "Timer expiry" as TimerExpiry + state "Account Code /* from Calling User */" as AccountCode + state "Calling User Disconnect" as Disconnect + TimerExpiry --> RetryAllowed + state "Retry allowed?" as RetryAllowed + RetryAllowed -- Yes --> Request + RetryAllowed -- No --> CodeValid + state "Code valid?" as CodeValid + CodeValid -- Yes --> [*] + CodeValid -- No --> PlayAnnouncement + state "Play Announcement To User" as PlayAnnouncement + PlayAnnouncement --> [*] + Disconnect --> [*] + note right of Disconnect: T0103540-94/d09 + +``` + +SDL diagram for GVNS service provider's view of GVNS. The process starts with an initial state, followed by 'Request Account Code from Calling User', then 'GVNS Waiting for Calling User Response'. It then branches into three events: 'Timer expiry', 'Account Code /\* from Calling User \*/', and 'Calling User Disconnect'. 'Timer expiry' leads to a decision 'Retry allowed?'. If 'Yes', it loops back to the request. If 'No', it leads to 'Code valid?'. If 'Code valid?' is 'Yes', it ends. If 'Code valid?' is 'No', it leads to 'Play Announcement To User' and then ends. 'Calling User Disconnect' also ends the process. A label T0103540-94/d09 is present. + +FIGURE 3/F.16 (sheet 6 of 7) +**SDL diagram-for GVNS service provider's view of GVNS** + +## PROCEDURE Collect Dialled Number + +![SDL diagram for GVNS service provider's view of GVNS. The flowchart starts with an initial state, followed by 'Request Dialled Number from Calling User', then 'GVNS Waiting for User Response'. It then branches into three events: 'Timer expiry', 'Dialled Number /* from user Calling /*', and 'Calling user Disconnect'. 'Timer expiry' leads to a decision 'Retry allowed?'. If 'Yes', it loops back to the request. If 'No', it leads to 'Number valid?'. If 'Number valid?' is 'Yes', it ends. If 'Number valid?' is 'No', it leads to 'Play Announcement to the Calling User', which then ends. 'Calling user Disconnect' ends the process. A reference code T0103550-94/d10 is present.](365b54f616aff249b4e6c0edafdcb9b3_img.jpg) + +``` + +stateDiagram-v2 + [*] --> RequestDialledNumber + RequestDialledNumber: Request Dialled Number from Calling User + state RequestDialledNumber + RequestDialledNumber --> Waiting + state Waiting { + direction LR + GVNS_Waiting: GVNS Waiting for User Response + } + Waiting --> TimerExpiry: Timer expiry + Waiting --> DialledNumber: Dialled Number /* from user Calling /* + Waiting --> Disconnect: Calling user Disconnect + + TimerExpiry --> RetryAllowed + state RetryAllowed <> + note right of RetryAllowed: 'Retry allowed?' + RetryAllowed --> RequestDialledNumber: 'Yes' + RetryAllowed --> NumberValid: 'No' + + DialledNumber --> NumberValid + state NumberValid <> + note right of NumberValid: 'Number valid?' + NumberValid --> [*]: 'Yes' + NumberValid --> PlayAnnouncement: 'No' + + PlayAnnouncement: Play Announcement to the Calling User + PlayAnnouncement --> [*] + + Disconnect --> [*] + + note bottom of Disconnect: T0103550-94/d10 + +``` + +SDL diagram for GVNS service provider's view of GVNS. The flowchart starts with an initial state, followed by 'Request Dialled Number from Calling User', then 'GVNS Waiting for User Response'. It then branches into three events: 'Timer expiry', 'Dialled Number /\* from user Calling /\*', and 'Calling user Disconnect'. 'Timer expiry' leads to a decision 'Retry allowed?'. If 'Yes', it loops back to the request. If 'No', it leads to 'Number valid?'. If 'Number valid?' is 'Yes', it ends. If 'Number valid?' is 'No', it leads to 'Play Announcement to the Calling User', which then ends. 'Calling user Disconnect' ends the process. A reference code T0103550-94/d10 is present. + +FIGURE 3/F.16 (sheet 7 of 7) +**SDL diagram for GVNS service provider's view of GVNS** \ No newline at end of file diff --git a/marked/F/T-REC-F.162-199607-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.162-199607-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..bb91db91fb24cacf9135b70f6b9c8cc1b19350e2 --- /dev/null +++ b/marked/F/T-REC-F.162-199607-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d9b60eca3bb8a9408f91cab114d9faf702a2d3ab136c3bfee43f65e41f3cf4dd +size 8252 diff --git a/marked/F/T-REC-F.162-199607-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg b/marked/F/T-REC-F.162-199607-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c09f150515ae9f00964ec875c7b6333e508b74e4 --- /dev/null +++ b/marked/F/T-REC-F.162-199607-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:30b6db48bdd09e2f5a2c65491abbe8fdeddad66102a472b41608b7facebc1d6f +size 15339 diff --git a/marked/F/T-REC-F.162-199607-I_PDF-E/raw.md b/marked/F/T-REC-F.162-199607-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..005039cc925633cecc4bf8144c2c1f6ec4d993a6 --- /dev/null +++ b/marked/F/T-REC-F.162-199607-I_PDF-E/raw.md @@ -0,0 +1,375 @@ + + +![ITU logo: A globe with a lightning bolt and the letters ITU.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with a lightning bolt and the letters ITU. + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.162** + +(07/96) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Telematic services – Public facsimile service + +--- + +**Service and operational requirements of +store-and-forward facsimile service** + +ITU-T Recommendation F.162 + +(Previously "CCITT Recommendation") + +--- + +# FOREWORD + +The ITU-T (Telecommunication Standardization Sector) is a permanent organ of the International Telecommunication Union (ITU). The ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Conference (WTSC), which meets every four years, establishes the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics. + +The approval of Recommendations by the Members of the ITU-T is covered by the procedure laid down in WTSC Resolution No. 1 (Helsinki, March 1-12, 1993). + +ITU-T Recommendation F.162 was revised by ITU-T Study Group 1 (1993-1996) and was approved under the WTSC Resolution No. 1 procedure on the 19th of July 1996. + +--- + +# NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +© ITU 1996 + +All rights reserved. No part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and microfilm, without permission in writing from the ITU. + +# CONTENTS + +*Page* + +| | | | +|-----|------------------------------------------------------|---| +| 1 | Introduction ..... | 1 | +| 2 | Scope ..... | 1 | +| 3 | Definitions ..... | 1 | +| 4 | Service description ..... | 2 | +| 5 | Service facilities ..... | 2 | +| 5.1 | Access management ..... | 2 | +| 5.2 | Message addressing ..... | 3 | +| 5.3 | Address list codes ..... | 3 | +| 5.4 | Message identification ..... | 3 | +| 5.5 | Classes of message delivery ..... | 3 | +| 5.6 | Message deferral ..... | 3 | +| 5.7 | Cover page ..... | 3 | +| 5.8 | Customer report ..... | 4 | +| 5.9 | Facsimile format conversion ..... | 5 | +| 6 | Access procedures ..... | 6 | +| 7 | Delivery procedures ..... | 6 | +| 7.1 | Automatic call retry ..... | 6 | +| 7.2 | Call records ..... | 6 | +| 7.3 | Recovery from interrupted message delivery ..... | 6 | +| 8 | Quality of service ..... | 6 | +| 8.2 | Quality of service targets ..... | 6 | +| 8.3 | Storage capacity ..... | 7 | +| 8.4 | Public network circuits ..... | 7 | +| | Annex A – Example of COMFAX service cover page ..... | 8 | +| | Annex B – Example of access procedures ..... | 9 | + + + +# SERVICE AND OPERATIONAL REQUIREMENTS OF STORE-AND-FORWARD FACSIMILE SERVICE + +*(revised in 1992 and in 1996)* + +# 1 Introduction + +1.1 The development and growth of store-and-forward equipment and the potential for the international interconnection of such equipment requires a common set of facilities and operational procedures to meet the needs of users. The COMFAX service may be accessed by a variety of terminals which can employ a variety of networks for access purposes. Access to a COMFAX service will be predominantly from facsimile terminals connected to its associated national public networks. However, there are advantages in defining features and procedures for the service to promote and assist the provision of such services in many countries. + +# 2 Scope + +2.1 This Recommendation defines the general principles and basic operational requirements of the COMFAX service provided by Administrations using computer-controlled facsimile store-and-forward units (Fax SFU). Although customer access will be, predominantly, from within the same country in which the Fax SFU is situated, this Recommendation sets out the minimum service requirements that should be provided to allow extension of the service, by the interconnection with one or more other Fax SFUs, across national boundaries, for message delivery purposes. + +2.2 Technical requirements of the service are covered in the relevant T- and X-Series Recommendations. + +2.3 Tariff and accounting aspects are covered in the D-Series Recommendations. Fax SFUs must record and store sufficient information for charging and settlement of accounts. + +2.4 The operational aspects of the interconnection of Fax SFUs are covered in Recommendation F.163. + +2.5 The interworking between COMFAX and other services using the Message Handling System (MHS) is a requirement; details are for further study. The interworking between the COMFAX service and other UIT-T defined services is for further study. + +# 3 Definitions + +This Recommendation uses the terms listed below. + +NOTE – Use of the terms “delivery”, “submission”, “transfer”, does not imply any specific protocol implementation regarding the X.400-Series Recommendations. + +3.1 **Fax SFU:** A Fax SFU refers to the boundary within which the facsimile store-and-forward service is provided by one or more facsimile store-and-forward units under the control of one Administration. A Fax SFU must provide the functions of message submission, message delivery and, in the case of interconnection, message transfer. + +3.2 **message delivery:** Message delivery is defined as the transfer of message data and responsibility for the message from a Fax SFU to a recipient address. + +3.3 **recipient address:** A recipient address is defined as the ultimate destination of a message. A recipient address may designate a facsimile terminal or a storage device empowered to receive messages for a FAX recipient. + +3.4 **message submission:** Message submission is defined as the transfer of message data and responsibility for the message from the originator to the COMFAX service (e.g. typically indicated by the successful completion of the T.30 protocol). + +3.5 **message transfer:** Message transfer is defined as the transfer of message data and responsibility for the message from one Fax SFU (or transfer system) to another Fax SFU (or transfer system). + +**3.6 transfer system:** A transfer system is defined as a system which accomplishes message transfer as defined above. A transfer system may or may not provide end user facsimile services. + +# 4 Service description + +**4.1** Customers may submit messages to the COMFAX service for delivery to one or more recipients. The service provides a range of facilities which may be used by the customer for message management purposes. Facilities such as cover page and deferred delivery allow the customer to specify how messages are handled in their transmission from source to destination. Other facilities, such as customer reports, may provide the customer with more complete messages statistics than are generally available in the real-time group 3 facsimile service. + +The service shall accept messages submitted in group 3 facsimile format. Acceptance of messages in other formats is optional. + +**4.2** Customers access the COMFAX service via public networks. Messages are also delivered to destination terminals via public networks. + +**4.3** Where the service is provided by means of the interconnection by two or more Administrations interworking between Fax SFUs is as defined in Recommendation F.163. Interworking using group 3 facsimile image format is mandatory. Interworking using other formats is optional. + +# 5 Service facilities + +The provision of service facilities to the facsimile customer is a local matter and may vary between Administrations. However, Administrations extending their service across international boundaries by interconnection with one or more Fax SFUs should regard the basic service facilities as mandatory or optional in accordance with Table 1. + +## 5.1 Access management + +Access management is the identification and validation of the originating customer. This may be achieved by passwords where pre-registration is involved. In cases where pre-registration is not employed, the identification of the originating customer may be achieved through the transfer of automatic number identification data from the access network to the Fax SFU. Where passwords are used, it should be possible for the customer to change the password periodically. The exact procedures for security of access should be a local matter. + +TABLE 1/F.162 +**Classification of COMFAX service facilities** + +| Service facility | Classification | Reference | +|-----------------------------|----------------|-----------| +| Access management | M | 5.1 | +| Single address message | M | 5.2.1 | +| Multiple address message | O | 5.2.2 | +| Address list codes | O | 5.3 | +| Message identification | M | 5.4 | +| Class of message delivery | O | 5.5 | +| Message deferral | O | 5.6 | +| Cover page | O | 5.7 | +| Message failure report | M | 5.8.1 | +| Message delivery report | O | 5.8.2 | +| Facsimile format conversion | O | 5.9 | +| M Mandatory | | | +| O Optional | | | + +## **5.2 Message addressing** + +The method of specifying the recipient address is a local matter. + +#### **5.2.1 Single address message** + +A single address message is a message with one recipient address. + +#### **5.2.2 Multiple address message** + +A multiple address message is a message with more than one recipient address. Fax SFUs should be able to accommodate messages addressed to 512 recipient addresses. + +## **5.3 Address list codes** + +An address list code is a code which is used to reference a pre-registered list of recipient addresses. A customer may use an address list code at the time of message submission. + +## **5.4 Message identification** + +The message reference is an identification which can be used to unambiguously identify a message relative to the message originator. When interconnection between Fax SFUs is required, the message reference must be linked to the message identifier (see 6.3/F.163). + +## **5.5 Classes of message delivery** + +There may be three classes of message delivery: + +- a) urgent; +- b) nominal; +- c) non-urgent. + +The process by which the class of message delivery is selected should be a local matter. Quality of Service objectives for the above classes of delivery are shown in clause 8. + +### **5.6 Message deferral** + +The process by which the message deferral is selected is a local matter. There may be two types of message deferrals: + +- a) to a time specified by the originator; +- b) to a time determined by the service provider. + +#### **NOTES** + +- 1 In a) the customer may choose any of the classes of delivery supported to take effect at the specified time. +- 2 Deferral b) is only provided when a customer has selected the non-urgent class of delivery. + +### **5.7 Cover page** + +A cover page is a service provided facsimile page which is automatically added to the front of a customer provided message. The size of the cover page is the same as the first page of the message. All of the data presented on the cover page should be displayable within this size. The cover page may contain information concerning the originator, recipient, and the message. A customer may select the cover page facility on a per message basis. + +#### **5.7.1 Multiple address cover page** + +Cover page generation for multiple address messages is supported. A cover page request applies to all recipients of the multiple address message. The service should have the capability to provide a different cover page per recipient of a multiple address message. + +#### **5.7.2 Cover page selection** + +A customer of the COMFAX service should have three options in selecting a cover page: + +- a) no cover page; +- b) COMFAX service cover page; +- c) customer defined cover page. + +#### **5.7.3 COMFAX service cover page** + +In this case, the COMFAX service is responsible for providing a cover page. The contents of the cover page and selection are described in the following subclause. + +##### **5.7.3.1 Cover page contents** + +The contents of the cover page are found in Table 2. + +##### **5.7.3.2 Selection and display of contents in the COMFAX service cover page** + +All of the information fields listed in Table 2 which are provided in a message shall be displayed on the cover page. Fields which are required by the protocol will, therefore, always be displayed. Fields which are optional will be displayed if the field is present. Control and suppression of the display of fields is for further study. + +A suggested layout can be found in Annex A. + +#### **5.7.4 Customer defined cover page** + +The method of providing customer defined cover pages is a local matter. + +### **5.8 Customer report** + +Message status information provided in level 1 and level 2 notifications described in Recommendation F.163 may be used to provide originators with customer reports, as described in the following subclauses. The presentation of the content of reports and the handling of multiple address reports is a local matter. The method of delivery of the reports is a local matter, e.g. forced – facsimile or queried – delivery. + +#### **5.8.1 Message failure report** + +The service should report to the originator the failure to deliver a message to the destination address. + +The report to the customer may provide at least the following information: + +- message reference; +- recipient address; +- last attempt time. + +NOTE – Provision of time zone indication is a local matter. + +#### **5.8.2 Message delivery report** + +The service may report to the originator the successful delivery of a message to the recipient address. + +The report to the customer may provide at least the following information: + +- message reference; +- recipient address; +- time of delivery to recipient address; +- number of pages delivered. + +NOTE – Provision of time zone indication is a local matter. + +TABLE 2/F.162 + +#### **COMFAX service cover page contents** + +| Field | Length | PM/PA | +|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------------------------|--------------------------------------------------------| +| Image field | Maximum 8 cm | PM (Note 1) | +| Recipient information
NAME: choice of
Personal name
Common name
Free form name
ORGANIZATION
ORGANIZATIONAL UNIT(S)
PHYSICAL ADDRESS LINES
FAX NETWORK ADDRESS
TELEPHONE NUMBER |
(Note 3)
(Note 3)
(Note 3)
(Note 3)
(Note 3)
(Note 3)
(Note 3)
(Note 3)
(Note 3) |
PA
PA
PA
PA
PA
PA
PA
PA
PA | +| Message information
Pages
Priority
Submission date and time
Message reference
Subject |
3 bytes
1 byte
(Note 5)
(Note 4)
(Note 3) |
PM
PM
PM
PM/PA (Note 2)
PM | +| Originator information
NAME: choice of
Personal name
Common name
Free form name
ORGANIZATION
ORGANIZATIONAL UNIT(S)
PHYSICAL ADDRESS LINES
FAX NETWORK ADDRESS
TELEPHONE NUMBER |
(Note 3)
(Note 3)
(Note 3)
(Note 3)
(Note 3)
(Note 3)
(Note 3)
(Note 3)
(Note 3) |
PM
PM
PM
PM
PM
PM
PM
PM
PM | +| Additional recipients | (Note 6) | PM/PA | +| PA Per address
PM Per message
NOTES
1 The image field is either customer provided or service provided with a maximum length of 8 cm and fall within the guaranteed reproducible area of the group 3 facsimile page, in accordance with Appendix I/T.4.
2 The message reference shall be unique for each message, an additional three-digit part may be optionally used to signify each recipient.
3 Same value as in Recommendation F.401 or X.402.
4 The message reference is described in 5.4.
5 The date and time of submission of the message is to be displayed in the local time; a time zone indication should be provided. Translation of submission time to time zones other than the Fax SFU time zone is a local matter.
6 Display of additional recipients is an item for further study. A method of requesting additional recipient fields is to be defined. If necessary, additional cover pages may be used to provide additional recipient information. | | | + +## **5.9 Facsimile format conversion** + +The facsimile format conversion service facility enables the content (image) of a facsimile message to be converted from one format to another. Conversion may be performed on the paper size, the resolution of the message and so on. + +If a Fax SFU supports optional values of several features of Group 3, it shall either support conversion of the image of a message to deliver it or shall not deliver a message whenever it detects that unsupported conversion is required. + +NOTE – It is preferred to support conversion to increase success rate of delivery in cases that loss of information due to conversion is not significant. + +If a Fax SFU supports facsimile terminals other than Group 3, then it shall support conversion between that group and Group 3. + +Conversion shall be an automatic facility and shall not require additional procedures for users. When conversion is supported in a Fax SFU, it shall be performed in a way that the image of the original message is best preserved. That is, conversion is performed only on the unsupported features trying to use values closer to the basic ones. + +Conversion that may result in loss of information may optionally be prohibited by the originator. The details of specifying conversion prohibition at origination are a local matter. + +The details of conversion rules and possible combinations of conversions are defined in the relevant T-Series Recommendations. + +# **6 Access procedures** + +Access procedures are a local matter but should make provision for security of access, destination address selection, delivery instructions and message submission. An example of access procedures is given in Annex B. + +# **7 Delivery procedures** + +## **7.1 Automatic call retry** + +Procedures for automatic call retry on terminal busy, ringing no reply or incomplete calls, subject to national regulatory requirements should be provided. Where the service is extended across national boundaries for the delivery of messages, then the target delivery times specified in clause 8 should be taken into account when setting the call retry periods. + +## **7.2 Call records** + +A call record confirmation successful or unsuccessful delivery to the recipient address should be maintained, which will include the relevant information provided by notifications. The storage of call records is a local matter. + +## **7.3 Recovery from interrupted message delivery** + +If a message is interrupted during the delivery from the destination Fax SFU to the destination terminal, subsequent attempts to deliver the message should commence with the first page not successfully delivered. Delivery should be attempted as soon as possible to avoid significant time periods occurring between receipt of the parts of the message. The recipient must be given sufficient information to correlate the parts of the message. + +# **8 Quality of service** + +**8.1** The introduction of a Fax SFU should not increase, on a per address basis, the probability of message loss or corruption. + +In the event of system failure, all accepted messages should be traceable. If messages have to be cancelled, because of system failure, the originating customers should be informed. + +## **8.2 Quality of service targets** + +**8.2.1** When the service is extended across international boundaries for delivery purposes, it is recommended that the Fax SFU design be such that quality of service targets shown in Figure 1 and Table 3 are maintained. + +![Diagram illustrating quality of service times. It shows a timeline with three vertical lines: 'End of submission', 'Start of first delivery attempt', and 'Forced negative L2N'. A horizontal double-headed arrow labeled T_A spans from 'End of submission' to 'Start of first delivery attempt'. A horizontal double-headed arrow labeled T_B spans from 'End of submission' to 'Forced negative L2N'. Below the diagram is the text 'T0102450-92/d01'.](cfef993dcc8fb513de79eb1f93cf26ae_img.jpg) + +Diagram illustrating quality of service times. It shows a timeline with three vertical lines: 'End of submission', 'Start of first delivery attempt', and 'Forced negative L2N'. A horizontal double-headed arrow labeled T\_A spans from 'End of submission' to 'Start of first delivery attempt'. A horizontal double-headed arrow labeled T\_B spans from 'End of submission' to 'Forced negative L2N'. Below the diagram is the text 'T0102450-92/d01'. + +$T_A$ Time from the end of submission of the message or the time specified for deferred delivery to the start of the first call attempt + +$T_B$ Time from the end of submission of the message or the time specified for deferred delivery to the forced generation of a negative 2nd level notification and termination of message delivery attempt (unless transmission has already commenced) + +FIGURE 1/F.162 +**Illustration of quality of service times** + +TABLE 3/F.162 +**Quality of service targets** + +| Class of message delivery | $T_A$ | $T_B$ | +|---------------------------|------------|----------| +| Urgent | 0.25 hours | 1 hour | +| Normal | 1 hour | 4 hours | +| Non-urgent | 8 hours | 12 hours | + +**8.2.2** The Fax SFU should be capable of meeting the quality of service targets at least 95% of traffic received during any one hour period. + +**8.2.3** While delivery is being attempted the Fax SFU may send interim reports to the customer showing that delivery of the message is still outstanding. + +## 8.3 Storage capacity + +Storage capacity will vary from one Fax SFU to another according to the volume of traffic. However, sufficient storage capacity should be provided to cater for an average message size of two pages, and up to maximum message size of 128 pages. With associated notifications and possible cover page requirements, this could represent a total data content of 6 megabytes. + +## 8.4 Public network circuits + +**8.4.1** The provision of network circuits for the reception and delivery of traffic by the Fax SFU will vary between Administrations. However, sufficient outgoing public network circuits should be provided to meet the quality of service targets of 8.2. + +**8.4.2** Administrations should make observations to evaluate the performance of the Fax SFU to ensure, in particular, that public network provisions are such that quality of service targets of 8.2 are being achieved. + + + +# Annex B + +## Example of access procedures + +(This annex is not an integral part of this Recommendation) + +- B.1** Access to the COMFAX system will normally be via the Public Switched Telephone Network (PSTN). +- B.2** After normal voice call establishment procedures, the COMFAX system should answer with a tone or voice prompt requesting the user to input their respective user identity (ID) code by means of either the facsimile terminal Mediation Function (MF) keypad or an associated system compatible device. +- B.3** The COMFAX system should then request the input of a Personal Identity Number (PIN) to ensure security of use. The request for the PIN may be prompted by either a tone or voice message. +- B.4** When the ID and PIN codes have been successfully validated, the COMFAX system should request the input of the required destination address or addresses. These may be input in the form of the full national and international telephone number, including national access code requirements or code representing a pre-stored address list. The request for address input should be prompted by a tone or voice message. The destination address information should be entered by using either the facsimile terminal MF keypad, or an associated system compatible device. +- B.5** Delivery instructions, i.e. deferred delivery, may be entered by selection of the appropriate code. The request for delivery instructions may be prompted by a tone or voice message. The required code should be entered by using either the facsimile terminal MF keypad or an associated system compatible device. +- B.6** When destination address input and delivery instructions are completed, the COMFAX system should request the submission of the facsimile message. The request for message input may be prompted by a tone or voice message. +- B.7** At the end of message input the PSTN call should be cleared using normal group 3 facsimile procedures. +- B.8** The choice of tone or voice message prompts is a local matter. Where voice prompts are used, they should be in the language of the originator. The support of more than one language on a COMFAX system is a local matter. \ No newline at end of file diff --git a/marked/F/T-REC-F.163-199607-I_PDF-E/1be8e9cad5f38fa47bdb81e549a3bec9_img.jpg b/marked/F/T-REC-F.163-199607-I_PDF-E/1be8e9cad5f38fa47bdb81e549a3bec9_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..704bc4272f5a71ea8c7dfa14e5acc0e0a2d07ba5 --- /dev/null +++ b/marked/F/T-REC-F.163-199607-I_PDF-E/1be8e9cad5f38fa47bdb81e549a3bec9_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f3ce2bf8bb20c399f03be7cd207f9289b6b994537cb5b3ae3e6f0d09d90f0d82 +size 13466 diff --git a/marked/F/T-REC-F.163-199607-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.163-199607-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..808a172e767145666d18916a681beb8edf7d5626 --- /dev/null +++ b/marked/F/T-REC-F.163-199607-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:45c6dff9053d8a7e288da732dba26e17495b18ed53976902395111cbbc38e6cb +size 8288 diff --git a/marked/F/T-REC-F.163-199607-I_PDF-E/d4af765160d04ecef538e5066006dc77_img.jpg b/marked/F/T-REC-F.163-199607-I_PDF-E/d4af765160d04ecef538e5066006dc77_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3a9106fbd416cb6ff304225bb849b6022205083d --- /dev/null +++ b/marked/F/T-REC-F.163-199607-I_PDF-E/d4af765160d04ecef538e5066006dc77_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e9340c29194277dc841e9026f330dfe4dbf19bd9ef0a0d9d0aad727a5d761a7e +size 93033 diff --git a/marked/F/T-REC-F.163-199607-I_PDF-E/raw.md b/marked/F/T-REC-F.163-199607-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..160781c5cb05a53c1d7bb5022184d5a0241f294e --- /dev/null +++ b/marked/F/T-REC-F.163-199607-I_PDF-E/raw.md @@ -0,0 +1,463 @@ + + +![ITU logo: A globe with a lightning bolt and the letters ITU.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with a lightning bolt and the letters ITU. + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.163** + +(07/96) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Telematic services – Public facsimile service + +--- + +**Operational requirements of the interconnection +of facsimile store-and-forward units** + +ITU-T Recommendation F.163 + +(Previously "CCITT Recommendation") + +--- + +# ITU-T F-SERIES RECOMMENDATIONS NON-TELEPHONE TELECOMMUNICATION SERVICES + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| AUDIOVISUAL SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to ITU-T List of Recommendations.* + +## FOREWORD + +The ITU-T (Telecommunication Standardization Sector) is a permanent organ of the International Telecommunication Union (ITU). The ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Conference (WTSC), which meets every four years, establishes the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics. + +The approval of Recommendations by the Members of the ITU-T is covered by the procedure laid down in WTSC Resolution No. 1 (Helsinki, March 1-12, 1993). + +ITU-T Recommendation F.163 was revised by ITU-T Study Group 1 (1993-1996) and was approved under the WTSC Resolution No. 1 procedure on the 19th of July 1996. + +--- + +### NOTE + +In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +© ITU 1996 + +All rights reserved. No part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and microfilm, without permission in writing from the ITU. + +## CONTENTS + +| | Page | +|-------------------------------------------------------------|-------------| +| 1 Introduction..... | 1 | +| 2 Scope..... | 1 | +| 3 Definitions..... | 1 | +| 4 Overview..... | 2 | +| 5 Interconnection requirements ..... | 2 | +| 5.1 General ..... | 2 | +| 5.2 Information transfer unit..... | 2 | +| 5.3 Transit arrangements ..... | 4 | +| 6 Service interworking principles..... | 4 | +| 6.1 Message addressing ..... | 5 | +| 6.2 Address list codes ..... | 5 | +| 6.3 Message identification ..... | 5 | +| 6.4 Class of message delivery..... | 5 | +| 6.5 Message deferral..... | 5 | +| 6.6 Cover page..... | 5 | +| 6.7 Advice of successful/unsuccessful message transfer..... | 5 | +| 6.8 Advice of successful/unsuccessful message delivery..... | 6 | +| 6.9 Facsimile format conversion..... | 7 | +| 7 Quality of service ..... | 7 | +| 7.6 Observations on the quality of service..... | 7 | +| 7.7 Duration of service ..... | 7 | + +# OPERATIONAL REQUIREMENTS OF THE INTERCONNECTION OF FACSIMILE STORE-AND-FORWARD UNITS + +*(revised in 1996)* + +## 1 Introduction + +A growing number of Administrations are offering store-and-forward facsimile (COMFAX) services to their customers. For the efficient and economical extension of these services across international boundaries, there is a need for the interconnection of the facilities used to provide these services. + +## 2 Scope + +2.1 This Recommendation describes the general principles and operational aspects of the COMFAX service when operated and interconnected between Administrations. + +2.2 This Recommendation should be implemented in conjunction with Recommendation F.162 which describes the COMFAX service, and the relevant technical requirements of the T- and X-Series Recommendations. This Recommendation may also be used in connection with the provision of the bureaufax service as described in Recommendations F.170 and F.171. + +## 3 Definitions + +This Recommendation uses the terms listed below. + +NOTE – Use of the terms “delivery”, “submission”, “transfer”, does not imply any specific protocol implementation regarding the X.400-Series of Recommendations. + +3.1 **Fax SFU:** A Fax SFU refers to the boundary within which the facsimile store-and-forward service is provided by one or more facsimile store-and-forward units under the control of one Administration. A Fax SFU must provide the functions of message submission, message delivery and, in the case of interconnection, message transfer. + +3.2 **message delivery:** Message delivery is defined as the transfer of message data and responsibility for the message from a Fax SFU to a recipient address. + +3.3 **recipient address:** A recipient address is defined as the ultimate destination of a message. A recipient address may designate a facsimile terminal or a storage device empowered to receive messages for a FAX recipient. + +3.4 **message submission:** Message submission is defined as the transfer of message data and responsibility for the message from the originator to the COMFAX service (e.g. typically indicated by the successful completion of the T.30 protocol). + +3.5 **message transfer:** Message transfer is defined as the transfer of message data and responsibility for the message from one Fax SFU (or transfer system) to another Fax SFU (or transfer system). + +3.6 **transfer system:** A transfer system is defined as a system which accomplishes message transfer as defined above. A transfer system may or may not provide end user facsimile services. + +3.7 **transit facility:** A transit facility is either a Fax SFU or a transfer system. + +## 4 Overview + +4.1 The interconnection of Fax SFUs is provided on the basis of bilateral agreement between the Administrations concerned. + +4.2 The following two types of interconnection as illustrated in Figure 1 have been identified. + +4.2.1 Interconnected store and forward is where Fax SFUs operated by two Administrations are directly connected. + +4.2.2 Transit store and forward is where the Fax SFU transfers the facsimile message between Fax SFUs via other Fax SFUs or transfer systems. + +4.3 Interconnection between COMFAX and other services using the Message Handling System (MHS) will be provided by the transfer system. Details are for further study. + +![Diagram illustrating COMFAX interworking scenarios. It shows two main paths between an SFU in country A and an SFU in country B. Path (4.2.1) is a direct connection. Path (4.2.2) goes from the SFU in country A through an SFU transit system to the SFU in country B. A reference code T0105980-96/d01 is shown near the transit system.](1be8e9cad5f38fa47bdb81e549a3bec9_img.jpg) + +``` +graph LR; A["SFU country A"] -- "(4.2.1)" --> B["SFU country B"]; A -- "(4.2.2)" --> T["SFU transit"]; T -- "T0105980-96/d01" --> B; +``` + +Diagram illustrating COMFAX interworking scenarios. It shows two main paths between an SFU in country A and an SFU in country B. Path (4.2.1) is a direct connection. Path (4.2.2) goes from the SFU in country A through an SFU transit system to the SFU in country B. A reference code T0105980-96/d01 is shown near the transit system. + +FIGURE 1/F.163 +**COMFAX interworking scenarios** + +## 5 Interconnection requirements + +### 5.1 General + +5.1.1 In the interconnection of Fax SFUs, the responsibility to deliver single and multi-address messages is transferred from the originating Fax SFU to one or more destination Fax SFUs. + +5.1.2 In the interconnection of Fax SFUs, it is mandatory to provide two levels of notification to support the COMFAX service as defined in Recommendation F.162. + +### 5.2 Information transfer unit + +5.2.1 The basic unit of the inter-SFU communication is the Information Transfer Unit (ITU). This unit is classified as either a user ITU or a service ITU allowing easy identification of the function(s) for which cooperation is required. + +5.2.1.1 There is only one type of user ITU, which is the user message. This user message consists of the customer's facsimile message and some additional information (e.g. address). + +5.2.1.2 There are two types of service ITUs: + +- a) Level 1 notification (LIN) – Indicates the successful and/or unsuccessful transfer of responsibility for the user message to the destination SFU. An LIN conveys the information associated with the successful or unsuccessful transfer of responsibility on a per recipient basis. Table 1 lists the contents of the LIN. + +TABLE 1/F.163 + +#### **Contents of level 1 notification (LIN)** + +| | +|-------------------------------------| +| Common field | +| original message identifier | +| recipient address | +| original encoded information types | +| converted encoded information types | +| Successful transfer field | +| last transfer time | +| Unsuccessful transfer field | +| reason code | +| diagnostic code | + +- b) Level 2 notification (L2N) – Indicates the successful or unsuccessful delivery of the responsibility for and delivery of the user message to the recipient address. Responsibility is transferred to a user by transmission of the message to either a facsimile terminal or a storage device empowered to receive messages for a facsimile recipient. Table 2 lists the contents of the L2N. + +TABLE 2/F.163 + +#### **Contents of level 2 notification (L2N)** + +| | +|---------------------------------------| +| Common field | +| original message identifier | +| recipient address | +| last encoded information types | +| SFU reference | +| Successful delivery field | +| time of delivery to recipient address | +| receiving terminal identifier | +| number of pages | +| duration of delivery | +| Unsuccessful delivery field | +| reason code | +| diagnostic code | +| last attempt time | +| number of attempts | +| delivered | + +**5.2.2** Information transfer units should be transferred between Fax SFUs using Protocol Data Units (PDUs) in accordance with the X.400-Series Recommendations. + +**5.2.3** The encoded message part within the user information transfer unit should be indicated as group 3 facsimile for interconnection. The use of other facsimile formats and encoded information types is subject to bilateral agreement. + +**5.2.4** Fax SFUs shall transfer recipient addresses using the originator/recipient (O/R) address form defined in Recommendation F.401. + +### **5.3 Transit arrangements** + +**5.3.1** The use of transit facilities for message transfer is subject to agreement of the service providers involved. + +**5.3.2** A facsimile message may pass through a transit facility consisting of a Fax SFU or some other transfer system. Where messages are routed via transit facilities, the transit facility is required to support the first level notification. + +**5.3.3** Where a message contains a multiple recipient address, it shall be possible for the transfer facility to accept responsibility for some recipient addresses and to onward transfer to one or more Fax SFUs or transfer systems the responsibility for other recipient addresses. + +## **6 Service interworking principles** + +The classification of the interworking requirements is shown in Table 3. + +TABLE 3/F.163 + +### **Interworking requirements of service facilities** + +| Service Facility Requirement | ORIG | TRANS | DEST | Reference | +|--------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------|-------|------|-----------| +| Single address message | M | M1 | M | 6.1.1 | +| Multiple address message | C | M1 | M | 6.1.2 | +| Address list codes | C | N/A | N/A | 6.2 | +| Message identification | M | M | M | 6.3 | +| Class of message delivery | C | M1 | M | 6.4 | +| Message deferral | C | N/A | N/A | 6.5 | +| Cover page | C | N/A | M | 6.6 | +| Advice of successful message transfer (positive L1N) | C | M | M | 6.7 | +| Advice of unsuccessful message transfer (negative L1N) | M | M | M | 6.7 | +| Advice of successful message delivery (positive L2N) | M | M1 | M | 6.8 | +| Advice of unsuccessful message delivery (negative L2N) | M | M1 | M | 6.8 | +| Facsimile format conversion | C | N/A | C | 6.9 | +| M | Mandatory | | | | +| M1 | Mandatory. In the case that the transit facility does not take responsibility for any message delivery, the service facility shall be conveyed. | | | | +| C | Conditional | | | | +| N/A | Not applicable | | | | +| ORIG | Originating Fax SFU | | | | +| TRANS | Transit facility | | | | +| DEST | Destination Fax SFU | | | | + +### **6.1 Message addressing** + +#### **6.1.1 Single address message** + +In the interconnection of Fax SFUs a single address message should be supported as defined in Recommendation F.162. + +#### **6.1.2 Multiple address message** + +In the interconnection of Fax SFUs a multiple address message should be supported as defined in Recommendation F.162. + +### **6.2 Address list codes** + +Expansion of address lists should be performed at the originating Fax SFU. Expansion of address lists at other Fax SFUs is for further study. + +### **6.3 Message identification** + +On successful submission of a message, the originating Fax SFU shall assign a globally unique identification to the message. The message identifier is used to identify the message in notifications between Fax SFUs. + +### **6.4 Class of message delivery** + +In the interconnection of Fax SFUs, the three classes of message delivery, urgent, normal and non-urgent, as defined in Recommendation F.162, should be supported. + +### **6.5 Message deferral** + +The storage of messages to the customer specified time for message delivery, as described in Recommendation F.162, should be carried out by the originating Fax SFU. + +### **6.6 Cover page** + +Cover page is described in Recommendation F.162. The customer choice of cover page shall be transferred from the originating Fax SFU to the destination Fax SFU. + +#### **6.6.1 Support of no cover page request** + +When no cover page is requested, no cover page shall be provided. + +#### **6.6.2 Support of the COMFAX service cover page** + +The destination Fax SFU generates the COMFAX service cover page. The layout of the page is a local matter. Table 4 specifies which Fax SFU provides the COMFAX service cover page contents. + +#### **6.6.3 Support of customer defined cover page** + +Use of a page description method of providing customer defined cover pages is for further study. + +### **6.7 Advice of successful/unsuccessful message transfer** + +**6.7.1** A LIN includes information indicating the successful and/or unsuccessful acceptance of responsibility for the message on a per recipient basis. + +**6.7.2** A LIN which only contains information regarding unsuccessful transfer shall be generated by either the destination Fax SFU or the transit facility as soon as it is determined that responsibility for the message on a per recipient basis cannot be accepted or transferred. + +6.7.3 A LIN for successful transfer shall only be returned when requested. The request shall be processed on a per recipient address basis. + +6.7.4 An unsuccessful message transfer is indicated by the generation of a negative LIN. (There is no requirement for the generation of a negative L2N in this case.) It shall be mandatory to return a LIN for unsuccessful transfer on a per recipient basis. + +TABLE 4/F.163 + +#### **Provision of COMFAX Service cover page contents** + +| Field | Provided by | +|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|----------------------------------------------| +| Image field | O/D (Note 1) | +| Recipient information
NAME: choice of
Personal name
Common name
Free form name
ORGANIZATION
ORGANIZATIONAL UNIT(S)
PHYSICAL ADDRESS LINES
FAX NETWORK ADDRESS
TELEPHONE NUMBER |

O
O
O
O
O
O
O
O | +| Message information
Pages
Priority
Submission date and time
Message Reference
Subject |
D
O
O
O
O | +| Originator information
NAME: choice of
Personal name
Common name
Free form name
ORGANIZATION
ORGANIZATIONAL UNIT(S)
PHYSICAL ADDRESS LINES
FAX NETWORK ADDRESS
TELEPHONE NUMBER |

O
O
O
O
O
O
O
O | +| Additional recipients | (Note 2) | +| O    Provided by originating Fax SFU
D    Provided by destination Fax SFU
NOTES
1    Provision of service providers' images are by bilateral agreement. If joint service providers' images are required no extension to the image field size is allowed.
2    Display of additional recipients is an item for further study. | | + +### **6.8 Advice of successful/unsuccessful message delivery** + +6.8.1 A positive L2N indicates successful message delivery and shall be returned to the originating Fax SFU on a per recipient basis. A L2N for successful message delivery shall only be returned when requested. The request shall be processed on a per recipient basis. + +6.8.2 A negative L2N indicates unsuccessful message delivery and shall be returned to the originating Fax SFU on a per recipient basis. It is mandatory to return a L2N for unsuccessful delivery. + +### **6.9 Facsimile format conversion** + +The facsimile format conversion service facility enables the content (image) of a facsimile message to be converted from one format to another. Conversion may be performed on the paper size, the resolution of the message and so on. + +When a message is transferred to another Fax SFU, the message shall be in a basic format of group 3 using facsimile format conversion unless there is a bilateral agreement. If such a bilateral agreement exists, the destination Fax SFU shall either support conversion of the image of a message to deliver it or shall not deliver a message whenever it detects that unsupported conversion is required. + +Conversion shall be an automatic facility and shall not require additional procedures for users. When conversion is supported in a Fax SFU, it shall be performed in a way that the image of the original message is best preserved. That is, conversion is performed only on the unsupported features trying to use values closer to the basic ones. + +Conversion that may result in loss of information may optionally be prohibited by the originator. The details of transferring information on conversion prohibition are for further study. + +The details of conversion rules and possible combinations of conversions are defined in relevant T-Series Recommendations. + +## **7 Quality of service** + +**7.1** The quality of the image, as received from the originating facsimile, terminal should not be adversely affected by any Fax SFU, transit facility, or any transmission facility between Fax SFUs through which the image may pass. + +**7.2** The following quality of services objectives are defined for messages passed across the international boundary for delivery by a destination Fax SFU. The overall service delivery objectives are specified in Table 3/F.162. In order that the destination Fax SFU is able to meet the objectives as stated in Table 3/F.162, the originating Fax SFU should transfer the message according to the objectives illustrated in Figure 2 and described in Table 5. + +The objectives for times $T_A$ and $T_B$ are defined in Recommendation F.162. + +The objectives for times $T_1$ , $T_2$ , and $T_3$ should be met for at least 95% of messages during any 1 hour period. + +The number of attempts to deliver a message to the destination terminal and the interval between attempts are a local matter. + +**7.3** The quality of service targets specified in Table 5 should not be affected by any transit or transmission facilities. + +**7.4** If a message is interrupted during the delivery from the destination Fax SFU to the destination terminal, subsequent attempts to deliver the message should commence with the first page not successfully delivered. The recipient shall be given sufficient information to correlate the parts of the message. In addition, the number of pages delivered in each partial transmission are required to be returned to the originating Fax SFU. + +**7.5** System design and dimensioning should be such that the quality of service objectives are met for at least 95% of messages received during any 1 hour period. + +### **7.6 Observations on the quality of service** + +Administrations shall make observations to evaluate the quality of the COMFAX service across national boundaries. + +### **7.7 Duration of service** + +The national and international COMFAX facilities shall be open continuously. + +![Sequence diagram of a fax transmission across an international boundary. The diagram shows the flow from an originating FAX terminal through a Fax SFU, across an international boundary to another Fax SFU, and finally to a destination FAX terminal. Key events include message submission, transfer, Level 1 and Level 2 notifications, and multiple delivery attempts. Timing intervals T1, T2, TA, TB, T3, and T4 are indicated on the left side.](d4af765160d04ecef538e5066006dc77_img.jpg) + +The diagram illustrates the sequence of operations for a fax transmission across an international boundary. The components involved are an Originating FAX terminal, a Fax SFU, an international boundary, another Fax SFU, and a Destination FAX terminal. The sequence of messages is as follows: + +- Message submission:** From the Originating FAX terminal to the first Fax SFU. +- Message transfer:** From the first Fax SFU to the second Fax SFU across the international boundary. +- Level 1 notification:** From the second Fax SFU back to the first Fax SFU. +- Message delivery (1st attempt):** From the second Fax SFU to the Destination FAX terminal. +- Message delivery (last attempt of retry algorithm):** From the second Fax SFU to the Destination FAX terminal. +- Level 2 notification:** From the second Fax SFU back to the first Fax SFU. +- Customer report:** From the first Fax SFU back to the Originating FAX terminal. + +Timing intervals are defined on the left side of the diagram: + +- $T_1$ : Time interval between message submission and message transfer. +- $T_2$ : Time interval between message transfer and Level 1 notification. +- $T_A$ : Time interval between message submission and Level 1 notification. +- $T_B$ : Time interval between message submission and the start of the first delivery attempt. +- $T_3$ : Time interval between the start of the first delivery attempt and the start of the last delivery attempt. +- $T_4$ : Time interval between the start of the last delivery attempt and the start of the Level 2 notification. + +T0105990-96/d02 + +Sequence diagram of a fax transmission across an international boundary. The diagram shows the flow from an originating FAX terminal through a Fax SFU, across an international boundary to another Fax SFU, and finally to a destination FAX terminal. Key events include message submission, transfer, Level 1 and Level 2 notifications, and multiple delivery attempts. Timing intervals T1, T2, TA, TB, T3, and T4 are indicated on the left side. + +#### NOTES + +- 1 The objectives for times $T_A$ and $T_B$ are defined in Recommendation F.162. +- 2 The objectives for times $T_1$ , $T_2$ , and $T_3$ should be met for at least 95% of messages during any 1 hour period. +- 3 The number of attempts to deliver a message to the destination terminal and the interval between attempts are a national matter. + +FIGURE 2/F.163 + +TABLE 5/F.163 + +#### **Quality of service targets** + +| Time | Description | Grade of delivery | Target (hours) | +|----------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------------|----------------| +| T A | Time from the end of submission of the message, or the time specified for the deferred delivery, to the start of the first call attempt | Urgent | 0.25 | +| | | Normal | 1.0 | +| | | Non-Urgent | 8.0 | +| T B | Time from the end of submission of the message, or the time specified for deferred delivery, to the forced generation of a negative second level notification and termination of message delivery attempt (unless transmission has already commenced) | Urgent | 1.0 | +| | | Normal | 4.0 | +| | | Non-Urgent | 212.0 | +| T 1 | Time from the end of the submission of the message, or the time specified for deferred delivery, to the end of the transfer of the message from the originating Fax SFU to the destination Fax SFU | Urgent | 0.15 | +| | | Normal | 0.6 | +| | | Non-Urgent | 4.8 | +| T 2 | Time from the end of the transfer of the message from the originating Fax SFU to the destination Fax SFU, to the end of the transfer of the first notification from the destination Fax SFU to the originating Fax SFU | All | 0.15 | +| T 3 | Time from the end of message delivery, or the forced generation of a negative second level notification, to the end of the transfer of a second level notification from the destination fax SFU to the originating Fax SFU | All | 0.15 | +| T 4 | Time from the end of transfer of a second level notification from the destination Fax SFU to the originating Fax SFU, to the generation of a customer report | All | Local matter | + + + +# ITU-T RECOMMENDATIONS SERIES + +- Series A Organization of the work of the ITU-T +- Series B Means of expression +- Series C General telecommunication statistics +- Series D General tariff principles +- Series E Telephone network and ISDN +- Series F Non-telephone telecommunication services** +- Series G Transmission systems and media +- Series H Transmission of non-telephone signals +- Series I Integrated services digital network +- Series J Transmission of sound-programme and television signals +- Series K Protection against interference +- Series L Construction, installation and protection of cables and other elements of outside plant +- Series M Maintenance: international transmission systems, telephone circuits, telegraphy, facsimile and leased circuits +- Series N Maintenance: international sound-programme and television transmission circuits +- Series O Specifications of measuring equipment +- Series P Telephone transmission quality +- Series Q Switching and signalling +- Series R Telegraph transmission +- Series S Telegraph services terminal equipment +- Series T Terminal equipments and protocols for telematic services +- Series U Telegraph switching +- Series V Data communication over the telephone network +- Series X Data networks and open system communication +- Series Z Programming languages \ No newline at end of file diff --git a/marked/F/T-REC-F.17-199208-I_PDF-E/raw.md b/marked/F/T-REC-F.17-199208-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..3858bc2a684f03fa8059a9fb0c98383d96267833 --- /dev/null +++ b/marked/F/T-REC-F.17-199208-I_PDF-E/raw.md @@ -0,0 +1,89 @@ + + +![ITU logo: A globe with a lightning bolt and the letters ITU.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with a lightning bolt and the letters ITU. + +INTERNATIONAL TELECOMMUNICATION UNION + +**CCITT** + +THE INTERNATIONAL +TELEGRAPH AND TELEPHONE +CONSULTATIVE COMMITTEE + +**F.17** + +(08/92) + +**TELEGRAPH AND MOBILE SERVICES +OPERATIONS AND QUALITY OF SERVICE** + +--- + +**OPERATIONAL ASPECTS OF SERVICE +TELECOMMUNICATIONS** + +**Recommendation F.17** + +--- + +![CCITT logo: A globe with the letters CCITT.](84a1d09fb489061482111515543b60dc_img.jpg) + +CCITT logo: A globe with the letters CCITT. + +Geneva, 1992 + +# FOREWORD + +The CCITT (the International Telegraph and Telephone Consultative Committee) is a permanent organ of the International Telecommunication Union (ITU). CCITT is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The Plenary Assembly of CCITT which meets every four years, establishes the topics for study and approves Recommendations prepared by its Study Groups. The approval of Recommendations by the members of CCITT between Plenary Assemblies is covered by the procedure laid down in CCITT Resolution No. 2 (Melbourne, 1988). + +Recommendation F.17 was prepared by Study Group I and was approved under the Resolution No. 2 procedure on the 4th of August 1992. + +## --- CCITT NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized private operating agency. + +# **OPERATIONAL ASPECTS OF SERVICE TELECOMMUNICATIONS** + +(1992) + +## **1 Scope** + +- 1.1 This Recommendation lays down the general operational principles concerning service telecommunications. +- 1.2 Additional operational provisions relating to service telecommunications in the different telecommunication services (telegram, telephone, telex, etc.) may be found in the Recommendations (and Instructions) specific to those services (F.1, E.105, F.60, etc.). +- 1.3 The principles for charging and accounting for service telecommunications are given in Recommendation D.192. + +## **2 Definition and basic principles** + +- 2.1 Article 2 of the International Telecommunication Regulations (Melbourne, 1988) defines **service telecommunications** as: + +A telecommunication that relates to public international telecommunications and that is exchanged among the following: + +- Administrations; + - recognized private operating agencies; and + - the Chairman of the Administrative Council, the Secretary-General, the Deputy Secretary-General, the Directors of the International Consultative Committees, the members of the International Frequency Registration Board, other representatives or authorized officials of the International Telecommunication Union, including those working on official matters outside the seat of the Union. +- 2.2 Appendix 3 of the same Regulations states that service telecommunications may be provided free of charge1). + - 2.3 Service telecommunications should be established outside the busiest hours as far as possible, and should be kept to a minimum in number and length consistent with efficient operation. + +## **3 Content of service telecommunications** + +- 3.1 Service telecommunications as identified in § 2.1 must relate to the business of providing international telecommunication services, i.e. the doing of all things needful to provide, to administer and to maintain services, such as the establishment and maintenance of circuits, network management, accounting, handling of customer complaints, etc. +- 3.2 Service telecommunications may only be requested or established by persons authorized to do so by their respective Administration. + +## **4 Corresponding arrangements between Administrations** + +- 4.1 An Administration that routes service telecommunications over facilities owned or operated by another Administration in a given relation, should allow its facilities to be used in a similar manner to carry service telecommunications originated by the latter Administration. +- 4.2 This principle may be applied not only to a given telecommunication service in a relation but also between services in that relation. + +--- + +1) Service telecommunications are normally, but not automatically, provided free of charge. + +## 5 Communications with and via the ITU + +5.1 Where appropriate, service telecommunications may be used for official communications between Administrations and the ITU, e.g. for the purposes described in Article 8 of the International Telecommunication Regulations and in Resolution No. 7 of the World Administrative Telegraph and Telephone Conference (Melbourne, 1988). + +*Note* — Privilege telecommunications (see Recommendation D.193) are not service telecommunications, although the treatment of this optional class of traffic may be similar. \ No newline at end of file diff --git a/marked/F/T-REC-F.170-199803-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.170-199803-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..808a172e767145666d18916a681beb8edf7d5626 --- /dev/null +++ b/marked/F/T-REC-F.170-199803-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:45c6dff9053d8a7e288da732dba26e17495b18ed53976902395111cbbc38e6cb +size 8288 diff --git a/marked/F/T-REC-F.170-199803-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.170-199803-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7e9bbade3b5fa7944e6d2b27a9087061b2824d82 --- /dev/null +++ b/marked/F/T-REC-F.170-199803-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dda899e1b71927c55d77fda707f8840b1bead45137664fc225f5a67566e48969 +size 85226 diff --git a/marked/F/T-REC-F.170-199803-I_PDF-E/raw.md b/marked/F/T-REC-F.170-199803-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..b7e3c3773165ceb518c12abf6f121ae5a673e44a --- /dev/null +++ b/marked/F/T-REC-F.170-199803-I_PDF-E/raw.md @@ -0,0 +1,465 @@ + + +![ITU logo](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +The logo of the International Telecommunication Union (ITU) features the letters 'ITU' in a bold, sans-serif font, superimposed on a stylized globe with intersecting lines. + +ITU logo + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.170** + +(03/98) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Telematic services – Public facsimile service + +--- + +**Operational provisions for the international +public facsimile service between public bureaux +(Bureaufax)** + +ITU-T Recommendation F.170 + +(Previously CCITT Recommendation) + +--- + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +## **TELEGRAPH SERVICE** + +| | | +|-----------------------------------------------------------------|-------------| +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | + +## **MOBILE SERVICE** + +| | | +|--------------------------------------------------------|-------------| +| Mobile services and multdestination satellite services | F.110–F.159 | +|--------------------------------------------------------|-------------| + +## **TELEMATIC SERVICES** + +| | | +|-------------------------------------------|--------------------| +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | + +## **MESSAGE HANDLING SERVICES** + +F.400–F.499 + +## **DIRECTORY SERVICES** + +F.500–F.549 + +## **DOCUMENT COMMUNICATION** + +| | | +|--------------------------------------|-------------| +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | + +## **DATA TRANSMISSION SERVICES** + +F.600–F.699 + +## **AUDIOVISUAL SERVICES** + +F.700–F.799 + +## **ISDN SERVICES** + +F.800–F.849 + +## **UNIVERSAL PERSONAL TELECOMMUNICATION** + +F.850–F.899 + +## **HUMAN FACTORS** + +F.900–F.999 + +*For further details, please refer to ITU-T List of Recommendations.* + +## **ITU-T RECOMMENDATION F.170** + +# **OPERATIONAL PROVISIONS FOR THE INTERNATIONAL PUBLIC FACSIMILE SERVICE BETWEEN PUBLIC BUREAUX (BUREAUFAX)** + +## **Summary** + +In accordance with the general conditions of Recommendation F.160, this Recommendation describes an international public facsimile service between public bureaux, provided that terminals are compatibles or compatibility is ensured by the network from the standpoint of the characteristics specified in the relevant Recommendations and in accordance with bilateral or multilateral agreements. This service, known internationally as Bureaufax, may be provided by telecommunication Recognized Operating Agencies (ROA) or by postal Administrations. + +###### **Source** + +ITU-T Recommendation F.170 was revised by ITU-T Study Group 2 (1997-2000) and was approved under the WTSC Resolution No. 1 procedure on the 9th of March 1998. + +## FOREWORD + +ITU (International Telecommunication Union) is the United Nations Specialized Agency in the field of telecommunications. The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of the ITU. The ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Conference (WTSC), which meets every four years, establishes the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics. + +The approval of Recommendations by the Members of the ITU-T is covered by the procedure laid down in WTSC Resolution No. 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +## INTELLECTUAL PROPERTY RIGHTS + +The ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. The ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, the ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementors are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database. + +© ITU 1998 + +All rights reserved. No part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and microfilm, without permission in writing from the ITU. + +## CONTENTS + +| | Page | +|------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------| +| 1 General provisions ..... | 1 | +| 2 Conditions for acceptance ..... | 1 | +| 3 Composition of a facsimile transmission..... | 2 | +| 3.2 Transmittal sheet..... | 2 | +| 3.3 Sender's document..... | 2 | +| 4 Acceptance ..... | 2 | +| 5 Transmission ..... | 4 | +| 6 Delivery..... | 4 | +| 7 Service facsimile correspondence ..... | 5 | +| 8 Archives ..... | 5 | +| 9 Charging, refunds and accounting..... | 5 | +| 10 Cancellation at the request of the sender ..... | 6 | +| 11 Bureaufax table ..... | 6 | +| 12 Bureaufax service via the public switched telephone network, via a public data network, via ISDN or via facilities dedicated to Bureaufax service ..... | 6 | +| Annex A – Definitions of delivery modes ..... | 7 | +| A.1 Letter-post item (mode A) ..... | 7 | +| A.2 Delivery by special messenger (mode B)..... | 7 | +| A.3 EMS (Express Mail Service) (mode C) ..... | 7 | +| A.4 Counter collection (mode D) ..... | 7 | +| A.5 Counter collection with advice (modes E, G) ..... | 7 | +| A.6 Telefax (mode F) ..... | 7 | +| Bibliography..... | 7 | + + + +## OPERATIONAL PROVISIONS FOR THE INTERNATIONAL PUBLIC FACSIMILE SERVICE BETWEEN PUBLIC BUREAUX (BUREAUFAX) + +*(revised in 1998)* + +## 1 General provisions + +1.1 In accordance with the general conditions of Recommendation F.160, this Recommendation describes an international public facsimile service between public bureaux, provided that terminals are compatible or compatibility is ensured by the network from the standpoint of the characteristics specified in the relevant Recommendations and in accordance with bilateral or multilateral agreements. This service, known internationally as Bureaufax, may be provided by telecommunication Recognized Operating Agencies (ROA) or by Postal Administrations. For the purposes of this Recommendation only, both are comprised in the term "service providers". + +NOTE 1 – The operation of the international public facsimile service between public bureaux and subscriber stations and vice versa is covered by Recommendation F.190. + +NOTE 2 – Where agreed between the relevant service providers, telegrams and teletext messages may also be transmitted directly via facsimile from the origin country to a public facsimile bureau of the destination country and delivered by the Bureaufax service in the destination country. + +1.2 Facsimile terminals should have automatic operation of the called station. The operational procedures applicable in relations between the different types of stations are set out in the relevant Recommendations in the T-Series. + +## 2 Conditions for acceptance + +2.1 Documents for facsimile transmission shall normally be accepted up to a maximum size of ISO format A4 (210 × 297 mm). Documents with larger sizes can be accepted only if the transmission can be done within the ISO format A4. Exceptionally, documents of a different size may be accepted by bilateral agreement between those service providers concerned, for example the North American "legal" size (216 × 356 mm). + +2.2 For faithful transmission, contents of documents must leave minimum blank margins of 8 mm on all four sides. With regard to the reproduction in formats other than A4 which have been bilaterally agreed, the margins to be left on all four sides of the documents shall be 15 mm. + +2.3 Where the area to be transmitted exceeds the area reproducible by facsimile, a document should be divided by the sender or his authorized representative. + +2.4 Care must be taken to ensure that there is no loss of the reproducible part transmitted. + +2.5 To ensure the satisfactory facsimile transmission of a document, senders should be advised not to submit documents with insufficiently sharp contrast or inadequate definition. + +2.6 If the sender or his authorized representative presents a document containing colours of half-tones, he should be informed that, with existing equipment, faithful reproduction at the distant end will not be possible as printing will be in black and white only. + +2.7 If, after having been informed that the quality of the document to be transmitted is unsuitable for satisfactory facsimile transmission (this includes all photographs and documents in colour), the sender of the document or his authorized representative still insists on its transmission, it shall be accepted by the public bureau only at the sender's own risk. In this event the service instruction **RISQUES EXPÉDITEUR** (abbreviated **RE**) must be inserted in box 6 on the transmittal sheet (see Figure 1). When the service instruction **RE** is not applicable to the whole document, the relevant pages must be designated in box 6 of the transmittal sheet, for example: **RE 3 + 5 + 12**. + +2.8 Administrations shall reserve the right to refuse the transmission of documents in the circumstances described in Articles 34 and 35 of the ITU Constitution [1] and in Article 26 of the Universal Postal Union Convention [2]. + +## 3 Composition of a facsimile transmission + +### 3.1 Every facsimile transmission shall include: + +- a transmittal sheet as the first page, in accordance with 3.2; and +- a sender's document in accordance with 3.3. + +### 3.2 Transmittal sheet + +#### 3.2.1 The transmittal sheet must permit the identification of the following information: + +- 1) originating office and optional, originating number of the document; +- 2) international transmitting bureau (sending office); +- 3) document sending number; +- 4) number of pages (including the transmittal sheet); +- 5) date and, if necessary, time of acceptance; +- 6) service instruction **RE (RISQUES EXPÉDITEUR)** in accordance with 2.7, if applicable; +- 7) delivery mode, in coded form, and other service instructions, if any and the call number for delivery modes **E**, **F** and **G**; +- 8) the addressee's address containing all the particulars necessary to ensure the routing and delivery of the facsimile document without enquiries or requests for information. As a general rule, it should indicate: + - i) the designation of the addressee; + - ii) his full postal address (including, if it exists, the postal code) and, if available, his telephone, and telex number; +- 9) sender's address (including, if it exists, the postal code) and, if available, his telephone and/or telex number. + +NOTE – Items 1) and 9) may be omitted from the transmittal sheet if this information can be traced from other information provided elsewhere on the form. + +#### 3.2.2 The various parts of the transmittal sheet shall be contained within ISO format A5 (210 × 148 mm). They may also be located in the upper half of a sheet of ISO format A4 so that the sender can use the lower half to write a document. + +NOTE – If a North American format is used for the transmittal sheet, the width of the sheet will be 216 mm (instead of 210 mm). + +#### 3.2.3 The various parts of the transmittal sheet shall be designated at least in French or in English or in Spanish. Other languages may be added by service providers. + +#### 3.2.4 Figure 1 shows the measurements, locations and designations of the various parts of the transmittal sheet. + +### 3.3 Sender's document + +#### 3.3.1 The document for facsimile transmission can contain written or printed matter, drawings, or any other graphic, subject to the limitations referred to under conditions for acceptance in clause 2 above. A signature is optional. + +## 4 Acceptance + +### 4.1 A document for facsimile transmission may be handed in at the counter of a public bureau. Other means of acceptance and the modes of returning the original to the sender after transmission may be designated by the service provider responsible for the public bureau. + +### 4.2 The sender may use transmittal sheets that consist of the A4 version in accordance with 3.2.2, if available, for writing out a document in the lower half of the sheet. + +### 4.3 The sender of a facsimile document shall be required to establish his identity if called upon to do so by the accepting public facsimile bureau. + +![Diagram of a transmittal sheet with measurements, locations, and designations of various parts. The sheet is 148 mm high and 210 mm wide. It contains several numbered boxes: 1. Originating office, Document originating No., Telephone No.; 2. Sending office; 3. Document sending number; 4. Pages; 5. Date (and time) with 'Acceptance-sending' and 'Receiving' options; 6. *) with 'RE' option; 7. Delivery mode and service instructions; 8. Addressee (including postal code); 9. Sender (including postal code). Dimensions are marked: 45 mm for boxes 2-5, 15 mm for box 7, 42 mm for box 8, 15 mm for the bottom section, 80 mm for the left part of box 9, 10 mm for the middle part of box 9, 90 mm for the right part of box 9, and 8.2 mm for the bottom right corner of box 8.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +Diagram of a transmittal sheet with measurements, locations, and designations of various parts. The sheet is 148 mm high and 210 mm wide. It contains several numbered boxes: 1. Originating office, Document originating No., Telephone No.; 2. Sending office; 3. Document sending number; 4. Pages; 5. Date (and time) with 'Acceptance-sending' and 'Receiving' options; 6. \*) with 'RE' option; 7. Delivery mode and service instructions; 8. Addressee (including postal code); 9. Sender (including postal code). Dimensions are marked: 45 mm for boxes 2-5, 15 mm for box 7, 42 mm for box 8, 15 mm for the bottom section, 80 mm for the left part of box 9, 10 mm for the middle part of box 9, 90 mm for the right part of box 9, and 8.2 mm for the bottom right corner of box 8. + +a) North American standard: 216 mm. + +\*) Insert "RE" in box 6 in case the message is accepted at the sender's own risk. + +NOTE 1 – The information in boxes 1 and 9 need not necessarily be transmitted and may be inserted elsewhere on the sheet. When the information in box 9 (sender) is given on the front of the sheet, it must appear in the left-hand part. + +NOTE 2 – In box 5, the time and words "acceptance-sending" and "receiving" are optional. + +NOTE 3 – Concerning box 7, the following delivery modes may be supplied on the basis of agreements between the service providers concerned (for definitions, see Annex A): + +| | | | | +|---|-----------------------------------------------------------------------|----------------------------------------------------------------------------|----------------------------------------------------------------------------| +| A | Letter-post item | Courrier ordinaire | Correo ordinario | +| B | Delivery by special messenger (express item) | Porteur spécial | Urgente (entrega efectuada por un mensajero especial) | +| C | EMS item (Express Mail Service) | EMS (service de courrier exprès) | EMS (servicio de correo urgente) | +| D | Counter collection | Retrait au guichet | Recogida en ventanilla | +| E | Counter collection with telephone advice to the indicated call number | Retrait au guichet avec avis téléphonique au numéro indiqué | Recogida en ventanilla con aviso telefónico al número indicado | +| F | Telefax (with call number and ITU-T Group when known) | Téléfax (avec numéro d'appel et, s'il est connu, groupe de l'UIT-T) | Telefax (con número de llamada y grupo del UIT-T cuando se conozca) | +| G | Counter collection with telex advice to the indicated call number | Retrait au guichet avec avis télex au numéro indiqué | Recogida en ventanilla con aviso télex al número indicado | +| T | Telegram | Télégramme | Telegrame | + +Other modes of delivery may be agreed between service providers. + +**Figure 1/F.170 – Measurements, locations and designations of the various parts of the transmittal sheet in accordance with 3.2** + +## 5 Transmission + +5.1 In principle, documents shall be transmitted by a public bureau in the order in which they are accepted, except where a priority system is established. + +NOTE – In the case of facsimile transmission of telegrams and telemessages, the priorities for transmission of these services have to be considered. + +5.2 At the request of the international transmitting public bureau, on a case-by-case basis and immediately after the transmission, the international receiving public bureau will acknowledge the satisfactory receipt of a facsimile document. + +5.3 A transmission prevented by adverse conditions shall be repeated as soon as conditions allow. + +In case of unsatisfactory receipt, the international receiving public bureau should indicate service instruction **RPT PAGE** ... in box 7 of the transmittal sheet. + +After a maximum of three attempts, the international transmitting public bureau should, if practicable, use another transmission means among those listed in 3.4/F.18. If this is not practicable, the sender will be informed of the situation by the transmitting public bureau in his country. + +5.4 If transmission cannot be effected by the international transmitting public bureau within four hours of acceptance by the public facsimile bureau, the sender should be informed as soon as possible by the transmitting bureau in his country. + +5.5 On no account will the international receiving public bureau request repeat transmissions in an attempt to improve the quality of facsimile reproduction of pages designated as unsuitable by the service instruction **RE**. + +5.6 Any two service providers may agree to apply special operational procedures in their mutual relation when there is a large volume of messages being sent to a given office, or in other special cases. + +## 6 Delivery + +6.1 The service providers determine the mode of delivery they use for the facsimile documents they receive, according to Figure 1, Note 3. The modes of delivery used by each service provider should figure in the Bureaufax table (see clause 11). + +NOTE – If a telegram1 or a telemessage2 is transmitted to a public facsimile bureau for delivery, the delivery modes of the Bureaufax service laid down in 6.4 will apply: + +- urgent telegrams are delivered according to the delivery mode B or C; +- ordinary telegrams are delivered according to the delivery mode A; +- letter telegrams and telemessages are delivered according to the delivery mode A. + +6.2 The facsimile documents received by an international receiving public bureau are, as far as possible, delivered to the addressee according to the mode of delivery indicated in box 7 of the facsimile transmittal sheet. + +6.3 The receiving public bureau shall, if required, record the date and time of receipt and the mode of delivery of each facsimile document. + +6.4 Facsimile documents should be delivered to the addressee within the following times from the moment when the international receiving public bureau is able to process the documents received (the hours when the receiving public bureau is closed are not taken into account for delivery times): + +Mode of delivery A:                      delivery next working day after receipt; + +Mode of delivery B:                      places of delivery having a Bureaufax bureau: four hours; + +places of destination without a Bureaufax bureau: as provided by the service provider responsible for the Bureaufax special delivery service; + +1 For details, see Recommendation F.1 for telegrams. + +2 For details, see Recommendations F.40 and F.41 for telemessages. + +Mode of delivery C:                places of delivery having a Bureaufax bureau and places of destination without a Bureaufax bureau: documents received before 1200 noon on working days, delivery the same day; documents received after 1200 noon on working days, delivery the next working day; + +Mode of delivery D:              collection from the counter of the receiving public bureau: one hour after receipt; + +Method of delivery E, F, G:    the addressee is called within one hour by the receiving public bureau. If this cannot reach the addressee within three hours, the destination service provider may apply another method of delivery, to deliver the facsimile document to the addressee as soon as possible. + +NOTE – If the destination service provider cannot comply with the established times, other times may be agreed between service providers in relation to the ordinary postal service times. + +**6.5**        The addressee of a facsimile document must establish his identity if required to do so by the receiving public bureau. + +**6.6**        The addressee of a facsimile document with the service instruction **RE** in accordance with 2.7 above shall be obliged to accept the quality of facsimile reproduction offered. + +**6.7**        When a facsimile document cannot be delivered to the addressee, the destination service provider must advise the international transmitting public bureau of the origin service provider, giving the reason for non-delivery. Box 7 of the Bureaufax transmittal sheet can be used to this effect. The service providers can agree to transmit this information through other telecommunication services like telex, etc. The origin service provider should notify the sender giving the reason why a document was not delivered. + +NOTE – If an origin country transmits a telegram or a telemessage to the destination country by facsimile this subclause will also apply. + +## **7        Service facsimile correspondence** + +**7.1**        By agreement between service providers (see Recommendation F.17), service facsimile correspondence may be exchanged between the service providers concerned for the purposes mentioned below: + +- to exchange communications between the service providers concerned for the efficient operation of the Bureaufax service as necessary, including communications concerning enquiries or claims from customers in connection with the Bureaufax service; +- to exchange communications between the service providers concerned in connection with other telecommunication services provided mutually by these service providers, in particular urgent communications that cannot be effected by other telecommunication means, e.g. by international telex or telegram services, because they contain diagrammatic material or other material where it is important for the original to be exactly reproduced. + +NOTE – The indication **SERVICE** should be clearly shown on the transmittal sheet. + +**7.2**        The service providers should reply to the requests for information and claims by correspondence within 15 days of receipt of the request. + +## **8        Archives** + +**8.1**        The maintenance of archives shall be at the discretion of the individual service providers concerned which will decide on the most suitable arrangements to answer any subsequent queries, in particular on the international accounts. + +## **9        Charging, refunds and accounting** + +**9.1**        Charging principles, refund of charges and international accounting for the transmission of documents in the international public facsimile service between public bureaux are governed by Recommendations, agreements and/or Conventions such as are given in Recommendation D.70. + +## **10 Cancellation at the request of the sender** + +**10.1** Where permitted by the service provider concerned, the sender of a facsimile document, or his authorized representative, may, upon establishing his identity or authority, cancel his document provided its international transmission has not begun. + +**10.2** In the case of multipage facsimile documents, pages of which the international transmission has not yet begun may be cancelled. If pages have already been transmitted, the international transmitting public bureau must amend accordingly the number of pages cited on the transmittal sheet and request the receiving public bureau to do likewise. + +This amendment goes at the end of the transmission, through a new transmittal sheet, giving the correct number of pages effectively transmitted. This second transmittal sheet shall contain the necessary information to identify the sheet initially transmitted. + +**10.3** At the sender's request, pages already received by the receiving public bureau may be either delivered to the addressee or destroyed according to the instructions of the sender. + +## **11 Bureaufax table** + +**11.1** This table contains the information required for routing traffic as well as useful information concerning operational aspects of the service offered by the service provider concerned. + +**11.2** With the authorization and/or coordination of the relevant Administration, each service provider participating in the Bureaufax service should supply information required for the Bureaufax table to the ITU, either directly or through the International Bureau of the UPU. All subsequent amendments should be communicated by the Administration concerned in the same manner as they arise. + +**11.3** The ITU publishes this information periodically in the Bureaufax table, with subsequent amendments appearing in the Operational Bulletin. + +NOTE – The number of gateway bureaux must be kept to a reasonable limit. For guidance, see Bureaufax table. + +## **12 Bureaufax service via the public switched telephone network, via a public data network, via ISDN or via facilities dedicated to Bureaufax service** + +**12.1** The special provisions applicable in the case of international public facsimile service via the public switched telephone network are normally those indicated in Recommendation F.182. + +**12.2** Special provisions applicable in the case of international public facsimile service via a public data network are being studied. + +**12.3** Special provisions applicable in the case of international public facsimile service via the integrated services digital network are for further study. + +**12.4** Special provisions applicable in the case of international public facsimile service via dedicated circuits are for further study. + +**12.5** Special provisions applicable in the case of international public facsimile service using a store-and-forward switching facilities are those indicated in Recommendation F.171. + +**12.6** Special provisions applicable in the case of international public facsimile using MH services are for further study. + +## Annex A + +### Definitions of delivery modes (see Figure 1) + +#### **A.1 Letter-post item (mode A)3** + +This delivery mode enables the receiving service provider to transport and deliver the Bureaufax message through the letter-mail service in the country of destination. This mode is utilized for the transport and delivery of a physical message in case no other mode is requested (default action). In countries of destination where letter-post items are classified according to speed, Bureaufax messages are sent as priority items (1st class, "A" mail, etc.). + +#### **A.2 Delivery by special messenger (mode B)4** + +This delivery mode allows an originating user to instruct the receiving service provider to transport the Bureaufax message through the ordinary letter-mail circulation system and to deliver it by special messenger delivery. + +#### **A.3 EMS (Express Mail Service) (mode C)5** + +This delivery mode allows an originating user to instruct the receiving service provider to transport and deliver the Bureaufax message through the quickest postal service by physical means available in the destination country [Express Mail Service (EMS) or the equivalent domestic service]. Whilst delivery mode B only provides for fast delivery at the place of destination, delivery mode C means both fast transport from the receiving public bureau to the place of destination as well as fast delivery. + +#### **A.4 Counter collection (mode D)** + +This delivery mode allows an originating user to instruct the receiving service provider to keep the Bureaufax message ready for counter collection at the Bureaufax office specified by the originator, or at the Bureaufax office which offers counter collection service closest to the given recipient's address. + +#### **A.5 Counter collection with advice (modes E, G)** + +This delivery mode allows an originating user to instruct the receiving service provider to keep the Bureaufax message ready for counter collection at the Bureaufax office specified by the originator, or at the Bureaufax office which offers counter collection service closest to the given recipient's address, and to inform the recipient via telephone, or telex, using the number provided by the originator. + +#### **A.6 Telefax (mode F)** + +This delivery mode allows an originating user to instruct the accepting service provider to transmit the message to the telefax terminal of a private subscriber. + +These definitions are in alignment with the respective definitions of physical delivery elements of service as given in Annex B/F.400. + +## **Bibliography** + +- [1] International Telecommunication Constitution, Geneva, 1992. +- [2] Universal Postal Union Convention, Seoul, 1994. + +--- + +3 The corresponding service on international postal level is described in Article 9 of the Universal Postal Convention (Seoul 1994) (Article 19 of the 1989 Washington Convention). + +4 The corresponding express delivery service on international postal level is described in Article 19 of the Universal Postal Convention (Seoul 1994) (Article 35 of the 1989 Washington Convention). + +5 The corresponding service on international postal level is described in Article 57 of the Universal Postal Convention (Seoul 1994) (Article 92 of the 1989 Washington Convention). + + + +# ITU-T RECOMMENDATIONS SERIES + +| | | +|-----------------|--------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of the ITU-T | +| Series B | Means of expression: definitions, symbols, classification | +| Series C | General telecommunication statistics | +| Series D | General tariff principles | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Construction, installation and protection of cables and other elements of outside plant | +| Series M | TMN and network maintenance: international transmission systems, telephone circuits, telegraphy, facsimile and leased circuits | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks and open system communications | +| Series Y | Global information infrastructure | +| Series Z | Programming languages | \ No newline at end of file diff --git a/marked/F/T-REC-F.171-198811-I_PDF-E/raw.md b/marked/F/T-REC-F.171-198811-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..bec20baf04f8214fc40f4b6e7b218e7601b99fcd --- /dev/null +++ b/marked/F/T-REC-F.171-198811-I_PDF-E/raw.md @@ -0,0 +1,137 @@ + + +![ITU logo](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +The logo of the International Telecommunication Union (ITU) features the letters 'ITU' in a bold, sans-serif font, superimposed on a stylized globe with intersecting lines. + +ITU logo + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +**F.171** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**OPERATIONS AND QUALITY OF SERVICE +TELEMATIC SERVICE** + +--- + +**OPERATIONAL PROVISIONS RELATING TO +THE USE OF STORE - AND - FORWARD +SWITCHING NODES WITHIN THE +BUREAUFAX SERVICE** + +**ITU-T Recommendation F.171** + +(Extract from the *Blue Book*) + +--- + +# NOTES + +1 ITU-T Recommendation F.171 was published in Fascicle II.5 of the *Blue Book*. This file is an extract from the *Blue Book*. While the presentation and layout of the text might be slightly different from the *Blue Book* version, the contents of the file are identical to the *Blue Book* version and copyright conditions remain unchanged (see below). + +2 In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +## **OPERATIONAL PROVISIONS RELATING TO THE USE OF STORE-AND-FORWARD SWITCHING NODES WITHIN THE BUREAUFAX SERVICE** + +### **1 General provisions** + +1.1 In accordance with Recommendations F.160 and F.170, Administrations may operate an international public facsimile service between public bureaux. Additionally, in accordance with Recommendation F.190. Administrations may operate a service between public bureaux and private subscribers' stations. + +1.2 In the bureaufax service, it is not uncommon for the transmitted document to be scanned and transmitted several times as it progresses from the office of origin to the office of destination. With each rescan and retransmission, the quality of the document, as finally delivered to the addressee is progressively degraded. By the replacement of manual retransmission practices with switching nodes, employing store-and-forward techniques, document quality may be preserved. Quality may even be improved over current transmission practices through the use of error handling techniques between nodes and the use of data networks, rather than the telephone network, as the transmission medium, + +1.3 This Recommendation states the operational provisions peculiar to the bureaufax service when store-and-forward facsimile switching nodes are employed within the international transmission of documents. + +### **2 Conditions of acceptance** + +2.1 Documents for transmission within the bureaufax service and between subscriber stations and public bureaux shall conform to § 2 of Recommendation F.170. + +### **3 Composition of a facsimile transmission** + +3.1 Every facsimile transmission shall include a transmittal sheet and the customer's document as defined in § 3 of Recommendation F.170 + +### **4 Acceptance** + +#### *4.1 Acceptance at a public bureau* + +The acceptance of documents at the counter of a public bureau will conform to § 4 of Recommendation F.170. + +#### *4.2 Acceptance from subscribers' stations* + +4.2.1 A document for delivery through the bureaufax service may be accepted directly from the subscriber's station into store and forward node in the country of origination. + +4.2.2 The customer will access the store and forward node in conformance with the provisions of Recommendation F.162. At the discretion of the administration in the country of origin, the customer must be a registered user of the service. + +4.2.3 The Administration in the country of origin will provide users of the service with the information necessary to route the document through the store and forward node, and any interconnected store and forward node, to the appropriate public bureau in the country of destination. The addressing of such facsimile messages should use abbreviated dialling codes rather than explicit telephone numbers for the facsimile terminals in the destination bureau. + +4.2.4 The correct addressing of the facsimile message will be the responsibility of the originating subscriber. + +4.2.5 As for transmissions originating from a public bureau, messages entered into the bureaufax service by a subscriber must contain a transmittal sheet, in accordance with F.170, § 3.2, as the first page. The Administration in the country of origin will ensure that transmittal forms are available to its subscribers. + +4.2.6 Acceptance of the facsimile document by the node will constitute acceptance into the bureaufax service. + +4.2.7 Facsimile documents to be entered into the bureaufax service may, optionally, be accepted from text terminals in cases where the node can provide the necessary conversion to the appropriate facsimile format. In such cases, the transmittal sheet may be generated either by the originating terminal or by the node. + +### 5 Transmission + +5.1 International bureaufax transmissions may involve store and forward switching nodes, typically at gateway locations, in three different ways: + +- nodes in both the country of origin and country of destination, +- node located in the country of origin, +- node located in the country of destination. + +In any case, each node will be programmed with abbreviated codes for delivering documents to inland public bureaux. These abbreviated codes will be made available to other administrations for the efficient routing of traffic. + +#### 5.1.1 *Nodes in both countries* + +5.1.1.1 Facsimile documents will be entered into the node both by public bureaux and by private subscribers. + +5.1.1.2 Documents will be transmitted between nodes in accordance with Recommendation F.162, § 4.6. Wherever possible, the routing between nodes will employ networks designed specifically for the transmission of data to preserve a maximum quality of delivered document. This is important since the end-to-end transmission may still involve multiple scans and retransmissions. + +5.1.1.3 If the quality of the document received at the public bureau in the country of destination is not satisfactory, the receiving bureau will contact the assistance operator at the node in the country of destination, who, following confirmation that the document held in store in the node is acceptable, will initiate a retransmission. + +5.1.1.4 If the quality of the document held in store by the node in the country of destination is not satisfactory, the operator will contact the assistance operator at the node in the country of origin who, following confirmation that the document held in store there is acceptable, will initiate a retransmission. + +5.1.1.5 If the quality of the document held in store by the node in the country of origin is not satisfactory, the operator will request retransmission from the originating public bureau or subscriber. + +#### 5.1.2 *Node in the country of origin* + +5.1.2.1 The international transmission from the node in the country of origin to the public bureau in the country of destination will comply with Recommendation F.170, § 5, with the following exceptions: + +5.1.2.2 If a document received at the international receiving public station is not satisfactory, the international receiving public station will contact the assistance operator at the node in the originating country to request retransmission. The assistance operator will act in accordance with § 5.1.1.5 above. + +#### 5.1.3 *Node in the destination country* + +5.1.3.1 Except by bilateral agreement, the node in the destination country will accept facsimile transmissions from the country of origin in a manner emulating a facsimile terminal, i.e. the originating country's operator need not be knowledgeable of any special procedures to deposit the document into the node. In this case, transmission is in conformance with § 5 of Recommendation F.170 except that the node acts as the international receiving public facsimile station. + +5.1.3.2 Documents received by the node in the above manner will be routed to the destination public bureau through manual intervention. + +### 6 Delivery + +6.1 Delivery of facsimile messages by the destination public facsimile bureau will generally comply with § 6 of Recommendation F.170. + +6.2 Delivery of bureaufax messages to subscribers' stations will generally comply with § 2 of Recommendation F. 190. + +6.2.1 In cases where a store and forward node exists in the country of destination, the message will be transmitted to the subscribers' stations directly from the node, without recourse to an additional stage of printing and rescanning the document. + +6.2.2 By bilateral agreement, the facsimile transmitting station in the country of origin may provide the node with sufficient information to effect automatic switching and retransmission of the message directly to subscribers' stations. Otherwise, transmission to subscribers' stations will be effected by manual intervention of the node operator. + +6.2.3 By bilateral agreement in cases where a node exists only in the country of origin, facsimile messages may be transmitted directly from the node to subscribers' stations in the country of destination. + +### **7 Other provisions** + +Provisions relating to: + +- Service facsimile correspondence +- Archives +- Charging, refunds and accounting +- Cancellation at the request of the sender +- List of public bureaux + +will conform to the appropriate section of Recommendation F.170. \ No newline at end of file diff --git a/marked/F/T-REC-F.18-199803-I_PDF-E/raw.md b/marked/F/T-REC-F.18-199803-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..156e21406a007f339bdcfacae5d3a98a418e4dbe --- /dev/null +++ b/marked/F/T-REC-F.18-199803-I_PDF-E/raw.md @@ -0,0 +1,376 @@ + + +![ITU logo: A globe with a lightning bolt and the letters ITU.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with a lightning bolt and the letters ITU. + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.18** + +(03/98) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Telegraph service – Operating methods for the +international public telegram service + +--- + +**Guidelines on harmonization of international +public bureau services** + +ITU-T Recommendation F.18 + +(Previously CCITT Recommendation) + +--- + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|------------------------------------------------------------------------|-----------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| AUDIOVISUAL SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to ITU-T List of Recommendations.* + +# **ITU-T RECOMMENDATION F.18** + +# **GUIDELINES ON HARMONIZATION OF INTERNATIONAL PUBLIC BUREAU SERVICES** + +## **Summary** + +This Recommendation contains the general concept for the harmonized application of international public bureau services. It is intended to guide Administrations/ROAs in implementing Recommendation F.11 (Continued availability of traditional services). + +###### **Source** + +ITU-T Recommendation F.18 was revised by ITU-T Study Group 2 (1997-2000) and was approved under the WTSC Resolution No. 1 procedure on the 9th of March 1998. + +## FOREWORD + +ITU (International Telecommunication Union) is the United Nations Specialized Agency in the field of telecommunications. The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of the ITU. The ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Conference (WTSC), which meets every four years, establishes the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics. + +The approval of Recommendations by the Members of the ITU-T is covered by the procedure laid down in WTSC Resolution No. 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +### INTELLECTUAL PROPERTY RIGHTS + +The ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. The ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, the ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementors are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database. + +© ITU 1998 + +All rights reserved. No part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and microfilm, without permission in writing from the ITU. + +## CONTENTS + +| | Page | +|---------------------------------------------------------------------------------------------|-------------| +| 1 Introduction..... | 1 | +| 2 Terminology..... | 1 | +| 3 Scope..... | 1 | +| 4 Basic requirements..... | 2 | +| 5 Mode of operation..... | 2 | +| Annex A – Harmonization of international public bureau services – Detailed guidelines ..... | 2 | +| A.1 Introduction ..... | 2 | +| A.2 Possible solutions for a harmonization of bureau services ..... | 3 | +| A.3 Routing and retransmission of messages (Selection modes)..... | 5 | + + + +## **GUIDELINES ON HARMONIZATION OF INTERNATIONAL PUBLIC BUREAU SERVICES** + +*(revised in 1998)* + +## **1 Introduction** + +*considering* + +- a) the reasonable needs of customers for a service of last resort; +- b) ITU-T Study Group 1's work on harmonization of the telegram, telemessage and bureaufax services; +- c) Resolution No. 6 (World Administrative Telegraph and Telephone Conference, WATTC, Melbourne 1988) and Recommendation F.11 on continued availability of traditional services in accordance with which the classical telegram service is no longer obligatory for international relations where a reasonable alternative such as bureaufax or telemessage exists; +- d) the moral obligation on ITU Members to provide at least a form of international service of last resort; +- e) the possibilities of newly introduced transport mechanisms or services which could play an important role for efficient operation of traditional services; +- f) that there is no basis in the Regulations of the ITU to require any particular form of transmission for bureau services; +- g) that a harmonization and utilization of positive elements of the individual bureau services could improve flexibility, simplify operation, and increase cost-effectiveness; +- h) that bureaufax, in view of its much greater flexibility, simple operation and cost-effectiveness, might be encouraged in other countries to allow a transfer from telegram transmitted in accordance with the 1998 version of Recommendation F.1 to facsimile applications over a wide area; +- i) that other technical solutions may be developed, + +this Recommendation provides broad guidelines on the minimum requirements for a harmonized operation of existing bureau services. + +## **2 Terminology** + +For the purpose of this Recommendation, a **bureau service** is a message service for those who are permanently or temporarily not subscribers to other services or who wish to use this service for any other reason whereby it will be perceived as simply transporting a message regardless of the mechanism used. + +## **3 Scope** + +**3.1** This Recommendation contains the general concept for the harmonized application of international public bureau services. It is intended to guide Administrations/ROAs in implementing Recommendation F.11. + +**3.2** The international public bureau services should provide the capability of operation in conjunction with each other to the extent practicable. + +**3.3** The tariff principles for the international public bureau services are laid down in the relevant D-Series Recommendations. + +**3.4** The provisions set out below shall apply to the operation of public bureau services via telecommunication networks: + +- a) between public facsimile bureaux (bureaufax: see Recommendations F.170 and F.171); +- b) between public facsimile bureaux and subscribers' facsimile stations, and vice versa (see Recommendation F.190); +- c) between telegram offices (see Recommendations F.1 and F.31); +- d) between telemessage offices (see Recommendation F.40); +- e) between telegram and bureaufax offices (see Recommendations F.1, F.31 and F.170); +- f) between telegram offices and subscriber facsimile stations (see Recommendation F.1); +- g) between telegram offices and telex subscribers (see Recommendation F.1); +- h) between telegram and telemessage offices and vice versa (see Recommendation F.41). + +### **3.5 Classes of service** + +Classes of service in each category mentioned in 3.4 are handled in the relevant F-Series Recommendations. + +## **4 Basic requirements** + +**4.1** The basic requirements for international public bureau services are: + +- a) reasonable access to the national service in the country of origin for lodging messages (e.g. over a counter, by telephone and perhaps by other telecommunication links to a public office, which may need to include format conversion); +- b) international transmission (e.g. by facsimile, telegram retransmission, gentex, telex, MHS, ...); +- c) format conversion, store-and-forward at a gateway if required; +- d) national transmission and/or format conversion as necessary; +- e) delivery by mail, or electronically as an option where practicable. + +**4.2** The more likely and preferred harmonized applications will be identified in the appropriate service Recommendations as necessary, including the operational procedures, and service parameters. + +**4.3** Any effects on charging and accounting will be covered in the D-Series Recommendations with the objective of operational simplicity. + +## **5 Mode of operation** + +**5.1** The mode of operation should be in accordance with the mechanism used for transporting the message [see 4.1, b)]. + +**5.2** Specific details will be found in the individual service Recommendations. + +**5.3** The quality of service aspects are also handled in the relevant F-Series Recommendations. + +## **Annex A** + +### **Harmonization of international public bureau services – Detailed guidelines** + +### **A.1 Introduction** + +This annex represents a more detailed approach for possible solutions for a harmonization of bureau services, taking into account the different situations in the various countries. Different steps or phases seem to be necessary and/or possible to introduce improvements in the area of bureau services. + +A number of problems/solutions depend on national organization/structure of bureau services. + +The following subclauses propose arrangements where facsimile transmission is used to carry telegrams nationally and/or internationally. + +Specific operational provisions applying to facsimile transmissions can be found in Recommendations F.1, Part E, and F.170. + +If specific provisions do not exist in Recommendation F.1, a number of problems could be solved by bilateral agreements. + +### **A.2 Possible solutions for a harmonization of bureau services** + +#### **A.2.1 Possible first approach (Domestic telegram and bureaufax services; see Figure A.1)** + +The harmonization of both services could start nationally simply by common use of telegram and bureaufax facilities. No technical changes are necessary. + +The way these services are named, charged and marketed is a national matter. + +National innovation should not be restrained. Which organization (postal authority, telecommunication ROA, etc.) operates any bureau service is also a national matter. + +At this stage, at the international interface, a clear distinction should be made between telegrams and bureaufax. + +Internationally the character-coded telegram service could be replaced by a fax-based bureau service. + +Where suitable replacements exist, the international telegram will gradually disappear. Abolition in low-volume relations should be encouraged. + +This stage could be characterised by: + +- The existing technical facilities will be used further, only necessary improvements will be implemented. +- The transmission via facsimile will be preferred. +- Reasonable access to the national service for the input of messages will be provided. +- Domestic messages handed in over the counter will be transmitted via facsimile to the appropriate public bureau. Messages can contain written matter or printed matter, drawings or any other graphics. +- Telegrams and bureaufax for abroad will be transmitted, by the accepting public bureau, to the appropriate national gateway for further transmission (character-coded or image-coded). +- Where agreed, the office of origin may route telegrams directly to a telefax subscriber in another country. + +#### **A.2.2 Possible next step (International telegram and bureaufax service; see Figure A.2)** + +All aspects under A.2.1 should be considered. + +The domestic Telegram Retransmission System will be improved (character-coded/image-coded interface to a fax store-and-forward system). + +Direct transmission and delivery to fax machines nationwide and abroad will be possible. + +By bilateral agreement it becomes possible to transmit (F.1) telegrams from the origin country to a public bureau abroad by fax. + +The character set (ITA2) used for telegrams remains unchanged. + +Direct international transmission via point-to-point (p.t.p.) and gentex will no longer be the sole means of international transmission. + +Some countries may need to continue an F.1-type service nationally for longer. In this case a gateway to the international bureaufax service would facilitate these international transmissions. + +![Figure A.1/F.18: Possible solutions for a harmonization of bureaux services. This flowchart illustrates various transmission paths from a 'CUSTOMER (Sender)' to an 'Addressee'. The sender can choose between VTX, FAX, COUNTER, TLF, and TLX. VTX and FAX lead to a 'Bureaufax Office (Gateway)', which connects to 'Foreign countries' via 'Facsimile' and to an 'Appropriate Public Bureau (destination)' via 'FAX'. The COUNTER path leads to a 'Public Bureau (origin)', which connects to the 'National Telegraph Office' via 'FAX' and to the 'Appropriate Public Bureau (destination)' via 'FAX'. The TLF and TLX paths lead to the 'National Telegraph Office' and 'TRS Telegram Retransmission System', which both connect to 'Foreign countries' via 'Point-to-Point', 'Gentex', and 'Facsimile transmission'. The 'Appropriate Public Bureau (destination)' connects to the 'Addressee' via 'FAX', 'TLF', and 'Messenger Counter collection Mail P.O. Box'. The 'National Telegraph Office' and 'TRS' also connect to the 'Addressee' via 'TLF' and 'TLX'. A label 'T0206800-97/d01' is present in the bottom right.](a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg) + +Figure A.1/F.18: Possible solutions for a harmonization of bureaux services. This flowchart illustrates various transmission paths from a 'CUSTOMER (Sender)' to an 'Addressee'. The sender can choose between VTX, FAX, COUNTER, TLF, and TLX. VTX and FAX lead to a 'Bureaufax Office (Gateway)', which connects to 'Foreign countries' via 'Facsimile' and to an 'Appropriate Public Bureau (destination)' via 'FAX'. The COUNTER path leads to a 'Public Bureau (origin)', which connects to the 'National Telegraph Office' via 'FAX' and to the 'Appropriate Public Bureau (destination)' via 'FAX'. The TLF and TLX paths lead to the 'National Telegraph Office' and 'TRS Telegram Retransmission System', which both connect to 'Foreign countries' via 'Point-to-Point', 'Gentex', and 'Facsimile transmission'. The 'Appropriate Public Bureau (destination)' connects to the 'Addressee' via 'FAX', 'TLF', and 'Messenger Counter collection Mail P.O. Box'. The 'National Telegraph Office' and 'TRS' also connect to the 'Addressee' via 'TLF' and 'TLX'. A label 'T0206800-97/d01' is present in the bottom right. + +**Figure A.1/F.18 – Possible solutions for a harmonization of bureaux services** + +The number of fax machines in public bureaux will be increased, gentex transmission to teleprinters (domestic) will be substituted by facsimile transmission. + +**Advantages:** Cost reduction (fax machines and transmission over the public telephone network is cheaper than the costs of teleprinters and the use of the national gentex network). + +The handling of messages takes less time (no word count necessary, easy transmission by facsimile etc.). + +In the absence of new international accounting provisions, accounting could be solved by bilateral agreements. + +#### **A.2.3 Possible second approach (International telegram and bureaufax services; see Figure A.3)** + +All aspects under A.2.1 and A.2.2 should be considered. + +Implementation of a domestic fax store-and-forward-system with automatic routing capabilities and centralised accounting and statistics. + +In addition, where bilaterally agreed, the form accepted from the originator can be transmitted by facsimile internationally. Where agreed, characters outside ITA2 may be used. + +The treatment of such messages, including international accounting, can be agreed bilaterally. + +#### **A.2.4 Possible final solution; see Figure A.4** + +The aim is to replace the classical telegram transmission technique completely by another transport mechanism. + +### A.3 Routing and retransmission of messages (Selection modes) + +It should be noted that the choice of international transmission mode is not necessarily the same in each direction. For example, in any international relation it may be agreed that facsimile be used in one direction and printing telegraph in the other. + +![Figure A.2/F.18: Possible next step (International telegram and bureaufax service). This flowchart shows message routing from various sources (Counter/Fax, Customer/Fax A, Customer/Fax B, Counter/TF, Customer/TF, Counter/TF/Fax, Customer/TF/Fax) through a Manual operation centre or Intercept to a Telegram retransmission centre. From there, messages are routed through a National gateway (F.31/fax store and forward) or an International gateway (F.31/fax store and forward). The National gateway leads to a National delivery point (outputting Messenger and TF) or a Fax subscriber. The International gateway leads to an International delivery point (outputting Messenger and TF) or a Fax subscriber. All international paths are labeled 'charactercoded and/or picle coded'. A code T0206810-97/d02 is present at the bottom right.](e9314c83043183351ed74908e9bf2f90_img.jpg) + +Figure A.2/F.18: Possible next step (International telegram and bureaufax service). This flowchart shows message routing from various sources (Counter/Fax, Customer/Fax A, Customer/Fax B, Counter/TF, Customer/TF, Counter/TF/Fax, Customer/TF/Fax) through a Manual operation centre or Intercept to a Telegram retransmission centre. From there, messages are routed through a National gateway (F.31/fax store and forward) or an International gateway (F.31/fax store and forward). The National gateway leads to a National delivery point (outputting Messenger and TF) or a Fax subscriber. The International gateway leads to an International delivery point (outputting Messenger and TF) or a Fax subscriber. All international paths are labeled 'charactercoded and/or picle coded'. A code T0206810-97/d02 is present at the bottom right. + +Figure A.2/F.18 – Possible next step (International telegram and bureaufax service) + +![Figure A.3/F.18: Possible second approach (International telegram and bureaufax services). This flowchart shows message routing from sources (Counter/Fax (National Tel), Customer/Fax A (National Tel), Customer/Fax B (National Tel), Counter/TF (National Tel), Customer/TF (National Tel), Counter/TF/Fax (International Tel), Customer/TF/Fax (International Tel)). National sources go to a Manual operation centre or Intercept A (National traffic), then to a Fax store and forward unit. International sources go to a Manual operation centre or Intercept B (International traffic), then to a Telegram retransmission centre. The Fax store and forward unit connects to a National gateway (leading to National delivery point: Messenger, TF, or Fax subscriber) and an International gateway (leading to International delivery point: Messenger, TF, or Fax subscriber). The Telegram retransmission centre connects to an International gateway (leading to Point-to-Point or Gentex). A code T0206820-97/d03 is present at the bottom right.](d26959f4514c26ca19c3d6f00da85956_img.jpg) + +Figure A.3/F.18: Possible second approach (International telegram and bureaufax services). This flowchart shows message routing from sources (Counter/Fax (National Tel), Customer/Fax A (National Tel), Customer/Fax B (National Tel), Counter/TF (National Tel), Customer/TF (National Tel), Counter/TF/Fax (International Tel), Customer/TF/Fax (International Tel)). National sources go to a Manual operation centre or Intercept A (National traffic), then to a Fax store and forward unit. International sources go to a Manual operation centre or Intercept B (International traffic), then to a Telegram retransmission centre. The Fax store and forward unit connects to a National gateway (leading to National delivery point: Messenger, TF, or Fax subscriber) and an International gateway (leading to International delivery point: Messenger, TF, or Fax subscriber). The Telegram retransmission centre connects to an International gateway (leading to Point-to-Point or Gentex). A code T0206820-97/d03 is present at the bottom right. + +Figure A.3/F.18 – Possible second approach (International telegram and bureaufax services) + +![Figure A.4/F.18 – Possible final solution. A flow diagram showing various input types entering a central unit and branching out to national and international delivery points.](d4af765160d04ecef538e5066006dc77_img.jpg) + +``` + +graph LR + subgraph Inputs + C1[Counter/Fax] --> FSFU + C2[Customer/Fax A] --> FSFU + C3[Customer/Fax B] --> FSFU + C4[Counter/TF] --> FSFU + C5[Customer/TF] --> FSFU + C6[Counter/TF/Fax] --> FSFU + C7[Customer/TF/Fax] --> FSFU + end + + subgraph Central_Units + FSFU[Fax store and forward unit] + MOC[Manual operation centre Intercept] + FSFU --- MOC + end + + FSFU --> NGW[National gateway] + FSFU --> IGW[International gateway] + + NGW --> NDP[National delivery point] + NGW --> FS1[Fax subscriber] + NDP --> M1[Messenger] + NDP --> TF1[TF] + + IGW --> IDP[International delivery point] + IGW --> FS2[Fax subscriber] + IDP --> M2[Messenger] + IDP --> TF2[TF] + +``` + +T0206830-97/d04 + +Figure A.4/F.18 – Possible final solution. A flow diagram showing various input types entering a central unit and branching out to national and international delivery points. + +**Figure A.4/F.18 – Possible final solution** + +![](042733dc5e8e7f5f30b60adba3266cde_img.jpg) + +Operational Modes of Bureau-Services + +| Single operational mode | More than one single operational mode | +|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +|
Acceptance counter
(in any form), TF, TLX, etc.
National system
National gateway country A
to foreign countries
Image Character
Coded transmission

a)

|
Acceptance counter
(in any form), TF, TLX, etc.
National system
National gateway country A
to foreign countries
International gateway in country B
Image Character
coded transmission

b)

Image Character
coded transmission
International gateway in country C
| + +T0206840-97/d05 + +- a) Conversion is in all parts a national matter. +- b) International conversion between gateways B and C. + +**Figure A.5/F.18 – Routing and retransmission of messages (selection modes)** + +# ITU-T RECOMMENDATIONS SERIES + +| | | +|-----------------|--------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of the ITU-T | +| Series B | Means of expression: definitions, symbols, classification | +| Series C | General telecommunication statistics | +| Series D | General tariff principles | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Construction, installation and protection of cables and other elements of outside plant | +| Series M | TMN and network maintenance: international transmission systems, telephone circuits, telegraphy, facsimile and leased circuits | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks and open system communications | +| Series Y | Global information infrastructure | +| Series Z | Programming languages | \ No newline at end of file diff --git a/marked/F/T-REC-F.182bis-199610-I_PDF-E/raw.md b/marked/F/T-REC-F.182bis-199610-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..2acdeed02c99335a5957af7a9c5178c23323be48 --- /dev/null +++ b/marked/F/T-REC-F.182bis-199610-I_PDF-E/raw.md @@ -0,0 +1,294 @@ + + +![ITU logo](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +The logo of the International Telecommunication Union (ITU) features a globe with a lightning bolt superimposed on it, and the letters 'ITU' written across the globe. + +ITU logo + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.182 *bis*** + +(10/96) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Telematic services – Public facsimile service + +--- + +**Guidelines for the support of the +communication of documents using Group 3 +facsimile between user terminals via public +networks** + +ITU-T Recommendation F.182 *bis* + +(Previously CCITT Recommendation) + +--- + +# ITU-T F-SERIES RECOMMENDATIONS NON-TELEPHONE TELECOMMUNICATION SERVICES + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1–F.109 | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | F.110–F.159 | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | F.160–F.399 | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | F.550–F.599 | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| AUDIOVISUAL SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to ITU-T List of Recommendations.* + +## FOREWORD + +The ITU-T (Telecommunication Standardization Sector) is a permanent organ of the International Telecommunication Union (ITU). The ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Conference (WTSC), which meets every four years, establishes the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics. + +The approval of Recommendations by the Members of the ITU-T is covered by the procedure laid down in WTSC Resolution No. 1 (Helsinki, March 1-12, 1993). + +ITU-T Recommendation F.182 *bis*, was prepared by ITU-T Study Group 1 (1993-1996) and was approved by the WTSC (Geneva, 9-18 October 1996). + +--- + +## NOTE + +In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +© ITU 1997 + +All rights reserved. No part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and microfilm, without permission in writing from the ITU. + +## CONTENTS + +| | Page | +|-----------------------------------------------------|-------------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Abbreviations..... | 1 | +| 4 General provisions ..... | 2 | +| 4.1 General..... | 2 | +| 4.2 Basic requirements..... | 2 | +| 4.3 Standardized options..... | 2 | +| 5 Network requirements..... | 3 | +| 6 Procedures for the transmission of documents..... | 3 | +| 7 Coding scheme..... | 3 | +| 8 Operation..... | 3 | +| 9 Terminal identification..... | 3 | +| 10 User terminals ..... | 4 | +| 11 Directories ..... | 4 | + +# **SUMMARY** + +This Recommendation defines the guidelines for the support of the communication of documents using Group 3 facsimile between user terminals via public networks and is intended to apply in those cases where an ROA or service provider does not wish to offer a Telefax 3 service as defined in Recommendation F.182. This Recommendation is based upon existing Recommendation F.182 but has been developed to reflect the changing telecommunications environment with regard to the provision of regulated "end-to-end" services. + + + +# **GUIDELINES FOR THE SUPPORT OF THE COMMUNICATION OF DOCUMENTS USING GROUP 3 FACSIMILE BETWEEN USER TERMINALS VIA PUBLIC NETWORKS** + +*(Geneva, 1996)* + +## **1 Scope** + +**1.1** This Recommendation defines the guidelines which may be considered by ROAs or service providers in supporting the communication of documents using Group 3 facsimile. + +**1.2** The communication of documents using Group 3 facsimile enables users to exchange documents either manually or automatically via telecommunication networks across international boundaries. + +**1.3** The basic element of the correspondence between people is the page, as the smallest unit of text treated as an entity. No restrictions shall exist as far as the user procedures for generation of the text or the positioning of text within the reproducible area on a page are concerned. + +**1.4** Questions of an essentially technical nature concerning the communication of documents using Group 3 facsimile are dealt with in the T-Series Recommendations. + +**1.5** Provision for communication of documents using Group 3 facsimile may vary from one ROA to another. In some networks, the communication may be provided on an unassured basis and have no guarantee of the quality of transmission, transmission speed, etc. For example, in some cases the call is treated in the same way as a telephone call. + +NOTE – If an ROA does not wish to follow the guidelines defined in this Recommendation, then consideration could be given to the use of Recommendation F.182. + +## **2 References** + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; all users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. + +- ITU-T Recommendation F.182 (1996), *Operational provisions for the international public facsimile service between subscribers with Group 3 facsimile terminals (Telefax 3)*. +- ITU-T Recommendation F.180 (1996), *General operational provisions for the international public facsimile service between subscribers' terminals (Telefax)*. +- ITU-T Recommendation T.4 (1996), *Standardization of Group 3 facsimile terminals for document transmission*. +- ITU-T Recommendation T.30 (1996), *Procedures for document facsimile transmission in the general switched telephone network*. +- ITU-T Recommendation T.505 (1994), *Document application profile PM-26 for the interchange of enhanced structure, mixed content documents in processable and formatted forms*. + +## **3 Abbreviations** + +This Recommendation uses the following abbreviations: + +ITU-T International Telecommunication Union – Telecommunication Standardization Sector + +ROA Recognized Operating Agency + +## 4 General provisions + +### 4.1 General + +The essential characteristic of the communication of documents using Group 3 facsimile is that it provides a basic level of compatibility between all terminals supporting the Group 3 capabilities. + +### 4.2 Basic requirements + +- a) The basic level of compatibility is provided between any two terminals internationally so that they may communicate image coded information to each other. This is achieved by terminals complying with Recommendations T.4 and T.30; +- b) it is for each ROA or service provider to decide on the network(s) on which the communication of documents using Group 3 facsimile will be carried. There is no restriction on the type of network to be used; +- c) it should be possible to extend the communication of documents using Group 3 facsimile to any number of countries; +- d) to permit private use options, for example encryption, there is no technical limitation on the bit sequence of the user's information that may be transmitted; +- e) a received Group 3 document can be printed or displayed as decided by the recipient and the terminal characteristics. If the document is printed, the receiving user will be furnished with a document that is identical with that produced by the sending user as far as its contents, layout and format are concerned; +- f) it is intended that support for the communication of documents using Group 3 facsimile requires no changes to the Recommendations for existing services or networks. + +### 4.3 Standardized options + +**4.3.1** It is recognized that some users may need to use their Group 3 facsimile terminals to communicate internationally using features that are not included in the basic requirements. A number of ITU-T standardized options have therefore been defined. However, the provision of any option may lead to some degree of incompatibility and the number of standardized options has been restricted to those features which are listed below and for which a clear international need can be foreseen. + +The sending terminal ensures the transmission of documents using only those options that have been indicated as being available at the receiving terminal. + +**4.3.2** The standardized options provide means for: + +- a) higher data signalling rates (Recommendation T.4); +- b) subscriber identification (Recommendation T.30); +- c) Error Correction Mode (ECM) (Recommendation T.4); +- d) higher resolutions (Recommendation T.4); +- e) optional coding schemes (Recommendation T.4); +- f) optional page sizes (Recommendation T.4); +- g) polling (Recommendation T.30); +- h) Selective Polling (SEP) (Recommendation T.30); +- i) Password for polling (PWD) (Recommendation T.30); +- j) Sender Identification (SID) (Recommendation T.30); +- k) Sub-addressing (SUB) (Recommendation T.30); +- l) coded character mode (Recommendation T.4); +- m) Binary File Transfer mode (BFT) (Recommendation T.4); +- n) Basic Transfer Mode (BTM) (Recommendation T.4); +- o) Document Transfer Mode (DTM) (Recommendation T.4); +- p) Electronic Document Interchange (EDI) (Recommendation T.4); + +- q) simple mixed mode (Recommendation T.4); +- r) processable mode 26 (Recommendation T.505); +- s) digital network operation (Recommendation T.30). + +NOTE – As the communication of documents using Group 3 facsimile evolves and further market and user requirements are identified, there is a continuing need for further study and consequently changes may be required to the above list. + +**4.4** Communication between terminals takes place at the level of compatibility at, or nearest to, the parameters initially selected by the originating user. The determination of this level is carried out automatically via the premessage procedures. This takes into account aspects of the quality of transmission media available in the participating networks and the options available on the receiving terminal. + +## **5 Network requirements** + +**5.1** It is the responsibility of each ROA or service provider to decide on which network(s) the communication of documents using Group 3 facsimile is to be provided. + +**5.2** The communication of documents using Group 3 facsimile may be supported over all public networks using the appropriate ITU-T technical Recommendations. + +If the Integrated Services Digital Network is used, the communication of documents using Group 3 facsimile will be supported as appropriate. + +**5.3** Terminals on any public network which support the communication of documents using Group 3 facsimile are able to communicate automatically with terminals supporting the communication of documents using Group 3 Facsimile on any other public network. + +## **6 Procedures for the transmission of documents** + +The technical procedures for transmitting documents using the communication of documents using Group 3 facsimile are given in Recommendation T.30. + +## **7 Coding scheme** + +**7.1** The basic coding scheme for the communication of documents using Group 3 facsimile is given in Recommendation T.4. + +**7.2** The optional image coding schemes are given in Recommendation T.4. + +**7.3** The optional character coding repertoire is given in Recommendation T.4. + +## **8 Operation** + +**8.1** Support for the communication of documents using Group 3 facsimile in each country and the interconnection between countries or networks should use fully automatic switching so that it is possible for any user to reach any other user using fully automatic selection. + +**8.2** It is recommended to allow the through connection of a call between terminals connected to a private automatic branch exchange (or similar systems) and those connected to public exchanges used to provide support for the communication of documents using Group 3 facsimile. + +## **9 Terminal identification** + +**9.1** On the PSTN, an answering tone of 2100 Hz identifies connection to a non-voice terminal. Additionally, an automatic identification may be provided, see 9.3. ROAs and service providers may wish to consider recommending to users that the terminal identity is entered in the format described in 9.3. + +**9.2** If the pages are printed, it is desirable for the identification of the transmitting terminal to appear at the receiving terminal, in the form of an identification line printed at the top of each page received, in an area 10 mm deep extending across the whole width of the page. This area may be either inside or outside the nominal A4 page, in accordance with the option selected by the user. + +**9.3** The terminal identification consists of up to 20 characters (including only digits and spaces). For the PSTN and the ISDN, the identification is the international telephone number and the format is as follows: plus sign, country code, space, area code, space, user's number. The coding arrangements are defined in Recommendation T.30. + +The form of the terminal identification on other networks is for further study. + +**9.4** At the transmitting terminal, the identification of the receiving terminal may be displayed or printed. + +## **10 User terminals** + +**10.1** A range of data signalling rates has been defined as follows: + +### **10.1.1 Public switched telephone network** + +Terminals on the public switched network are able to operate at the basic data signalling rate of 4800 bit/s with fall-back to 2400 bit/s. In addition, higher data signalling rates may be provided. Details are given in Recommendation T.30. + +### **10.1.2 Integrated Services Digital Network (ISDN)** + +If the Integrated Services Digital Network is used, the communication of documents using Group 3 facsimile will be provided as appropriate. + +### **10.1.3 Public data networks** + +For further study. + +**10.2** The facilities required in terminals participating in the international communication of documents using Group 3 facsimile are listed in the following subclauses: + +**10.2.1** Group 3 facsimile terminals have the ability to send, receive and display documents encoded with the image coding scheme defined in Recommendation T.4. In addition, they may have the ability to utilize the optional coding schemes defined in Recommendation T.4. + +**10.2.2** No constraints shall be placed upon the type of presentation technology employed. + +## **11 Directories** + +As a means of stimulating the use of the network(s), some ROAs or service providers may wish to consider publishing or encouraging other organizations to publish a directory of users. + +The format of the directory is outside the scope of this Recommendation. + +# ITU-T RECOMMENDATIONS SERIES + +- Series A Organization of the work of the ITU-T +- Series B Means of expression: definitions, symbols, classification +- Series C General telecommunication statistics +- Series D General tariff principles +- Series E Overall network operation, telephone service, service operation and human factors +- Series F Non-telephone telecommunication services** +- Series G Transmission systems and media, digital systems and networks +- Series H Audiovisual and multimedia systems +- Series I Integrated services digital network +- Series J Transmission of television, sound programme and other multimedia signals +- Series K Protection against interference +- Series L Construction, installation and protection of cables and other elements of outside plant +- Series M Maintenance: international transmission systems, telephone circuits, telegraphy, facsimile and leased circuits +- Series N Maintenance: international sound programme and television transmission circuits +- Series O Specifications of measuring equipment +- Series P Telephone transmission quality, telephone installations, local line networks +- Series Q Switching and signalling +- Series R Telegraph transmission +- Series S Telegraph services terminal equipment +- Series T Terminals for telematic services +- Series U Telegraph switching +- Series V Data communication over the telephone network +- Series X Data networks and open system communication +- Series Z Programming languages \ No newline at end of file diff --git a/marked/F/T-REC-F.19-199610-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.19-199610-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..898e6b9d9694e2a087ecc69cc2c45083051a3f77 --- /dev/null +++ b/marked/F/T-REC-F.19-199610-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a74b051ef0f222b2f19f144ee2b23a0ca3aff81e673aafd6626c095c4fffcc0d +size 8265 diff --git a/marked/F/T-REC-F.190-199208-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.190-199208-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..0b78b9ad33df9b1f824d02a4743f4c7b6346ad12 --- /dev/null +++ b/marked/F/T-REC-F.190-199208-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bc0726a4e83123da3189bfa26dffd0080e2d16d49d0f5d83886537342f91bf4c +size 8217 diff --git a/marked/F/T-REC-F.190-199208-I_PDF-E/547f726730e589392f239257a833ede3_img.jpg b/marked/F/T-REC-F.190-199208-I_PDF-E/547f726730e589392f239257a833ede3_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a9ea8fb2f85b28e68e4757f1b760a7178db74199 --- /dev/null +++ b/marked/F/T-REC-F.190-199208-I_PDF-E/547f726730e589392f239257a833ede3_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:01b6a171d26aed03ce0dcb9c91e19a8466d50c813bd155fbdd713aedf1b6ebee +size 92752 diff --git a/marked/F/T-REC-F.190-199208-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.190-199208-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..daf0982965c0e50f68caa686800bcedb8083a285 --- /dev/null +++ b/marked/F/T-REC-F.190-199208-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f4e5c5a414fb4a88d02e74f360667a904b0b6ba0d3ef075045394b00a2a2685c +size 5780 diff --git a/marked/F/T-REC-F.2-198811-I_PDF-E/raw.md b/marked/F/T-REC-F.2-198811-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..5956def7dab9c2d44c1ac534f48ec82c30b95b31 --- /dev/null +++ b/marked/F/T-REC-F.2-198811-I_PDF-E/raw.md @@ -0,0 +1,53 @@ + + +![ITU logo](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +The logo of the International Telecommunication Union (ITU) features the letters 'ITU' in a bold, sans-serif font, superimposed on a stylized globe with intersecting lines. + +ITU logo + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.2** + +# **OPERATIONS AND QUALITY OF SERVICE TELEGRAPH SERVICES** --- + +## **OPERATIONAL PROVISIONS FOR THE COLLECTION OF TELEGRAM CHARGES** + +**ITU-T Recommendation F.2** + +(Extract from the *Blue Book*) + +--- + +## NOTES + +1 ITU-T Recommendation F.2 (new numbering of Rec. F.42) was published in Fascicle II.4 of the *Blue Book*. This file is an extract from the *Blue Book*. While the presentation and layout of the text might be slightly different from the *Blue Book* version, the contents of the file are identical to the *Blue Book* version and copyright conditions remain unchanged (see below). + +2 In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +## **OPERATIONAL PROVISIONS FOR THE COLLECTION OF TELEGRAM CHARGES** + +### **1 Collection of charges** + +1.1 The charges shall normally be collected from the sender. + +### **2 Telegrams payable by the addressee or a third party – the TA service** + +2.1 Administrations may, either by participating in the transferred account telegraph and telematic service (the TA service) provided for in Recommendations F.41 and D.98, or by special agreement and at the express request of the addressee or other party undertaking payment, admit telegrams of all classes without payment of charges in the origin country. These charges shall be collected from the addressee or other party undertaking payment. + +### **3 Prohibition of rebates** + +3.1 Administrations shall not grant rebates on the rates appearing in the applicable official tariff lists to senders or addressees of telegrams in any form whatsoever (for example, per word, per telegram, by means of discounts, etc.). + +### **4 Errors in collection** + +4.1 If, owing to an error, a telegram is undercharged, the balance necessary to make up the full charge must be collected from the sender, in accordance with the internal regulations of each country. + +4.2 Amounts overcharged in error shall be refunded to the sender in accordance with the internal regulations of each country. \ No newline at end of file diff --git a/marked/F/T-REC-F.23-198811-I_PDF-E/raw.md b/marked/F/T-REC-F.23-198811-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..935bf19214843e1c004abe87a20a9964b2384433 --- /dev/null +++ b/marked/F/T-REC-F.23-198811-I_PDF-E/raw.md @@ -0,0 +1,49 @@ + + +![ITU logo](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +The logo of the International Telecommunication Union (ITU) features the letters 'ITU' in a bold, sans-serif font, superimposed on a stylized globe with latitude and longitude lines. + +ITU logo + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +**F.23** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**OPERATIONS AND QUALITY OF SERVICE +TELEGRAPH SERVICES** + +--- + +**GRADE OF SERVICE FOR LONG - DISTANCE +INTERNATIONAL GENTEX CIRCUITS** + +**ITU-T Recommendation F.23** + +(Extract from the *Blue Book*) + +--- + +# NOTES + +1 ITU-T Recommendation F.23 was published in Fascicle II.4 of the *Blue Book*. This file is an extract from the *Blue Book*. While the presentation and layout of the text might be slightly different from the *Blue Book* version, the contents of the file are identical to the *Blue Book* version and copyright conditions remain unchanged (see below). + +2 In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +# **GRADE OF SERVICE FOR LONG-DISTANCE INTERNATIONAL GENTEX CIRCUITS** + +The CCITT, + +*considering* + +that the main purpose of the gentex service is to ensure that traffic shall be passed without delay, whilst also ensuring a sufficient use of groups of long-distance international circuits intended to carry traffic, + +*unanimously declares* + +that the grade of service corresponding to a loss probability of 1 in 50, as set out in Table 2/F.64, should apply to the groups of long-distance international circuits used in the gentex service. \ No newline at end of file diff --git a/marked/F/T-REC-F.24-198811-I_PDF-E/raw.md b/marked/F/T-REC-F.24-198811-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..1f7098b11a8a7c5101e9ba15ffacabe937f7303d --- /dev/null +++ b/marked/F/T-REC-F.24-198811-I_PDF-E/raw.md @@ -0,0 +1,52 @@ + + +![ITU logo](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +The logo of the International Telecommunication Union (ITU) features the letters 'ITU' in a bold, sans-serif font, superimposed on a stylized globe with latitude and longitude lines. + +ITU logo + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +**F.24** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +# **OPERATIONS AND QUALITY OF SERVICE TELEGRAPH SERVICES** --- + +## **AVERAGE GRADE OF SERVICE FROM COUNTRY TO COUNTRY IN THE GENTEX SERVICE** + +**ITU-T Recommendation F.24** + +(Extract from the *Blue Book*) + +--- + +## NOTES + +1 ITU-T Recommendation F.24 was published in Fascicle II.4 of the *Blue Book*. This file is an extract from the *Blue Book*. While the presentation and layout of the text might be slightly different from the *Blue Book* version, the contents of the file are identical to the *Blue Book* version and copyright conditions remain unchanged (see below). + +2 In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +### **Recommendation F.24** + +## **AVERAGE GRADE OF SERVICE FROM COUNTRY TO COUNTRY IN THE GENTEX SERVICE** + +The CCITT, + +*considering* + +- (a) that Recommendation F.23 gives a recommended grade of service for groups of long-distance international circuits used in the gentex service but; +- (b) that it would be helpful for outgoing countries to be certain that the gentex calls can be put through with a loss probability sufficient to maintain the grade of gentex service without delay working; +- (c) that small offices connected to the gentex network cannot ensure, at the incoming end, a very high grade of service, otherwise their equipment would be uneconomically used; +- (d) that it is sufficient for an outgoing country to be able to count on an average grade of service for all gentex calls to a given incoming country, + +*unanimously declares* + +- (1) that it is helpful to define an average grade of service between countries for gentex calls; +- (2) that this grade of service should be expressed as the proportion of calls that reach the incoming country participating in the gentex service, but that fail to get through to its gentex stations; and +- (3) that this grade of service should not involve more than an average of 1 lost call in 10 during the busy hour on a normal day. Calls routed to an overflow position are considered as successful calls. \ No newline at end of file diff --git a/marked/F/T-REC-F.30-199303-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.30-199303-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d996f27b9f80c5bf6afbf778e209f526aea14595 --- /dev/null +++ b/marked/F/T-REC-F.30-199303-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dbadc5eb897c2a8dfd9329a01063aaaa4360532af77ddf3ef91dcc2fbf58253f +size 8242 diff --git a/marked/F/T-REC-F.30-199303-I_PDF-E/raw.md b/marked/F/T-REC-F.30-199303-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..300e15066f0ce85e0b2eff3fa9f0ece14fe98f39 --- /dev/null +++ b/marked/F/T-REC-F.30-199303-I_PDF-E/raw.md @@ -0,0 +1,84 @@ + + +![ITU logo: A globe with a lightning bolt and the letters ITU.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with a lightning bolt and the letters ITU. + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.30** + +(03/93) + +**OPERATIONS AND QUALITY OF SERVICE +TELEGRAPH SERVICES** + +--- + +**USE OF VARIOUS SEQUENCES OF +COMBINATIONS FOR SPECIAL PURPOSES** + +**ITU-T Recommendation F.30** + +(Previously "CCITT Recommendation") + +--- + +# FOREWORD + +The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of the International Telecommunication Union. The ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Conference (WTSC), which meets every four years, established the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics. + +ITU-T Recommendation F.30 was revised by the ITU-T Study Group I (1988-1993) and was approved by the WTSC (Helsinki, March 1-12, 1993). + +# --- NOTES + +1 As a consequence of a reform process within the International Telecommunication Union (ITU), the CCITT ceased to exist as of 28 February 1993. In its place, the ITU Telecommunication Standardization Sector (ITU-T) was created as of 1 March 1993. Similarly, in this reform process, the CCIR and the IFRB have been replaced by the Radiocommunication Sector. + +In order not to delay publication of this Recommendation, no change has been made in the text to references containing the acronyms "CCITT, CCIR or IFRB" or their associated entities such as Plenary Assembly, Secretariat, etc. Future editions of this Recommendation will contain the proper terminology related to the new ITU structure. + +2 In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +# **USE OF VARIOUS SEQUENCES OF COMBINATIONS FOR SPECIAL PURPOSES** + +*(Melbourne, 1988; revised Helsinki, 1993)* + +The CCITT, + +*unanimously declares that* + +- 1) when it is necessary to provide for switching of telegrams to different routes in semi-automatic or fully automatic telegram retransmission systems, the beginning and ending of each telegram are identified by the insertion of start-of-message and end-of-message signals; +- 2) the start-of-message signal consists of the sequence **ZCZC** in letter case; +- 3) the end-of-message signal consists of the sequence **NNNN**, in letter case; +- 4) the equipment that recognizes the start-of-message and end-of-message signals may be designed to do so by searching only for the sequence of four combinations corresponding to **ZCZC** or **NNNN** respectively (i.e. combinations 26, 3, 26, 3 or 14, 14, 14, 14 in International Telegraph Alphabet No. 2 regardless of whether they are in letter or figure case); +- 5) Table 1 lists other sequences of combinations used for special purposes. + +TABLE 1/F.30 + +| Purpose of sequence | Sequence of combination | Printed position | | Recommendation reference | +|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|----------------------------|-----------------------------------|------------------|--------------------------| +| | | Letter case | Figure case | | +| Connection of reperforator (or equivalent device) | 3 3 3 3 | CCCC | ::::: | S.4 | +| Disconnection of reperforator (or equivalent device) | 6 6 6 6 | FFFF | (Note 1) | S.4 | +| Suppression of delay signals | 8 8 8 8 | HHHH | (Note 1) | S.4, U.22 | +| Ready for test | 11 11 11 11 | KKKK | (((( | R.79, R.79 bis | +| Switching a reader (or equivalent device) into circuit by remote control | 11 12 11 12 | KLKL | ()() | S.4 | +| End of message or enabling of delay signals | 14 14 14 14 | NNNN | ,,,, | F.1, F.31, S.4, U.22 | +| Change of alphabet | 19 19 19 19 | SSSS | '''' | S.15 | +| Error signal | 24 24 24 24 24 | XXXXX | //// | F.1 (C111) | +| Start of message | 26 3 26 3 | ZCZC | +: +: | F.1, F.31 | +| End of input | 26 26 26 26 | ZZZZ | ++++ | F.200 | +| Request for automatic advice of parties cleared prematurely during the broadcast call followed by call cleared | 13 13 13 13
(Note 3) | MMMM
(Note 3) | (Note 2) | S.4, U.44 | +| WRU signal | 30 4 | | | S.6 | +| Operator recall over radiotelegraph circuit | 28 27 27 27 27 | | (Note 6) | U.21 | +| Stop transmission from distant auto terminal | 20 20 20 ...
(Note 4) | T T T ...
(Note 4) | 5 5 5 ... | S.4 | +| Interrogation of distant terminal | 10 ....
(Notes 4 and 5) | J . . .
(Notes 4 and 5) | | S.4, S.22 | +| Follow-on call | 12 12 12 12 | L L L L | ))) | S.4, U.43 | +| NOTES
1 As noted in C5/F.1, the figure case of combinations 6, 7 and 8 is available for the internal service of each Administration.
2 As noted in Recommendation U.44 this sequence is to be recognized in letter case only.
3 A minimum of 4 Ms would clear a telex broadcast call. The usage of a 5th or more Ms is a national matter.
4 Repeated if necessary.
5 Function only applicable in figure case.
6 See also Table 2/S.1. | | | | | \ No newline at end of file diff --git a/marked/F/T-REC-F.31-198811-I_PDF-E/raw.md b/marked/F/T-REC-F.31-198811-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..791c21f48b331157189afdcb5bea78f379ee0211 --- /dev/null +++ b/marked/F/T-REC-F.31-198811-I_PDF-E/raw.md @@ -0,0 +1,584 @@ + + +![ITU logo](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +The logo of the International Telecommunication Union (ITU) features a globe with a lightning bolt superimposed on it, and the letters 'ITU' in a bold, sans-serif font. + +ITU logo + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +**F.31** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**OPERATIONS AND QUALITY OF SERVICE +TELEGRAPH SERVICES** + +--- + +**TELEGRAM RETRANSMISSION SYSTEM** + +**ITU-T Recommendation F.31** + +(Extract from the *Blue Book*) + +--- + +# NOTES + +1 ITU-T Recommendation F.31 was published in Fascicle II.4 of the *Blue Book*. This file is an extract from the *Blue Book*. While the presentation and layout of the text might be slightly different from the *Blue Book* version, the contents of the file are identical to the *Blue Book* version and copyright conditions remain unchanged (see below). + +2 In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +## **TELEGRAM RETRANSMISSION SYSTEM** + +(a) For the routing of telegram traffic, Administrations can use the *Telegram Retransmission System*. This system comprises a network of interconnected telegram retransmission centres, which carry out the switching and retransmission functions. The indications needed to route the telegram are included with the telegram when it enters the system and are retransmitted with the telegram from one centre to another. + +(b) The offices where telegrams enter or leave the telegram retransmission system are linked to at least one retransmission centre; such offices are described as *linked* offices. In the case of a particular telegram, the linked office through which the telegram enters the system is known as the linked entry office; the linked office through which the telegram leaves the system is known as the linked exit office. + +(c) The present Recommendation has been drawn up for fully automatic working, but may be used for semi-automatic and manual working. + +(d) To facilitate world-wide operation of the telegram retransmission system, to simplify the transfer of telegrams between the retransmission network and other networks and to enable the switching and accounting equipment needed for the retransmission centres to be designed, the CCITT, + +*unanimously declares that* + +**1** Each telegram must be treated independently, even if several telegrams for the same destination arrive in series at a linked entry office. + +**2** The format to be used in preparing the telegram for transmission shall be as follows: + +**2.1** The heading of the telegram shall start with the *numbering line* preceded by a carriage-return, a line-feed and a letter-shift. The numbering line consists of: + +**2.1.1** *The start-of-message signal (SOM)* + +In accordance with Recommendation F.30 the start-of-message signal (SOM) is made up by the sequence of combinations Nos. 26, 3, 26, 3 of International Telegraph Alphabet No. 2 (**ZCZC**) followed by: + +**2.1.2** *The channel sequence number* + +Telegrams transmitted over a channel shall be numbered according to a series of numbers for each channel. The channel sequence number will therefore be composed of three letters characteristic of the channel used (channel indicator) followed by a number showing the order of this telegram in the series sent over this channel. The channel sequence numbers shall be in sequence from **001** to **999** and change automatically from **999** to **001** at the end of a numbering cycle. + +When the telegram passes from one channel to another in a network, each new channel sequence number shall be inserted immediately after the start-of-message signal. The channel sequence numbers will appear in the numbering line of the telegram in the opposite order to that in which the telegram passed through the network. + +The channel sequence number shall be transmitted as: + +- space; +- three letters constituting the channel indicator; +- figure-shift; +- three figures constituting the number in the series on the channel. + +Service advices will be numbered in the channel series unless agreed otherwise by the Administrations concerned. The channel sequence number(s) is followed by: + +### 2.1.3 *The telegram identification group (TIG)* + +The telegram identification group enables the office of origin to recognize the telegram. This group must not exceed 15 printing characters, which may be composed of any combination of letters and/or figures. In the special case of a telegram from the gentex network to the telegram retransmission system, the telegram identification group will be the only reference in the numbering line and will consist of two letters identifying the originating country or a particular network therein (in accordance with Recommendation F.96), the call number of the gentex office followed by one or two letters identifying the gentex position and the serial number of the telegram which that position sends. + +The telegram identification group shall be transmitted as: + +- space; +- letter-shift (if necessary); +- not more than 15 printing characters; + +followed by: + +### 2.1.4 *End of line* + +- carriage-return; +- line-feed; + +which will mark the end of the numbering line, followed by: + +2.2 The second line of the heading, the *pilot line*, which consists of: + +#### 2.2.1 *The destination indicator* + +This indicator is extracted from the *List of Destination Indicators and Telex Network Identification Codes* [1], and consists of four letters; the first two characterize the destination country (or a particular network in the destination country) and the following two letters characterize an office of that country (see also § 4 below). + +The destination indicator shall be transmitted as: + +- letter-shift; +- four letters; + +followed by: + +#### 2.2.2 *The priority and tariff indicator* + +This indicator consists of two letters chosen in such a way that the priority letters of the indicator differ from each other by at least two unit elements and the same letter does not appear twice in the same indicator, thus reducing the possibilities of error. + +2.2.2.1 The first letter will designate the priority of transmission according to the following table: + +- A** – Telegrams relating to safety of life (**SVH**) +- B** – Telegrams relating to the application of the United Nations Charter (**ETATPRIORITE**) + - Government telegrams with priority (**ETATPRIORITE**) + - Meteorological telegrams (**OBS**) + - Ordinary private telegrams with urgent transmission and delivery (**URGENT**) + - **RCT** telegrams with urgent transmission and delivery (**URGENT RCT**) + - Money orders and postal cheque telegrams with urgent transmission and delivery (**URGENT POSTFIN**) +- C** – Government telegrams (**ETAT**) + - Service telegrams (**A**) + - Service advices (**A**) + +- Ordinary private telegrams +- **RCT** telegrams (**RCT**) +- Money orders and postal cheque telegrams (**POSTFIN**) +- Telemessages +- H** – Letter telegrams (including government letter telegrams) (**LT, LTF**) + +This table takes account of the order of transmission given in Recommendation F.1, A136 to A144. + +2.2.2.2 A telegram that has been abnormally delayed can be upgraded to a higher priority group. Such promotion can be effected only in the linked entry office; in this case, a category **H** telegram could be marked **C** and a category **C** telegram marked **B**, but a telegram with lower priority than **A** can never be promoted to category **A**. + +2.2.2.3 The *second letter* will designate the tariff class to be used according to the following table: + +- F** Radiotelegrams routed via a foreign land station; +- J**2 **POSTFIN** telegrams1); +- K** Government telegrams where preferential rate is to be charged; +- L** Letter telegrams; +- M** Meteorological telegrams; +- N** Non-chargeable telegrams; +- O** Ordinary private telegrams; +- Q**2 Telegrams involving special accounting features1); +- R** **RCT** telegrams; +- T** Telemessages +- U** Telegrams (other than **RCT** telegrams) with urgent transmission and delivery; +- V**2 Divided telegrams1); +- X** For use by transit offices when inserting the pilot line on transit telegrams; +- Y** Government full-rate telegrams; +- Z** Government letter-rate telegrams. + +2.2.2.4 Letters **D, E, G, I, P, S** and **W** are available for assignment when required.2) + +The priority and tariff indicator shall be transmitted as: + +- space; +- two letters; + +followed by: + +### 2.2.3 *The origin indicator* + +The origin indicator consists of four letters; the first two will be the same as those used in the destination indicators for the originating Administration or network; the third and fourth will represent: + +- a) a particular city or town within that country or network; or +- b) the office or department to which service correspondence regarding the telegram should be addressed. + +1) Where there is a choice tariff indicator the one marked with a 2 takes precedence. + +2) Administrations may agree mutually to accept additional tariff indicators. + +Administrations may select the last two letters to satisfy the requirements of their internal organization but where there is more than one entry point to a country or network from any station the choice of letter combinations must be negotiated with that station where they differ from the destination indicator for the office in question. Administrations that take advantage of this facility should notify the CCITT of the origin indicators in use in their system. + +The origin indicator shall be transmitted as: + +- space; +- four letters; + +followed by: + +### 2.2.4 *The number of chargeable words* + +The number of chargeable words shall be transmitted in the form of a three-figure number (for example **009**) using the number of chargeable words shown in the preamble line for the telegram in question. For a non-chargeable telegram the number **000** will be shown. + +The number of chargeable words shall be transmitted as: + +- space; +- figure-shift; +- three figures; + +followed by: + +2.2.5 A *customer identification group* (optional): This group characterizes the customer for accounting purposes and can be composed of any combination of letters and/or figures. It will not be transmitted beyond the first retransmission centre. (See also Recommendation F.1 C108.) + +The customer identification group shall be transmitted as: + +- space; +- the customer identification group; + +followed by: + +### 2.2.6 *End of line* + +- carriage-return; +- line-feed; + +which will mark the end of the second line (pilot line), followed by: + +2.3 The third line of the heading which is the *preamble line* and which shall be transmitted as shown in Recommendation F.1, A140 to A144, including any service instructions, followed by: + +- carriage-return; +- three line-feeds; + +followed by: + +### 2.4 *The address part*3) + +#### 2.4.1 *The service indication line* (where required) + +Any service indications should be placed on a separate line immediately before the address. Each indication shall be transmitted as one word. If there are several indications each will be separated by a space. + +The service indication line is transmitted as: + +- letter-shift (if necessary); + +--- + +3) For postal addresses, the address part (including service indication line) should be composed of a maximum of six lines preferably not exceeding 30 printing/ spacing characters. An address part of five lines of up to 43 printing/spacing characters each be accepted. + +- the service indications; +- carriage-return; +- line-feed; + +followed by: + +#### 2.4.2 *The address lines* + +The address of a telegram shall be transmitted as: + +- letter-shift (if necessary); +- the address; +- the name of the office of destination isolated in the last line (see Recommendation F.1, A39 to A100); +- carriage-return; +- three line-feeds; + +followed by: + +### 2.5 *The text part* + +#### 2.5.1 *The text shall be transmitted as:* + +- letter-shift (if necessary); +- text; +- carriage-return; +- line-feed (if signature present) or – ten line-feeds (if no signature present). + +#### 2.5.2 *The signature shall be transmitted as:* + +- minimum five spaces; +- signature; +- carriage-return; +- ten line-feeds. + +### 2.6 *The end-of-message signal (EOM)* + +In accordance with Recommendation F.30 the *end-of-message signal (EOM)* shall be transmitted as: + +- letter-shift; +- NNNN; +- ten letter-shifts (except where not required). + +### 2.7 *Format examples* + +Annex A gives an example of the format of a typical ordinary private telegram showing all the functional signals. Formats of other telegrams will be found in Recommendation F.1. + +## 3 **Choice of destination indicator** + +3.1 The linked entry office of the telegram retransmission network selects the destination indicator to be entered in the pilot line of a telegram. + +3.2 Apart from the exceptions given in §§ 4.3, 4.4 and 4.5 the destination indicator shall be selected from the *List of Destination Indicators and Telex Network Identification Codes* [1] (see Recommendation F.96) according to the rules in Table I/F.31. + +3.3 Administrations wishing to do so may use the destination indicator *all others* (or one of the *all others* indicators) of the country of the office concerned for the telegrams to an office appearing in the *List of Indicators* [1], but not directly connected to the telegram retransmission network. + +3.4 By private agreement between the Administrations concerned, the destination indicator to be used for each town of a country may be selected from any of the destination indicators for the country according to the internal routing of the country. + +3.5 For return service advices, etc., relating to a telegram, the destination indicator shall be the origin indicator given in the telegram. + +## 4 Routing + +4.1 Within a telegram retransmission centre, a telegram shall be directed to the following channel in the chain of connections by the destination indicator shown on the pilot line of the telegram, in accordance with the traffic circulation scheme of the telegram retransmission centre. + +4.2 If the appropriate subsequent route for the telegram is not connected with the telegram retransmission equipment, the destination indicator shall direct the telegram to a place in the centre where it can be handled and reforwarded. + +TABLE 1/F.31 + +### Rules for selecting the destination indicator + +| | A
A single network in destination country | B | C | +|------------------------------------------------------------------------------------------------------------------------------|----------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------| +| | | Several networks in destination country | | +| | | Telegram showing routing | Telegram not showing routing | +| 1. Destination town directly connected with telegram retransmission system, or to which a destination indicator is allocated | Use destination indicator given in the List against the town concerned | Use destination indicator with the two letters for the network in the List followed by the two letters for the town in the List | Use destination indicator with the two letters for unrouted for the country concerned followed by the two letters for the town in the List | +| 2. Other destinations | Use destination indicator all others given in the List for destination country | Use destination indicator with the two letters for the network in the List , followed by the two letters corresponding to all others for the country concerned | Use destination indicator with the two letters for unrouted for the country concerned, followed by two letters for all others for the country | + +## 5 Tolerances as to the format and checking of format + +5.1 Switching equipment shall tolerate: + +- a space and letter-shift transposition or a space and figure-shift transposition in a sequence normally prescribed as having to be a space followed by a shift; +- the repetition of a function signal, except for the space between the destination indicator and the priority indicator; +- the reception of characters between successive end-of-message signals and start-of-message signals (for example: spurious signals, letter-shifts or other functional signals) without affecting the proper functioning of the equipment. (Any transmission from the sending end of a channel between an end-of-message signal and the subsequent start-of-message signal should be limited to those characters that have a function at the receiving end of the channel.) + +5.2 If a repetition or a transposition in the sequence carriage-return, line-feed, letter-shift separating the numbering line from the pilot line cannot be tolerated by the switching equipment, the telegram affected by such a defect will be directed towards a manual service position. + +5.3 Any deviation from the format that might be recognized by a centre and that goes beyond the acceptable tolerances as given herein shall as far as possible be corrected before retransmission to another switching centre. + +5.4 If the operator in preparing a telegram detects an error in the set-up of the numbering line or the pilot line, he must destroy the part already set up and start preparation of his telegram again. But if the transmission of these two lines has already started, the operator shall send the code expression **ANUL** space **ANUL** followed by ten line-feeds and the end-of-message signal. Any telegram so terminated shall not, if possible, be retransmitted by the first switching equipment to receive it. + +## 6 Protection against loss of telegrams + +### 6.1 *Transmission* + +6.1.1 Whenever a retransmission is made, a channel sequence number is sent, showing the channel used for retransmission and the sequence of the telegram on that channel. + +6.1.2 During the retransmission, a telegram interrupted by the code expression **ANUL ANUL** followed by the EOM sequence is considered as not transmitted. This telegram must be retransmitted with its original channel sequence number without other indication. + +6.1.3 A check must be made in every retransmission centre, either automatically or by an operator, to prove that every telegram received is retransmitted. + +### 6.2 *Reception* + +6.2.1 A check is made to verify the regular sequence of the channel numbers of telegrams received on each incoming channel. Should there be any irregularity, an alarm will warn the supervisory staff. + +6.2.2 During reception, a telegram interrupted by the code expression **ANUL ANUL** followed by the EOM sequence is considered as not received. This telegram must be received again in a complete way with its original channel sequence number without other indication. + +### 6.3 *Mutilation* + +6.3.1 If there is a mutilation of the text of a telegram, the incident will be dealt with by end-to-end servicing since the alteration will be noted in practice only at the linked exit office. + +6.3.2 If there is a mutilation of a channel sequence number, which may be noted automatically when entering an office, a request for re-run, by service advice, as applicable, indicating the queried sequence number, will be sent to the preceding office on the channel chain, which will then re-run the telegram in question. + +6.3.3 For enquiries about a telegram whose telegram identification group has been mutilated, investigation will be made by going back along the route from office to office and by identifying the telegram by means of operating information. + +## 7 Starting of motors + +7.1 In general the motors of terminal equipment will always be running, at least on intercontinental circuits. + +7.2 However, the two Administrations concerned may agree to operate a channel with terminal equipment fitted with time-delay devices to start and stop the motor. They will agree mutually on arrangements for controlling the start of the motor. The provisions of Recommendation S.7 seem to be the most applicable. + +## 8 Use of tape-printing equipment + +8.1 Recommendation S.5 should be followed with regard to the use of tape-printing equipment in the telegram retransmission system. + +## 9 Offices operated semi-automatically or manually + +9.1 For offices that are connected to fully automatic systems, Administrations should follow as closely as possible the format recommended in this Recommendation. + +9.2 Other offices should also follow the recommended format on any telegram that will enter the telegram retransmission system on a second or subsequent link. + +## 10 Automatic service procedures + +### 10.1 *Re-runs and Put-backs* + +10.1.1 A re-run is the repetition, between two directly connected centres of offices, of one or more telegrams that have previously been sent. + +10.1.2 A put-back involves stopping a transmission on a channel, recommencing at a particular telegram previously transmitted and continuing from there. + +10.1.3 Re-runs and Put-backs should only consist of information that was initially transmitted. Request for re-runs will only occur between the directly connected centres concerned. Telegrams so repeated shall be transmitted under their original channel sequence numbers only. + +10.1.4 If a re-run of any telegram is given automatically by a telegram retransmission centre, such re-run should be restricted to that telegram retransmission centre or linked office to which the traffic was initially transmitted, or a centre or linked office to which the traffic has been diverted. + +### 10.2 *Automatic service notes* + +10.2.1 Automatic service notes are designed to initiate an automatic action at a telegram retransmission centre. They may or may not be generated automatically, but they shall be numbered in the normal sequence. + +10.2.2 Automatic service notes should only be sent from stations that work directly to the distant automatic retransmission centre at which the action is to be taken. + +10.2.3 Where automatic action is required, the numbering line may also be composed of: + +**ZCZC ABC000 or ZCZC XQ** + +10.2.4 Automatic service notes shall contain a pilot line with a special destination indicator composed of the country code followed by **ZZ**; the **B** priority indicator must be used. + +10.2.5 A preamble line may appear in these notes, but it may be in abbreviated form, e.g. a date and time group only. + +10.2.6 The text must commence with a four-letter code designating the action required: + +**RRUN** for re-run, + +**PUTB** for put-back (retransmission of all the telegrams, starting from a given number), + +**BKBK** for break, + +**SITU** for situation, + +**GAGA** for go ahead (after an interruption in the traffic). + +10.2.7 Automatic service notes and responses shall conform to the layout in Annex A. + +### 10.3 *Automatic re-runs and put-backs (commands RRUN and PUTB)* + +10.3.1 Where a centre can automatically generate re-runs and put-backs, these should normally be available on request from another directly connected centre during the 24 hours following the original transmission of the telegrams in question. + +10.3.2 Automatic re-run of a maximum of ten telegrams may be requested in a single service note. + +10.3.3 Put-back of a maximum of 50 telegrams may be requested in a single service note (**PUTB**). + +10.3.4 The telegrams referred to in a request for automatic re-run or put-back shall all bear the same original channel indicator. + +10.3.5 Automatic re-runs and put-backs should take place over the channel used for the transmission of the original telegram(s). However, they may also be sent over the original route, but not necessarily over the original channel, if the two Administrations concerned have agreed to this course in advance. + +10.3.6 Where automatic answer is not given immediately, the reply must be made manually subsequently. + +10.3.7 The same format should also be used by centres that have access to automatic repetition facilities at a distant centre but that have to prepare failure notes manually. + +### 10.4 *Automatic stoppages and re-starts of transmission (commands BKBK and GAGA)* + +10.4.1 Automatic stoppage and re-start of transmission should be available on request from any directly connected centre. + +10.4.2 Both facilities should be possible, either stopping or re-starting on one specified channel or on all channels of a relation. + +10.4.3 In the case of a general break command from a telegram retransmission centre, a service note including a numbering line consisting only of **ZCZC XQ** and having a special destination indicator **XQXQ** in the pilot line shall be used to notify all connected Administrations. + +### 10.5 *Automatic situation requests (command SITU)* + +10.5.1 Where a Centre can automatically provide the channel sequence numbers of the last telegram received and the last telegram sent, this should normally be available on request from another directly connected Centre. + +10.5.2 The information shall contain the time to which the situation refers. + +10.5.3 The information given should include the situation for all channels on the same route. + +### 10.6 *Automatic continuity checks (LRLS)* + +10.6.1 Sixty minutes after the last message received or sent on a circuit, a continuity check message shall be sent. + +10.6.2 The format of this message is identical to the reply (**LRLS**) to a situation request. + +# ANNEX A + +(to Recommendation F.31) + +## **Format examples** + +### A.1 *Ordinary private telegram with service indication (showing all functional signals)* + +``` +<≡ +↓ ZCZC→AOE↑262→↓LDB↑814→↓PLD↑606<≡ +↓ AASD→CO→GBLD→↑018<≡ +↓ LONDON↑/↓LD→↑18/16→22→1430<≡≡≡ + +↓ LX<≡ +HARRIS<≡ +↑ 2462→↓SOUTHERNHIGHWAY<≡ +SYDNEY<≡≡≡ + +CONGRATULATIONS→ON→YOUR→PROMOTION→AND<≡ +BEST→WISHES→FOR→THE→FUTURE<≡ +→→→→→JOHN<≡≡≡≡≡≡≡≡≡≡ + +↓ NNNN↓↓↓↓↓↓↓↓↓ +``` + +### A.2.1 *Service note requesting automatic re-run of one message* + +**ZCZC ABC000 <= (or ZCZC XQ)** +**DPZZ BN FRZZ 000 <= (ZZ if request is generated automatically, XQ otherwise)** +**PARIS 21 0926 <= (optional or abbreviated in case of request only)** + +**RRUN BCA165 <=** + +NNNN + +#### A.2.2 *Service note requesting automatic re-run of several messages in a sequence (maximum ten messages)* + +**ZCZC XQ <=** +**DPZZ BN FRZZ 000 <=** +**PARIS 21 0926 <=** + +**RRUN BCA286/293 <=** + +NNNN + +#### A.2.3 *Service note requesting automatic re-run of several non-consecutive messages (maximum five)* + +3 first lines identical <= + +**RRUN BCA123 141 162 173 <=** + +NNNN + +### A.3 *Service note requesting put-backs (maximum 50 messages)* + +3 first lines identical <= + +**PUTB BCA123 <=** + +NNNN + +### A.4.1 *Service note with break command on a specific circuit* + +3 first lines identical <= + +**BKBK BCA <=** + +NNNN + +#### A.4.2 *Service note with break command on all circuits in a relation* + +3 first lines identical **<= = =** + +**BKBK <= = = = = = = = = =** + +NNNN + +#### A.4.3 *Service note with break command consecutive to a system failure* + +**ZCZC XQ <=** + +**XQQX BN FRXQ 000 <=** + +Preamble line optional **<= = =** + +**BKBK <= = = = = = = = = =** + +NNNN + +### A.5.1 *Service advice to restart the traffic on a specific circuit* + +**ZCZC ABC000 <=** + +**DPZZ BN FRZZ 000 <=** + +Preamble line optional **<= = =** + +**GAGA BCA <= = = = = = = = = =** + +NNNN + +#### A.5.2 *Service note to restart the traffic on all circuits of a relation* + +**ZCZC XQ <=** + +**DPZZ BN FRZZ 000 <=** + +Preamble line optional **<= = =** + +**GAGA <= = = = = = = = = =** + +NNNN + +#### A.6.1 *Service note for a situation request* + +**ZCZC LPA000 <=** + +**FRZZ BN GXZZ 000 <=** + +**LONDON 9 1027 <= = = (optional)** + +**SITU <= = = = = = = = = =** + +NNNN + +#### A.6.2 *Reply format to the § A.6.1 request* + +**ZCZC PLA000 <=** + +**GXZZ BN FRZZ 000 <=** + +**PARIS 9 1031 <= <= <= (complete preamble compulsory)** + +**LRLS LPA074 PLA444 <=** + +**LPB570 PLB009 <=** + +**MISSING LPA040/043 <=** + +**LPB551 554 560 <= <= <= <= <= <= <= <= <= <=** + +**NNNN** + +# **References** + +- [1] *List of indicators for the telegram retransmission system and the telex network identification codes*, ITU, Geneva. +- [2] *List of telegraph offices open for international service*, ITU, Geneva. \ No newline at end of file diff --git a/marked/F/T-REC-F.32-199510-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.32-199510-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7ef7a100d54c2e021e5cf616aaff6ebdd7f862d4 --- /dev/null +++ b/marked/F/T-REC-F.32-199510-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6432d4de673b586a292f985ebffe1cce156a6e2045a04e688100dd93cf5f7d70 +size 8293 diff --git a/marked/F/T-REC-F.32-199510-I_PDF-E/fc46871d72c65d3381d9201646d23439_img.jpg b/marked/F/T-REC-F.32-199510-I_PDF-E/fc46871d72c65d3381d9201646d23439_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..486c076f4b1574afeba4c1f6934aae87be07d4cd --- /dev/null +++ b/marked/F/T-REC-F.32-199510-I_PDF-E/fc46871d72c65d3381d9201646d23439_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ca7df535bec29180f07ce2774c57dc49ccff1eefcff53625da2a7bf2df18d22c +size 134110 diff --git a/marked/F/T-REC-F.32-199510-I_PDF-E/raw.md b/marked/F/T-REC-F.32-199510-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..9db236852cb9ecbeb62c84e71df8d4ce87361c93 --- /dev/null +++ b/marked/F/T-REC-F.32-199510-I_PDF-E/raw.md @@ -0,0 +1,199 @@ + + +![ITU logo: A globe with the letters ITU and a lightning bolt.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with the letters ITU and a lightning bolt. + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +**F.32** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +(10/95) + +**OPERATIONS AND QUALITY OF SERVICE +TELEGRAPH SERVICES** + +--- + +**TELEGRAM DESTINATION INDICATORS** + +**ITU-T Recommendation F.32** + +(Previously "CCITT Recommendation") + +--- + +# FOREWORD + +The ITU-T (Telecommunication Standardization Sector) is a permanent organ of the International Telecommunication Union (ITU). The ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Conference (WTSC), which meets every four years, establishes the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics. + +The approval of Recommendations by the Members of the ITU-T is covered by the procedure laid down in WTSC Resolution No. 1 (Helsinki, March 1-12, 1993). + +ITU-T Recommendation F.32 was prepared by ITU-T Study Group 1 (1993-1996) and was approved under the WTSC Resolution No. 1 procedure on the 3rd of October 1995. + +--- + +# NOTE + +In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +# CONTENTS + +| | Page | +|-------------------------------------------------|-------------| +| 1 Background and scope ..... | 1 | +| 2 References and terminology ..... | 1 | +| 3 Code structure ..... | 1 | +| 4 Allocation of telegram network codes..... | 2 | +| 5 Particular combinations for office codes..... | 4 | +| 6 Allocation of office codes ..... | 4 | +| 7 Registration and publication..... | 4 | + +# **SUMMARY** + +The operation of the “Telegram Retransmission System”, described in Recommendation F.31, depends on the use of a standard format in telegrams. The format contains routing and other operational information in a heading part. This includes a four-letter “destination indicator”, which is one of the key elements used for routing a telegram at switching centres in the origin country, and in any transit country(ies). They are also used for routing in the destination country. + +This Recommendation lays down the general principles applying to the selection of destination indicators, the allocation of the indicators’ four letters, the application of analogous two- and four-letter codes for certain other purposes in operating the telegram service, procedures for changing or adding to code allocations and the timely promulgation of the information on codes to Recognized Operating Agencies (ROAs). + +# **TELEGRAM DESTINATION INDICATORS** + +*(Geneva, 1995)* + +# **1 Background and scope** + +**1.1** The operation of the “Telegram Retransmission System”, described in Recommendation F.31, depends on the use of a standard format in telegrams. The format contains routing and other operational information in a heading part. This includes a four-letter “destination indicator”, which is one of the key elements used for routing a telegram at switching centres in the origin country, and in any transit country(ies). They are also used for routing in the destination country. Depending on the arrangements in individual centres, other elements used for automatic or manual routing include the origin indicator, the outgoing or incoming channel indicator and the office of destination. These elements are also used to provide the basis for the billing and accounting processes. For this and other operational reasons, the “F.31” format or its component parts are used when handling telegrams other than via the Telegram Retransmission System. Destination indicators have wide applicability. + +**1.2** This Recommendation lays down the general principles applying to the selection of destination indicators, the allocation of the indicators’ four letters, the application of analogous two- and four-letter codes for certain other purposes in operating the telegram service, procedures for changing or adding to code allocations and the timely promulgation of the information on codes to Recognized Operating Agencies (ROAs). + +## **2 References and terminology** + +**2.1** The Recommendations relevant to the allocation of codes, the use of destination indicators in the telegram service and their relation with other codes used in the international public telegram and telex services are: + +- ITU-T Recommendation C.2 (1993), *Collection and publication of official service information*. +- CCITT Recommendation F.31 (1988), *Telegram retransmission system*. +- CCITT Recommendation F.68 (1988), *Establishment of the automatic intercontinental telex network*. +- CCITT Recommendation F.69 (1988), *Plan for telex destination codes*. + +**2.2** Various technical and operational arrangements may be used to support the international public telegram service. The service itself and the main operational arrangements are defined in Recommendation F.1. + +**2.3** In accordance with the provisions in Recommendation F.31, the terms used in this Recommendation to identify the two-letter parts of destination indicators are “Telegram Network Code” and “Office Code” (see clause 3). + +## **3 Code structure** + +**3.1** Each destination indicator consists of four letters taken from the Latin alphabet. + +**3.2** The first two letters (the “telegram network code”) fulfil a function analogous to a telex destination code (see Recommendation F.69). The two letters characterize either: + +- a) a particular destination country or geographical area; or +- b) a particular network in a destination country or geographical area, for cases where the origin office wishes to route a telegram to a particular network; or + +- c) where the office of origin has no special preference for the destination network, an “unrouted” combination, which allows a gateway or transit centre to select the actual network to be used, e.g. on a quota basis. + +**3.3** In accordance with Recommendation F.68, where a ROA has been, or is to be, allocated a two-letter telex network identification code, this code should be the same as its telegram network code described in 3.2 above. + +**3.4** The third and fourth letters (the “office code”) characterize, principally for routing purposes, either: + +- a) the destination office shown in the address part of the telegram; or +- b) a centre that serves a number of destination offices in a region or district; or +- c) an “all others” indicator to cover those destinations not covered by a) and b) above; or +- d) certain applications related to the operation of the telegram service (see clause 4 below); or +- e) a single code covering all destinations [except for the applications in d) above]. + +**3.5** As far as possible the whole of the four-letter destination indicators should be selected such that any indicator differs in at least two letters from any other indicator. + +**3.6** The same four-letter code structure is used for destination and origin indicators (see Recommendation F.31). + +## **4 Allocation of telegram network codes** + +**4.1** Any administration of an ROA wishing to bring a new network into service (or to change the designation of an existing network, country or geographical area) should apply to the Director of the TSB for the assignment of a two-letter network code (see 3.2 above). In its request it may indicate the available code preferred, but see 3.3 above. + +**4.2** In all cases such requests (and any request for a change in the telegram network code of an existing network) shall be referred to the Chairman of ITU-T Study Group 1 for guidance on technical and operational issues, taking account of possible future requirements. + +**4.3** It is the responsibility of the TSB: + +- to carry out the necessary discussions with the applicant Administration and/or its ROA; and +- to ensure that the Chairman of Study Group 1 is satisfied that any technical and operational requirements have been covered before allocation of a new telegram network code. + +**4.4** Discontinued telegram network codes shall not be reallocated for a period of at least two years. + +**4.5** Figure 1 illustrates the procedure specific to telegram network codes. + +**4.6** Where there are competing ROAs, an “unrouted” telegram network code must be allocated in addition to a code for each ROA’s network [see 3.2 c) above]. + +**4.7** The letter **X** should not be allocated as the first letter of a telegram network code for any specific destination country. The destination indicator **XQXQ** is reserved for use in emergency broadcast messages (see clause 10/F.31). + +**4.8** The procedure in 4.1 to 4.5 above shall also be applied for two-letter indicators used to designate international organizations in the preamble line of telegrams pursuant to Recommendation F.1, A218. + +![Flowchart for the allocation and publication of telegram network codes. The process starts with a request received in TSB, followed by discussions, passing details to Study Group 1, seeking expert guidance, identifying issues, replying to TSB, reaching an agreement, informing the applicant, and finally publishing the updated list.](fc46871d72c65d3381d9201646d23439_img.jpg) + +``` + +graph TD + A[Request received in TSB for allocation/change of code(s)] --> B[Details passed by TSB to Study Group 1 Chairmana)] + A --- C[TSB/Applicant discussions] + B --> D[Study Group 1 Chairmana) seeks guidance from Study Group 1 expert group(s) on proposalb)] + D --> E{Have any technical/operational issues been identified by expert group?} + E -- Yes --> F[Study Group 1 Chairmana) reverts to TSB with guidance on resolution of any identified operational or technical issues] + E -- No --> G[Study Group 1 Chairmana) replies accordingly to the TSB] + F -- "Normally 15 working days" --> C + F --> H[TSB/Applicant discussions] + G --> I{Agreement?} + H --> I + I -- Yes --> J[TSB informs applicant and announces code allocation in Operational Bulletin] + I -- No --> C + J --> K[TSB to publish updated list of allocated and spare codes periodically in Operational Bulletin] + J --> L[Relevant Study Group 1 expert group to inform the next meeting of Study Group 1 (Working Party) on work] + +``` + +T0104920-95/d01 + +Flowchart for the allocation and publication of telegram network codes. The process starts with a request received in TSB, followed by discussions, passing details to Study Group 1, seeking expert guidance, identifying issues, replying to TSB, reaching an agreement, informing the applicant, and finally publishing the updated list. + +a) Or Chairman's delegated representative. + +b) To be done in consultation with expert group nominated by Study Group 1 for such tasks. + +FIGURE 1/F.32 +Scheme for the allocation and publication of telegram network codes + +## 5 Particular combinations for office codes + +- 5.1 The second letter of any “all others” office code should be **X**. +- 5.2 Where a single office code is used to cover all destinations in a given network, it should be **XX**. +- 5.3 The combinations **SV**, **MV**, **XQ** and **YQ** are reserved for the segregation of particular types of messages (service advices, service notes) at gateway centres or major offices. (See 2.2.3/F.31 and 3.5/F.31 on the use of such combinations in origin and destination indicators for return service advices.) +- 5.4 The combination **ZZ** is strictly reserved for automatic service notes, which are designed to trigger an automatic reaction in a connected telegram retransmission centre. (See clause 10/F.31.) +- 5.5 The combinations **RQ**, **BQ**, **XQ**, if used at all, should only be associated with service notes (i.e. messages between connected switching centres). + +## 6 Allocation of office codes + +- 6.1 The office codes (i.e. the third and fourth letters in a destination indicator) allocated for each network code should be kept to a minimum number in each network. (This does not exclude the use of additional unregistered office codes for national purposes.) Keeping the number of office codes to a minimum eases the task of operators in distant countries who have to select the appropriate code in their outgoing telegrams. A very short list is particularly helpful for offices handling a small number of outgoing international telegrams. + +Where the destination network is capable of carrying out the national routing without an office code (e.g. by analysing the address part of the telegram as well as the pilot line), a single office code (**XX**) is preferred. + +- 6.2 Administrations should coordinate the requirements of their individual ROAs, where applicable, for any changes to the office codes within a given country or geographical area. Any proposed changes should respect the guidelines in clause 5 above. +- 6.3 The Administration, or a ROA if so authorized by the Administration, should then advise the Director of the TSB of the proposed changes. +- 6.4 Where ROAs use special office codes within their origin indicators in accordance with 2.2.3/F.31, any proposed changes to these shall be handled in the same manner as for office codes within destination indicators. +- 6.5 The Director of the TSB should consult the Chairman of Study Group 1 for guidance on any technical and operational issues when proposed changes to office codes: + - do not conform with the requirements in this Recommendation; and/or + - propose a major re-arrangement of a network's office codes; or + - may impinge on possible future requirements and/or known or anticipated developments in other networks. + +## 7 Registration and publication + +- 7.1 All additions and changes to destination codes that are accepted in accordance with clauses 4 and 6 above shall be entered into a register maintained by the TSB. All special origin indicators (6.4 above) used internationally shall also be included. +- 7.2 The fact that changes are being made in a given network or networks shall also be published in the next available Operation Bulletin, together with dates of entry into effect. The Operational Bulletin shall also give the details of any new or changed telegram network codes. However, the Administration or ROA(s) initiating the changes shall circulate details of the approved changes to office codes to all ROAs needing this operational information. + +**7.3** In accordance with Recommendation F.69 (1994), a list of telex network identification codes is published periodically in the Operational Bulletin. For convenience, the two-letter telegram network codes corresponding to one-letter telex network identification codes should be appended to this list. + +**7.4** The date of entry into effect of changes (additions and deletions) in destination indicators should normally be the first day of the third month following publication. However a shorter delay, but normally not less than a month, may be acceptable in the case of changes in office codes. + +**7.5** The TSB shall publish a list of the current four-letter destination indicators in the Operational Bulletin at approximately three-yearly intervals, or as the number of changes may justify it. The current list on the TSB's database should also be readily accessible to Administrations and ROAs. \ No newline at end of file diff --git a/marked/F/T-REC-F.353-198811-I_PDF-E/raw.md b/marked/F/T-REC-F.353-198811-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..8a49a87a69de55acb6cb4adf687acf440555c82f --- /dev/null +++ b/marked/F/T-REC-F.353-198811-I_PDF-E/raw.md @@ -0,0 +1,281 @@ + + +![ITU logo](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +The logo of the International Telecommunication Union (ITU) features a globe with a lightning bolt superimposed on it, and the letters 'ITU' in a bold, sans-serif font. + +ITU logo + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +**F.353** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**OPERATIONS AND QUALITY OF SERVICE +TELEMATIC SERVICE** + +--- + +**PROVISION OF TELEMATIC AND DATA +TRANSMISSION SERVICES ON INTEGRATED +SERVICES DIGITAL NETWORK (ISDN)** + +**ITU-T Recommendation F.353** + +(Extract from the *Blue Book*) + +--- + +# NOTES + +1 ITU-T Recommendation F.353 was published in Fascicle II.5 of the *Blue Book*. This file is an extract from the *Blue Book*. While the presentation and layout of the text might be slightly different from the *Blue Book* version, the contents of the file are identical to the *Blue Book* version and copyright conditions remain unchanged (see below). + +2 In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +# **PROVISION OF TELEMATIC AND DATA TRANSMISSION SERVICES ON INTEGRATED SERVICES DIGITAL NETWORK (ISDN)** + +The CCITT, + +*considering* + +- (a) that the I-series of Recommendations apply to the general concept and to the network capabilities of an ISDN. Specifically, Recommendations I.210, I.211 and I.212 describe, in a general way, the principles of telecommunication services, the bearer services and the teleservices supported by an ISDN; +- (b) that the F-series of Recommendations describe the operations and quality of Telematic services; +- (c) that many countries wish to adopt a common strategy for introducing the non-voice Telematic services on ISDN; +- (d) that there is a need to harmonize the service approach for Telematic services, specially the requirements for supplementary services; +- (e) that intercommunication is required between the existing non- voice services on the dedicated networks and non-voice services in the ISDN; +- (f) that there is a need to identify the priority for introducing non-voice Telematic services in the ISDN, + +*recommends* + +that the guidelines laid down in this Recommendation be followed in providing Telematic and data transmission services on ISDN. + +## **1 Existing services provided on ISDN** + +When Telematic and data services are provided on ISDN, it is necessary to refer to the Recommendation of each service for service description: + +- F.160 General operational provisions for the international facsimile services +- F.162 Operational requirements of an international store-and-forward facsimile switching service (COMFAX) +- F.170 Operational provisions for the international public facsimile service between public bureaux (bureaufax) +- F.180 General operational provisions for the international public facsimile service between subscribers' stations +- F.184 Operational provisions for the international public facsimile service between subscribers stations with Group 4 facsimile machines (Telefax 4) (formerly Recommendation F.161) +- F.190 Operational provisions for the international facsimile service between public bureaux and subscriber station and vice versa (Bureaufax-Telefax and vice versa) +- F.200 Teletex service +- F.201 Interworking between the Teletex service and the telex service – General principles +- F.202 Interworking between the telex service and the Teletex service – General procedures and operational requirements for the international interconnection of telex/Teletex conversion facilities. +- F.203 Network based storage for the Teletex service +- F.220 Service requirements unique to the processable mode number one (PM1) used within the Teletex service +- F.230 Service requirements unique to the mixed mode (MM) used within the Teletex service +- F.300 Videotex service + +F.400 Message Handling System and Service overview + +F.500 International public directory services + +Existing services that may be provided on ISDN have to intercommunicate with the same services on existing networks including existing conversion facilities (see Table 1/F.353). + +TABLE 1/F.353 + +| Existing services on ISDN | Services on existing network | +|---------------------------|-------------------------------------------| +| Teletex | Teletex (PSTN, CSPDN, PSPDN)
Telex | +| Telex (see Note 1) | Telex
Teletex (PSTN, CSPDN, PSPDN) | +| Telefax 2/3 | Telefax 2/3 (PSTN) | +| Videotex | Videotex (PSTN, PSPDN) | +| MHS services | MHS S (PSTN, CSPDN, PSPDN) | +| Data transmission | Data transmission
(PSTN, CSPDN, PSPDN) | + +ISDN Integrated Services Digital Network + +PSTN Public Switched Telephone Network + +CSPDN Circuit Switched Public Data Network + +PSPDN Packet Switched Public Data Network + +MHSS Message Handling System Services + +*Note 1* – Telex on the ISDN + +*Note 2* – The migration from non-ISDN terminals to ISDN is a national matter. + +*Note 3* – If existing terminals migrate to ISDN, their compatibility with terminals on existing networks shall not be degraded. + +*Note 4* – Compatibility between ISDN terminals and existing terminals must be retained. + +Where administrations offer the telex service on the ISDN, the following principles should be followed: + +- Operational procedures should be in accordance with Recommendation F.60. +- The telex terminal on the ISDN should be addressable, from international telex networks, by the F.69 telex destination code and a national telex number. The maximum length of the national telex number should be in accordance with the telex signalling requirements of the CCITT U-series of Recommendations. + +## 2 New services1) provided on ISDN + +New services that may be provided on ISDN with intercommunication requirements with other services on ISDN and on existing networks (see Table 2/F.353). + +1) A new service is one that is not yet generally available on existing networks. + +## 3 Teleservices classification + +### 3.1 General principles + +Teleservices classification allows structured approach of the compatibility problem between terminals of the Telematic services. + +To ensure stability of the classification in the future, the list has been divided into major categories. + +Each category contains a limited representative set of ISDN teleservices. The subsequent development of this list will take into account the new Telematic services which could be defined by the CCITT. + +TABLE 2/F.353 + +| New services on ISDN | Services on ISDN and existing networks | +|-------------------------------------|-------------------------------------------------------------------------------------------| +| Teletex with mixed mode | Teletex (PSTN, CSPDN, PSPDN, ISDN)
Telefax 4 (ISDN, CSPON)
Telex | +| Teletex with processable mode (PM1) | Teletex (PSTN, CSPDN, PSPDN, ISDN)
Telefax 4 (ISDN, CSPDN)
(Class II, III)
Telex | +| Telefax 4 Class I | Telefax 3 (PSTN)
Telefax 4 (CSPDN) | +| Telefax 4 Class II | Telefax 3 (PSTN)
Telefax 4 (CSPDN)
Teletex (PSTN, CSPDN, PSPDN, ISDN) | +| Telefax 4 Class III | Telefax 3
Telefax 4
Teletex (PSTN, CSPDN, PSPDN, ISDN) | +| Videotex 64
(see Note 1) | Videotex (PSTN, PSPDN, ISDN) | + +Note 1 – Videotex service using the full capacity of the 64 kbit/s B-channel of the ISDN access. + +Note 2 – Additional intercommunication requirements may be identified subsequently. + +A Telematic service is defined by its specific operational rules, by the list of supported standardized functionalities (applications or modes) and by the relevant communication protocols. + +Each application or mode handles presentation schemes based on the user of one medium or several media which can be synchronized. A medium conveys the characteristics of information presented to the user. + +The simplest and the most discriminant form of classification has to consider the media as the basic elements based upon which the categories are elaborated. + +### 3.2 Teleservices list + +The following list will be completed progressively taking into account the new services as they will appear in the next years. + +#### *Audio services* + +- Telephony +- Audio conference +- . . . (other audio service) + +#### *Text and data services* + +- Telex +- Teletex +- Telefax 3 +- Telefax 4 +- Videotex +- Telewriting +- Data transmission +- MHS +- . . . (other text and data service) + +#### *Video services (see Note)* + +- Videophone +- Video conference +- . . . (other video service) + +#### *Multimedia services* + +- Audiovideography +- Audiography +- Telematic audio conference +- Telematic video conference +- Teleaction +- . . . (other multimedia service) + +*Note* – In this category, the sound is always and implicitly included. + +## **4 Interworking requirements** + +### **4.1 General** + +Interworking of Telematic terminals with compatible terminals on ISDN is mandatory. + +In order to maintain the defined quality of service, suitable selection mechanisms have to be implemented to guarantee access to compatible terminal. + +### **4.2 Interworking with the Teletex service** + +Interworking between all terminals connected to different networks must be possible. + +Real-time connection between Teletex terminals operating at different speeds is required for the duration of the call. The information on the successful transmission should be given by the receiving terminal to the sending terminal within the call. + +The procedures for call set-up to terminals connected to different networks should be as similar as possible. + +Interworking with international telex is mandatory. + +### **4.3 Interworking within the Telefax 4 service** + +Interworking between all terminals connected to different types of networks must be possible. + +Real-time connection between Telefax 4 terminals operating at different speeds is required for the duration of the call. The information on the successful transmission should be given by the receiving terminal to the sending terminal within the call. + +The procedures for call set-up to terminals connected to different networks should be as similar as possible. + +### 4.4 *Interworking from Telefax 4 to Telefax 3* + +Interworking between Telefax 4 terminals and Telefax 3 terminals should be possible even if they are connected to different types of networks. The preferred method of interworking is to be provided by the inclusion of a Group 3 capability in the Group 4 terminal. + +Real-time connection between Telefax terminals (Group 3 and Group 4) operating at different speeds is required for the duration of the call. The information on the successful transmission should be given by the receiving terminal to the sending terminal within the call. Interworking between Group 4 only terminals and Group 3 terminals could be resolved by using network units (store-and-forward or MH system). Further study on this is needed. + +The procedure for call set-up to terminals connected to different networks shall be as similar as possible. + +### 4.5 *Additional interworking configurations* + +Other interworking configurations are for further study. + +## 5 **Assignment of priorities** + +The current assignment of priorities to the future work is: + +- a) Priority for non-voice teleservices: + - 1) existing Telematic services on ISDN, + - 2) new non-voice Teleservices. +- b) Priority for bearer services: + - 1) new bearer services on ISDN, + - 2) use of bearer services of ISDN by existing terminals. + +## 6 **Introductory strategy guideline** + +These are strategic guidelines for the introduction of existing services on ISDN. The implementation is strictly a national matter; however, a measure of co-ordination is necessary internationally in order to provide users worldwide attractive telecommunication services, appropriate intercommunication possibilities, and thereby to ensure financial success for the service providers. + +Any existing service introduced on ISDN should have immediate access to the same service on existing networks. + +There should be minimum changes in the user access procedures. + +Any transition from existing to future international tariffs and accounting should be orderly. + +The suggested provisional list of priorities for the implementation of existing services on ISDN is: + +- 1) Teletex, Telefax, Videotex; +- 2) Data transmission; +- 3) Message Handling System Services; +- 4) Telex. + +It is stressed once again that the priority of implementation is a national matter but it has significant international relevance. + +Simultaneous implementation of existing services on ISDN in all countries is deemed desirable and efforts should be made to achieve this goal. + +## 7 **Bearer services for non-voice communications** + +Recommendation I.211 describes and defines a recommended set of bearer services and their bearer capabilities to be supported by an ISDN as a basis for defining the network capabilities required. + +The purpose of this chapter is to indicate the recommended bearer services associated to non-voice Telematic teleservices: + +### 7.1 *Circuit mode bearer services* + +- 64 kbit/s unrestricted, +- 64 kbit/s (usable for speech information transfer), +- 64 kbit/s (usable for 3.1 kHz audio information transfer), +- 384 kbit/s unrestricted, +- 1536 kbit/s unrestricted, +- 1920 kbit/s unrestricted, +- $2 \times 64$ kbit/s unrestricted. + +### 7.2 *Packet-mode bearer services* + +- Virtual call and permanent virtual circuit (B or D channel); +- Connectionless packet bearer service on a D channel (further study needed); +- Use signalling bearer service on a D channel (further study needed). \ No newline at end of file diff --git a/marked/F/T-REC-F.4-198811-I_PDF-E/raw.md b/marked/F/T-REC-F.4-198811-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..1dd37e8460be8f1a9d8b3f2115a4ad6ff52ea5aa --- /dev/null +++ b/marked/F/T-REC-F.4-198811-I_PDF-E/raw.md @@ -0,0 +1,108 @@ + + +![ITU logo](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +The logo of the International Telecommunication Union (ITU) features the letters 'ITU' in a bold, sans-serif font, superimposed on a stylized globe with intersecting lines. + +ITU logo + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +**F.4** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**OPERATIONS AND QUALITY OF SERVICE +TELEGRAPH SERVICES** + +--- + +**PLAIN AND SECRET LANGUAGE** + +**ITU-T Recommendation F.4** + +(Extract from the *Blue Book*) + +--- + +# NOTES + +1 ITU-T Recommendation F.4 was published in Fascicle II.4 of the *Blue Book*. This file is an extract from the *Blue Book*. While the presentation and layout of the text might be slightly different from the *Blue Book* version, the contents of the file are identical to the *Blue Book* version and copyright conditions remain unchanged (see below). + +2 In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +## Recommendation F.4 + +### PLAIN AND SECRET LANGUAGE + +The CCITT, + +*considering* + +(a) Article 27 of the International Telecommunication Convention (Nairobi, 1982) concerning the use of secret language in telegrams; + +(b) the general provisions in Recommendation F.1 on the preparation and handing in of telegrams; + +(c) the CCITT studies on simplifying the international public telegram service and increasing its cost-effectiveness; + +(d) that the provisions governing the international telex, telematic and data transmission services do not restrict the language or content of messages transmitted by these services, nor would it be practicable to apply restrictions in subscriber-operated services without withdrawing automatic operation or otherwise disrupting the quality of service; + +(e) that, nonetheless, it may be necessary for an ITU Member to impose restrictions on the content of telegrams transmitted from its territory or received from another country; + +(f) that where an ITU Member elects to impose language restrictions on telegrams received from another country, this should not unduly burden the operation of the service in that country, + +*declares the view* + +that the following procedures shall apply regarding the use of plain and secret language in the international telegram service. + +### 1 Definitions + +1.1 **plain language** consists of words that present an intelligible meaning in one or more of the languages admitted for international telegrams, which include at least French, English and Spanish in every relation. Each word and each expression has a meaning normally assigned to it in the language to which it belongs. A text in plain language may contain: + +- numbers written in letters or figures; +- proper names or abbreviated addresses; +- groups comprising letters, figures, signs or any combination of them providing that they have no secret meaning. + +1.2 **secret language** comprises words in which one or more consist of: + +- groups of letters, figures, signs or any combination of them that have a secret meaning; +- words in plain language that are not used with the meaning normally assigned to them; +- any other words not fulfilling the conditions laid down for plain language. + +### 2 General principles + +2.1 All ITU Members shall accept in all relations the use of plain language in messages sent or received by any public international telecommunication service. + +2.2 At least the three working languages of the Union (French, English and Spanish) shall be admitted as plain language in all relations. + +2.3 In order to facilitate efficient working and convenience for the users of telecommunication services, all ITU Members should also normally admit telegrams wholly or partly in secret language. In any case, government telecommunications and service telecommunications may be expressed in secret language in all relations. Except in the case of suspension of services defined in Article 20 of the Convention, any telecommunication containing secret language shall also be allowed to pass in transit from one country to another. + +2.4 Exceptionally, where an ITU Member finds it necessary to impose language restrictions on international telecommunications, for example pursuant to Article 19 of the Convention (stoppage of telecommunications): + +- a) the Member concerned shall make all necessary arrangements regarding any restrictions on messages originating within its own territory; +- b) except where appropriate bilateral agreements are reached, it shall also accept the final responsibility for stopping any non-compliant messages originating in other countries, although origin Administrations should assist to the extent reasonable and practicable; +- c) the Secretary-General should be advised of the restrictions so that he can then advise all Administrations promptly through the *Operational Bulletin*. + +### **3 Telegrams with secret language** + +3.1 If requested by the origin Administration, the sender of a telegram in secret language must produce the code or identify the dictionary language used in drafting the telegram. Administrations may also require the sender to produce a translation of the telegram into plain language or a language acceptable to the Administration. This provision shall not apply to government telegrams. + +3.2 If the origin Administration considers it appropriate, or, in relations where so agreed to meet the requirements of the destination Administration, the office of origin should insert the name of the code and/or language used in such a telegram at the end of the preamble line as a service instruction, which is not chargeable. It may not be convenient for this procedure to be applied with some types of lodgement. + +3.3 In cases where a destination Administration receives a telegram (other than a government telegram) wholly or partly in non admissable language, that Administration: + +- a) may require the addressee to produce a translation of the telegram; or +- b) shall make every effort to identify the commercial code or language used and check that the translated text is then acceptable; and +- c) shall advise the origin Administration where a) or b) causes late delivery as defined in Recommendation D.42 (which shall nevertheless not be grounds for a refund of charges to the originator); or +- d) shall inform the origin Administration by service advice when either the original telegram cannot be translated or when the translated text contravenes national law (neither case shall be grounds for a refund of charges to the originator). + +3.4 Further to point c) of § 2.4 above, any Administrations wishing to impose language restrictions shall advise the Secretary-General of their requirements both for lodgements within their country and for reception from other countries in terms of: + +- a) languages other than French, English and Spanish admitted as plain language; +- b) commercial or other standard codes admitted; +- c) whether identification in the preamble line of the code or language used is desired; +- d) whether secret language as defined in § 1.2, other than b) above is not admitted. \ No newline at end of file diff --git a/marked/F/T-REC-F.400-199906-I_PDF-E/raw.md b/marked/F/T-REC-F.400-199906-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..8fb1a4fb31fc15f94eb6ac091668922008aec16e --- /dev/null +++ b/marked/F/T-REC-F.400-199906-I_PDF-E/raw.md @@ -0,0 +1,4132 @@ + + +![ITU logo: A globe with the letters ITU and a lightning bolt.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with the letters ITU and a lightning bolt. + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +**F.400/X.400** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +(06/1999) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Message handling services + +SERIES X: DATA NETWORKS AND OPEN SYSTEM +COMMUNICATIONS + +Message handling systems + +--- + +**Message handling system and service overview** + +ITU-T Recommendation F.400/X.400 + +--- + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| AUDIOVISUAL SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# **Message handling system and service overview** + +## **Summary** + +This revision of ITU-T Rec. F.400/X.400 is a consolidation of Recommendation F.400/X.400 (07/96) and Amendment 1 (09/98). This Recommendation provides an overview of system and service of Message Handling Systems (MHS). It should be considered in conjunction with the F.400-series and X.400-series of Recommendations. + +## **Source** + +Following ITU-T decision to publish new editions of the set of Message Handling Recommendations this edition of ITU-T Recommendation F.400/X.400, dated 18 June 1999, consolidates F.400/X.400 (07/96) and F.400/X.400 Amendment 1 (09/98). + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications. The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Conference (WTSC), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSC Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure e.g. interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementors are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database. + +© ITU 2004 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## CONTENTS + +*Page* + +| | | | +|----|------------------------------------------------------------------|----| +| 1 | Scope ..... | 1 | +| 2 | Normative references ..... | 1 | +| 3 | Definitions ..... | 3 | +| | 3.1 Open Systems Interconnection ..... | 4 | +| | 3.2 Directory systems ..... | 4 | +| 4 | Abbreviations ..... | 4 | +| 5 | Conventions ..... | 5 | +| 6 | Purpose ..... | 5 | +| 7 | Functional model of MHS ..... | 6 | +| | 7.1 Description of the MHS model ..... | 6 | +| | 7.2 Structure of messages ..... | 7 | +| | 7.3 Application of the MHS model ..... | 8 | +| | 7.3.1 Physical mapping ..... | 8 | +| | 7.3.2 Organizational mapping ..... | 8 | +| | 7.3.3 Administration management domain ..... | 8 | +| | 7.3.4 Private management domain ..... | 9 | +| | 7.4 Message store (MS) ..... | 10 | +| | 7.4.1 Physical configurations ..... | 13 | +| | 7.4.2 Organizational configurations ..... | 13 | +| 8 | Message transfer service ..... | 13 | +| | 8.1 Submission and delivery ..... | 13 | +| | 8.2 Transfer ..... | 13 | +| | 8.3 Notifications ..... | 13 | +| | 8.4 User agent ..... | 13 | +| | 8.5 Message store ..... | 14 | +| | 8.6 Access unit ..... | 14 | +| | 8.7 Use of the MTS in the provision of various services ..... | 14 | +| 9 | IPM service ..... | 14 | +| | 9.1 IPM service functional model ..... | 14 | +| | 9.2 Structure of IP-messages ..... | 15 | +| | 9.3 IP-notifications ..... | 16 | +| 10 | Intercommunication with physical delivery services ..... | 16 | +| | 10.1 Introduction ..... | 16 | +| | 10.2 Organizational configurations ..... | 17 | +| 11 | Specialized access ..... | 18 | +| | 11.1 Introduction ..... | 18 | +| | 11.2 Telex access ..... | 18 | +| | 11.2.1 Registered access to the IPM service ..... | 18 | +| | 11.2.2 Non-registered (public) access to the IPM service ..... | 18 | +| | 11.3 Facsimile access ..... | 18 | +| | 11.3.1 Non-registered (public) access from the IPM service ..... | 18 | +| 12 | Naming and addressing ..... | 19 | +| | 12.1 Introduction ..... | 19 | +| | 12.2 Directory names ..... | 19 | +| | 12.3 OR-names ..... | 19 | +| | 12.4 OR-addresses ..... | 19 | +| 13 | MHS use of directory ..... | 20 | +| | 13.1 Introduction ..... | 20 | +| | 13.2 Functional model ..... | 20 | +| | 13.3 Physical configurations ..... | 20 | + +| | | Page | +|---------|--------------------------------------------------------------------|-------------| +| 14 | Distribution lists in MHS..... | 21 | +| 14.1 | Introduction..... | 21 | +| 14.2 | Properties of a DL..... | 21 | +| 14.3 | Submission..... | 22 | +| 14.4 | DL use of a directory..... | 22 | +| 14.5 | DL expansion..... | 22 | +| 14.6 | Nesting..... | 22 | +| 14.7 | Recursion control..... | 23 | +| 14.8 | Delivery..... | 23 | +| 14.9 | Routing loop control..... | 23 | +| 14.10 | Notifications..... | 23 | +| 14.11 | DL-handling policy..... | 23 | +| 15 | Security capabilities of MHS..... | 23 | +| 15.1 | Introduction..... | 23 | +| 15.2 | MHS-security threats..... | 24 | +| 15.2.1 | Access threats..... | 24 | +| 15.2.2 | Inter-message threats..... | 24 | +| 15.2.3 | Intra-message threats..... | 24 | +| 15.2.4 | Data-store threats..... | 24 | +| 15.3 | Security model..... | 24 | +| 15.3.1 | Secure access management and administration..... | 24 | +| 15.3.2 | Secure messaging..... | 25 | +| 15.4 | MHS-security capabilities..... | 26 | +| 15.5 | Security management..... | 26 | +| 15.6 | MHS-security dependencies..... | 27 | +| 15.7 | IPM security..... | 27 | +| 16 | Conversion in MHS..... | 28 | +| 17 | Use of the MHS in provision of public services..... | 28 | +| 18 | Elements of service – Purpose..... | 29 | +| 19 | Elements of service – Classification..... | 33 | +| 19.1 | Purpose of classification..... | 33 | +| 19.2 | Basic message transfer service..... | 33 | +| 19.3 | MT service optional user facilities..... | 33 | +| 19.4 | Base MH/PD service intercommunication..... | 35 | +| 19.5 | Optional user facilities for MH/PD service intercommunication..... | 35 | +| 19.6 | Base message store..... | 35 | +| 19.7 | MS optional user facilities..... | 36 | +| 19.8 | Basic interpersonal messaging service..... | 36 | +| 19.9 | IPM service optional user facilities..... | 37 | +| Annex A | – Glossary of terms..... | 41 | +| A.1 | access unit (AU)..... | 41 | +| A.2 | actual recipient..... | 41 | +| A.3 | Administration..... | 41 | +| A.4 | administration domain name..... | 41 | +| A.5 | administration management domain (ADMD)..... | 41 | +| A.6 | alternate recipient..... | 41 | +| A.7 | attribute..... | 42 | +| A.8 | attribute list..... | 42 | +| A.9 | attribute type..... | 42 | +| A.10 | attribute value..... | 42 | +| A.11 | basic service..... | 42 | +| A.12 | body..... | 42 | +| A.13 | body part..... | 42 | +| A.14 | common name..... | 42 | + +| | Page | +|-------------------------------------------------------------|-------------| +| A.15 content ..... | 43 | +| A.16 content type..... | 43 | +| A.17 conversion..... | 43 | +| A.18 country name..... | 43 | +| A.19 delivery ..... | 43 | +| A.20 delivery report..... | 43 | +| A.21 direct submission..... | 43 | +| A.22 directory ..... | 44 | +| A.23 directory name..... | 44 | +| A.24 directory system agent (DSA)..... | 44 | +| A.25 directory user agent (DUA)..... | 44 | +| A.26 direct user..... | 44 | +| A.27 distribution list (DL) ..... | 44 | +| A.28 distribution list expansion ..... | 44 | +| A.29 distribution list name..... | 45 | +| A.30 domain ..... | 45 | +| A.31 domain-defined attributes..... | 45 | +| A.32 element of service ..... | 45 | +| A.33 encoded information type (EIT)..... | 45 | +| A.34 envelope ..... | 45 | +| A.35 explicit conversion ..... | 45 | +| A.36 extension of physical delivery address components..... | 45 | +| A.37 extension of postal OR-address components..... | 46 | +| A.38 file transfer body part..... | 46 | +| A.39 formatted postal OR-address..... | 46 | +| A.40 general text body part..... | 46 | +| A.41 heading ..... | 46 | +| A.42 immediate recipient..... | 46 | +| A.43 implicit conversion..... | 46 | +| A.44 indirect submission..... | 47 | +| A.45 indirect user..... | 47 | +| A.46 preferred recipient ..... | 47 | +| A.47 intercommunication..... | 47 | +| A.48 interpersonal messaging service..... | 47 | +| A.49 IP-message ..... | 47 | +| A.50 local postal attributes..... | 47 | +| A.51 management domain (MD) ..... | 48 | +| A.52 management domain name..... | 48 | +| A.53 members..... | 48 | +| A.54 message ..... | 48 | +| A.55 message handling (MH) ..... | 48 | +| A.56 message handling environment (MHE)..... | 48 | +| A.57 message handling service..... | 49 | +| A.58 message handling system (MHS)..... | 49 | +| A.59 message storage..... | 49 | +| A.60 message store (MS)..... | 49 | +| A.61 message transfer (MT) ..... | 49 | +| A.62 message transfer agent (MTA)..... | 49 | +| A.63 message transfer service..... | 49 | +| A.64 message transfer system (MTS)..... | 50 | +| A.65 messaging system..... | 50 | +| A.66 mnemonic OR-address..... | 50 | +| A.67 naming authority ..... | 50 | +| A.68 network address..... | 50 | +| A.69 non-delivery ..... | 50 | + +| | Page | +|--------------------------------------------------------|-------------| +| A.70 non-registered access ..... | 50 | +| A.71 numeric OR-address..... | 51 | +| A.72 numeric user identifier ..... | 51 | +| A.73 OR-address..... | 51 | +| A.74 OR-name ..... | 51 | +| A.75 optional user facilities ..... | 51 | +| A.76 organization name ..... | 51 | +| A.77 organizational unit name ..... | 51 | +| A.78 originator..... | 52 | +| A.79 personal name ..... | 52 | +| A.80 physical delivery (PD)..... | 52 | +| A.81 physical delivery access unit (PDAU)..... | 52 | +| A.82 physical delivery address components ..... | 52 | +| A.83 physical delivery country name..... | 52 | +| A.84 physical delivery domain ..... | 52 | +| A.85 physical delivery office address components ..... | 53 | +| A.86 physical delivery office name ..... | 53 | +| A.87 physical delivery office number..... | 53 | +| A.88 physical delivery organization name..... | 53 | +| A.89 physical delivery personal name ..... | 53 | +| A.90 physical delivery service..... | 53 | +| A.91 physical delivery service name ..... | 53 | +| A.92 physical delivery system (PDS) ..... | 54 | +| A.93 physical message..... | 54 | +| A.94 physical rendition..... | 54 | +| A.95 postal code..... | 54 | +| A.96 postal OR-address ..... | 54 | +| A.97 postal OR-address components ..... | 54 | +| A.98 post office box address (P.O. Box address)..... | 54 | +| A.99 poste restante address ..... | 55 | +| A.100 potential recipient..... | 55 | +| A.101 private domain name ..... | 55 | +| A.102 private management domain (PRMD) ..... | 55 | +| A.103 probe ..... | 55 | +| A.104 public message handling service..... | 55 | +| A.105 public services..... | 55 | +| A.106 receipt ..... | 56 | +| A.107 recipient ..... | 56 | +| A.108 recursion..... | 56 | +| A.109 redirection ..... | 56 | +| A.110 registered access..... | 56 | +| A.111 report ..... | 56 | +| A.112 retrieval ..... | 56 | +| A.113 security capabilities..... | 57 | +| A.114 specialized access..... | 57 | +| A.115 standard attribute..... | 57 | +| A.116 street address ..... | 57 | +| A.117 subject ..... | 57 | +| A.118 subject message..... | 57 | +| A.119 subject probe ..... | 57 | +| A.120 submission..... | 57 | +| A.121 substitute recipient ..... | 58 | +| A.122 terminal identifier..... | 58 | +| A.123 terminal OR-address..... | 58 | +| A.124 terminal type ..... | 58 | + +| | Page | +|---------------------------------------------------------------------|-------------| +| A.125 transfer ..... | 58 | +| A.126 transfer system ..... | 58 | +| A.127 transmittal..... | 58 | +| A.128 unformatted postal OR-address ..... | 59 | +| A.129 unique postal name..... | 59 | +| A.130 user ..... | 59 | +| A.131 user agent (UA)..... | 59 | +| A.132 voice body part..... | 59 | +| Annex B – Definitions of elements of service ..... | 60 | +| B.1 access management MT ..... | 60 | +| B.2 additional physical rendition PD PR ..... | 60 | +| B.3 alternate recipient allowed MT ..... | 60 | +| B.4 alternate recipient assignment MT ..... | 60 | +| B.5 authorization time indication IPM..... | 61 | +| B.6 authorizing users indication IPM ..... | 61 | +| B.7 auto-acknowledgement of IP-messages IPM MS-94 ..... | 61 | +| B.8 auto-action log MS-94..... | 61 | +| B.9 auto-advise IPM MS ..... | 61 | +| B.10 auto-assignment of annotations MS-94 ..... | 61 | +| B.11 auto-assignment of group names MS-94 ..... | 61 | +| B.12 auto-assignment of storage period MS-94 ..... | 61 | +| B.13 auto-correlation of IP-messages IPM MS-94 ..... | 62 | +| B.14 auto-correlation of IP-notifications IPM MS-94 ..... | 62 | +| B.15 auto-correlation of reports MS-94..... | 62 | +| B.16 auto-deletion after storage period MS-94..... | 62 | +| B.17 auto-discarding of IP-messages IPM MS-94..... | 62 | +| B.18 auto-forwarded indication IPM ..... | 62 | +| B.19 auto-forwarding of IP-messages IPM MS..... | 63 | +| B.20 auto-submitted indication IPM..... | 63 | +| B.21 basic physical rendition PD PR..... | 63 | +| B.22 blind copy recipient indication IPM PR..... | 63 | +| B.23 body part authentication and integrity IPM..... | 63 | +| B.24 body part encryption IPM ..... | 63 | +| B.25 circulation list recipients indication IPM ..... | 64 | +| B.26 content confidentiality MT..... | 64 | +| B.27 content integrity MT PR..... | 64 | +| B.28 content type indication MT ..... | 64 | +| B.29 conversion prohibition MT..... | 64 | +| B.30 conversion prohibition in case of loss of information MT ..... | 64 | +| B.31 converted indication MT PR ..... | 64 | +| B.32 counter collection PD PR ..... | 64 | +| B.33 counter collection with advice PD PR..... | 64 | +| B.34 cover page suppression MT PR ..... | 65 | +| B.35 cross-referencing indication IPM..... | 65 | +| B.36 deferred delivery MT ..... | 65 | +| B.37 deferred delivery cancellation MT ..... | 65 | +| B.38 delivery log MS-94 ..... | 65 | +| B.39 delivery notification MT PR ..... | 65 | +| B.40 delivery time stamp indication MT PR ..... | 65 | +| B.41 delivery via bureaufax service PD PR..... | 65 | +| B.42 designation of recipient by directory name MT PR ..... | 66 | +| B.43 disclosure of other recipients MT..... | 66 | +| B.44 distribution codes indication IPM ..... | 66 | +| B.45 DL-exempted recipients MT ..... | 66 | +| B.46 DL-expansion history indication MT ..... | 66 | + +| | Page | +|--------------------------------------------------------------|-------------| +| B.47 DL-expansion prohibited MT..... | 66 | +| B.48 EMS (Express Mail Service) PD PR..... | 66 | +| B.49 expiry date indication IPM..... | 66 | +| B.50 explicit conversion MT PR ..... | 66 | +| B.51 forwarded IP-message indication IPM..... | 67 | +| B.52 grade of delivery selection MT ..... | 67 | +| B.53 hold for delivery MT..... | 67 | +| B.54 implicit conversion MT..... | 67 | +| B.55 importance indication IPM..... | 67 | +| B.56 incomplete copy indication IPM ..... | 67 | +| B.57 information category indication IPM ..... | 68 | +| B.58 IP-message action status IPM MS-94 ..... | 68 | +| B.59 IP-message identification IPM..... | 68 | +| B.60 IP-message security labelling IPM..... | 68 | +| B.61 language indication IPM..... | 68 | +| B.62 latest delivery designation MT ..... | 68 | +| B.63 manual handling instructions indication IPM..... | 68 | +| B.64 message flow confidentiality MT..... | 69 | +| B.65 message identification MT ..... | 69 | +| B.66 message origin authentication MT PR ..... | 69 | +| B.67 message security labelling MT..... | 69 | +| B.68 message sequence integrity MT PR ..... | 69 | +| B.69 MS register MS ..... | 69 | +| B.70 multi-destination delivery MT PR..... | 69 | +| B.71 multi-part body IPM..... | 69 | +| B.72 non-delivery notification MT PR ..... | 70 | +| B.73 non-receipt notification request indication IPM PR..... | 70 | +| B.74 non-repudiation of content received IPM PR..... | 70 | +| B.75 non-repudiation of delivery MT PR..... | 70 | +| B.76 non-repudiation of IP-notification IPM PR..... | 71 | +| B.77 non-repudiation of origin MT PR..... | 71 | +| B.78 non-repudiation of submission MT ..... | 71 | +| B.79 obsoleting indication IPM..... | 71 | +| B.80 ordinary mail PD PR ..... | 71 | +| B.81 original encoded information types indication MT ..... | 71 | +| B.82 originator indication IPM..... | 71 | +| B.83 originator reference indication IPM ..... | 72 | +| B.84 originator requested alternate recipient MT PR ..... | 72 | +| B.85 physical delivery notification by MHS PD PR..... | 72 | +| B.86 physical delivery notification by PDS PD PR..... | 72 | +| B.87 physical forwarding allowed PD PR ..... | 72 | +| B.88 physical forwarding prohibited PD PR ..... | 72 | +| B.89 precedence indication IPM PR..... | 73 | +| B.90 prevention of non-delivery notification MT PR..... | 73 | +| B.91 primary and copy recipients indication IPM ..... | 73 | +| B.92 probe MT..... | 73 | +| B.93 probe origin authentication MT..... | 73 | +| B.94 proof of content received IPM PR..... | 74 | +| B.95 proof of delivery MT PR..... | 74 | +| B.96 proof of IP-notification IPM PR..... | 74 | +| B.97 proof of submission MT..... | 74 | +| B.98 receipt notification request indication IPM PR ..... | 74 | +| B.99 redirection disallowed by originator MT..... | 75 | +| B.100 redirection of incoming messages MT ..... | 75 | +| B.101 registered mail PD PR..... | 75 | + +| | Page | +|-----------------------------------------------------------------------------------------|-------------| +| B.102 registered mail to addressee in person PD PR..... | 75 | +| B.103 reply request indication IPM PR..... | 75 | +| B.104 replying IP-message indication IPM..... | 75 | +| B.105 report origin authentication MT..... | 75 | +| B.106 request for forwarding address PD PR..... | 76 | +| B.107 request for non-repudiation of content received IPM PR..... | 76 | +| B.108 request for non-repudiation of IP-notification IPM PR..... | 76 | +| B.109 request for proof of content received IPM PR..... | 76 | +| B.110 request for proof of IP-notification IPM PR..... | 76 | +| B.111 requested preferred delivery method MT PR..... | 76 | +| B.112 restricted delivery MT..... | 77 | +| B.113 return of content MT..... | 77 | +| B.114 secure access management MT..... | 77 | +| B.115 sensitivity indication IPM..... | 77 | +| B.116 special delivery PD PR..... | 77 | +| B.117 storage of draft messages MS-94..... | 77 | +| B.118 storage on submission MS-94..... | 78 | +| B.119 storage period assignment MS-94..... | 78 | +| B.120 stored message alert MS..... | 78 | +| B.121 stored message annotation MS-94..... | 78 | +| B.122 stored message deletion MS..... | 78 | +| B.123 stored message fetching MS..... | 78 | +| B.124 stored message grouping MS-94..... | 78 | +| B.125 stored message listing MS..... | 79 | +| B.126 stored message summary MS..... | 79 | +| B.127 subject indication IPM..... | 79 | +| B.128 submission log MS-94..... | 79 | +| B.129 submission of IP-messages incorporating stored messages IPM MS-94..... | 79 | +| B.130 submission time stamp indication MT..... | 79 | +| B.131 typed body IPM..... | 79 | +| B.132 undeliverable mail with return of physical message PD PR..... | 80 | +| B.133 use of distribution list MT PR..... | 80 | +| B.134 user/UA capabilities registration MT..... | 80 | +| Annex C – Elements of service changes from 1992..... | 81 | +| C.1 New elements of service in 1996 version..... | 81 | +| C.2 Classification of new elements of service..... | 82 | +| Annex D – Differences between ITU-T Recommendation F.400/X.400 and ISO/IEC 10021-1..... | 83 | +| D.1 Differences..... | 83 | +| Annex E – Multinational PRMD..... | 84 | + + + +## Message handling system and service overview + +### 1 Scope + +This Recommendation defines the overall system and service of an MHS and serves as a general overview of MHS. + +Other aspects of message handling systems and services are defined in other Recommendations. The layout of Recommendations defining the message handling system and services is shown in Table 1. The public services built on MHS, as well as access to and from the MHS for public services are defined in the F.400-series Recommendations. + +The technical aspects of MHS are defined in the X.400-series Recommendations. The overall system architecture of MHS is defined in ITU-T Rec. X.402 | ISO/IEC 10021-2. + +### 2 Normative references + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +- CCITT Recommendation F.60 (1992), *Operational provisions for the international telex service*. +- ITU-T Recommendation F.160 (1993), *General operational provisions for the international public facsimile services*. +- ITU-T Recommendation F.300 (1993), *Videotex service*. +- CCITT Recommendation F.401 (1992), *Naming and addressing for public message handling services*. +- CCITT Recommendation F.410 (1992), *The public message transfer service*. +- CCITT Recommendation F.415 (1988), *Intercommunication with public physical delivery services*. +- CCITT Recommendation F.420 (1992), *The public interpersonal messaging service*. +- CCITT Recommendation F.421 (1988), *Intercommunication between the IPM service and the telex service*. +- CCITT Recommendation F.423 (1992), *Intercommunication between the interpersonal messaging service and the telefax service*. +- ITU-T Recommendation F.435 (1999), *Electronic data interchange messaging service*. +ISO/IEC 10021-8:1999, *Information technology – Text Communication – Message Handling Systems (MHS) – Part 8: Electronic Data Interchange Messaging Service*. +- CCITT Recommendation F.440 (1992), *The voice messaging service*. +- CCITT Recommendation T.330 (1988), *Telematic access to interpersonal message system*. +- ITU-T Recommendation X.200 (1994) | ISO/IEC 7498-1:1994, *Information technology – Open Systems Interconnection – Basic Reference Model: The Basic Model*. +- ITU-T Recommendation X.207 (1993) | ISO/IEC 9545:1994, *Information technology – Open Systems Interconnection – Application layer structure*. +- ITU-T Recommendation X.217 (1995) | ISO/IEC 8649:1996, *Information technology – Open Systems Interconnection – Service definition for the Association Control Service Element*. +- ITU-T Recommendation X.218 (1993), *Reliable transfer: Model and service definition*. +ISO/IEC 9066-1:1989, *Information processing systems – Text communication – Reliable Transfer – Part 1: Model and service definition*. +- CCITT Recommendation X.219 (1988), *Remote Operations: Model, notation and service definition*. + +- ISO/IEC 9072-1:1989, *Information processing systems – Text communication – Remote Operations – Part 1: Model, notation and service definition.* +- ITU-T Recommendation X.402 (1999) | ISO/IEC 10021-2:2003, *Information technology – Message Handling Systems (MHS): Overall architecture.* + - CCITT Recommendation X.407 (1988), *Message handling systems: Abstract service definition conventions.* + - ISO/IEC 10021-3:1990, *Information technology – Text communication – Message-Oriented Text Interchange Systems (MOTIS) – Part 3: Abstract Service Definition Conventions.* + - CCITT Recommendation X.408 (1988), *Message handling systems: Encoded information type conversion rules.* + - ITU-T Recommendation X.411 (1999) | ISO/IEC 10021-4:2003, *Information technology – Message Handling Systems (MHS): Message transfer system: Abstract service definition and procedures.* + - ITU-T Recommendation X.413 (1999) | ISO/IEC 10021-5:1999, *Information technology – Message Handling Systems (MHS): Message store – Abstract service definition.* + - ITU-T Recommendation X.419 (1999) | ISO/IEC 10021-6:2003, *Information technology – Message Handling Systems (MHS): Protocol specifications.* + - ITU-T Recommendation X.420 (1999) | ISO/IEC 10021-7:2003, *Information technology – Message Handling Systems (MHS): Interpersonal messaging system.* + - ITU-T Recommendation X.435 (1999) | ISO/IEC 10021-9:1999, *Information technology – Message Handling Systems (MHS): Electronic data interchange messaging system.* + - ITU-T Recommendation X.440 (1999), *Message handling systems: Voice messaging system.* + - ITU-T Recommendation X.460 (1995) | ISO/IEC 11588-1:1996, *Information technology – Message Handling Systems (MHS) management: Model and architecture.* + - ITU-T Recommendation X.500 (1997) | ISO/IEC 9594-1:1998, *Information technology – Open Systems Interconnection – The Directory: Overview of concepts, models and services.* + - ITU-T Recommendation X.501 (1997) | ISO/IEC 9594-2:1998, *Information technology – Open Systems Interconnection – The Directory: Models.* + - ITU-T Recommendation X.509 (1997) | ISO/IEC 9594-8:1998, *Information technology – Open Systems Interconnection – The Directory: Authentication framework.* + - ITU-T Recommendation X.511 (1997) | ISO/IEC 9594-3:1998, *Information technology – Open Systems Interconnection – The Directory: Abstract service definition.* + - ITU-T Recommendation X.518 (1997) | ISO/IEC 9594-4:1998, *Information technology – Open Systems Interconnection – The Directory: Procedures for distributed operation.* + - ITU-T Recommendation X.519 (1997) | ISO/IEC 9594-5:1998, *Information technology – Open Systems Interconnection – The Directory: Protocol specifications.* + - ITU-T Recommendation X.520 (1997) | ISO/IEC 9594-6:1998, *Information technology – Open Systems Interconnection – The Directory: Selected attribute types.* + - ITU-T Recommendation X.521 (1997) | ISO/IEC 9594-7:1998, *Information technology – Open Systems Interconnection – The Directory: Selected object classes.* + - ITU-T Recommendation X.525 (1997) | ISO/IEC 9594-9:1998, *Information technology – Open Systems Interconnection – The Directory: Replication.* + - ITU-T Recommendation X.680 (1997) | ISO/IEC 8824-1:1998, *Information technology – Abstract Syntax Notation One (ASN.1): Specification of basic notation.* + - ITU-T Recommendation X.690 (1997) | ISO/IEC 8825-1:1998, *Information technology – ASN.1 encoding rules: Specification of Basic Encoding Rules (BER), Canonical Encoding Rules (CER) and Distinguished Encoding Rules (DER).* + - ITU-T Recommendation X.880 (1994) | ISO/IEC 13712-1:1995, *Information technology – Remote Operations: Concepts, model and notation.* + - ITU-T Recommendation X.881 (1994) | ISO/IEC 13712-2:1995, *Information technology – Remote Operations: OSI realizations – Remote Operations Service Element (ROSE) service definition.* + - ITU-T Recommendation X.882 (1994) | ISO/IEC 13712-3:1995, *Information technology – Remote Operations: OSI realizations – Remote Operations Service Element (ROSE) protocol specification.* + +**Table 1/F.400/X.400 – Structure of ITU-T Recommendations and ISO/IEC International Standards for Message Handling Systems (MHSs)** + +| Short title | Joint MHS | | Joint support | | ITU-T only | | +|------------------------------------------------------------------------------------------------------------------------------------------|-----------------|----------|---------------|---------|----------------|---------| +| | ITU-T | ISO/IEC | ITU-T | ISO/IEC | System | Service | +| MHS: System and service overview | F.400/
X.400 | 10021-1 | | | | | +| MHS: Overall architecture | X.402 | 10021-2 | | | X.408 | | +| MHS: Encoded information type conversion rules | | | | | | | +| MHS: MTS: Abstract service definition and procedures | X.411 | 10021-4 | | | | | +| MHS: MS: Abstract service definition | X.413 | 10021-5 | | | | | +| MHS: Protocol specifications | X.419 | 10021-6 | | | | | +| MHS: Interpersonal messaging system | X.420 | 10021-7 | | | | | +| MHS: Electronic data interchange messaging system | X.435 | 10021-9 | | | | | +| MHS: Voice messaging system
Telematic access to interpersonal message system | | | | | X.440
T.330 | | +| MHS: MHS routing | X.412 | 10021-10 | | | | | +| MHS: MHS routing: Guide for messaging systems managers | X.404 | 10021-11 | | | | | +| MHS: Naming and addressing for public message handling services | | | | | | F.401 | +| MHS: The public message transfer service | | | | | | F.410 | +| MHS: Intercommunication with public physical delivery services | | | | | | F.415 | +| MHS: The public interpersonal messaging service | | | | | | F.420 | +| MHS: Intercommunication between the IPM service and the telex service | | | | | | F.421 | +| MHS: Intercommunication between the interpersonal messaging service and the telefax service | | | | | | F.423 | +| MHS: Electronic data interchange messaging service | | 10021-8 | | | | F.435 | +| MHS: The voice messaging service | | | | | | F.440 | +| OSI: Basic reference model: The basic model | | | X.200 | 7498-1 | | | +| ASN.1: Specification of basic notation | | | X.680 | 8824-1 | | | +| ASN.1 encoding rules: Specification of Basic Encoding Rules (BER), Canonical Encoding Rules (CER) and Distinguished Encoding Rules (DER) | | | X.690 | 8825-1 | | | +| OSI: Service definition for the association control service element | | | X.217 | 8649 | | | +| OSI: Reliable transfer: Model and service definition | | | X.218 | 9066-1 | | | +| Remote Operations: Concepts, model and notation | | | X.880 | 13712-1 | | | +| Remote Operations: OSI realizations – Remote Operations Service Element (ROSE) service definition | | | X.881 | 13712-2 | | | +| OSI: Connection-oriented protocol for the association control service element: Protocol specification | | | X.227 | 8650-1 | | | +| OSI: Reliable transfer: Protocol specification | | | X.228 | 9066-2 | | | +| Remote Operations: OSI realizations – Remote Operations Service Element (ROSE) protocol specification | | | X.882 | 13712-3 | | | + +### 3 Definitions + +For the purposes of this Recommendation, the definitions given in Annex A and the following definitions apply. + +Definitions of the elements of service applicable to MHS are contained in Annex B. + +### **3.1 Open Systems Interconnection** + +This Recommendation makes use of the following terms defined in ITU-T Rec. X.200 | ISO/IEC 7498-1: + +- a) application layer; +- b) application-process; +- c) Open Systems Interconnection; +- d) OSI reference model. + +### **3.2 Directory systems** + +This Recommendation makes use of the following terms defined in ITU-T Rec. X.500 | ISO/IEC 9594-1: + +- a) directory entry; +- b) directory system agent; +- c) directory system; +- d) directory user agent. + +This Recommendation makes use of the following terms defined in ITU-T Rec. X.501 | ISO/IEC 9594-2: + +- e) attribute; +- f) group; +- g) name. + +## **4 Abbreviations** + +This Recommendation uses the following abbreviations: + +| | | +|------|----------------------------------| +| A | Additional | +| ADMD | Administration Management Domain | +| AU | Access Unit | +| CA | Contractual Agreement | +| DL | Distribution List | +| DSA | Directory System Agent | +| DUA | Directory User Agent | +| E | Essential | +| EDI | Electronic Data Interchange | +| EIT | Encoded Information Type | +| EMS | Express Mail Service | +| I/O | Input/Output | +| IP | Interpersonal | +| IPM | Interpersonal Messaging | +| IPMS | Interpersonal Messaging System | +| MD | Management Domain | +| MH | Message Handling | +| MHS | Message Handling System | +| MS | Message Store | +| MT | Message Transfer | +| MTA | Message Transfer Agent | + +| | | +|--------|-------------------------------| +| MTS | Message Transfer System | +| N/A | Not applicable | +| OR | Originator/Recipient | +| OSI | Open Systems Interconnection | +| PD | Physical Delivery | +| PDAU | Physical Delivery Access Unit | +| PDS | Physical Delivery System | +| PFAXAU | Public Telefax Access Unit | +| PM | Per-message | +| PR | Per-recipient | +| PRMD | Private Management Domain | +| PTLXAU | Public Telex Access Unit | +| ROA | Recognized Operating Agency | +| TLMA | Telematic Agent | +| TLXAU | Telex Access Unit | +| UA | User Agent | + +## 5 Conventions + +In this Recommendation the expression "Administration" is used for shortness to indicate a telecommunication administration, a recognized operating agency, and, in the case of intercommunication with public physical delivery service, a postal administration. + +## 6 Purpose + +This Recommendation is one of a set of Recommendations which describes the system model and elements of service of the Message Handling System (MHS) and services. This Recommendation overviews the capabilities of an MHS that are used by Service providers for the provision of public message handling (MH) services to enable users to exchange messages on a store-and-forward basis. + +The message handling system is designed in accordance with the principles of the reference model of Open Systems Interconnection (OSI Reference Model) for ITU-T applications (ITU-T Rec. X.200) and uses the presentation layer services and services offered by other, more general, application service elements. An MHS can be constructed using any network fitting in the scope of OSI. The message transfer service provided by the MTS is application independent. An example of a standardized application is the IPM service. End systems can use the message transfer (MT) service for specific applications that are defined bilaterally. + +Message handling services provided by service providers belong to the group of telematic services defined in the F-series Recommendations. + +Various telematic services and telex (see ITU-T Recs. F.60, F.160, F.200, F.300, etc.), data transmission services (see ITU-T Rec. X.1), or physical delivery services (see ITU-T Rec. F.415) gain access to, and intercommunicate with, the IPM service or intercommunicate with each other, via access units. + +Elements of service are the service features provided through the application processes. The elements of service are considered to be components of the services provided to users and are either elements of a basic service or they are *optional user facilities*, classified either as *essential optional user facilities*, or as *additional optional user facilities*. + +## 7 Functional model of MHS + +The MHS functional model serves as a tool to aid in the development of Recommendations for MHS, and aids in describing the basic concepts that can be depicted graphically. It comprises several different functional components that work together to provide MH services. The model can be applied to a number of different physical and organizational configurations. + +### 7.1 Description of the MHS model + +A functional view of the MHS model is shown in Figure 1. In this model, a user is either a person or a computer process. Users are either direct users (i.e. engage in message handling by direct use of MHS), or are indirect users [i.e. engage in message handling through another communication system (e.g., a physical delivery system) that is linked to MHS]. A user is referred to as either an originator (when sending a message) or a recipient (when receiving a message). Message handling elements of service define the set of message types and the capabilities that enable an originator to transfer messages of those types to one or more recipients. + +An originator prepares messages with the assistance of his user agent. A user agent (UA) is an application process that interacts with the message transfer system (MTS) or a message store (MS), to submit messages on behalf of a single user. The MTS delivers the messages submitted to it, to one or more recipient UAs, access units (AUs), or MSs, and can return notifications to the originator. Functions performed solely by the UA and not standardized as part of the message handling elements of service are called local functions. A UA can accept delivery of messages directly from the MTS, or it can use the capabilities of a MS to receive delivered messages for subsequent retrieval by the UA. + +The MTS comprises a number of message transfer agents (MTAs). Operating together, in a store-and-forward manner, the MTAs transfer messages and deliver them to the intended recipients. + +Access by indirect users of MHS is accomplished by AUs. Delivery to indirect users of MHS is accomplished by AUs, such as in the case of physical delivery, by the physical delivery access unit (PDAU). + +The message store (MS) is an optional general purpose capability of MHS that acts as an intermediary between the UA and the MTA. The MS is depicted in the MHS functional model as shown in Figure 1. The MS is a functional entity whose primary purpose is to store delivered, and, optionally, submitted messages and permit their retrieval by the MS-user (UA). The MS also allows for submission from, and alerting to the MS-user. + +The collection of UAs, MSs, AUs and MTAs is called the message handling system (MHS). + +![Figure 1/F.400/X.400 – MHS functional model. This diagram illustrates the functional architecture of the Message Handling System (MHS). At the top, 'Other telematic services' are shown with two 'User' boxes connected to an oval. Below this is the 'MHS' box, which contains an 'AU' (Accounting Unit) oval at the top. The 'AU' is connected to a central 'MTS' (Message Transfer System) box, which is enclosed in a dashed line. Inside the 'MTS' box, there are three 'MTA' (Message Transfer Agent) rectangles. One 'MTA' is at the top, connected to the 'AU'. Two 'MTA's are at the bottom, connected to each other and to a 'PDUA' (Physical Delivery User Agent) oval at the bottom of the 'MHS' box. On the left side of the 'MTS' box, two 'UA' (User Agent) circles are shown, each connected to a 'User' box and to one of the bottom 'MTA's. On the right side, a 'MS' (Message Store) circle is connected to a 'UA' circle, which is in turn connected to a 'User' box. The 'MS' is also connected to the right 'MTA'. At the bottom of the 'MHS' box, the 'PDUA' is connected to 'Physical delivery services', which are shown with two 'User' boxes connected to an oval. A '(Note)' is placed between the 'PDUA' and the 'Physical delivery services'. The reference 'T0102780-93/d01' is located in the bottom right corner of the diagram area.](9c6461e1e94afae4dec455e69a2ce152_img.jpg) + +Figure 1/F.400/X.400 – MHS functional model. This diagram illustrates the functional architecture of the Message Handling System (MHS). At the top, 'Other telematic services' are shown with two 'User' boxes connected to an oval. Below this is the 'MHS' box, which contains an 'AU' (Accounting Unit) oval at the top. The 'AU' is connected to a central 'MTS' (Message Transfer System) box, which is enclosed in a dashed line. Inside the 'MTS' box, there are three 'MTA' (Message Transfer Agent) rectangles. One 'MTA' is at the top, connected to the 'AU'. Two 'MTA's are at the bottom, connected to each other and to a 'PDUA' (Physical Delivery User Agent) oval at the bottom of the 'MHS' box. On the left side of the 'MTS' box, two 'UA' (User Agent) circles are shown, each connected to a 'User' box and to one of the bottom 'MTA's. On the right side, a 'MS' (Message Store) circle is connected to a 'UA' circle, which is in turn connected to a 'User' box. The 'MS' is also connected to the right 'MTA'. At the bottom of the 'MHS' box, the 'PDUA' is connected to 'Physical delivery services', which are shown with two 'User' boxes connected to an oval. A '(Note)' is placed between the 'PDUA' and the 'Physical delivery services'. The reference 'T0102780-93/d01' is located in the bottom right corner of the diagram area. + +NOTE – Message input from PDS to MHS is for further study. Flow from PD services to the PDUA shown is for the purpose of notifications. + +**Figure 1/F.400/X.400 – MHS functional model** + +## 7.2 Structure of messages + +The basic structure of messages conveyed by the MTS is shown in Figure 2. A message is made up of an envelope and a content. The envelope carries information that is used by the MTS when transferring the message within the MTS. The content is the piece of information that the originating UA wishes to be delivered to one or more recipient UAs. The MTS neither modifies nor examines the content, except for conversion (see clause 16). + +![Figure 2/F.400/X.400 – Basic message structure. This diagram shows the components of a message. On the left is an 'Envelope', represented by a rectangle with a triangle on top, containing the word 'Envelope'. To its right is a plus sign '+'. To the right of the plus sign is a 'Content' block, represented by a solid grey rectangle containing the word 'Content'. The reference 'T0102790-93/d02' is located below the 'Content' block.](898fb89a50d9ec1dfb4e425c816976a7_img.jpg) + +Figure 2/F.400/X.400 – Basic message structure. This diagram shows the components of a message. On the left is an 'Envelope', represented by a rectangle with a triangle on top, containing the word 'Envelope'. To its right is a plus sign '+'. To the right of the plus sign is a 'Content' block, represented by a solid grey rectangle containing the word 'Content'. The reference 'T0102790-93/d02' is located below the 'Content' block. + +**Figure 2/F.400/X.400 – Basic message structure** + +### 7.3 Application of the MHS model + +#### 7.3.1 Physical mapping + +Users access UAs for message processing purposes, for example, to create, present, or file messages. A user can interact with a UA via an input/output (I/O) device or process (e.g., keyboard, display, printer etc.). A UA can be implemented as a (set of) computer process(es) in an intelligent terminal. + +A UA and MTA can be co-located in the same system, or a UA/MS can be implemented in physically separate systems. In the first case, the UA accesses the MT elements of service by interacting directly with the MTA in the same system. In the second case, the UA/MS will communicate with the MTA via standardized protocols specified for MHS. It is also possible for an MTA to be implemented in a system without UAs or MSs. + +Some possible physical configurations are shown in Figures 3 and 4. The different physical systems can be connected by means of dedicated lines or switched network connections. + +![Figure 3/F.400/X.400 – Co-resident UA and MTA. This diagram shows a central 'Processing system' box containing two 'UA' boxes and one 'MTA' box. The 'MTA' is connected to both 'UAs'. Outside the box, on the left, a person icon is connected to an 'I/O device' box, which is in turn connected to the left 'UA'. On the right, another 'I/O device' box is connected to the right 'UA', which is then connected to another person icon. The diagram is labeled T0102800-93/d03.](3da1a07cb87051bf616c9876db958cf0_img.jpg) + +Figure 3/F.400/X.400 – Co-resident UA and MTA. This diagram shows a central 'Processing system' box containing two 'UA' boxes and one 'MTA' box. The 'MTA' is connected to both 'UAs'. Outside the box, on the left, a person icon is connected to an 'I/O device' box, which is in turn connected to the left 'UA'. On the right, another 'I/O device' box is connected to the right 'UA', which is then connected to another person icon. The diagram is labeled T0102800-93/d03. + +Figure 3/F.400/X.400 – Co-resident UA and MTA + +![Figure 4/F.400/X.400 – Stand-alone UA and co-resident MS/MTA and UA/MTA. This diagram shows three main components connected in a line. On the left, a person icon is connected to a box labeled 'Intelligent terminal' containing a 'UA'. This is connected to a 'Processing system' box containing an 'MS' and an 'MTA' connected to each other. This 'Processing system' is connected to another 'Processing system' box containing a 'UA' and an 'MTA' connected to each other. This second 'Processing system' is connected to an 'I/O device' box, which is then connected to a person icon. The diagram is labeled T0102810-93/d04.](e636d7ccca0ad14c6b95201404324823_img.jpg) + +Figure 4/F.400/X.400 – Stand-alone UA and co-resident MS/MTA and UA/MTA. This diagram shows three main components connected in a line. On the left, a person icon is connected to a box labeled 'Intelligent terminal' containing a 'UA'. This is connected to a 'Processing system' box containing an 'MS' and an 'MTA' connected to each other. This 'Processing system' is connected to another 'Processing system' box containing a 'UA' and an 'MTA' connected to each other. This second 'Processing system' is connected to an 'I/O device' box, which is then connected to a person icon. The diagram is labeled T0102810-93/d04. + +Figure 4/F.400/X.400 – Stand-alone UA and co-resident MS/MTA and UA/MTA + +#### 7.3.2 Organizational mapping + +An Administration or organization can play various roles in providing message handling services. An organization in this context can be a company or a non-commercial enterprise. + +The collection of at least one MTA, zero or more UAs, zero or more MSs, and zero or more AUs operated by a service provider or organization constitutes a management domain (MD). An MD managed by a service provider is called an administration management domain (ADMD). An MD managed by an organization other than a service provider is called a private management domain (PRMD). An MD provides message handling services in accordance with the classification of elements of service as described in clause 19. The relationships between management domains are shown in Figure 5. + +#### 7.3.3 Administration management domain + +In one country, one or more ADMDs can exist. An ADMD is characterized by its provision of relaying functions between other management domains and the provision of the message transfer service for the applications provided within the ADMD. + +A service provider can provide access for its users to the ADMD in one or more of the following ways: + +- user to service provider-provided UA; +- private UA to service provider MTA; +- private UA to service provider MS; + +- private MTA to service provider MTA; +- user to service provider-provided AU. + +See also the examples of configurations shown in Figures 3 and 4. + +Service provider-provided UAs can exist as part of an intelligent terminal that the user can use to access the MHS. They can also exist as part of service provider resident equipment which itself is part of the MHS. In this case the user obtains access to the UA via an I/O device. + +In the case of a private UA, the user has a private stand-alone UA which interacts with the service provider-provided MTA or MS, using submission, delivery and retrieval functions. A private, stand-alone UA can be associated with one or more MDs, provided that the required naming conventions are preserved. + +A private MTA as part of a PRMD can access one or more ADMDs in a country, following national regulations. + +Access can also be provided by service provider-provided AUs described in clauses 10 and 11. + +### **7.3.4 Private management domain** + +An organization other than a service provider can have one or more MTA(s), and zero or more UAs, AUs and MSs forming a PRMD which can interact with an ADMD (or other PRMD) on an MD-to-MD (MTA-to-MTA) basis. A PRMD is characterized by the provision of messaging functions within that management domain. + +A PRMD can have access to one or more ADMDs as shown in Figure 5. However, in the case of a specific interaction between a PRMD and an ADMD (such as when a message is transferred between MDs), the PRMD is considered to be associated only with that ADMD. A PRMD may act as a relay to other MDs if national regulations and bilateral agreements permit. + +In the interaction between a PRMD and an ADMD, the ADMD takes responsibility for the actions of the PRMD which are related to the interaction. In addition to ensuring that the PRMD properly provides the message transfer service, the ADMD is responsible for ensuring that the accounting, logging, quality of service, uniqueness of names, and related operations of the PRMD are correctly performed. As a national matter, the name of a PRMD can be either nationally unique or relative to the associated ADMD. If a PRMD is associated with more than one ADMD, the PRMD can have more than one name. + +See Annex E for guidance in the case of multinational PRMDs. + +![Diagram showing relationships between management domains (ADMDs and PRMDs) in Country A and Country B. The diagram illustrates various interconnections between MTAs, UAs, and MSs within and across these domains. Country A contains ADMD 1, ADMD 2, PRMD 1, PRMD 2, and PRMD 4. Country B contains ADMD 3, PRMD 3, and PRMD 5. PRMD 2 is shown spanning the border between Country A and Country B.](e69b9188aa2c14ec6b21c83f711fef65_img.jpg) + +The diagram illustrates the following components and connections: + +- Country A:** + - ADMD 1:** Contains two UAs, an MTA, an MS, and another UA. Its MTA is connected to the MTA in ADMD 2 and the MTA in PRMD 1. + - ADMD 2:** Contains an MTA connected to the MTA in PRMD 2. + - PRMD 1:** Contains two UAs, an MTA, and two UAs. Its MTA is connected to the MTA in ADMD 1 and the MTA in PRMD 4. + - PRMD 2:** Spans the border. It contains an MTA connected to ADMD 2 (in Country A) and another MTA connected to PRMD 3 (in Country B). It also contains a UA and an MS connected to a UA. + - PRMD 4:** Contains an MTA connected to the MTA in PRMD 1. It also contains a UA, an MS, and a UA. +- Country B:** + - ADMD 3:** Contains an MTA connected to the MTA in PRMD 3. It also contains an MTA connected to an MS, which is connected to a UA. + - PRMD 3:** Contains two UAs, an MTA, two UAs, an MS, an MTA, two MSs, and two UAs. Its MTAs are connected to the MTA in ADMD 3 and the MTA in PRMD 5. + - PRMD 5:** Contains two UAs, an MTA, a UA, an MS, and a UA. Its MTAs are connected to the MTA in PRMD 3 and the MTA in PRMD 2. + +Labels: Country A, Country B, T0102820-93/d05 + +Diagram showing relationships between management domains (ADMDs and PRMDs) in Country A and Country B. The diagram illustrates various interconnections between MTAs, UAs, and MSs within and across these domains. Country A contains ADMD 1, ADMD 2, PRMD 1, PRMD 2, and PRMD 4. Country B contains ADMD 3, PRMD 3, and PRMD 5. PRMD 2 is shown spanning the border between Country A and Country B. + +NOTE 1 – This diagram gives examples of possible interconnections. It does not attempt to identify all possible configurations. This Recommendation places no restrictions on interconnections between MDs, although these may be the subject of regulatory agreements within and between countries. + +NOTE 2 – PRMD 1 has connections to two ADMDs within country A; + +PRMD 2 spans a country border, and has connections to an ADMD in each country; + +PRMD 3 has multiple connections to ADMD 3; + +PRMD 4 is only connected to other MDs by relaying through PRMD 1; + +PRMD 5 has connections to other PRMDs, both within the same country (to PRMD 3) and internationally (to PRMD 2) + +NOTE 3 – The lines between MTAs represent logical connections, which implies that the MTAs have the ability to establish associations between themselves when required, using supporting OSI layers over any physical medium. + +NOTE 4 – The shaded boxes surrounding logical components (e.g., UAs, MTAs) represent examples of physically co-located systems. + +**Figure 5/F.400/X.400 – Relationships between management domains** + +## 7.4 Message store (MS) + +Remote UAs can be implemented on a wide variety of equipment, including personal computers of varying capabilities. The MS service can complement a remote UA by providing continuously available storage and delivery services on behalf of a user, for example. + +One MS acts on behalf of only one user, i.e. it does not provide a common or shared MS capability to several users. See also PRMD 3 of Figure 5. + +The MS will store delivered messages and reports. As an option it may also store submitted messages, submitted probes, and draft messages. The MS may also keep a history of messages by storing extracts of previously and currently stored messages in logs. Messages may be grouped in a user-defined and potentially hierarchical structure. + +The MS retrieval capability provides users who subscribe to an MS with basic message retrieval capabilities potentially applicable to all information held by the MS. Figure 6 shows the delivery, and subsequent retrieval of messages that are delivered to an MS, and the submission of messages via the MS. + +When a user subscribes to an MS, all messages destined for the user are delivered to the MS only. The MS-user, if on line, may receive Alerts that announce the delivery of certain messages to the MS. Messages delivered to an MS are considered delivered from the MTS perspective. + +The basic MS is independent of application specific services (see 8.7) and may store messages with all types of content, the type of content being dependent on the type of service. However, it may provide additional capabilities depending on the type of content. + +When an MS-user submits a message, the MS conveys the submission request to the MTS and reports the outcome returned by the MTS to the MS-user. If requested by the MS-user, the MS may expand the message by forwarding parts of delivered or submitted messages that are currently stored in the MS before conveying the submission to the MTS. The MS may also store a copy of the message submitted to the MTS if the submission is successful. The MS service allows the user to transfer a message to the MS for storage as a draft message. The draft message may subsequently be retrieved, or the MS may include its body-parts in a message submitted to the MTS when requested in a message submitted by the MS-user. + +![Diagram illustrating the interaction between a User Agent (UA), a Message Store (MS), and a Message Transfer Agent (MTA). The UA sends an 'Indirect submission' to the MS. The MS sends a 'Submission' to the MTA. The MTA sends a 'Delivery' to the MS. The MS sends a 'Retrieval' to the UA. A reference code 'T0102830-92/d06' is shown below the MTA box.](ab846b81e78dbc8da2a6f9511e2f248a_img.jpg) + +``` +graph LR; UA[UA] -- "Indirect submission" --> MS[MS]; MS -- "Submission" --> MTA[MTA]; MTA -- "Delivery" --> MS; MS -- "Retrieval" --> UA; T0102830-92/d06 +``` + +Diagram illustrating the interaction between a User Agent (UA), a Message Store (MS), and a Message Transfer Agent (MTA). The UA sends an 'Indirect submission' to the MS. The MS sends a 'Submission' to the MTA. The MTA sends a 'Delivery' to the MS. The MS sends a 'Retrieval' to the UA. A reference code 'T0102830-92/d06' is shown below the MTA box. + +**Figure 6/F.400/X.400 – Submission and delivery with an MS** + +The MS-user may be provided with the capability to request the MS service to forward selected messages automatically upon delivery. The MS may also provide automatic deletion of messages after a user specified period of time, or when the message expires, or when the message is rendered obsolete by another message. + +The MS may automatically attach information to a previously submitted message concerning its delivery or non-delivery. The MS may also generate content-specific notifications, acknowledging receipt or acceptance when requested by the user or when the user has retrieved the message. + +The elements of service describing the features of the MS are defined in Annex B and classified in clause 19. Users are provided with the capability based on various criteria, to get counts and lists of messages, to fetch messages, and to delete messages, currently held in the MS. + +Figure 7 depicts a simplified model of the information types stored in the MS, and the functions the MS fulfils. + +![Figure 7/F.400/X.400 – Message Store functional model. This diagram illustrates the functional architecture of a Message Store (MS). At the top, an oval labeled 'MTS' (Message Transfer System) is connected via arrows to three top-level boxes: 'Administration', 'Delivery', and 'Submission'. These three boxes are grouped by a bracket on the right labeled 'MTS-Ports'. Below these, a large dashed rectangle contains the 'MS-Functions'. Inside this rectangle, on the left, are three ovals labeled 'Auto-Actions', a circle labeled 'Timer', and another oval labeled 'Auto-Actions'. Below these is a box labeled 'Registrations'. In the center is a large box containing 'Submitted Messages', 'Delivered Messages', and 'Draft Messages', with a 'Log' box above it. On the right side of the MS-Functions box are two more 'Auto-Actions' ovals. At the bottom of the diagram, an oval labeled 'MS-User' is connected via arrows to three bottom-level boxes: 'Administration', 'Retrieval', and 'MS-Submission'. These three boxes are grouped by a bracket on the right labeled 'MS-Ports'. Arrows with circled letters A through F indicate specific interactions: (A) from MS-User Administration to Draft Messages; (B) from MS-User Retrieval to Draft Messages; (C) from MS-User MS-Submission to Draft Messages; (D) from MS-User MS-Submission to an Auto-Action; (E) from an Auto-Action to Submitted Messages; and (F) from Registrations to an Auto-Action. A small text 'T0103830-94/d07' is located near the bottom right of the diagram.](c5452f95f3b28f1bfe29e84fbc2e1267_img.jpg) + +Figure 7/F.400/X.400 – Message Store functional model. This diagram illustrates the functional architecture of a Message Store (MS). At the top, an oval labeled 'MTS' (Message Transfer System) is connected via arrows to three top-level boxes: 'Administration', 'Delivery', and 'Submission'. These three boxes are grouped by a bracket on the right labeled 'MTS-Ports'. Below these, a large dashed rectangle contains the 'MS-Functions'. Inside this rectangle, on the left, are three ovals labeled 'Auto-Actions', a circle labeled 'Timer', and another oval labeled 'Auto-Actions'. Below these is a box labeled 'Registrations'. In the center is a large box containing 'Submitted Messages', 'Delivered Messages', and 'Draft Messages', with a 'Log' box above it. On the right side of the MS-Functions box are two more 'Auto-Actions' ovals. At the bottom of the diagram, an oval labeled 'MS-User' is connected via arrows to three bottom-level boxes: 'Administration', 'Retrieval', and 'MS-Submission'. These three boxes are grouped by a bracket on the right labeled 'MS-Ports'. Arrows with circled letters A through F indicate specific interactions: (A) from MS-User Administration to Draft Messages; (B) from MS-User Retrieval to Draft Messages; (C) from MS-User MS-Submission to Draft Messages; (D) from MS-User MS-Submission to an Auto-Action; (E) from an Auto-Action to Submitted Messages; and (F) from Registrations to an Auto-Action. A small text 'T0103830-94/d07' is located near the bottom right of the diagram. + +- (A) Fetch, list and summarize +- (B) Modify message attributes and delete +- (C) Draft storage +- (D) Submission to the MTS +- (E) Submission to the MTS with incorporation of parts of stored messages +- (F) Register-MS + +**Figure 7/F.400/X.400 – Message Store functional model** + +The scope of the MS services defined in CCITT Recommendation F.400 (1988) and (1992) and in ISO/IEC 10021-1:1990 was mainly limited to the storage of delivered messages and reports and their subsequent retrieval by the MS-user. The 1994 and later versions of F.400/X.400 and ISO/IEC 10021-1 define new extensions to provide a broader range of service facilities. These enhanced facilities particularly apply in those environments where the MS is used as a personal data base to store, retrieve, modify, and classify a user's messages, often with frequent and long-lasting interaction between the MS-user and MS. Examples of such environments might be found in local area networks, or in environments where the user employs different user agent implementations at different locations to access one MS. In other environments where the MS is used mainly as a temporary storage system, to take delivery of messages and reports and provide for their retrieval by infrequent and short-lasting interactions, these enhanced facilities may not be required. In this latter case, some enhanced facilities may be provided locally by the MS-user itself. + +Consequently, the basic and essential optional requirements defined for the MS in F.400/X.400 and ISO/IEC 10021-1 are the same as those defined in versions published prior to 1994. + +#### 7.4.1 Physical configurations + +The MS can be physically located with respect to the MTA in a number of ways. The MS can be collocated with the UA, collocated with the MTA, or stand-alone. From an external point of view, a collocated UA and MS are indistinguishable from a stand-alone UA. Colocating the MS with the MTA offers significant advantages which will probably make it the predominant configuration. + +#### 7.4.2 Organizational configurations + +Either ADMDs or PRMDs can operate MSs. In the case of service provider-supplied MSs, the subscriber either provides his own UA or makes use of a service provider-supplied UA via an I/O device. In either case, all the subscriber's messages are delivered to the MS for subsequent retrieval. + +The physical and organizational configurations described above are examples only and other equally valid cases can exist. + +### 8 Message transfer service + +The MTS provides the general, application independent, store and forward message transfer service. The elements of service describing the features of the MT service are defined in Annex B, and classified in clause 19. Provision of the public message transfer service by service providers is described in ITU-T Rec. F.410. + +#### 8.1 Submission and delivery + +The MTS provides the means by which UAs exchange messages. There are two basic interactions between MTAs and UAs, or AUs, or MSs: + +- 1) The submission interaction is the means by which an originating UA, AU or MS transfers to an MTA the content of a message and the submission envelope. The submission envelope contains the information that the MTS requires to provide the requested elements of service. +- 2) The delivery interaction is the means by which the MTA transfers to a recipient UA, AU or MS the content of a message plus the delivery envelope. The delivery envelope contains information related to delivery of the message. + +In the submission and delivery interactions, responsibility for the message is passed between the MTA and the UA, AU or MS. + +#### 8.2 Transfer + +Starting at the originator's MTA, each MTA transfers the message to another MTA until the message reaches the recipients' MTA, which then delivers it to the recipient UA or MS using the delivery interaction. + +The transfer interaction is the means by which one MTA transfers to another MTA the content of a message plus the transfer envelope. The transfer envelope contains information related to the operation of the MTS plus information that the MTS requires to provide elements of service requested by the originating UA. + +MTAs transfer messages containing any type of binary coded information. MTAs neither interpret nor alter the content of messages except when performing a conversion. + +#### 8.3 Notifications + +Notifications in the MT service comprise the delivery and non-delivery notifications. When a message, or probe, cannot be delivered by the MTS, a non-delivery notification is generated and returned to the originator in a report signifying this. In addition, an originator can specifically ask for acknowledgment of successful delivery through use of the delivery notification element of service on submission. + +#### 8.4 User agent + +The UA uses the MT service provided by the MTS. A UA is a functional entity by means of which a single direct user engages in message handling. + +UAs are grouped into classes based on the type of content of messages they can handle. The MTS provides a UA with the ability to identify its class when sending messages to other UAs. UAs within a given class are referred to as cooperating UAs since they cooperate with each other to enhance the communication amongst their respective users. + +NOTE – A UA can support more than one type of message content, and hence belong to several UA classes. + +## **8.5 Message store** + +The message store (MS) uses the MT Service provided by the MTS. An MS is a functional entity associated with a user's UA. The user may submit messages through the MS and retrieve messages that have been either delivered to the MS, or submitted by the user. + +## **8.6 Access unit** + +An access unit (AU) uses the MT service provided by the MTS. An AU is a functional entity associated with an MTA to provide for intercommunication between the MHS and another system or service. + +## **8.7 Use of the MTS in the provision of various services** + +The MTS is used by application specific services for the provision of message handling services of various types. The interpersonal messaging service, described in clause 9, is one example of this. Other examples are the Electronic Data Interchange (EDI) messaging service described in ITU-T Rec. F.435, and the voice messaging service described in ITU-T Rec. F.440. Other services (e.g., asynchronous group communications) can be built on the foundation of the MTS, either with corresponding Recommendations or as private applications. + +# **9 IPM service** + +The interpersonal messaging (IPM) service provides a user with features to assist in communicating with other IPM service users. The IPM service uses the capabilities of the MT service for sending and receiving interpersonal messages. The elements of service describing the features of the IPM service are defined in Annex B, and classified in clause 19. The provision of public interpersonal messaging service by service providers is described in ITU-T Rec. F.420. + +## **9.1 IPM service functional model** + +Figure 8 shows the functional model of the IPM service. The UAs used in the IPM service (IPM-UAs) comprise a specific class of cooperating UAs. The optional access units shown (TLMA, PTLXAU, PFAXAU) allow for telex, and telefax users to intercommunicate with the IPM service. The optional physical delivery access unit (PDAU) allows IPM users to send messages to users outside the IPM service who have no access to MHS. The message store can optionally be used by IPM users to take delivery of messages on their behalf. + +![Figure 8/F.400/X.400 – IPM service functional model. This diagram illustrates the functional components of the IPM service. At the top, an 'IPM user' connects to an 'IPM-UA'. Below this, a large oval labeled 'IPM service' contains several components: another 'IPM-UA', an 'MS' (Message Store), a central 'MT service' (Message Transfer Service), and a 'PDAU' (Presentation Data Unit). The 'MT service' is connected to four external services: 'Teletex service' (via 'TLMA'), 'IPM/TTX user' (via 'TLMA'), 'Telex service' (via 'PTLXAU'), and 'Telefax service' (via 'PFAXAU'). Below the 'IPM service' oval, a 'PDS' (Packet Data Service) is shown, which connects to a 'Recipient' at the bottom. A reference code 'T0102840-93/d08' is located in the bottom right corner.](aaf3e6e44cdeabd6d1df869c5f392ea1_img.jpg) + +Figure 8/F.400/X.400 – IPM service functional model. This diagram illustrates the functional components of the IPM service. At the top, an 'IPM user' connects to an 'IPM-UA'. Below this, a large oval labeled 'IPM service' contains several components: another 'IPM-UA', an 'MS' (Message Store), a central 'MT service' (Message Transfer Service), and a 'PDAU' (Presentation Data Unit). The 'MT service' is connected to four external services: 'Teletex service' (via 'TLMA'), 'IPM/TTX user' (via 'TLMA'), 'Telex service' (via 'PTLXAU'), and 'Telefax service' (via 'PFAXAU'). Below the 'IPM service' oval, a 'PDS' (Packet Data Service) is shown, which connects to a 'Recipient' at the bottom. A reference code 'T0102840-93/d08' is located in the bottom right corner. + +Figure 8/F.400/X.400 – IPM service functional model + +## 9.2 Structure of IP-messages + +The IPM class of UAs create messages containing a content specific to the IPM. The specific content that is sent from one IPM-UA to another is a result of an originator composing and sending a message, called an IP-message. The structure of an IP-message as it relates to the basic message structure of MHS is shown in Figure 9. The IP-message is conveyed with an envelope when being transferred through the MTS. + +![Figure 9/F.400/X.400 – IP-message structure. This diagram shows the internal structure of an IP-message. On the left, a vertical stack of components is labeled 'IP-message' with a double-headed arrow. These components are 'Envelope', 'Heading', and 'Body'. The 'Body' contains multiple 'Body part' elements. On the right, under the heading 'Basic message structure', there is a diagram showing an 'Envelope' and a '+' sign above a 'Content' block. Dashed lines connect the 'Envelope' and 'Heading' of the IP-message to the 'Envelope' of the basic message structure, and the 'Body' of the IP-message to the 'Content' block. A reference code 'T0102850-93/d09' is located in the bottom right corner.](e90987faabad6a6665cd8ed1151dc474_img.jpg) + +Figure 9/F.400/X.400 – IP-message structure. This diagram shows the internal structure of an IP-message. On the left, a vertical stack of components is labeled 'IP-message' with a double-headed arrow. These components are 'Envelope', 'Heading', and 'Body'. The 'Body' contains multiple 'Body part' elements. On the right, under the heading 'Basic message structure', there is a diagram showing an 'Envelope' and a '+' sign above a 'Content' block. Dashed lines connect the 'Envelope' and 'Heading' of the IP-message to the 'Envelope' of the basic message structure, and the 'Body' of the IP-message to the 'Content' block. A reference code 'T0102850-93/d09' is located in the bottom right corner. + +Figure 9/F.400/X.400 – IP-message structure + +Figure 10 shows an analogy between a typical office memo, and the corresponding IP-message structure. The IP-message contains information (e.g., To, cc, Subject) provided by the user which is transformed by the IPM-UA into the heading of the IP-message. The main information that the user wishes to communicate (the body of the memo) is contained within the body of the IP-message. In the example shown, the body contains two types of encoded information: text and facsimile, which form what are called, body parts. In general, an IP-message body can consist of a number of body parts, each of which can be of a different encoded information type, such as voice, text, files, facsimile and graphics. + +![Diagram illustrating the analogy between a typical office memo and an IP-message structure. The memo on the left is divided into a header section (To, From, cc, Subject) and a body section (text and attachments). The IP-message structure on the right is divided into a Heading and a Body. The Body is further divided into two 'Body part' boxes. A vertical double-headed arrow on the right indicates the total height of the IP-message. A small diagram at the bottom left shows a central oval with several smaller ovals around it, connected by lines, representing a network or system structure. The text 'T0102860-93/d10' is visible at the bottom right of the diagram.](d17aa1fcc3b86503ad1dd0717a6c34c2_img.jpg) + +The diagram illustrates the structure of an IP-message using a typical office memo as an analogy. On the left, a memo is shown with a header section containing 'To: Dave', 'From: Jim', 'cc: Walter', and 'Subject: Meeting'. Below a dashed line is the body section containing text about a product design review meeting, 'Regards', and 'Attachment'. On the right, the IP-message structure is shown with a 'Heading' section at the top and a 'Body' section below. The 'Body' section contains two 'Body part' boxes. A vertical double-headed arrow on the right is labeled 'IP-message'. At the bottom left, a small diagram shows a central oval with several smaller ovals around it, connected by lines, representing a network or system structure. The text 'T0102860-93/d10' is visible at the bottom right of the diagram. + +Diagram illustrating the analogy between a typical office memo and an IP-message structure. The memo on the left is divided into a header section (To, From, cc, Subject) and a body section (text and attachments). The IP-message structure on the right is divided into a Heading and a Body. The Body is further divided into two 'Body part' boxes. A vertical double-headed arrow on the right indicates the total height of the IP-message. A small diagram at the bottom left shows a central oval with several smaller ovals around it, connected by lines, representing a network or system structure. The text 'T0102860-93/d10' is visible at the bottom right of the diagram. + +Figure 10/F.400/X.400 – IP-message structure for a typical memo + +### 9.3 IP-notifications + +In the IPM service, a user can request a notification of receipt or non-receipt of a message by a recipient. These notifications are requested by an originator and are generated as a result of some recipient action (such as reading/not reading the message). In certain cases, the non-receipt notification is generated automatically by the recipient's UA. + +## 10 Intercommunication with physical delivery services + +### 10.1 Introduction + +The value of message handling systems can be increased by connecting them to physical delivery (PD) systems such as the traditional postal service. This will allow for the physical (e.g., hard copy) delivery of messages originated within MHS to recipients outside of MHS, and in some cases will allow for the return of notifications from the PD service to an MHS originator. The ability for origination of messages in the PD service for submission to MHS through the PDAU is for further study. The capability of intercommunication between PD and MH services is an optional capability of MHS, and is applicable to any application such as IPM. All users of MHS will have the ability to generate messages for subsequent physical delivery. Figure 11 shows the functional model of this interworking. Provision of intercommunication between public message handling services offered by service providers and PD services is described in ITU-T Rec. F.415. The elements of service describing the features of this intercommunication are defined in Annex B and classified in clause 19. + +A physical delivery system is a system, operated by a management domain, that transports and delivers physical messages. A physical message is a physical object comprising a relaying envelope and its content. An example of a physical delivery system (PDS) is the postal service. An example of a physical message is a paper letter and its enclosing paper envelope. + +A physical delivery access unit (PDAU) converts an MH user's message to physical form, a process called physical rendition. An example of this is the printing of a message and its automatic enclosure in a paper envelope. The PDAU passes the physically rendered message to a PDS for further relaying and eventual physical delivery. + +A PDAU can be viewed as a set of UAs, each UA being identified by a postal address. To perform its functions, a PDAU must support submission (Notifications) and delivery interactions with the MTS, and also cooperate with other UAs. MH/PD service intercommunication is thus provided as part of the message transfer service. + +To enable MH users to address messages to be delivered physically by a PDS, an appropriate address format is described in clause 12. + +![Functional model MHS-PDS interworking diagram](474a819357587e34949a3e110ff19b30_img.jpg) + +The diagram illustrates the functional model for MHS-PDS interworking. At the top, a rounded rectangle labeled 'PDS' is connected via a double-headed vertical arrow to a rectangle labeled 'PDAU'. The 'PDAU' is situated within a large grey rectangular area labeled 'MHS'. Below the 'PDAU' is a rectangle labeled 'MTA', connected by a double-headed vertical arrow. This 'MTA' is part of a sub-system labeled 'MTS' (indicated by a label inside the MHS box). Within the 'MTS' box, there are three 'MTA' rectangles: one at the top (connected to the 'PDAU'), one on the left, and one on the right. The top 'MTA' is connected to both the left and right 'MTA's by double-headed horizontal arrows. The left 'MTA' is connected to a 'UA' rectangle on its left and another 'UA' rectangle below it by double-headed arrows. The right 'MTA' is connected to a 'UA' rectangle on its right and an 'AU' rectangle below it by double-headed arrows. The 'AU' is connected to a circle labeled 'Other telematic services'. Below the 'MTS' box, within the 'MHS' area, are two more 'UA' rectangles, each connected to the right 'MTA' by a double-headed vertical arrow. At the bottom right of the diagram, the text 'T0102870-93/d11' is present. + +Functional model MHS-PDS interworking diagram + +Figure 11/F.400/X.400 – Functional model MHS-PDS interworking + +## 10.2 Organizational configurations + +Possible organizational mappings of the functional model described above are shown in Figure 12. In each model (A and B), the term PD domain denotes the domain of responsibility of an organization providing a PD service. In A, the PD domain comprises an MD and a PDS. The boundary between the PD domain and the rest of MHS is a boundary between MDs. In B, the PD domain comprises only the PDS; the PDAU is not part of the PD domain. The boundary between the PD domain and MHS lies at the point where the PDAU passes physical messages to the PDS. + +![Figure 12/F.400/X.400 – Configurations for MH/PD service intercommunication. The diagram shows two configurations, A and B, for intercommunication between a Message Handling System (MHS) and a PD domain. In configuration A, the MHS contains two separate stacks: one with two MTAs and another with a PDAU, MTA, and MD. These two stacks are connected at the MTA level. The PDAU is connected to a PDS in the PD domain. In configuration B, the MHS contains a single stack with a PDAU, MTA, and MD, which is connected to a PDS in the PD domain. A reference code T0102880-93/d12 is shown below configuration B.](11edb7fcedf09ac6a817f8d7b8c61eec_img.jpg) + +Figure 12/F.400/X.400 – Configurations for MH/PD service intercommunication. The diagram shows two configurations, A and B, for intercommunication between a Message Handling System (MHS) and a PD domain. In configuration A, the MHS contains two separate stacks: one with two MTAs and another with a PDAU, MTA, and MD. These two stacks are connected at the MTA level. The PDAU is connected to a PDS in the PD domain. In configuration B, the MHS contains a single stack with a PDAU, MTA, and MD, which is connected to a PDS in the PD domain. A reference code T0102880-93/d12 is shown below configuration B. + +**Figure 12/F.400/X.400 – Configurations for MH/PD service intercommunication** + +## 11 Specialized access + +### 11.1 Introduction + +The functional model of MHS (see Figure 1) contains access units (AUs) to allow access between MHS and other communication systems and services. The model shows a generic access unit between MHS and telematic services. + +Also shown is a physical delivery access unit to allow for physical delivery of MHS messages to recipients without the need for terminal access to MHS. The access to physical delivery services is available to any application carried by the MTS, through a PDAU described in clause 10. + +Other forms of access are described below. + +NOTE – The use of the word "public" in these descriptions refers only to the concept of unrestricted access by any user without advanced registration, in contrast to registered use. The term is not intended to imply that these access units are only provided as part of a public service; they may equally be provided within a private system. + +### 11.2 Telex access + +#### 11.2.1 Registered access to the IPM service + +A telex access unit (TLXAU) is defined in Recommendations to allow the intercommunication between IPM users and telex users. To provide a service with this type of AU is a national matter. + +#### 11.2.2 Non-registered (public) access to the IPM service + +A specialized access unit is defined to allow the intercommunication between IPM users and telex users. This AU provides for public access to the IPM service for telex users who are not registered users of the IPM service, and is called a public telex access unit (PTLXAU). This is shown in Figure 8. The telex users are not subscribers to the IPM service, but use some of the features of the IPM service to pass messages to IPM users. IPM users can also send messages to telex users via this AU. The intercommunication between the IPM service and the telex service is defined in ITU-T Rec. F.421. + +### 11.3 Facsimile access + +#### 11.3.1 Non-registered (public) access from the IPM service + +A specialized access unit is defined to allow the intercommunication between IPM users and telefax users. This AU provides for access from the IPM service to telefax users who are not registered users of the IPM service, and is called a public telefax access unit (PFAXAU). This is shown in Figure 8. IPM users can send messages to telefax users via this AU. Operation of the PFAXAU in the direction telefax to IPM is for further study. The intercommunication between the IPM service and telefax services is defined in ITU-T Rec. F.423. + +## 12 Naming and addressing + +### 12.1 Introduction + +In an MHS, the principal entity that requires naming is the user (the originator and recipient of messages). In addition, distribution lists (DLs) have names for use in MHS. Users of MHS and DLs are identified by originator/recipient (OR) names. OR-names are comprised of directory names and/or OR-addresses, all of which are described in this clause. + +### 12.2 Directory names + +Users of the MH service, and DLs, can be identified by a name, called a directory name. A directory name must be looked up in a directory to find out the corresponding OR-address. + +The structure and components of directory names are described in the ITU-T X.500-series of Recommendations | ISO/IEC 9594. + +A user can access a directory system directly to find the OR-address of a user, or OR-addresses of the members of a DL [both of which are outside the scope of the F.400 and X.400-series Recommendations and ISO/IEC 100211]. As an alternative, a user can use the directory name and have the MHS access a directory to resolve the corresponding OR-address or addresses automatically as described in clause 14. + +An MH user or DL will not necessarily have a directory name, unless they are registered in a directory. As directories become more prevalent, it is expected that directory names will be the preferred method of identifying MHS users to each other. + +### 12.3 OR-names + +Every MH user or DL will have one or more OR-name(s). + +An OR-name comprises a directory name, an OR-address, or both. + +Either or both components of an OR-name can be used on submission of a message. If only the directory name is present, MHS will access a directory to attempt to determine the OR-address, which it will then use to route and deliver the message. If a directory name is absent, it will use the OR-address as given. When both are given on submission, MHS will use the OR-address, but will carry the directory name and present both to the recipient. If the OR-address is invalid, it will then attempt to use the directory name as above. + +### 12.4 OR-addresses + +An OR-address contains information that enables MHS to uniquely identify a user to whom a message is delivered or to whom a notification is returned. (The prefix "OR" recognizes the fact that the user can be acting as either the originator or recipient of the message or notification in question.) + +An OR-address is a collection of information called attributes. ITU-T Rec. X.402 | ISO/IEC 10021-2 specifies a set of standard attributes from which OR-addresses can be constructed. Standard attributes mean that their syntax and semantics are defined in ITU-T Rec. X.402 | ISO/IEC 10021-2. In addition to standard attributes, and to cater for existing messaging systems, there are domain defined attributes whose syntax and semantics are defined by management domains. + +Various forms of OR-addresses are defined, each serving their own purpose. These forms and their purpose are as follows: + +- *Mnemonic OR-address*: Provides a user-friendly means of identifying users in the absence of a directory. It is also used for identifying a distribution list. +- *Terminal OR-address*: Provides a means of identifying users with terminals belonging to various networks. +- *Numeric OR-address*: Provides a means of identifying users by means of numeric keypads. +- *Postal OR-address*: Provides a means of identifying originators and recipients of physical messages. + +--- + +1 See Table 1 for the correspondence of ISO/IEC 10021 to F.400 and X.400-series Recommendations. + +## 13 MHS use of directory + +### 13.1 Introduction + +The directory defined by the ITU-T X.500-series of Recommendations | ISO/IEC 9594 provides capabilities useful in the use and provision of a variety of telecommunication services. This clause describes how a directory can be used in message handling. Details can be found in other X.400-series Recommendations and parts of ISO/IEC 10021. + +The directory capabilities used in message handling fall into the following four categories: + +- a) **User-friendly naming:** The originator or recipient of a message can be identified by means of his directory name, rather than his machine oriented OR-address. At any time MHS can obtain the latter from the former by consulting the directory. +- b) **distribution lists (DLs):** A group whose membership is stored in the directory can be used as a DL. The originator simply supplies the name of the list. At the DL's expansion point MHS can obtain the directory names (and then the OR-addresses) of the individual recipients by consulting the directory. +- c) **Recipient UA capabilities:** MHS capabilities of a recipient (or originator) can be stored in his directory entry. At any time MHS can obtain (and then act upon) those capabilities by consulting the directory. +- d) **Authentication:** Before two MHS functional entities (two MTAs, or a UA and an MTA) communicate with one another, each establishes the identity of the other. This can be done by using authentication capabilities of MHS based on information stored in the directory. + +In addition to the above, one user can directly access the directory, for example, to determine the OR-address or MHS capabilities of another. The recipient's directory name is supplied to the directory, which returns the requested information. + +### 13.2 Functional model + +Both UAs and MTAs can use the directory. A UA can present the directory with the directory name of the intended recipient, and obtain from the directory, the recipient's OR-address. The UA can then supply both the directory name and the OR-address to the MTS. Another UA can supply just the recipient's directory name to the MTS. The MTS would then itself ask the directory for the recipient's OR-address and add it to the envelope. The originating MTA normally carries out the name-to-OR-address look-up using access rights granted to the MTA. + +A functional model depicting the above is shown in Figure 13. + +![Functional model of MHS-directory interworking diagram](1c9a5a80a4ed18fdfda1c8ae915966bf_img.jpg) + +The diagram illustrates the functional model of MHS-directory interworking. On the left, a large circle represents the 'Message transfer system'. Inside it are two rectangular boxes labeled 'MTA' and a third box split diagonally containing 'DUA' in the top-left and 'MTA' in the bottom-right. Arrows show internal connections within the MTS. Outside the MTS circle, there is a box split diagonally containing 'UA' in the top-left and 'DUA' in the bottom-right. On the right, a large circle represents the 'Directory system', containing three oval boxes labeled 'DSA'. Bidirectional and unidirectional arrows show the flow of information between the MTS components, the UA/DUA, and the Directory system DSAs. A small identifier 'T0102890-93/d13' is at the bottom right. + +Functional model of MHS-directory interworking diagram + +**Figure 13/F.400/X.400 – Functional model of MHS-directory interworking** + +### 13.3 Physical configurations + +Some possible physical configurations of the above functional model are shown in Figure 14. Where a directory user agent (DUA) and directory system agent (DSA) reside in physically separate systems, a standard directory protocol, defined in the ITU-T X.500-series of Recommendations | ISO/IEC 9594, governs their interactions. It will often be desirable to physically co-locate a UA or MTA with a DUA/DSA. However, other physical configurations are also possible. + +![Figure 14/F.400/X.400 – Physical configurations for MHS-directory interworking. The diagram shows four physical configurations for MHS-directory interworking. 1) An Intelligent terminal (co-resident UA and DUA) containing DUA and UA. 2) A Processing system (stand-alone DSA) containing a DSA. 3) A Processing system (stand-alone MTA and DUA) containing DUA and MTA. 4) A Processing system (co-resident UA, DUA, MTA and DSA) containing DSA, DUA, MTA, and UA. Arrows show interconnections: UA to DUA, DUA to DSA, MTA to MTA, and UA to an I/O device. Reference code T0102900-93/d14 is in the bottom right.](08dce7ad4c512fdf0c0cde60415fade6_img.jpg) + +Figure 14/F.400/X.400 – Physical configurations for MHS-directory interworking. The diagram shows four physical configurations for MHS-directory interworking. 1) An Intelligent terminal (co-resident UA and DUA) containing DUA and UA. 2) A Processing system (stand-alone DSA) containing a DSA. 3) A Processing system (stand-alone MTA and DUA) containing DUA and MTA. 4) A Processing system (co-resident UA, DUA, MTA and DSA) containing DSA, DUA, MTA, and UA. Arrows show interconnections: UA to DUA, DUA to DSA, MTA to MTA, and UA to an I/O device. Reference code T0102900-93/d14 is in the bottom right. + +**Figure 14/F.400/X.400 – Physical configurations for MHS-directory interworking** + +## 14 Distribution lists in MHS + +### 14.1 Introduction + +The ability to make use of a distribution list (DL) is an optional capability of MHS provided through the MT service. DL expansion allows a sender to have a message transmitted to a group of recipients, by naming the group instead of having to enumerate each of the final recipients. + +### 14.2 Properties of a DL + +The properties of a DL can be described as follows: + +- **DL members:** Users and other DLs that will receive messages addressed to the DL. +- **DL-submit permission:** A list of users and other DLs which are allowed to make use of the DL to send messages to the DL’s members. +- **DL-expansion point:** Each DL has one or more OR-addresses. When a message is addressed to a DL, the OR-address is used to locate an expansion point, which is a domain or MTA where the names of the members of the DL are added to the recipient list. The message is transported to the expansion point before expansion, as shown in Figure 15. There may be more than one MTA capable of acting as the DL expansion point for a particular DL, especially if the Directory is used to store the membership of the DL. +- **DL owner:** A user who is responsible for the management of a DL. + +![Diagram illustrating Distribution list expansion in F.400/X.400. It shows a hierarchy of Message Transfer Agents (MTAs) and Distribution Lists (DLs). At the top, an MTA box has two arrows pointing to User Agents (UAs), labeled with circled numbers 2 and 4. Below this, a second MTA box contains a DL1 node. An arrow labeled 1 points from a UA to DL1. An arrow labeled 2 points from DL1 to a UA. An arrow labeled 3 points from DL1 to a DL2 node in a third MTA box. An arrow labeled 4 points from DL1 to a UA. From DL2, an arrow labeled 4 points to a UA, and another arrow labeled 4 points to another UA. A text label 'T0102910-93/d15' is present near the bottom right of the diagram. Below the diagram is a legend: 1 Submission, 2 Delivery after first DL expansion, 3 Transfer, 4 Delivery after second expansion.](7ae836e598020d937ed1478c2ef13025_img.jpg) + +① Submission + ② Delivery after first DL expansion + ③ Transfer + ④ Delivery after second expansion + +Diagram illustrating Distribution list expansion in F.400/X.400. It shows a hierarchy of Message Transfer Agents (MTAs) and Distribution Lists (DLs). At the top, an MTA box has two arrows pointing to User Agents (UAs), labeled with circled numbers 2 and 4. Below this, a second MTA box contains a DL1 node. An arrow labeled 1 points from a UA to DL1. An arrow labeled 2 points from DL1 to a UA. An arrow labeled 3 points from DL1 to a DL2 node in a third MTA box. An arrow labeled 4 points from DL1 to a UA. From DL2, an arrow labeled 4 points to a UA, and another arrow labeled 4 points to another UA. A text label 'T0102910-93/d15' is present near the bottom right of the diagram. Below the diagram is a legend: 1 Submission, 2 Delivery after first DL expansion, 3 Transfer, 4 Delivery after second expansion. + +**Figure 15/F.400/X.400 – Distribution list expansion** + +### 14.3 Submission + +Submission of a message to a DL is similar to the submission of a message to a user. The originator can include in the DL's OR-name, the directory name, the OR-address, or both (see clause 12 for details). The originator need not be aware that the OR-name used is that of a DL. The originator can, however, through use of the element of service, DL expansion prohibited, prohibit the MTS from expanding a message unknowingly addressed to a DL. + +### 14.4 DL use of a directory + +A directory may or may not be used to store information about the properties of a DL. Among the information that can be stored are the following: DL members, DL owner, DL-submit permission and the DL-expansion point. + +### 14.5 DL expansion + +At the expansion point, the MTA responsible for expanding the DL will: + +- Look up the information about the DL, e.g., in the directory, using access rights granted to the MTA. +NOTE – Since this is done by the MTA at the expansion point, support of DLs in MHS does not require a globally interconnected directory. +- Verify whether expansion is allowed by checking the identity of the sender against the DL's submit permission. +- If expansion is allowed, add the members of the DL to the list of recipients of the message and transmit the message to them. + +### 14.6 Nesting + +A member of a DL can be another DL as shown in Figure 15. In this case the message is forwarded from the expansion point of the parent DL for further expansion. Thus during each expansion, only the members of a single DL are added to the message. + +During expansion of a nested DL, the identity of the parent DL (e.g., DL1 in Figure 15) rather than that of the message originator is compared against the submit permission of the member DL (e.g., DL2 in Figure 15). + +NOTE – DL structures can be defined which reference a particular nested DL more than once at different levels of the nesting. Submission to such a parent DL can cause a recipient to receive multiple copies of the same message. The same result can occur if a message is addressed to multiple DLs which contain a common member. Correlation of such copies can be done at the recipient's UA, and/or in the MS. + +## 14.7 Recursion control + +If a certain DL is directly or indirectly a member of itself (a situation which can validly arise), or when DLs are combined with redirection, then a message might get back to the same list and potentially circulate infinitely. This is detected by the MTS and prevented from occurring. + +## 14.8 Delivery + +On delivery of the message, the recipient will find out that he received the message as a member of a DL, and through which DL, or chain of DLs he received the message. + +## 14.9 Routing loop control + +A message can be originated in one domain/MTA, expanded in a second domain/MTA, and then sent back to a DL member in the first domain/MTA. The MTS will not treat this as a routing loop error. + +## 14.10 Notifications + +Delivery and non-delivery notifications can be generated both at the DL-expansion point (e.g., if submit permission is denied), and at delivery to the ultimate recipient. + +When a message coming from a DL generates a notification, this notification is sent to the DL from which the message came. The DL will then, depending on the policy of the list, forward the notification to the owner of the list, to the DL or originator from which it received the message, or both, as shown in Figure 16. + +NOTE – When notifications are sent to the originator after DL expansion, the originator can receive many delivery/non-delivery notifications for one originator-specified recipient (the DL itself). The originator can even receive more than one notification from an ultimate recipient, if that recipient received the message more than once via different lists. + +![Diagram illustrating DL notifications. The top part shows 'Sending of a message' with a flow from Originator to DL1 to DL2 to Final recipient. The bottom part shows 'Return of a notification' with a flow from Final recipient back to DL2, then to DL1, and finally to Originator. Above DL1 and DL2 are 'Owner' boxes with upward arrows indicating notification paths to them.](df6b7ae63e53e5f93dbd0a70acdae4a6_img.jpg) + +``` + +graph LR + subgraph "Sending of a message:" + O1[Originator] --> DL1_1[DL1] --> DL2_1[DL2] --> FR1[Final recipient] + end + subgraph "Return of a notification:" + FR2[Final recipient] --> DL2_2[DL2] --> DL1_2[DL1] --> O2[Originator] + end + DL1_2 --> Owner1[Owner] + DL2_2 --> Owner2[Owner] + T0102920-93/d16 + +``` + +Diagram illustrating DL notifications. The top part shows 'Sending of a message' with a flow from Originator to DL1 to DL2 to Final recipient. The bottom part shows 'Return of a notification' with a flow from Final recipient back to DL2, then to DL1, and finally to Originator. Above DL1 and DL2 are 'Owner' boxes with upward arrows indicating notification paths to them. + +Figure 16/F.400/X.400 – DL notifications + +## 14.11 DL-handling policy + +An MTA may or may not provide different policies on DL handling. Such policies will control whether notifications generated at delivery to DL members should be propagated back through the previous DL, or to the originator if no such previous DL, and/or to the list owner. If the policy is such that notifications are to be sent only to the list owner, then the originator will receive notifications if requested, only during expansion of that DL. In order to accomplish this restriction, the MTS will, while performing the expansion, reset the notification requests according to the policy for the list. + +# 15 Security capabilities of MHS + +## 15.1 Introduction + +The distributed nature of MHS makes it desirable that mechanisms are available to protect against various security threats that can arise. The nature of these threats and the capabilities to counter them are highlighted below. + +## **15.2 MHS-security threats** + +### **15.2.1 Access threats** + +Invalid user access into MHS is one of the prime security threats to the system. If invalid users can be prevented from using the system, then the subsequent security threat to the system is greatly reduced. + +### **15.2.2 Inter-message threats** + +Inter-message threats arise from unauthorized agents who are external to the message communication, and can manifest themselves in the following ways: + +- Masquerade: A user who does not have proof of whom he is talking to can be easily misled by an impostor into revealing sensitive information. +- Message modification: A genuine message which has been modified by an unauthorized agent while it was transferred through the system can mislead the message recipient. +- Replay: Messages whose originators and contents are genuine can be monitored by an unauthorized agent and could be recorded to be replayed to the message's intended recipient at a later date. This could be done in order to either extract more information from the intended recipient or to confuse him. +- Traffic analysis: Analysis of message traffic between MH users can reveal to an eavesdropper how much data (if any) is being sent between users and how often. Even if the eavesdropper cannot determine the actual contents of the messages, he can still deduce a certain amount of information from the rate of traffic flow (e.g., continuous, burst, sporadic or none). + +### **15.2.3 Intra-message threats** + +Intra-message threats are those performed by the actual message communication participants themselves, and can manifest themselves in the following ways: + +- Repudiation of messages: One of the actual communication participants can deny involvement in the communication. This could have serious implications if financial transactions were being performed via MHS. +- Security level violation: If a management domain within MHS employs different security clearance levels (e.g., public, personal, private and company confidential), then users must be prevented from sending or receiving any messages for which they have an inadequate security clearance level if the management domain's security is not to be compromised. + +### **15.2.4 Data-store threats** + +An MHS has a number of data stores within it that must be protected from the following threats: + +- Modification of routing information: Unauthorized modification of the directory's contents could lead to messages being misrouted or even lost while unauthorized modification to the deferred delivery data store or the hold for delivery data store could mislead or confuse the intended recipient. +- Preplay: An unauthorized agent could make a copy of a deferred delivery message and send this copy to the intended recipient while the original was still being held for delivery in the MTA. This could fool the message recipient into replying to the message originator before the originator was expecting a reply or simply mislead or confuse the original intended message recipient. + +## **15.3 Security model** + +Security features can be provided by extending the capabilities of the components in the message handling system to include various security mechanisms. + +There are two aspects to security in message handling: secure access management and administration, and secure messaging. + +### **15.3.1 Secure access management and administration** + +The capabilities in this subclause cover the establishment of an authenticated association between adjacent components, and the setting up of security parameters for that association. This can be applied to any pair of components in the message handling system: UA/MTA, MTA/MTA, MS/MTA, etc. + +### 15.3.2 Secure messaging + +The capabilities in this subclause cover the application of security features to protect messages in the message handling system in accordance with a defined security policy. This includes elements of service enabling various components to verify the origin of messages and the integrity of their content, and elements of service to prevent unauthorized disclosure of the message content. + +The capabilities in this subclause cover the application of security features to protect messages directly submitted to the message transfer system by a user agent, message store, or an access unit. They do not cover the application of security features to protect communication between users and the message handling system, or MH user-to-MH user communication (a large part of MH user-to-MH user communication is protected between two UAs). Thus they do not apply, for example, to communication between a remote user's terminal and its UA, or to communication between these users' terminal equipment and other users in the MHS. Security capabilities to protect MH user-to-MH user communication are for further study. + +Many of the secure messaging elements of service provide an originator-to-recipient capability, and require the use of user agents with security capabilities. They do not require the use of a message transfer system with security features. [As an example, content confidentiality can be applied by enciphering the message content by the originator, and deciphering by the recipient, with various security parameters transferred within the message envelope. Such a message can be transferred by any MTS which can handle the format of the content (unformatted octets), and transparently handle the security fields in the envelope.] + +Some of the secure messaging elements of service involve an interaction with the message transfer system, and require the use of message transfer agents with security capabilities. (As an example, non-repudiation of submission requires the MTA, to which the message is submitted, to contain mechanisms to generate a proof of submission field.) + +Some of the secure messaging elements of service apply to the MS as well as UAs and MTAs, such as message security labelling. In general, however, the MS is transparent to security features that apply between the originators' and the recipient's UAs. + +The scope of the secure messaging elements of service is given in Table 2. This describes the elements of service in terms of which the MHS component is the "provider" or the "user" of the security service. For example, probe-origin authentication is provided by the originating UA, and can be used by the MTAs through which the probe passes. An overview of these elements of service is given in 15.4. + +This overview describes the use of security services by the UA, MS and the MTA. How these features are applied to access units is for further study. + +**Table 2/F.400/X.400 – Provision and use of secure messaging elements of service by MHS components** + +| Elements of service | Originating MTS user | MTS | Recipient MTS user | +|-------------------------------------------------|----------------------|-----|--------------------| +| Message origin authentication | P | U | U | +| Report origin authentication | U | P | – | +| Probe origin authentication | P | U | – | +| Proof of delivery | U | – | P | +| Proof of submission | U | P | – | +| Secure access management | P | U | P | +| Content integrity | P | – | U | +| Content confidentiality | P | – | U | +| Message flow confidentiality | P | – | – | +| Message sequence integrity | P | – | U | +| Non-repudiation of origin | P | – | U | +| Non-repudiation of submission | U | P | – | +| Non-repudiation of delivery | U | – | P | +| Message security labelling | P | U | U | +| Request for non-repudiation of content received | R | – | U | +| Non-repudiation of content received | U | – | P | +| Request for non-repudiation of IP-notification | R | – | U | + +**Table 2/F.400/X.400 – Provision and use of secure messaging elements of service by MHS components** + +| Elements of service | Originating
MTS user | MTS | Recipient
MTS user | +|---------------------------------------|------------------------------------------------|-----|-----------------------| +| Non-repudiation of IP-notification | U | – | P | +| Request for proof of content received | R | – | U | +| Proof of content received | U | – | P | +| Request for proof of IP-notification | R | – | U | +| Proof of IP-notification | U | – | P | +| P | The MHS component is a provider of the service | | | +| U | The MHS component is a user of the service | | | + +## 15.4 MHS-security capabilities + +The elements of service describing the security features of MHS are defined in Annex B, and classified in clause 19. An overview of these capabilities is as follows: + +**15.4.1 message origin authentication:** Enables the recipient, or any MTA through which the message passes, to authenticate the identity of the originator of a message. + +**15.4.2 report origin authentication:** Allows the originator to authenticate the origin of a delivery/non-delivery report. + +**15.4.3 probe origin authentication:** Enables any MTA through which the probe passes to authenticate the origin of the probe. + +**15.4.4 proof of delivery:** Enables the originator of a message to authenticate the delivered message and its content, and the identity of the recipient(s). + +**15.4.5 proof of submission:** Enables the originator of a message to authenticate that the message was submitted to the MTS for delivery to the originally specified recipient(s). + +**15.4.6 secure access management:** Provides for authentication between adjacent components, and the setting up of the security context. + +**15.4.7 content integrity:** Enables the recipient to verify that the original content of a message has not been modified. + +**15.4.8 content confidentiality:** Prevents the unauthorized disclosure of the content of a message to a party other than the intended recipient. + +**15.4.9 message flow confidentiality:** Allows the originator of a message to conceal the message flow through MHS. + +**15.4.10 message sequence integrity:** Allows the originator to provide to a recipient proof that the sequence of messages has been preserved. + +**15.4.11 non-repudiation of origin:** Provides the recipient(s) of a message with proof of origin of the message and its content which will protect against any attempt by the originator to falsely deny sending the message or its content. + +**15.4.12 non-repudiation of delivery:** Provides the originator of a message with proof of delivery of the message which will protect against any attempt by the recipient(s) to falsely deny receiving the message or its content. + +**15.4.13 non-repudiation of submission:** Provides the originator of a message with proof of submission of the message, which will protect against any attempt by the MTS to falsely deny that the message was submitted for delivery to the originally specified recipient(s). + +**15.4.14 message security labelling:** Provides a capability to categorize a message, indicating its sensitivity, which determines the handling of a message in line with the security policy in force. + +## 15.5 Security management + +Aspects of an asymmetric key management scheme to support the above features are provided by the directory system authentication framework, described in ITU-T Rec. X.509 | ISO/IEC 9594-8. The Directory stores certified copies of public keys for MH users which can be used to provide authentication and to facilitate key exchange for use in data confidentiality and data integrity mechanisms. The certificates can be read from the Directory using the directory access protocol described in ITU-T Rec. X.519 | ISO/IEC 9594-5. + +Other types of key management schemes, including symmetric encryption, to support the security features are for further study. + +## 15.6 MHS-security dependencies + +If, as a result of using MHS security capabilities, there are any dependencies, consequences or restrictions on other MHS capabilities (e.g., on distribution lists or conversion), then these shall be defined by the security policy. + +The abstract security model for message transfer is described in clause 10/X.402. In particular, 10.1/X.402 describes the concept of security policy. + +## 15.7 IPM security + +The elements of service describing the additional security features of IPMS are defined in Annex B, and classified in clause 19. An overview of these capabilities is as follows: + +| | | +|-------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Request for non-repudiation of content received | Enables the originator of an IP-message to request that the recipient of the IP-message provides irrevocable proof that the recipient received and verified the integrity of the IP-message content. This element of service provides only an indication of the originator's request. Fulfilment of the request requires support of the non-repudiation of content received element of service. | +| Non-repudiation of content received | Provides the originator of an IP-message with irrevocable proof that the recipient validated the security features of the IP-message. This provides a proof with non-repudiation properties of the authenticity and integrity of the contents of the IP-message as it was received by the recipient. The recipient is required to fulfil the request for this element of service only when the UA is subject to a security policy which mandates the support of this Element of Service. | +| Request for non-repudiation of IP-notification | Enables the originator of an IP-message to request that the recipient of the IP-message provides irrevocable proof that the recipient received the IP-message and the recipient of the IP-message originated the IP-notification. This element of service provides only an indication of the originator's request. Fulfilment of the request requires support of the non-repudiation of IP-notification element of service. | +| Non-repudiation of IP-notification | Provides the originator of an IP-message with irrevocable proof that the IP-message was received by the recipient and that the recipient originated the resulting IP-notification. This provides a signed receipt of the IP-message identifier which has non-repudiation properties. The recipient is required to fulfil the request for this Element of Service only when the UA is subject to a security policy which mandates the support of this element of service. | +| Request for proof of content received | Enables the originator of an IP-message to request that the recipient of the IP-message provides proof that the recipient received and verified the integrity of the IP-message contents. This element of service provides only an indication of the originator's request. Fulfilment of the request requires support of the proof of content received element of service. | +| Proof of content received | Provides the originator of an IP-message with proof that the recipient validated the security features of the IP-message. This element of service provides proof of the authenticity and integrity of the contents of the IP-message as it was received by the recipient. The recipient is required to fulfil the request for this element of service only when the UA is subject to a security policy which mandates the support of this element of service. | +| Request for proof of IP-notification | Enables the originator of an IP-message to request that the recipient of the IP-message provides proof that the recipient received the IP-message and the recipient of the IP-message originated the IP-notification. This element of service provides only an indication of the originator's request. Fulfilment of the request requires support of the proof of IP-notification element of service. | +| Proof of IP-notification | Provides the originator of an IP-message with proof that the IP-message was received by the recipient and that the recipient originated the resulting IP-notification. This provides a signed receipt of the IP-message identifier. The recipient is required to fulfil the request for this element of service only when the UA is subject to a security policy which mandates the support of this element of service. | + +The application of security features between the IPM-UA and the MHS user, such as user authentication and local access control, is a local matter. Information required to implement local access control may be conveyed between the originator and recipient IPM-UAs using the message security labelling element of service. The syntax and semantics of private and local access control information may be defined by registration of security policies and security categories. + +The above IPM-UA security features use the MHS security capabilities defined in ITU-T Rec. X.411 | ISO/IEC 10021-4 provided by the user of the MTS. They do not require any extensions to the MTS defined in ITU-T Rec. X.411 | ISO/IEC 10021-4. IPM-UAs which support the above security features are required to support the related + +protocol extensions and associated procedures defined in ITU-T Rec. X.420 | ISO/IEC 10021-7. The IPM-UA security features use the MTS-user security mechanisms defined in ITU-T Rec. X.411 | ISO/IEC 10021-4 such as **Content-integrity-check**, **Message-origin-authentication-check**, and **Message-token**. The IPM-UA security features do not define any additional security mechanisms. + +NOTE – In case of the use of a notarizing function, non-repudiation can be provided implicitly, and is not reflected in any specific protocol elements. + +**Table 3/F.400/X.400 – Provision and use of additional secure messaging elements of service by IPM-UAs** + +| Elements of service | IP-message originator | MTS | IP-message recipient | +|-------------------------------------------------|-----------------------|-----|----------------------| +| Request for non-repudiation of content received | Requester | – | User | +| Non-repudiation of content received | User | – | Provider | +| Request for non-repudiation of IP-notification | Requester | – | User | +| Non-repudiation of IP-notification | User | – | Provider | +| Request for proof of content received | Requester | – | User | +| Proof of content received | User | – | Provider | +| Request for proof of IP-notification | Requester | – | User | +| Proof of IP-notification | User | – | Provider | + +## 16 Conversion in MHS + +The MTS provides conversion functions to allow users to input messages in one or more encoded formats, called encoded information types (EITs) and have them delivered in other EITs to cater to users with various UA capabilities and terminal types. This capability is inherent in the MTS and increases the possibility of delivery by tailoring the message to the recipient's terminal capabilities. The EITs standardized in MHS are listed in ITU-T Rec. X.411 | ISO/IEC 10021-4. Conversions and the use of the elements of service relating to conversion are available for EITs not defined in ITU-T Rec. X.411 | ISO/IEC 10021-4, but supported by certain domains, either bilaterally between these domains or within a domain itself. + +MH users have some control over the conversion process through various elements of service as described in Annex B. These include the ability for a user to explicitly request the conversion required or as a default to let the MTS determine the need for conversion, and the type of conversion performed. Users also have the ability to request that conversion not be performed or that conversion not be performed if loss of information will result. When the MTS performs conversion on a message, it informs the UA to whom the message is delivered that conversion took place and what the original EITs were. + +The conversion process for IP-messages can be performed on body parts of specific types if they are present in a message. The general aspects of conversion and the specific conversion rules for conversion between different EITs are detailed in ITU-T Rec. X.408. ITU-T Rec. X.408 deals with conversion including the following: IA5 Text, G3Fax, G4 Class1, and Videotex. + +## 17 Use of the MHS in provision of public services + +The message handling system is used in the provision of public MH services that are offered by service providers for use by their subscribers. These public MH services are defined in the F.400-series of Recommendations and include: + +- the public message transfer service (ITU-T Rec. F.410); +- the public interpersonal messaging service (ITU-T Rec. F.420). + +In addition, complementary public services are offered by service providers to allow for the intercommunication between ITU-T services and the public MH services mentioned above, as follows: + +- intercommunication between the IPM service and the telex service (ITU-T Rec. F.421); +- intercommunication between the IPM service and telefax services (ITU-T Rec. F.423); +- intercommunication with public physical delivery services (ITU-T Rec. F.415). + +A Recommendation describing the naming and addressing aspects for public MH services exists as follows: + +- naming and addressing for public message handling services (ITU-T Rec. F.401). + +See also ITU-T Recs F.435 and F.440. + +## 18 Elements of service – Purpose + +Elements of service are particular features, functions, or capabilities of MHS. All the elements of service applicable for MHS are defined in Annex B, where they are listed in alphabetical order with a corresponding reference number. The realization of these elements of service in MHS is described in other Recommendations in the X.400-Series and parts of ISO/IEC 10021. + +Elements of service are associated with the various services provided in MHS. There are elements of service for the message transfer service which provide for a basic capability for sending and receiving messages between UAs. There are elements of service for the interpersonal messaging service which provide for the sending and receiving of messages between a particular class of UAs called IPM-UAs. There are elements of service for the physical delivery service, enabling MH users to send messages and have them delivered in a physical medium to non-MH users. There are elements of service specifically available for the use of message stores. + +The elements of service for the IPM service include those available for the MT service, the PD service and the message store as well as specific ones applicable to the IPM service. + +Table 4 lists all the elements of service available in MHS, shows what services they are associated with of the presently defined services, MT service, IPM service, PD service, and MS service, and gives the corresponding reference number to the definition in Annex B. Elements of service relevant to the IPM message store are marked on both the IPM and MS columns. + +**Table 4/F.400/X.400 – MHS elements of service** + +| Elements of service | MT | IPM | PD | MS | Annex B reference | +|----------------------------------------|----|-----|----|----|-------------------| +| Access management | X | | | | B.1 | +| Additional physical rendition | | | X | | B.2 | +| Alternate recipient allowed | X | | | | B.3 | +| Alternate recipient assignment | X | | | | B.4 | +| Authorization time indication | | X | | | B.5 | +| Authorizing users indication | | X | | | B.6 | +| Auto-acknowledgment of IP-messages | | X | | X | B.7 | +| Auto-action log | | | | X | B.8 | +| Auto-advise | | X | | X | B.9 | +| Auto-assignment of annotations | | | | X | B.10 | +| Auto-assignment of group names | | | | X | B.11 | +| Auto-assignment of storage period | | | | X | B.12 | +| Auto-correlation of IP-messages | | X | | X | B.13 | +| Auto-correlation of IP-notifications | | X | | X | B.14 | +| Auto-correlation of reports | | | | X | B.15 | +| Auto-deletion after storage period | | | | X | B.16 | +| Auto-discarding of IP-messages | | X | | X | B.17 | +| Auto-forwarded indication | | X | | | B.18 | +| Auto-forwarding of IP-messages | | X | | X | B.19 | +| Auto-submitted indication | | X | | | B.20 | +| Basic physical rendition | | | X | | B.21 | +| Blind copy recipient indication | | X | | | B.22 | +| Body part authentication and integrity | | X | | | B.23 | +| Body part encryption | | X | | | B.24 | + +**Table 4/F.400/X.400 – MHS elements of service** + +| Elements of service | MT | IPM | PD | MS | Annex B reference | +|-------------------------------------------------------|-----------|------------|-----------|-----------|--------------------------| +| Circulation list recipients indication | | X | | | B.25 | +| Content confidentiality | X | | | | B.26 | +| Content integrity | X | | | | B.27 | +| Content type indication | X | | | | B.28 | +| Conversion prohibition | X | | | | B.29 | +| Conversion prohibition in case of loss of information | X | | | | B.30 | +| Converted indication | X | | | | B.31 | +| Counter collection | | | X | | B.32 | +| Counter collection with advice | | | X | | B.33 | +| Cover page suppression | X | | | | B.34 | +| Cross-referencing indication | | X | | | B.35 | +| Deferred delivery | X | | | | B.36 | +| Deferred delivery cancellation | X | | | | B.37 | +| Delivery log | | | | X | B.38 | +| Delivery notification | X | | | | B.39 | +| Delivery time stamp indication | X | | | | B.40 | +| Delivery via bureaufax service | | | X | | B.41 | +| Designation of recipient by directory name | X | | | | B.42 | +| Disclosure of other recipients | X | | | | B.43 | +| Distribution codes indication | | X | | | B.44 | +| DL-exempted recipients | X | | | | B.45 | +| DL-expansion history indication | X | | | | B.46 | +| DL-expansion prohibited | X | | | | B.47 | +| EMS (Express Mail Service) | | | X | | B.48 | +| Expiry date indication | | X | | | B.49 | +| Explicit conversion | X | | | | B.50 | +| Forwarded IP-message indication | | X | | | B.51 | +| Grade of delivery selection | X | | | | B.52 | +| Hold for delivery | X | | | | B.53 | +| Implicit conversion | X | | | | B.54 | +| Importance indication | | X | | | B.55 | +| Incomplete copy indication | | X | | | B.56 | +| Information category indication | | X | | | B.57 | +| IP-message action status | | X | | X | B.58 | +| IP-message identification | | X | | | B.59 | +| IP-message security labelling | | X | | | B.60 | +| Language indication | | X | | | B.61 | + +**Table 4/F.400/X.400 – MHS elements of service** + +| Elements of service | MT | IPM | PD | MS | Annex B reference | +|-----------------------------------------------|-----------|------------|-----------|-----------|--------------------------| +| Latest delivery designation | X | | | | B.62 | +| Manual handling instructions indication | | X | | | B.63 | +| Message flow confidentiality | X | | | | B.64 | +| Message identification | X | | | | B.65 | +| Message origin authentication | X | | | | B.66 | +| Message security labelling | X | | | | B.67 | +| Message sequence integrity | X | | | | B.68 | +| MS register | | | | X | B.69 | +| Multi-destination delivery | X | | | | B.70 | +| Multi-part body | | X | | | B.71 | +| Non-delivery notification | X | | | | B.72 | +| Non-receipt notification request indication | | X | | | B.73 | +| Non-repudiation of content received | | X | | | B.74 | +| Non-repudiation of delivery | X | | | | B.75 | +| Non-repudiation of IP-notification | | X | | | B.76 | +| Non-repudiation of origin | X | | | | B.77 | +| Non-repudiation of submission | X | | | | B.78 | +| Obsoleting indication | | X | | | B.79 | +| Ordinary mail | | | X | | B.80 | +| Original encoded information types indication | X | | | | B.81 | +| Originator indication | | X | | | B.82 | +| Originator reference indication | | X | | | B.83 | +| Originator requested alternate recipient | X | | | | B.84 | +| Physical delivery notification by MHS | | | X | | B.85 | +| Physical delivery notification by PDS | | | X | | B.86 | +| Physical forwarding allowed | | | X | | B.87 | +| Physical forwarding prohibited | | | X | | B.88 | +| Precedence indication | | X | | | B.89 | +| Prevention of non-delivery notification | X | | | | B.90 | +| Primary and copy recipients indication | | X | | | B.91 | +| Probe | | X | | | B.92 | +| Probe origin authentication | X | | | | B.93 | +| Proof of content received | | X | | | B.94 | +| Proof of delivery | X | | | | B.95 | +| Proof of IP-notification | | X | | | B.96 | +| Proof of submission | X | | | | B.97 | +| Receipt notification request indication | | X | | | B.98 | + +**Table 4/F.400/X.400 – MHS elements of service** + +| Elements of service | MT | IPM | PD | MS | Annex B reference | +|---------------------------------------------------------|-----------|------------|-----------|-----------|--------------------------| +| Redirection disallowed by originator | X | | | | B.99 | +| Redirection of incoming messages | X | | | | B.100 | +| Registered mail | | | X | | B.101 | +| Registered mail to addressee in person | | | X | | B.102 | +| Reply request indication | | X | | | B.103 | +| Replying IP-message indication | | X | | | B.104 | +| Report origin authentication | X | | | | B.105 | +| Request for forwarding address | | | X | | B.106 | +| Request for non-repudiation of content received | | X | | | B.107 | +| Request for non-repudiation of IP-notification | | X | | | B.108 | +| Request for proof of content received | | X | | | B.109 | +| Request for proof of IP-notification | | X | | | B.110 | +| Requested preferred delivery method | X | | | | B.111 | +| Restricted delivery | X | | | | B.112 | +| Return of content | X | | | | B.113 | +| Secure access management | X | | | | B.114 | +| Sensitivity indication | | X | | | B.115 | +| Special delivery | | | X | | B.116 | +| Storage of draft messages | | | | X | B.117 | +| Storage on submission | | | | X | B.118 | +| Storage period assignment | | | | X | B.119 | +| Stored message alert | | | | X | B.120 | +| Stored message annotation | | | | X | B.121 | +| Stored message deletion | | | | X | B.122 | +| Stored message fetching | | | | X | B.123 | +| Stored message grouping | | | | X | B.124 | +| Stored message listing | | | | X | B.125 | +| Stored message summary | | | | X | B.126 | +| Subject indication | | X | | | B.127 | +| Submission log | | | | X | B.128 | +| Submission of IP-messages incorporating stored messages | | X | | X | B.129 | +| Submission time stamp indication | X | | | | B.130 | +| Typed body | | X | | | B.131 | +| Undeliverable mail with return of physical message | | | X | | B.132 | +| Use of distribution list | X | | | | B.133 | +| User/UA capabilities registration | X | | | | B.134 | + +## 19 Elements of service – Classification + +### 19.1 Purpose of classification + +The elements of service of MHS are classified either as belonging to a basic (also called base for PD and MS) service, or as optional user facilities. Elements of service belonging to a basic service are inherently part of that service; they constitute the basic service and are always provided and available for use of that service. + +Other elements of service, called optional user facilities, can be selected by the subscriber or user, either on a per-message basis, or for an agreed contractual period of time. Each optional user facility is classified as either essential or additional. Essential (E) optional user facilities are to be made available to all MH users. Additional (A) optional user facilities can be made available for national use, and for international use on the basis of bilateral agreement. + +### 19.2 Basic message transfer service + +The basic MT service enables a UA to submit and to have messages delivered to it. If a message cannot be delivered, the originating UA is so informed through a non-delivery notification. Each message is uniquely and unambiguously identified. To facilitate meaningful communication, a UA can specify the encoded information type(s) that can be contained in messages which are delivered to it. The content type and original encoded information type(s) of a message and an indication of any conversions that have been performed, and the resulting encoded information type(s), are supplied with each delivered message. In addition, the submission time and delivery time are supplied with each message. The MT elements of service belonging to the basic MT service are listed in Table 5. + +**Table 5/F.400/X.400 – Elements of service belonging to the basic MT service** + +| Elements of service | Annex B reference | +|-----------------------------------------------|-------------------| +| Access management | B.1 | +| Content type indication | B.28 | +| Converted indication | B.31 | +| Delivery time stamp indication | B.40 | +| Message identification | B.65 | +| Non-delivery notification | B.72 | +| Original encoded information types indication | B.81 | +| Submission time stamp indication | B.130 | +| User/UA capabilities registration | B.134 | + +### 19.3 MT service optional user facilities + +Optional user facilities for the MT service can be selected on a per-message basis, or for an agreed period of time. Each optional user facility is classified as either essential or additional as described in 19.1. Table 6 lists the elements of service comprising the optional user facilities of the MT service with their classification and their availability [per-message (PM); contractual agreement (CA)]. Optional user facilities for the PD service and the message store, while forming a part of the MT service optional user facilities, are not listed in Table 6 because they are subject to either a PDAU or an MS being supplied, and are given separate classifications in Tables 7 to 10. A security policy will define when the security related MT service optional user facilities are invoked. + +**Table 6/F.400/X.400 – MT service optional user facilities** + +| Elements of service | Classification | Available | Annex B reference | +|-------------------------------------------------------|-----------------------|------------------|--------------------------| +| Alternate recipient allowed | E | PM | B.3 | +| Alternate recipient assignment | A | CA | B.4 | +| Content confidentiality | A | PM | B.26 | +| Content integrity | A | PM | B.27 | +| Conversion prohibition | E | PM | B.29 | +| Conversion prohibition in case of loss of information | A | PM | B.30 | +| Cover page suppression | A | PM | B.34 | +| Deferred delivery | E | PM | B.36 | +| Deferred delivery cancellation | E | PM | B.37 | +| Delivery notification | E | PM | B.39 | +| Designation of recipient by directory name | A | PM | B.42 | +| Disclosure of other recipients | E | PM | B.43 | +| DL-exempted recipients | A | PM | B.45 | +| DL-expansion history indication | A | PM | B.46 | +| DL-expansion prohibited | A | PM | B.47 | +| Explicit conversion | A | PM | B.50 | +| Grade of delivery selection | E | PM | B.52 | +| Hold for delivery | A | CA | B.53 | +| Implicit conversion | A | CA | B.54 | +| Latest delivery designation | A | PM | B.62 | +| Message flow confidentiality | A | PM | B.64 | +| Message origin authentication | A | PM | B.66 | +| Message security labelling | A | PM | B.67 | +| Message sequence integrity | A | PM | B.68 | +| Multi-destination delivery | E | PM | B.70 | +| Non-repudiation of delivery | A | PM | B.75 | +| Non-repudiation of origin | A | PM | B.77 | +| Non-repudiation of submission | A | PM | B.78 | +| Originator requested alternate recipient | A | PM | B.84 | +| Prevention of non-delivery notification | A | PM | B.90 | +| Probe | A | PM | B.92 | +| Probe origin authentication | A | PM | B.93 | +| Proof of delivery | A | PM | B.95 | +| Proof of submission | A | PM | B.97 | +| Redirection disallowed by originator | A | PM | B.99 | +| Redirection of incoming messages | A | PM | B.100 | +| Report origin authentication | A | PM | B.105 | +| Requested preferred delivery method | A (Note) | PM | B.111 | +| Restricted delivery | A | PM | B.112 | +| Return of content | A | PM | B.113 | +| Secure access management | A | CA | B.114 | +| Use of distribution list | A | PM | B.133 | + +NOTE – Does not imply the provision of all delivery methods which may be requested. + +## 19.4 Base MH/PD service intercommunication + +The base MH/PD service intercommunication can be supplied, to enhance the MT service, and to enable messages to be delivered to recipients in a physical (typically hard copy) format via a physical delivery service such as the postal service. This capability is applicable for use by any application making use of the MT service. The MH/PD elements of service belonging to the base MH/PD service intercommunication are available on a per-recipient basis and are listed in Table 7. When this intercommunication is provided, through a PDAU, all the elements of service shown in Table 7 shall be supported. + +**Table 7/F.400/X.400 – Elements of service belonging to the base MH/PD service intercommunication** + +| Elements of service | Annex B reference | +|----------------------------------------------------|-------------------| +| Basic physical rendition | B.21 | +| Ordinary mail | B.80 | +| Physical forwarding allowed | B.87 | +| Undeliverable mail with return of physical message | B.132 | + +## 19.5 Optional user facilities for MH/PD service intercommunication + +Base MH/PD elements of service (see 19.4) together with the optional user facilities can be used for the provision of the MH/PD service intercommunication. This capability is applicable for use by any application making use of the enhanced MT service. These optional user facilities can be selected on a per-recipient basis and are listed in Table 8. + +**Table 8/F.400/X.400 – Optional user facilities for MH/PD service intercommunication** + +| Elements of service | Classification | Annex B reference | +|----------------------------------------|----------------|-------------------| +| Additional physical rendition | A | B.2 | +| Counter collection | E | B.32 | +| Counter collection with advice | A | B.33 | +| Delivery via bureaufax service | A | B.41 | +| EMS (Express Mail Service) (Note) | E | B.48 | +| Physical delivery notification by MHS | A | B.85 | +| Physical delivery notification by PDS | A | B.86 | +| Physical forwarding prohibited | A | B.88 | +| Registered mail | A | B.101 | +| Registered mail to addressee in person | A | B.102 | +| Request for forwarding address | A | B.106 | +| Special delivery (Note) | E | B.116 | + +NOTE – At least one or the other element of service shall be supported by the PDAU and the associated PDS. + +## 19.6 Base message store + +The base message store is optionally available to provide for storage and management of incoming messages acting as an intermediary between a UA and an MTA. The MS is applicable for use in any application making use of the MT service. The elements of service belonging to the base message store are listed in Table 9. When an MS is provided, each element of service shown in Table 9 shall be supported for every type of message (delivered-message, submission-log, draft-messages, etc.) stored in the MS to which that element of service is applicable. + +**Table 9/F.400/X.400 – Base message store** + +| Elements of service | Annex B reference | +|-------------------------|-------------------| +| MS register | B.69 | +| Stored message deletion | B.122 | +| Stored message fetching | B.123 | +| Stored message listing | B.125 | +| Stored message summary | B.126 | + +## 19.7 MS optional user facilities + +Base MS elements of service (see 19.6) together with the optional user facilities can be used for enhanced use of a message store. The enhanced MS is applicable for use in any application making use of the MT service. The elements of service comprising the MS optional user facilities are listed in Table 10. + +**Table 10/F.400/X.400 – MS optional user facilities** + +| Elements of service | Classification | Annex B reference | +|------------------------------------|----------------|-------------------| +| Auto-action log | A | B.8 | +| Auto-assignment of annotations | A | B.10 | +| Auto-assignment of group names | A | B.11 | +| Auto-assignment of storage period | A | B.12 | +| Auto-correlation of reports | A | B.15 | +| Auto-deletion after storage period | A | B.16 | +| Delivery log | A | B.38 | +| Storage of draft messages | A | B.117 | +| Storage on submission | A | B.118 | +| Storage period assignment | A | B.119 | +| Stored message alert | A | B.120 | +| Stored message annotation | A | B.121 | +| Stored message grouping | A | B.124 | +| Submission log | A | B.128 | + +## 19.8 Basic interpersonal messaging service + +The basic IPM service, which makes use of the MT service, enables a user to send and receive IP-messages. A user prepares IP-messages with the assistance of his user agent (UA). User agents cooperate with each other to facilitate communication between their respective users. To send an IP-message, the originating user submits the message to his UA specifying the OR-name of the recipient who is to receive the IP-message. The IP-message, which has an identifier conveyed with it, is then sent by the originator's UA to the recipient's UA via the message transfer service. + +Following a successful delivery to the recipient's UA, the IP-message can be received by the recipient. To facilitate meaningful communication, a recipient can specify the encoded information type(s) contained in IP-messages that he will allow to be delivered to his UA. The original encoded information type(s) and an indication of any conversions that have been performed and the resulting encoded information type(s) are supplied with each delivered IP-message. In addition, the submission time and delivery time are supplied with each IP-message. Non-delivery notification is provided with the basic service. The IPM elements of service belonging to the basic IPM service are listed in Table 11. + +**Table 11/F.400/X.400 – Elements of service belonging to the basic IPM service** + +| Elements of service | Annex B reference | +|-----------------------------------------------|-------------------| +| Access management | B.1 | +| Content type indication | B.28 | +| Converted indication | B.31 | +| Delivery time stamp indication | B.40 | +| IP-message identification | B.59 | +| Message identification | B.65 | +| Non-delivery notification | B.72 | +| Original encoded information types indication | B.81 | +| Submission time stamp indication | B.130 | +| Typed body | B.131 | +| User/UA capabilities registration | B.134 | + +### 19.9 IPM service optional user facilities + +A set of the elements of service of the IPM service are optional user facilities. The optional user facilities of the IPM service, which can be selected on a per-message basis or for an agreed contractual period of time, are listed in Tables 12 and 13, respectively. Local user facilities can be usefully provided in conjunction with some of these user facilities. + +The optional user facilities of the IPM service that are selected on a per-message basis are classified for both origination and reception by UAs. If an MD offers these optional user facilities for origination by UAs, then a user is able to create and send IP-messages according to the procedures defined for the associated element of service. If an MD offers these optional user facilities for reception by UAs, MSs and AUs, then the receiving UA, MS and PDAU will be able to receive and recognize the indication associated with the corresponding element of service and to inform the user of the requested optional user facility. Each optional user facility is classified as additional (A) or essential (E) for UAs from these two perspectives. A security policy will define and determine when the security related IPM service optional user facilities are invoked. + +**Table 12/F.400/X.400 – IPM optional user facilities selectable on a per-message basis** + +| Elements of service | Origination | Reception | Annex B reference | +|-------------------------------------------------------|-------------|-----------|-------------------| +| Additional physical rendition | A | A | B.2 | +| Alternate recipient allowed | A | A | B.3 | +| Authorization time indication | A | A | B.5 | +| Authorizing users indication | A | E | B.6 | +| Auto-forwarded indication | A | E | B.18 | +| Auto-submitted indication | A | E | B.20 | +| Basic physical rendition | A | E* | B.21 | +| Blind copy recipient indication | A | E | B.22 | +| Body part authentication and integrity | A | A | B.23 | +| Body part encryption | A | E | B.24 | +| Circulation list recipients indication | A | A | B.25 | +| Content confidentiality | A | A | B.26 | +| Content integrity | A | A | B.27 | +| Conversion prohibition | E | E | B.29 | +| Conversion prohibition in case of loss of information | A | A | B.30 | +| Counter collection | A | E* | B.32 | +| Counter collection with advice | A | A | B.33 | +| Cover page suppression | A | A | B.34 | +| Cross-referencing indication | A | E | B.35 | +| Deferred delivery | E | N/A | B.36 | +| Deferred delivery cancellation | A | N/A | B.37 | + +**Table 12/F.400/X.400 – IPM optional user facilities selectable on a per-message basis** + +| Elements of service | Origination | Reception | Annex B reference | +|---------------------------------------------|-------------|-----------|-------------------| +| Delivery notification | E | N/A | B.39 | +| Delivery via bureaufax service | A | A | B.41 | +| Designation of recipient by directory name | A | N/A | B.42 | +| Disclosure of other recipients | A | E | B.43 | +| Distribution codes indication | A | A | B.44 | +| DL-exempted recipients | A | A | B.45 | +| DL-expansion history indication | N/A | E | B.46 | +| DL-expansion prohibited | A | N/A | B.47 | +| EMS (Express Mail Service) (Note) | A | E* | B.48 | +| Expiry date indication | A | E | B.49 | +| Explicit conversion | A | N/A | B.50 | +| Forwarded IP-message indication | A | E | B.51 | +| Grade of delivery selection | E | E | B.52 | +| Importance indication | A | E | B.55 | +| Incomplete copy indication | A | A | B.56 | +| Information category indication | A | A | B.57 | +| IP-message security labelling | A | A | B.60 | +| Language indication | A | A | B.61 | +| Latest delivery designation | A | N/A | B.62 | +| Manual handling instructions indication | A | A | B.63 | +| Message flow confidentiality | A | N/A | B.64 | +| Message origin authentication | A | A | B.66 | +| Message security labelling | A | A | B.67 | +| Message sequence integrity | A | A | B.68 | +| Multi-destination delivery | E | N/A | B.70 | +| Multi-part body | A | E | B.71 | +| Non-receipt notification request indication | A | E | B.73 | +| Non-repudiation of content received | A | A | B.74 | +| Non-repudiation of delivery | A | A | B.75 | +| Non-repudiation of IP-notification | A | A | B.76 | +| Non-repudiation of origin | A | A | B.77 | +| Non-repudiation of submission | A | N/A | B.78 | +| Obsoleting indication | A | E | B.79 | +| Ordinary mail | A | E* | B.80 | +| Originator indication | E | E | B.82 | +| Originator reference indication | A | A | B.83 | +| Originator requested alternate recipient | A | N/A | B.84 | +| Physical delivery notification by MHS | A | A | B.85 | +| Physical delivery notification by PDS | A | E* | B.86 | +| Physical forwarding allowed | A | E* | B.87 | +| Physical forwarding prohibited | A | E* | B.88 | +| Precedence indication | A | A | B.89 | +| Prevention of non-delivery notification | A | N/A | B.90 | +| Primary and copy recipients indication | E | E | B.91 | +| Probe | A | N/A | B.92 | +| Probe origin authentication | A | N/A | B.93 | +| Proof of content received | A | A | B.94 | +| Proof of delivery | A | A | B.95 | + +**Table 12/F.400/X.400 – IPM optional user facilities selectable on a per-message basis** + +| Elements of service | Origination | Reception | Annex B reference | +|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------|-----------|-------------------| +| Proof of IP-notification | A | A | B.96 | +| Proof of submission | A | N/A | B.97 | +| Receipt notification request indication | A | A | B.98 | +| Redirection disallowed by originator | A | N/A | B.99 | +| Registered mail | A | A | B.101 | +| Registered mail to addressee in person | A | A | B.102 | +| Reply request indication | A | E | B.103 | +| Replying IP-message indication | E | E | B.104 | +| Report origin authentication | A | A | B.105 | +| Request for forwarding address | A | A | B.106 | +| Request for non-repudiation of content received | A | A | B.107 | +| Request for non-repudiation of IP-notification | A | A | B.108 | +| Request for proof of content received | A | A | B.109 | +| Request for proof of IP-notification | A | A | B.110 | +| Requested preferred delivery method | A | A | B.111 | +| Return of content | A | N/A | B.113 | +| Sensitivity indication | A | E | B.115 | +| Special delivery (Note) | A | E* | B.116 | +| Storage of draft messages | N/A | A | B.117 | +| Storage on submission | N/A | A | B.118 | +| Storage period assignment | N/A | A | B.119 | +| Stored message annotation | N/A | A | B.121 | +| Stored message deletion | N/A | E*** | B.122 | +| Stored message fetching | N/A | E*** | B.123 | +| Stored message grouping | N/A | A | B.124 | +| Stored message listing | N/A | E** | B.125 | +| Stored message summary | N/A | E** | B.126 | +| Subject indication | E | E | B.127 | +| Submission of IP-messages incorporating stored messages | N/A | A | B.129 | +| Undeliverable mail with return of physical message | A | E* | B.132 | +| Use of distribution list | A | N/A | B.133 | +| E    Essential optional user facility has to be provided
E*   Essential optional user facility only applying to PDAUs
E** Essential optional user facility applying to MSs. Additional optional user facility applying to UAs (which connect to MSs)
E*** Essential optional user facility applying to MSs and UAs
A    Additional optional user facility can be provided
N/A Not applicable
NOTE – At least EMS or special delivery shall be supported by the PDAU and associated PDS. | | | | + +**Table 13/F.400/X.400 – IPM optional user facilities agreed for a contractual period of time** + +| Elements of service | Classification | Annex B reference | +|--------------------------------------|-----------------------|--------------------------| +| Alternate recipient assignment | A | B.4 | +| Auto-acknowledgment of IP-messages | A | B.7 | +| Auto-action log | A | B.8 | +| Auto-advise | A | B.9 | +| Auto-assignment of annotations | A | B.10 | +| Auto-assignment of group names | A | B.11 | +| Auto-assignment of storage period | A | B.12 | +| Auto-correlation of IP-messages | A | B.13 | +| Auto-correlation of IP-notifications | A | B.14 | +| Auto-correlation of reports | A | B.15 | +| Auto-deletion after storage period | A | B.16 | +| Auto-discarding of IP-messages | A | B.17 | +| Auto-forwarding of IP-messages | A | B.19 | +| Delivery log | A | B.38 | +| Hold for delivery | A | B.53 | +| Implicit conversion | A | B.54 | +| IP-message action status | A | B.58 | +| MS register | A | B.69 | +| Redirection of incoming messages | A | B.100 | +| Restricted delivery | A | B.112 | +| Secure access management | A | B.114 | +| Stored message alert | A | B.120 | +| Submission log | A | B.128 | + +## Annex A + +### Glossary of terms + +(This annex does not form an integral part of this Recommendation) + +NOTE – The explanations given are not necessarily definitions in the strict sense. See also the definitions in Annex B and those provided in the other X.400-series Recommendations and parts of ISO/IEC 10021 (especially ITU-T Rec. X.402 | ISO/IEC 10021-2), where many entries are sourced. The terms have, depending on the source, varying levels of abstraction. + +#### A.1 access unit (AU) + +*F: unité d'accès* + +*S: unidad de acceso* + +In the context of a message handling system, the functional object, a component of MHS, that links another communication system (e.g., a physical delivery system or the telex network) to the MTS and via which its patrons engage in message handling as indirect users. + +In the context of message handling services, the unit which enables users of one service to intercommunicate with message handling services, such as the IPM service. + +#### A.2 actual recipient + +*F: receveur effectif* + +*S: recipiente real* + +In the context of message handling, a potential recipient for which delivery or affirmation takes place. + +#### A.3 Administration + +*F: Administration* + +*S: Administración* + +In the context of ITU-T, an Administration (member of ITU) or a Recognized Operating Agency (ROA). + +#### A.4 administration domain name + +*F: nom d'un domaine d'administration* + +*S: nombre de dominio de administración* + +In the context of message handling, a standard attribute of a name form that identifies an ADMD relative to the country denoted by a country name. + +#### A.5 administration management domain (ADMD) + +*F: domaine de gestion d'administration* + +*S: dominio de gestión de administración* + +A management domain that comprises messaging systems managed (operated) by a service provider. + +#### A.6 alternate recipient + +*F: destinataire suppléant* + +*S: recipiente alternativo* + +In the context of message handling, a user or a distribution list to which a message or probe may be conveyed if, and only if, it cannot be conveyed to a particular preferred recipient. The alternate recipient may be specified by the originator (see B.84), by the recipient (see B.100), or by the recipient MD (see B.4). + +## **A.7 attribute** + +*F: attribut* + +*S: atributo* + +In the context of message handling, an information item, a component of an attribute list, that describes a user or distribution list and that can also locate it in relation to the physical or organizational structure of MHS (or the network underlying it). + +## **A.8 attribute list** + +*F: liste d'attributs* + +*S: lista de atributos* + +In the context of message handling, a data structure, an ordered set of attributes, that constitutes an OR-address. + +## **A.9 attribute type** + +*F: type d'attribut* + +*S: tipo de atributo* + +An identifier that denotes a class of information (e.g., personal names). It is a part of an attribute. + +## **A.10 attribute value** + +*F: valeur d'attribut* + +*S: valor de atributo* + +An instance of the class of information an attribute type denotes (e.g., a particular personal name). It is a part of an attribute. + +## **A.11 basic service** + +*F: service de base* + +*S: servicio básico* + +In the context of message handling, the sum of features inherent in a service. + +## **A.12 body** + +*F: corps* + +*S: cuerpo* + +Component of the content of a message. Another component is the heading. + +## **A.13 body part** + +*F: partie du corps* + +*S: parte del cuerpo* + +Component of the body of a message. + +## **A.14 common name** + +*F: nom courant* + +*S: nombre común* + +In the context of message handling, a standard attribute of an OR-address form that identifies a user or distribution list relative to the entity denoted by another attribute (e.g., an organizational name). + +## **A.15 content** + +*F: contenu* + +*S: contenido* + +In the context of message handling, an information object, part of a message, that the MTS neither examines nor modifies, except for conversion, during its conveyance of the message. + +## **A.16 content type** + +*F: type de contenu* + +*S: tipo de contenido* + +In the context of message handling, an identifier, on a message envelope, that identifies the type (i.e. syntax and semantics) of the message content. + +## **A.17 conversion** + +*F: conversion* + +*S: conversión* + +In the context of message handling, a transmittal event in which an MTA transforms parts of a message's content from one encoded information type to another, or alters a probe so it appears that the described messages were so modified. + +## **A.18 country name** + +*F: nom de pays* + +*S: nombre de país* + +In the context of message handling, a standard attribute of a name form that identifies a country. A country name is a unique designation of a country for the purpose of sending and receiving messages. + +NOTE – In the context of physical delivery, additional rules apply (see also *physical delivery country name* and ITU-T Rec. F.415). + +## **A.19 delivery** + +*F: remise* + +*S: entrega* + +In the context of message handling, a transmittal step in which an MTA conveys a message or report to the MS, UA or AU of a potential recipient of the message or of the originator of the report's subject message or probe. + +## **A.20 delivery report** + +*F: rapport de remise* + +*S: informe de entrega* + +In the context of message handling, a report that acknowledges delivery, non-delivery, export, or affirmation of the subject message or probe, or distribution list expansion. + +## **A.21 direct submission** + +*F: dépôt direct* + +*S: depósito directo* + +In the context of message handling, a transmittal step in which the originator's UA or MS conveys a message or probe to an MTA. + +## **A.22 directory** + +*F: annuaire* + +*S: directorio* + +A collection of open systems cooperating to provide directory services. + +## **A.23 directory name** + +*F: nom d'annuaire* + +*S: nombre de directorio* + +Name of an entry in a directory. + +NOTE – In the context of message handling, the entry in the directory will enable the OR-address to be retrieved for submission of a message. + +## **A.24 directory system agent (DSA)** + +*F: agent de système d'annuaire* + +*S: agente de sistema de directorio* + +An OSI application process which is part of the directory, and whose role is to provide access to the directory information base to DUAs and/or other DSAs. + +## **A.25 directory user agent (DUA)** + +*F: agent d'utilisateur d'annuaire* + +*S: agente de usuario de directorio* + +An OSI application process which represents a user in accessing the directory. Each DUA serves a single user so that the directory can control access to directory information on the basis of the DUA names. DUAs can also provide a range of local facilities to assist users to compose requests (queries) and interpret the responses. + +## **A.26 direct user** + +*F: utilisateur direct* + +*S: usuario directo* + +In the context of message handling, a user that engages in message handling by direct use of the MTS. + +## **A.27 distribution list (DL)** + +*F: liste de distribution* + +*S: lista de distribución* + +In the context of message handling, the functional object, a component of the message handling environment, that represents a pre-specified group of users and other distribution lists and that is a potential destination for the information objects an MHS conveys. Membership can contain OR-names identifying either users or other distribution lists. + +## **A.28 distribution list expansion** + +*F: expansion de liste de distribution* + +*S: expansión de una lista de distribución* + +In the context of message handling, a transmittal event in which an MTA resolves a distribution list, among a message's immediate recipients, to its members. + +#### **A.29 distribution list name** + +*F: nom de liste de distribution* + +*S: nombre de lista de distribución* + +OR-name allocated to represent a collection of OR-addresses and directory names. + +#### **A.30 domain** + +*F: domaine* + +*S: dominio* + +See *management domain*. + +#### **A.31 domain-defined attributes** + +*F: attributs définis sur un domaine* + +*S: atributos definidos por el dominio* + +Optional attributes of an OR-address allocated to names in the responsibility of a management domain. + +#### **A.32 element of service** + +*F: élément de service* + +*S: elemento de servicio* + +Functional unit for the purpose of segmenting and describing message handling features. + +#### **A.33 encoded information type (EIT)** + +*F: type d'information codée* + +*S: tipo de información codificada* + +In the context of message handling, an identifier, on a message envelope, that identifies one type of encoded information represented in the message content. It identifies the medium and format (e.g., T.51 text, group 3 facsimile) on an individual portion of the content. + +#### **A.34 envelope** + +*F: enveloppe* + +*S: sobre* + +In the context of message handling, an information object, part of a message, whose composition varies from one transmittal step to another and that variously identifies the message originator and potential recipients, documents its past and directs its subsequent conveyance by the MTS, and characterizes its content. + +#### **A.35 explicit conversion** + +*F: conversion explicite* + +*S: conversión explícita* + +In the context of message handling, a conversion in which the originator selects both the initial and final encoded information types. + +#### **A.36 extension of physical delivery address components** + +*F: développement de composantes d'adresse de remise physique* + +*S: ampliación de los componentes de una dirección de entrega física* + +Standard attribute of a postal OR-address as a means to give further information about the point of physical delivery in a postal address, e.g., the name of a hamlet, room and floor numbers in a large building. + +### **A.37 extension of postal OR-address components** + +*F: développement de composantes d'adresse postale OR* + +*S: ampliación de los componentes de una dirección postal OR* + +Standard attribute of a postal OR-address as a means to give further information to specify the addressee in a postal address, e.g., by organizational unit. + +### **A.38 file transfer body part** + +*F: partie du corps de transfert de fichier* + +*S: parte de cuerpo de transferencia de fichero* + +A body part for conveying the contents of a stored file, and other information associated with the file, from originator to recipient. The other information includes attributes which are typically stored along with the file content, information on the environment from which the transfer originated, and references to existing stored files or previous messages. + +### **A.39 formatted postal OR-address** + +*F: adresse postale OR formatée* + +*S: dirección postal OR formatada* + +Component of the content of a message. Another component is the body. + +### **A.40 general text body part** + +*F: partie de corps de texte général* + +*S: parte de cuerpo de texto general* + +A body part that represents character text of a general nature, using 8-bit-encoding. It has parameters and data components. The parameter component identifies the character sets that are present in the data component. The data component comprises a single general string. + +### **A.41 heading** + +*F: en-tête* + +*S: encabezamiento* + +Component of an IP-message. Other components are the envelope and the body. + +### **A.42 immediate recipient** + +*F: destinataire direct* + +*S: recipiente inmediato; recipiente directo* + +In the context of message handling, one of the potential recipients assigned to a particular instance of a message or probe (e.g., an instance created by splitting). + +### **A.43 implicit conversion** + +*F: conversion implicite* + +*S: conversión implícita* + +In the context of message handling, a conversion in which the MTA selects both the initial and final encoded information types. + +#### **A.44 indirect submission** + +*F: dépôt indirect* + +*S: depósito indirecto* + +In the context of message handling, a transmittal step in which an originator's UA conveys a message or probe to an MTA via an MS. + +#### **A.45 indirect user** + +*F: utilisateur indirect* + +*S: usuario indirecto* + +In the context of message handling, a user that engages in message handling by indirect use of MHS, i.e. through another communication system (e.g., a physical delivery system or the telex network) to which MHS is linked. + +NOTE – Indirect users communicate via access units with direct users of MHS. + +#### **A.46 preferred recipient** + +*F: destinataire préféré* + +*S: recipiente preferido* + +In the context of message handling, one of the users and distribution lists that the originator selects as a message's or probe's preferred destination. + +#### **A.47 intercommunication** + +*F: intercommunication* + +*S: intercomunicación* + +In the context of message handling, a relationship between services, where one of the services is a message handling service, enabling the user of the message handling service to communicate with users of other services. + +NOTE – Examples are the intercommunication between the IPM service and the telex service, and the intercommunication between message handling services and physical delivery services. + +#### **A.48 interpersonal messaging service** + +*F: service de messagerie de personne à personne* + +*S: servicio de mensajería interpersonal* + +Messaging service between users belonging to the same management domain or to different management domains by means of message handling, based on the message transfer service. + +#### **A.49 IP-message** + +*F: message de personne à personne; message IP* + +*S: mensaje interpersonal; mensaje IP* + +The content of a message in the IPM Service. + +#### **A.50 local postal attributes** + +*F: attributs postaux locaux* + +*S: atributos postales locales* + +Standard attributes of a postal OR-address as a means to distinguish between places with the same name (e.g., by state name, county name, or geographical attribute) in a postal address. + +### **A.51 management domain (MD)** + +*F: domaine de gestion* + +*S: dominio de gestión* + +In the context of message handling, a set of messaging systems – at least one of which contains, or realizes, an MTA – that is managed by a single organization. It is a primary building block used in the organizational construction of MHS. It refers to an organizational area for the provision of services. + +NOTE – A management domain may or may not necessarily be identical with a geographical area. + +### **A.52 management domain name** + +*F: nom d'un domaine de gestion* + +*S: nombre de dominio de gestión* + +Unique designation of a management domain for the purpose of sending and receiving messages. + +### **A.53 members** + +*F: membres* + +*S: miembros* + +In the context of message handling, the set of users and distribution lists implied by a distribution list name. + +### **A.54 message** + +*F: message* + +*S: mensaje* + +An instance of the primary class of information object conveyed by means of message transfer, and comprising an envelope and content. + +### **A.55 message handling (MH)** + +*F: messagerie* + +*S: tratamiento de mensajes* + +A distributed information processing task that integrates the intrinsically related subtasks of message transfer and message storage. + +### **A.56 message handling environment (MHE)** + +*F: environnement de messagerie* + +*S: entorno de tratamiento de mensajes* + +The environment in which message handling takes place, comprising MHS, users, and distribution lists. + +The sum of all components of message handling systems. + +NOTE – Examples of components are: + +- message transfer agents; +- user agents; +- message stores; +- users. + +#### **A.57 message handling service** + +*F: service de messagerie* + +*S: servicio de tratamiento de mensajes* + +Service provided by the means of message handling systems. + +NOTE 1 – Service may be provided through administration management domains or private management domains. + +NOTE 2 – Examples of message handling services are: + +- interpersonal messaging service (IPM service); +- message transfer service (MT service). + +#### **A.58 message handling system (MHS)** + +*F: système de messagerie* + +*S: sistema de tratamiento de mensajes* + +The functional object, a component of the message handling environment, that conveys information objects from one party to another. + +#### **A.59 message storage** + +*F: mémorisation des messages* + +*S: almacenamiento de mensajes* + +The automatic storage for later retrieval of information objects conveyed by means of message transfer. It is one aspect of message handling. + +#### **A.60 message store (MS)** + +*F: mémoire de messages* + +*S: memoria de mensajes; almacenador de mensajes* + +The functional object, a component of MHS, that provides a single direct user with capabilities for message storage. + +#### **A.61 message transfer (MT)** + +*F: transfert de messages* + +*S: transferencia de mensajes* + +The non-real-time carriage of information objects between parties using computers as intermediaries. It is one aspect of message handling. + +#### **A.62 message transfer agent (MTA)** + +*F: agent de transfert de messages* + +*S: agente de transferencia de mensajes* + +A functional object, a component of the MTS, that actually conveys information objects to users and distribution lists. + +#### **A.63 message transfer service** + +*F: service de transfert de messages* + +*S: servicio de transferencia de mensajes* + +Service that deals with the submission, transfer and delivery of messages for other messaging services. + +#### **A.64 message transfer system (MTS)** + +*F: système de transfert de messages* + +*S: sistema de transferencia de mensajes* + +The functional object consisting of one or more message transfer agents which provides store-and-forward message transfer between user agents, message stores and access units. + +#### **A.65 messaging system** + +*F: système de messagerie* + +*S: sistema de mensajería* + +A computer system (possibly but not necessarily an open system) that contains, or realizes, one or more functional objects. It is a building block used in the physical construction of MHS. + +#### **A.66 mnemonic OR-address** + +*F: adresse OR mnémonique* + +*S: dirección OR nemotécnica* + +An OR-address that mnemonically identifies a user or distribution list relative to the ADMD through which the user is accessed or the distribution list is expanded. It identifies an ADMD, and a user or distribution list relative to that ADMD. + +#### **A.67 naming authority** + +*F: autorité dénominatrice* + +*S: autoridad de denominación* + +An authority responsible for the allocation of names. + +#### **A.68 network address** + +*F: adresse réseau* + +*S: dirección de red* + +In the context of message handling, a standard attribute of an OR-address form that gives the network address of a terminal. It is comprising the numbering digits for network access points from an international numbering plan. + +#### **A.69 non-delivery** + +*F: non-remise* + +*S: no entrega* + +In the context of message handling, a transmittal event in which an MTA determines that the MTS cannot deliver a message to one or more of its immediate recipients, or cannot deliver a report to the originator of its subject message or probe. + +#### **A.70 non-registered access** + +*F: accès non homologué* + +*S: acceso no registrado* + +In the context of message handling services, access to the service through publicly available telecommunications means by users who have neither been explicitly registered by the service provider, nor been allocated an OR-address. + +#### **A.71 numeric OR-address** + +*F: adresse OR numérique* + +*S: dirección OR numérica* + +In the context of message handling, an OR-address that numerically identifies a user relative to the ADMD through which the user is accessed. It identifies an ADMD, and a user relative to that ADMD. It is identifying a user of message handling services by means of a numeric keypad. + +#### **A.72 numeric user identifier** + +*F: identificateur numérique d'utilisateur* + +*S: identificador de usuario numérico* + +Standard attribute of an OR-address as a unique sequence of numeric information for identifying a user. + +#### **A.73 OR-address** + +*F: adresse OR* + +*S: dirección OR* + +In the context of message handling, an attribute list that distinguishes one user or DL from another and identifies the user's point of access to MHS or the distribution list's expansion point. + +#### **A.74 OR-name** + +*F: nom OR* + +*S: nombre OR* + +In the context of message handling, an information object by means of which a user can be designated as the originator, or a user or distribution list designated as a potential recipient of a message or probe. An OR-name distinguishes one user or distribution list from another and can also identify its point of access to MHS. + +#### **A.75 optional user facilities** + +*F: fonctionnalités optionnelles d'utilisateur* + +*S: facilidades facultativas de usuario* + +In the context of message handling services the elements of service which are selectable by the user either on a contractual basis (agreed period of time) or on a per-message basis. + +NOTE 1 – Optional user facilities are classified as either essential or additional. + +NOTE 2 – Essential optional user facilities are to be made available to all message handling users. + +NOTE 3 – Additional optional user facilities can be made available for national and international use on the basis of bilateral agreement between the service providers. + +#### **A.76 organization name** + +*F: nom d'organisation* + +*S: nombre de organización* + +Standard attribute of an OR-address as a unique designation of an organization for the purpose of sending and receiving of messages. + +#### **A.77 organizational unit name** + +*F: nom d'une unité organisationnelle* + +*S: nombre de unidad organizacional* + +Standard attribute of an OR-address as a unique designation of an organizational unit of an organization for the purpose of sending and receiving of messages. + +#### **A.78 originator** + +*F: expéditeur* + +*S: originador* + +In the context of message handling, the user (but not distribution list) that is the ultimate source of a message or probe. + +#### **A.79 personal name** + +*F: nom personnel* + +*S: nombre personal* + +In the context of message handling, a standard attribute of an OR-address form that identifies a person relative to the entity denoted by another attribute (e.g., an organization name). + +NOTE – Components are for example: + +- surname; +- given name; +- initials; +- generation qualifier. + +#### **A.80 physical delivery (PD)** + +*F: remise physique* + +*S: entrega física* + +The delivery of a message in physical form, such as a letter, through a physical delivery system. + +#### **A.81 physical delivery access unit (PDAU)** + +*F: unité d'accès de remise physique* + +*S: unidad de acceso de entrega física* + +An access unit that subjects messages (but neither probes nor reports) to physical rendition. + +#### **A.82 physical delivery address components** + +*F: componentes d'une adresse de remise physique* + +*S: componentes de dirección de entrega física* + +In a postal address they contain the information necessary for the local physical delivery within the physical delivery area of the physical delivery office, i.e. a street address, a P.O. Box address, a poste restante address or a unique name alternatively. + +NOTE – The information is generally restricted to one line with up to 30 printable graphic characters. Additional information may be supplied by using the attribute type "extension of physical delivery address component". + +#### **A.83 physical delivery country name** + +*F: nom du pays de remise physique* + +*S: nombre de país para entrega física* + +In the context of physical delivery, a unique description of the country of the final destination. + +#### **A.84 physical delivery domain** + +*F: domaine de remise physique* + +*S: dominio de entrega física* + +The domain of responsibility of an organization providing a physical delivery system and optionally an MTA/PDAU. + +#### **A.85 physical delivery office address components** + +*F: composantes d'adresse de bureau de remise physique* + +*S: componentes de dirección de oficina de entrega física* + +In a postal address they contain the information to specify the office which is responsible for the local physical delivery. + +NOTE – The information is generally restricted to one line with up to 30 printable graphic characters. In some countries the postal code will follow the physical delivery office address components in a separate line (possibly together with the country name). + +#### **A.86 physical delivery office name** + +*F: nom du bureau de remise physique* + +*S: nombre de oficina de entrega física* + +Standard attribute of a postal OR-address, in the context of physical delivery, specifying the name of the city, village, etc. where the physical delivery office is situated, or where the physical delivery is effected. + +#### **A.87 physical delivery office number** + +*F: numéro du bureau de remise physique* + +*S: número de oficina de entrega física* + +Standard attribute and in a postal OR-address a means to distinguish between more than one physical delivery office within a city, etc. + +#### **A.88 physical delivery organization name** + +*F: nom d'organisation de remise physique* + +*S: nombre de organización de entrega física* + +A free form name of the addressed entity in the postal address, taking into account the specified limitations in length. + +#### **A.89 physical delivery personal name** + +*F: nom personnel de remise physique* + +*S: nombre personal de entrega física* + +In a postal address a free form name of the addressed individual containing the family name and optionally the given name(s), the initial(s), title(s) and generation qualifier, taking into account the specified limitations in length. + +#### **A.90 physical delivery service** + +*F: service de remise physique* + +*S: servicio de entrega física* + +Service provided by a physical delivery system. + +#### **A.91 physical delivery service name** + +*F: nom du service de remise physique* + +*S: nombre del servicio de entrega física* + +Standard attribute of a postal OR-address in the form of the name of the service in the country electronically receiving the message on behalf of the physical delivery service. + +#### **A.92 physical delivery system (PDS)** + +*F: système de remise physique* + +*S: sistema de entrega física* + +A system that performs physical delivery. One important kind of physical delivery system is the postal system. + +#### **A.93 physical message** + +*F: message physique* + +*S: mensaje físico* + +A physical object comprising a relaying envelope and its content, e.g., a letter. + +#### **A.94 physical rendition** + +*F: conversion physique* + +*S: reproducción física* + +The transformation of an MHS message to a physical message, e.g., by printing the message on paper and enclosing it in a paper envelope. + +#### **A.95 postal code** + +*F: code postal* + +*S: código postal* + +Standard attribute of a postal OR-address to specify the geographical area, and in the context of MHS, used for routing of messages. + +#### **A.96 postal OR-address** + +*F: adresse OR postale* + +*S: dirección postal OR* + +In the context of message handling, an OR-address that identifies a user by means of its postal address. It identifies the physical delivery system through which the user is to be accessed and gives the user's postal address. + +#### **A.97 postal OR-address components** + +*F: composantes d'adresse OR postale* + +*S: componentes de la dirección postal OR* + +They contain in a postal address information to describe the sender or addressee by means of his name (personal name, organization name). + +NOTE – In a postal address the information is generally restricted to one line of 30 printable characters. Additional information may be supplied by using the attribute type "extension of postal OR-address components". + +#### **A.98 post office box address (P.O. Box address)** + +*F: adresse de boîte postale* + +*S: dirección de apartado de correos* + +A standard attribute in a postal address indicating that physical delivery through a P.O. Box is requested. It carries the P.O. Box number for distribution to the P.O. Box. + +#### **A.99 poste restante address** + +*F: adresse en poste restante* + +*S: dirección de lista de correos* + +A standard attribute in a postal address indicating that physical delivery at the counter is requested. It may also carry a code. + +#### **A.100 potential recipient** + +*F: destinataire potentiel* + +*S: recipiente potencial* + +In the context of message handling, any user or distribution list to which a message or probe is conveyed during the course of transmittal. Equivalently, a preferred member, alternate member, or substitute recipient. + +#### **A.101 private domain name** + +*F: nom de domaine privé* + +*S: nombre de dominio privado* + +In the context of message handling, a standard attribute of an OR-address form that identifies a PRMD relative to the ADM D denoted by an administration domain name. + +#### **A.102 private management domain (PRMD)** + +*F: domaine de gestion privé* + +*S: dominio de gestión privado* + +In the context of message handling, a management domain that comprises messaging system(s) managed (operated) by an organization other than a service provider. + +NOTE – This does not preclude a service provider from managing (operating) a PRMD. + +#### **A.103 probe** + +*F: message test* + +*S: sonda* + +In the context of message handling, an instance of a secondary class of information objects conveyed by means of message transfer that describes a class of messages and that is used to determine the deliverability of such messages. + +#### **A.104 public message handling service** + +*F: service public de messagerie* + +*S: servicio público de tratamiento de mensajes* + +Message handling service offered by a service provider. + +#### **A.105 public services** + +*F: services publics* + +*S: servicios públicos* + +In the context of telecommunication, the services offered by service providers. + +### **A.106 receipt** + +*F: réception* + +*S: recepción* + +In the context of message handling, a transmittal step in which either a UA conveys a message or report to its direct user, or the communication system that serves an indirect user conveys such an information object to that user. + +### **A.107 recipient** + +*F: destinataire* + +*S: recipiente* + +See *actual recipient*. + +### **A.108 recursion** + +*F: récursion* + +*S: repetición* + +In the context of message handling, the situation that a message gets back to the same distribution list of origin and potentially circulates infinitely. + +### **A.109 redirection** + +*F: réacheminement* + +*S: redireccionamiento* + +In the context of message handling, a transmittal event in which an MTA replaces a user among a message's immediate recipients with a user preselected for that message. + +### **A.110 registered access** + +*F: accès homologué* + +*S: acceso registrado* + +In the context of message handling services, access to the service performed by subscribers who have been registered by the service provider to use the service, and been allocated an OR-address. + +### **A.111 report** + +*F: rapport* + +*S: informe* + +In the context of message handling, an instance of a secondary class of information object conveyed by means of message transfer. It is generated by the MTS, it reports the outcome or progress of a message's or probe's transmittal to one or more potential recipients. + +### **A.112 retrieval** + +*F: extraction* + +*S: extracción* + +In the context of message handling, a transmittal step in which a user's message store conveys a message or report to the user's UA. The user is an actual recipient of the message or the originator of the subject message or probe. + +### **A.113 security capabilities** + +*F: capacités de sécurité* + +*S: capacidades de seguridad* + +In the context of message handling, the mechanisms that protect against various security threats. + +### **A.114 specialized access** + +*F: accès spécialisé* + +*S: acceso especializado* + +In the context of message handling, the involvement of specialized access units providing intercommunication between message handling services and other telecommunication services. + +### **A.115 standard attribute** + +*F: attribut normalisé* + +*S: atributo normalizado* + +An attribute whose type is bound to a certain class of information. + +### **A.116 street address** + +*F: adresse de rue* + +*S: dirección de calle* + +A standard attribute in a postal address giving information for the local distribution and physical delivery, i.e. the street name, the street identifier (like street, place, avenue) and the house number. + +### **A.117 subject** + +*F: objet* + +*S: asunto* + +In the context of message handling, the information, part of the header, that summarizes the content of the message as the originator has specified it. + +### **A.118 subject message** + +*F: message objet* + +*S: mensaje de asunto* + +The message that is the subject of a report. + +### **A.119 subject probe** + +*F: message test objet* + +*S: sonda de asunto* + +The probe that is the subject of a report. + +### **A.120 submission** + +*F: dépôt* + +*S: depósito* + +Direct submission or indirect submission. + +#### **A.121 substitute recipient** + +*F: destinataire substitut* + +*S: recibiente sustituto* + +In the context of message handling, the user or distribution list to which a preferred, alternate, or member (but not another substitute) recipient had elected to redirect messages (but not probes). + +#### **A.122 terminal identifier** + +*F: identificateur de terminal* + +*S: identificador de terminal* + +Standard attribute in an OR-address providing information for identifying a terminal amongst others. + +NOTE – An example is the telex answerback. + +#### **A.123 terminal OR-address** + +*F: adresse OR de terminal* + +*S: dirección OR de terminal* + +In the context of message handling, an OR-address that identifies a user by means of the network address of his terminal and that can identify the ADMD through which that terminal is accessed. The terminals identified can belong to different networks. + +#### **A.124 terminal type** + +*F: type de terminal* + +*S: tipo de terminal* + +Standard attribute of an OR-address that indicates the type of a terminal. + +NOTE – Examples: telex, G3 facsimile, G4 facsimile, ITU-T Rec. T.51, videotex terminal. + +#### **A.125 transfer** + +*F: transfert* + +*S: transferencia* + +In the context of message handling, a transmittal step in which one MTA conveys a message, probe, or report to another. + +#### **A.126 transfer system** + +*F: système de transfert* + +*S: sistema de transferencia* + +A messaging system that contains one MTA; optionally one or more access units, and neither a UA nor a message store. + +#### **A.127 transmittal** + +*F: transmission* + +*S: transmisión* + +The conveyance or attempted conveyance of a message from its originator to its potential recipients, or of a probe from its originator to MTAs able to affirm any described message's deliverability to its potential recipients. It also encompasses the conveyance or attempted conveyance, to the originator of the message or probe, of any reports it provokes. It is a sequence of transmittal steps and events. + +#### **A.128 unformatted postal OR-address** + +*F: adresse OR postale non formatée* + +*S: dirección postal OR no formatada* + +OR-address based on an unformatted postal address. + +#### **A.129 unique postal name** + +*F: nom postal unique* + +*S: nombre postal exclusivo* + +In a postal address a standard attribute describing the point of physical delivery by means of a unique name, e.g., that of a building. + +#### **A.130 user** + +*F: utilisateur* + +*S: usuario* + +In the context of message handling, a functional object (e.g., a person), a component of the message handling environment, that engages in (rather than provides) message handling and that is a potential source or destination for the information objects an MHS conveys. + +#### **A.131 user agent (UA)** + +*F: agent d'utilisateur* + +*S: agente de usuario* + +In the context of message handling, the functional object, a component of MHS, by means of which a single direct user engages in message handling. + +#### **A.132 voice body part** + +*F: partie de corps vocale* + +*S: parte de cuerpo de datos vocales* + +A body part sent or forwarded from an originator to a recipient which conveys voice encoded data and related information. The related information consists of parameters which are used to assist in the processing of the voice data. These parameters include information detailing the duration of the voice data, the voice encoding algorithm used to encode the voice data, and supplementary information. + +## Annex B + +### Definitions of elements of service + +(This annex forms an integral part of this Recommendation) + +NOTE – The abbreviations used in the title line have the following meanings: + +| | | +|-------|----------------------------------------------------| +| MT | Message transfer | +| IPM | Interpersonal messaging | +| PD | Physical delivery | +| MS | Message store | +| MS-94 | 1994 enhanced message store | +| PR | Per recipient (available on a per-recipient basis) | + +#### **B.1 access management** + +**MT** + +This element of service enables a UA and an MTA to establish access to one another and to manage information associated with access establishment. + +The element of service permits the UA and MTA to identify and validate the identity of the other. It provides a capability for the UA to specify its OR-address and to maintain access security. When access security is achieved through passwords, these passwords can be periodically updated. + +NOTE – A more secure form of access management is provided by the element of service secure access management. + +#### **B.2 additional physical rendition** + +**PD PR** + +This element of service allows an originating user to request the PDAU to provide the additional rendition facilities (e.g., kind of paper, coloured printing, etc.). Bilateral agreement is required to use this element of service. + +#### **B.3 alternate recipient allowed** + +**MT** + +This element of service enables an originating UA to specify that the message being submitted can be delivered to an alternate recipient as described below. + +A destination MD will interpret all of the user attributes in order to select a recipient UA. Three cases can be distinguished: + +- 1) All the attributes match precisely those of a subscriber UA. Delivery is attempted to that UA. +- 2) Either insufficient attributes are supplied or those supplied match those of more than one subscriber UA. The message cannot be delivered. +- 3) At least the minimum set of attributes required by the destination MD is supplied. Nevertheless, taking all of the other attributes into account, the attributes match those of no UA. + +In case 3, an MD that supports the alternate recipient assignment element of service can deliver the message to a UA that has been assigned to receive such messages. This UA will be notified of the OR-address of the intended recipient as specified by the originator. Delivery to this UA will be reported in a delivery notification, if requested by the originator. + +#### **B.4 alternate recipient assignment** + +**MT** + +This element of service enables a UA to be given the capability to have certain messages delivered to it for which there is not an exact match between the recipient attributes specified and the name of the user. Such a UA is specified in terms of one or more attributes for which an exact match is required, and one or more attributes for which any value is acceptable. For example, an organization can establish a UA to receive all messages for which country name, administration management domain name and organization name (for example, company name) are an exact match, but the personal name of the recipient does not correspond to an individual known by an MHS in that organization. This permits the organization to manually handle the messages to these individuals. + +In order for a message to be reassigned to an alternate recipient, the originator must have requested the alternate recipient allowed element of service. + +## **B.5 authorization time indication** + +**IPM** + +This element of service enables the originator to indicate to the recipient UA the date and time at which a message was formally authorized. Depending upon local requirements, this date and time stamp may vary from the date and time when the message was submitted to the MTS. This element of service may be used to augment the authorizing users indication element of service (see B.6) to provide supplementary information about the authorizing event. + +## **B.6 authorizing users indication** + +**IPM** + +This element of service allows the originator to indicate to the recipient the names of the one or more persons who authorized the sending of the message. For example, an individual can authorize a particular action which is subsequently communicated to those concerned by another person such as a secretary. The former person is said to authorize its sending while the latter person is the one who sent the message (originator). This does not imply signature-level authorization. + +## **B.7 auto-acknowledgement of IP-messages** + +**IPM MS-94** + +This element of service enables an MS-user to instruct the MS to generate a receipt notification automatically for each IP-message containing a receipt notification request which is delivered to the MS. The receipt notification is sent when the complete IP-message has been retrieved by the user or when the user indicates to the MS that he regards the message as having been retrieved. + +## **B.8 auto-action log** + +**MS-94** + +This element of service enables an MS-user to access a log that records details of selected auto-action executions performed by the MS. The MS-user is able to retrieve information from the auto-action log by means of the stored message listing and stored message fetching elements of service. The ability to delete auto-action log entries is subject to subscription. This log of information is available if, and only if, this element of service is subscribed to by the user of the MS. Support for an element of service which comprises an auto-action does not require support for the auto-action log element of service. For each type of auto-action that may generate log entries, it is a subscription option whether all auto-action executions are logged, or only those executions that result in an error, or no executions are logged for that auto-action. + +## **B.9 auto-advice** + +**IPM MS** + +This element of service enables an MS-user to instruct the MS to generate advice notifications automatically when selected IP-messages are delivered. The notification may inform the originator of the delivered IP-message that the MS-user is absent and, for the present, unable to take receipt of messages, or may intimate a change of address. The notification is generated only if so requested by the IP-message's originator. + +## **B.10 auto-assignment of annotations** + +**MS-94** + +This element of service enables an MS-user to instruct the MS to attach annotations to a selected message automatically, when the message is stored in the MS and satisfies specified criteria. The MS-user may specify, through registration, several sets of selection criteria each of which may indicate the attachment of a different value of annotation. Subscription to this element of service requires subscription to the stored message annotation element of service. + +## **B.11 auto-assignment of group names** + +**MS-94** + +This element of service enables an MS-user to instruct the MS to assign group names to a selected message automatically, when the message is stored in the MS and satisfies specified criteria. The MS-user may specify, through registration, several sets of selection criteria, each of which may indicate the assignment of a different group name. The MS will verify that only registered group names are assigned to messages. Subscription to this element of service requires subscription to the stored message grouping element of service. + +## **B.12 auto-assignment of storage period** + +**MS-94** + +This element of service enables an MS-user to instruct the MS to assign a storage period to a selected message automatically, when the message is stored in the MS and satisfies specified criteria. The MS-user may specify, through registration, several sets of selection criteria each of which may indicate the attachment of a different value of storage period. Subscription to this element of service requires subscription to the storage period assignment element of service. + +### **B.13 auto-correlation of IP-messages** + +**IPM MS-94** + +This element of service enables an MS-user to retrieve information, automatically generated by the MS, concerning the correlation between various related IP-messages. The following types of messages may be correlated: + +- 1) IP-messages received in reply to, or sent in reply to an IP-message; +- 2) the IP-messages which forwarded (or auto-forwarded) one or more messages; +- 3) the received or submitted IP-messages that obsolete an IP-message; +- 4) the received or submitted IP-messages that indicate that they are related to an IP-message. + +Besides identifying each IP-message related to a given message in the ways indicated, the MS provides a summary of all such responding IP-messages. + +### **B.14 auto-correlation of IP-notifications** + +**IPM MS-94** + +This element of service enables an MS-user to retrieve information, automatically generated by the MS, concerning the IP-notifications that have been received in response to a previously submitted IP-message. Information may also be retrieved concerning IP-notifications sent by the MS-user or the MS in response to delivered IP-messages. The MS identifies each IP-notification related to a given submitted or delivered message, and for submitted messages it also provides a summary of received IP-notifications. This enables the MS-user to access this information directly rather than perform an exhaustive search of all entries that could hold the information. This element of service is effective only if the submitted or delivered message that an IP-notification refers to is stored in the MS, or is recorded in the submission log or delivery log. Provision for the storage of submitted messages, and maintenance of the submission log and the delivery log are supported by separate elements of service. + +### **B.15 auto-correlation of reports** + +**MS-94** + +This element of service enables an MS-user to retrieve information, automatically generated by the MS, concerning the delivery and non-delivery reports that have been received in response to a previously submitted message. Successful cancellations of deferred delivery for submitted messages are also recorded. In addition to identifying each report related to a given submitted message, the MS provides a summary of these reports. This enables the MS-user to access this information directly rather than perform an exhaustive search of all entries that could hold the information. This element of service requires that at least one of the submission log or storage on submission elements of service has also been subscribed to. + +### **B.16 auto-deletion after storage period** + +**MS-94** + +This element of service enables an MS-user to instruct the MS to delete automatically any stored message whose storage period has elapsed. This registration remains in force until disabled by a subsequent registration. Messages that have not been listed or processed are not subject to auto-deletion. Equally, entries of the submission log, delivery log, and auto-action log are not subject to auto-deletion. Other content-specific message handling specifications may lay down additional rules for the performance of this element of service. Subscription to this element of service requires subscription to the storage period assignment element of service. + +### **B.17 auto-discarding of IP-messages** + +**IPM MS-94** + +This element of service enables an MS-user to instruct the MS to discard stored IP-messages automatically, if they satisfy criteria registered by the MS-user. An IP-message becomes a candidate for auto-discarding if a subsequently delivered IP-message renders it obsolete, or if it contains an Expiry Time that has been reached. The MS-user may control whether auto-discarding occurs for such IP-messages by specifying additional conditions which the IP-message must satisfy, e.g., that the message has been fetched by the MS-user, or that the obsoleting IP-message has the same originator as the obsoleted IP-message. Where the message has not been fetched by the MS-user before being auto-discarded, a non-receipt notification is generated if requested in the discarded IP-message. + +### **B.18 auto-forwarded indication** + +**IPM** + +This element of service allows a recipient to determine that a body of an incoming IP-message contains an IP-message that has been auto-forwarded. Thus the recipient can distinguish from that where an incoming IP-message contains a forwarded message (as described in B.51) in the body. As with a forwarded IP-message, an auto-forwarded IP-message can be accompanied by information (for example, time stamps, indication of conversion) associated with its original delivery. + +NOTE – The indication that auto-forwarding of an IP-message has occurred enables a recipient IPM UA, should it so choose, to prevent further auto-forwarding and thus the possibility of loops. In addition, a recipient IPM UA can choose whether or not to auto-forward based on other criteria (for example, sensitivity classification). + +When an IPM UA auto-forwards an IP-message, it designates it as auto-forwarded. If receipt/non-receipt notification has been requested for the IP-message being auto-forwarded, the IPM UA generates a non-receipt notification informing the originator of the auto-forwarding of the IP-message. The notification optionally includes a comment supplied by the originally intended recipient. No further notification applying to the auto-forwarded IP-message is generated by any IPM UA. + +#### **B.19 auto-forwarding of IP-messages** + +**IPM MS** + +This element of service enables an MS-user to instruct the MS to auto-forward selected IP-messages that are delivered to it. The MS-user may specify through registration several sets of criteria chosen from the attributes available in the MS, and IP-messages meeting each set of criteria will be auto-forwarded to one or more users or DLs. If requested by the message originator, a non-receipt notification is generated indicating that the IP-message was auto-forwarded even if the MS retains a copy of the forwarded message, unless the copy is retained as a new message. For each set of selection criteria, a body part may be specified, to be included as a "cover note" with each auto-forwarded IP-message. + +NOTE – In versions of F.400/X.400 and ISO/IEC 10021-1 published prior to 1994, this element of service was named stored message auto-forward, and classified as a general MS optional user facility; it has since been classified as IPM-specific. + +#### **B.20 auto-submitted indication** + +**IPM** + +This element of service allows the originator, or enables the UA or MS, to indicate to the recipient whether the message was or was not submitted automatically by a machine without either the direct or indirect control by a human of the submission, and to determine the nature of the submission, thus: + +- not auto-submitted; +- auto-generated; +- auto-replied. + +The absence of this indication yields no information as to whether the message submission involved human control or not. + +#### **B.21 basic physical rendition** + +**PD PR** + +This element of service enables the PDAU to provide the basic rendition facilities for converting the MHS message into a physical message. This is the default action to be taken by the PDAU. + +#### **B.22 blind copy recipient indication** + +**IPM PR** + +This element of service allows the originator to provide the OR-name of one or more additional users, or DLs, who are intended recipients of the IP-message being sent. These names are not disclosed to either the primary or copy recipients. Whether or not these additional recipients are disclosed to one another is a local matter. + +#### **B.23 body part authentication and integrity** + +**IPM** + +This element of service allows the originator of the message to provide the recipient with the means by which the recipient can verify that particular body parts of the message have not been modified and that their origin can be authenticated (i.e. a signature). + +#### **B.24 body part encryption** + +**IPM** + +This element of service allows the originator to indicate to the recipient that a particular body part of the IP-message being sent has been encrypted. Encryption can be used to prevent unauthorized inspection or modification of the body part. This element of service can be used by the recipient to determine that some body part(s) of the IP-message must be decrypted. The encrypted body part may retain the body part type information, or may be sent in a messaging-system independent format in which there is no information about the type of the information which has been encrypted. + +**B.25 circulation list recipients indication****IPM** + +This element of service enables the originator to indicate to the recipient a list of recipients to whom it is requested that the IP-message be distributed serially. The circulation list includes an indication of whether each recipient has already received the IP-message. In this context, recipients that have received the message are said to be "checked" in the circulation list. The circulation list should be updated by the recipient and included in an IP-message sent to the next recipient that has not been checked. + +**B.26 content confidentiality****MT** + +This element of service allows the originator of a message to protect the content of the message from disclosure to recipients other than the intended recipient(s). Content Confidentiality is on a per-message basis, and can use either an asymmetric or a symmetric encryption technique. + +**B.27 content integrity****MT PR** + +This element of service allows the originator of the message to provide to the recipient of the message a means by which the recipient can verify that the content of the message has not been modified. Content Integrity is on a per-recipient basis, and can use either an asymmetric or a symmetric encryption technique. + +**B.28 content type indication****MT** + +This element of service enables an originating UA to indicate the content type for each submitted message. A recipient UA can have one or more content types delivered to it. An example of a content type is the contents generated by the IPM class of cooperating UAs. + +**B.29 conversion prohibition****MT** + +This element of service enables an originating UA to instruct the MTS that implicit encoded information type conversion(s) shall not be performed for a particular submitted message. + +**B.30 conversion prohibition in case of loss of information****MT** + +This element of service enables an originating UA to instruct the MTS that encoded information type conversion(s) shall not be performed for a particular submitted message if such conversion(s) would result in loss of information. Loss of information is discussed in detail in ITU-T Rec. X.408. + +Should this and the conversion prohibition element of service both be selected, the latter shall take precedence. + +NOTE – This element of service will not protect against possible loss of information in certain cases where the recipient is using an I/O device whose capabilities are unknown to the MTA. + +**B.31 converted indication****MT PR** + +This element of service enables the MTS to indicate to a recipient UA that the MTS performed encoded information type conversion on a delivered message. The recipient UA is informed of the resulting types. + +**B.32 counter collection****PD PR** + +This element of service allows an originating user to instruct the PDS to keep the physical message ready for counter collection at the post office specified by the originator, or at the post office which offers counter collection service closest to the given recipient's address. + +**B.33 counter collection with advice****PD PR** + +This element of service allows an originating user to instruct the PDS to keep the physical message ready for counter collection at the post office specified by the originator, or at the post office which offers counter collection service closest to the given recipient's address, and to inform the recipient via telephone, or telex, using the number provided by the originator. + +**B.34 cover page suppression****MT PR** + +This element of service allows the originator to indicate to an Access Unit that a cover page should not be added to the message when it is rendered into physical form. This element of service is particularly intended for facsimile access units, but may also be applied to any other kind of access unit where the basic rendition calls for the AU to generate a cover page. + +**B.35 cross-referencing indication****IPM** + +This element of service allows the originator to associate with the IP-message being sent, the globally unique identifiers of one or more other IP-messages. This enables the recipient's IPM UA, for example, to retrieve from storage a copy of the referenced IP-messages. + +**B.36 deferred delivery****MT** + +This element of service enables an originating UA to instruct the MTS that a message being submitted shall be delivered no sooner than a specified date and time. Delivery will take place as close to the date and time specified as possible, but not before. The date and time specified for deferred delivery is subject to a limit which is defined by the originator's management domain. + +**B.37 deferred delivery cancellation****MT** + +This element of service enables an originating UA to instruct the MTS to cancel a previously submitted deferred delivery message. The cancellation attempt may or may not always succeed. Possible reasons for failure are: deferred delivery time has passed, or the message has already been forwarded within the MTS. + +**B.38 delivery log****MS-94** + +This element of service enables an MS-user to access a log that records details of the messages and reports delivered to the MS; these records persist even after the messages and reports have been deleted. A delivery log entry contains a subset of the information that may be stored for a delivered message. The quantity of information stored in the delivery log for each message is specified at subscription time. The MS-user is able to determine whether the delivered message corresponding to a delivery log entry has been deleted. The MS-user is able to retrieve information from the delivery log by means of the stored message listing, stored message fetching and stored message summary elements of service. The ability to delete delivery log entries is subject to subscription, and may be restricted to messages meeting certain criteria, e.g., messages stored longer than an agreed period of time. + +**B.39 delivery notification****MT PR** + +This element of service enables an originating UA to request that the originating UA be explicitly notified when a submitted message has been successfully delivered to a recipient UA, or, in the case of access units, may indicate that the message has been successfully received by the destination terminal. The notification is related to the submitted message by means of the message identifier and includes the date and time of delivery. In the case of a multi-destination message, the originating UA can request this element of service on a per-recipient basis. + +When a message is delivered after distribution list expansion, then, depending on the policy of the distribution list, the notification can be sent to either the list owner, the message originator, or both. + +Delivery notification carries no implication that any UA or user action, such as examination of the message's content, has taken place. + +**B.40 delivery time stamp indication****MT PR** + +This element of service enables the MTS to indicate to a recipient UA the date and time at which the MTS delivered a message. In the case of physical delivery, this element of service indicates the date and time at which the PDAU has taken responsibility for printing and further delivery of the physical message. + +**B.41 delivery via bureaufax service****PD PR** + +This element of service allows an originating user to instruct the PDAU and associated PDS to use the bureaufax service for transport and delivery. + +**B.42 designation of recipient by directory name****MT PR** + +This element of service enables an originating UA to use a directory name in place of an individual recipient's OR-address. + +**B.43 disclosure of other recipients****MT** + +This element of service enables the originating UA to instruct the MTS when submitting a multi-recipient message, to disclose the OR-names of all other recipients to each recipient UA, upon delivery of the message. The OR-names disclosed are as supplied by the originating UA. If distribution list expansion has been performed, then only the originator specified DL name will be disclosed, and not the names of its members. + +**B.44 distribution codes indication****IPM** + +This element of service enables the originator to provide the recipient with information to support its distribution of the IP-message either within the MHS (e.g., auto-forwarding) or external to the MHS (e.g., hard copy distribution). A specific definition of distribution code semantics should be mutually supported by the originator and recipients. Note that this element of service may provide information to auto-actions such as auto-forward and auto-alert. + +**B.45 DL-exempted recipients****MT** + +This element of service enables the originator to specify the OR-names of recipients that are requested to be excluded from the set of intended recipients generated as a result of DL expansion. Exclusion is performed at the point of DL expansion. The names of exempted list members are also provided to the remaining recipients. This service does not guarantee that the exempted recipients will not receive the message as the result of other services (e.g., forwarding, redirection). + +**B.46 DL-expansion history indication****MT** + +This element of service provides to a recipient, at delivery, information about the distribution list(s) through which the message has arrived. It is a local matter as to how much of this information is presented to the recipient. + +**B.47 DL-expansion prohibited****MT** + +This element of service allows an originating user to specify that if any of the recipients can directly or via reassignment refer to a distribution list, then no expansion shall occur. Instead, a non-delivery notification will be returned to the originating UA, unless prevention of non-delivery notification has been requested. + +**B.48 EMS (Express Mail Service)****PD PR** + +This element of service allows an originating user to instruct the PDS to transport and deliver the physical message produced from the MHS message through accelerated letter circulation and delivery service (such as EMS or the equivalent domestic service) in the destination country. + +**B.49 expiry date indication****IPM** + +This element of service allows the originator to indicate to the recipient the date and time after which he considers the IP-message to be invalid. The intent of this element of service is to state the originator's assessment of the current applicability of an IP-message. The particular action on behalf of a recipient by his IPM UA, or by the recipient himself, is unspecified. Possible actions might be to file or delete the IP-message after the expiry date has passed. + +**B.50 explicit conversion****MT PR** + +This element of service enables an originating UA to request the MTS to perform a specified conversion, such as required when interworking between different telematic services. When a message is delivered after conversion has been performed, the recipient UA is informed of the original encoded information types as well as the current encoded information types in the message. + +NOTE 1 – This element of service is intended to support interworking with telematic terminals/services. + +NOTE 2 – When DL names are used in conjunction with this element of service, conversion will apply to all members of the DL. + +#### **B.51 forwarded IP-message indication** + +**IPM** + +This element of service allows a forwarded IP-message, or a forwarded IP-message plus its "delivery information" to be sent as the body (or as one of the body parts) of an IP-message. An indication that the body part is forwarded is conveyed along with the body part. In a multi-part body, forwarded body parts can be included along with body parts of other types. "Delivery information" is information which is conveyed from the MTS when an IP-message is delivered (for example, time stamps and indication of conversion). However, inclusion of this delivery information along with a forwarded IP-message in no way guarantees that this delivery information is validated by the MTS. + +The receipt notification request indication and the non-receipt notification request elements of service are not affected by the forwarding of an IP-message. + +#### **B.52 grade of delivery selection** + +**MT** + +This element of service enables an originating UA to request that transfer through the MTS be urgent or non-urgent, rather than normal. The time periods defined for non-urgent and urgent transfer are longer and shorter, respectively, than that defined for normal transfer. This indication is also sent to the recipient with the message. + +#### **B.53 hold for delivery** + +**MT** + +This element of service enables a recipient UA to request that the MTS hold its messages and returning notifications for delivery until a later time. The UA can indicate to the MTS when it is unavailable to take delivery of messages and notifications, and also, when it is again ready to accept delivery of messages and notifications from the MTS. The MTS can indicate to the UA that messages are waiting due to the criteria the UA established for holding messages. Responsibility for the management of this element of service lies with the recipient MTA. + +Criteria for requesting a message to be held for delivery are: encoded information type, content type, maximum content length and priority. The message will be held until the maximum delivery time for that message expires, unless the recipient releases the hold prior to its expiry. + +NOTE – The hold for delivery element of service is distinct from the message store facility. The hold for delivery element of service provides temporary storage to facilitate delivery and only after a message has been transferred to the recipient's UA is delivery notification returned. The message store facility augments the storage of a UA and can be used to store messages for an extended period of time. Unlike the hold for delivery element of service, delivery notifications are returned as soon as the message is placed in (that is, delivered to) the message store. + +#### **B.54 implicit conversion** + +**MT** + +This element of service enables a recipient UA to have the MTS perform for a period of time any necessary conversion on messages prior to delivery. Neither the originating nor recipient UA explicitly requests this element of service on a per-message basis. If the encoded information type capabilities of the recipient UA are such that more than one type of conversion can be performed, the most appropriate conversion is performed. When a message is delivered after conversion has been performed, the recipient UA is informed of the original encoded information types as well as the current encoded information types in the message. + +#### **B.55 importance indication** + +**IPM** + +This element of service allows the originator to indicate to the recipients his assessment of the importance of the IP-message being sent. Three levels of importance are defined: *low*, *normal*, and *high*. + +This element of service is not related to the grade of delivery selection element of service provided by the MTS. The particular action taken by the recipient or his IPM UA based on the importance categorization is unspecified. It is the intent to allow the recipient IPM UA, for example, to present IP-messages in order of their importance or to alert the recipient of the arrival of IP-messages of high importance. + +#### **B.56 incomplete copy indication** + +**IPM** + +This element of service allows an originator to indicate that this IP-message is an incomplete copy of an IP-message with the same IP-message identification in that one or more body parts, and/or heading fields of the original IP-message are absent. + +#### **B.57 information category indication** + +**IPM** + +This element of service enables the originator to indicate to the recipient the character of the information contained in the IP-message. The service can provide a registered identifier for each particular category, or free form information describing the nature of the communication. The recipients may use the information provided by this service to affect the presentation of messages to the recipient, or to affect any other local processing functions. A specific definition of information category values and semantics should be mutually supported by the originator and the recipient. Examples of information categories include: *draft message, press release, contractual commitment, policy statement*. + +#### **B.58 IP-message action status** + +**IPM MS-94** + +This element of service enables an MS-user to determine whether a reply or a receipt notification has been requested of the user in an IP-message which the user has received. It allows the user to record in the MS (and subsequently retrieve the information) that the reply (or IP-notification) has been sent. In addition, the user may set a reminder that a reply is intended even if no reply was explicitly requested. + +#### **B.59 IP-message identification** + +**IPM** + +This element of service enables co-operating IPM UAs to convey a globally unique identifier for each IP-message sent or received. The IP-message identifier is composed of an OR-name of the originator and an identifier that is unique with respect to that name. IPM UAs and users use this identifier to refer to a previously sent or received IP-message (for example, in receipt notifications). + +#### **B.60 IP-message security labelling** + +**IPM** + +This element of service augments the message security labelling service (see B.67) by allowing the originator of an IP-message to convey to all recipients an indication of the security classification of the IP-message content, or optionally, of the component heading and body parts of an IP-message. This service enables the implementation of security policies in which the security labels associated with local objects (e.g., files) derived from component parts of the IP-message may be assigned values provided by the originating IPM user. The integrity of the IP-message security labelling may be provided by the content integrity or body part authentication and integrity security service, and confidentiality of the IP-message security labelling may be provided by the content confidentiality security service. Authentication of the originator of the IP-message security labelling may be provided by the message origin authentication service or the body part authentication and integrity service. + +NOTE 1 – Unless both end systems have mutual trust in each end system's ability to process and separate information based on security labels, this label should not be used to implement mandatory access control. + +NOTE 2 – The meaning of the term "security classification" in this context is relative to the specific security policy in force. + +#### **B.61 language indication** + +**IPM** + +This element of service enables an originating UA to indicate the language type(s) of a submitted IP-message. + +#### **B.62 latest delivery designation** + +**MT** + +This element of service enables an originating UA to specify the latest time by which the message is to be delivered. If the MTS cannot deliver by the time specified, the message is not delivered and is cancelled. On multi-recipient messages, the latest delivery time can expire prior to delivery to all recipients, but this will not negate any deliveries which have already occurred. + +#### **B.63 manual handling instructions indication** + +**IPM** + +This element of service enables the originator to indicate to the recipient instructions for manual handling of the IP-message, following its delivery. The service can provide instructions consisting of free form text. Examples of manual handling instructions include special recipient handling requests (e.g., "Please pass to...", "Please DO NOT pass to..."), and instructions on how to process body data. + +NOTE – Instructions indicated by this element of service may apply either to the IP-message as a whole or to specific components of the IP-message. Where necessary, the content of the instructions should indicate the scope of the instructions or the part(s) of the IP-message to which the instruction applies. + +**B.64 message flow confidentiality****MT** + +This element of service allows the originator of the message to protect information which might be derived from observation of the message flow. + +NOTE – Only a limited form of this is supported. + +**B.65 message identification****MT** + +This element of service enables the MTS to provide a UA with a unique identifier for each message or probe submitted or delivered by the MTS. UAs and the MTS use this identifier to refer to a previously submitted message in connection with elements of service such as delivery and non-delivery notification. + +**B.66 message origin authentication****MT PR** + +This element of service allows the originator of a message to provide to the recipient(s) of the message, and any MTA through which the message is transferred, a means by which the origin of the message can be authenticated (i.e., a signature). Message origin authentication can be provided to the recipient(s) of the message, and any MTA through which the message is transferred, on a per-message basis using an asymmetric encryption technique, or can be provided only to the recipient(s) of the message, on a per-recipient basis using either an asymmetric or a symmetric encryption technique. + +**B.67 message security labelling****MT** + +This element of service allows the originator of a message (or probe) to associate with the message (and any reports on the message or probe) an indication of the sensitivity of the message (a security label). The message security label may be used by the MTS and the recipient(s) of the message to determine the handling of the message in line with the security policy in force. + +**B.68 message sequence integrity****MT PR** + +This element of service allows the originator of the message to provide to a recipient of the message a means by which the recipient can verify that the sequence of messages from the originator to the recipient has been preserved (without message loss, re-ordering, or replay). Message sequence integrity is on a per-recipient basis, and can use either an asymmetric or a symmetric encryption technique. + +**B.69 MS register****MS** + +This element of service enables an MS-user to register various items of information with the MS in order to modify certain aspects of its behaviour, such as: + +- 1) the performance of automatic actions; +- 2) the default set of information retrieved when using the stored message fetching and stored message listing elements of service. One set of information may be registered per UA employed by the user; +- 3) the credentials used by the message store to authenticate the MS-user. + +If a user employs more than one UA implementation, then as a subscription option the MS may store a separate set of registration information for each UA. The user may retrieve the registered information from the MS. + +NOTE – The capability to store separate sets of registration information and to retrieve registered information was not defined in versions of F.400/X.400 and ISO/IEC 10021-1 published prior to 1994. + +**B.70 multi-destination delivery****MT PR** + +This element of service enables an originating UA to specify that a message being submitted is to be delivered to more than one recipient UA. Simultaneous delivery to all specified UAs is not implied by this element of service. + +**B.71 multi-part body****IPM** + +This element of service allows an originator to send to a recipient or recipients an IP-message with a body that is partitioned into several parts. The nature and attributes, or type, of each body part are conveyed along with the body part. + +## **B.72 non-delivery notification** + +**MT PR** + +This element of service enables the MTS to notify an originating UA if a submitted message was not delivered to the specified recipient UA(s), or, in the case of access units, may indicate that the message was not received by the destination terminal. The reason the message was not delivered is included as part of the notification. For example, the recipient UA can be unknown to the MTS. + +In the case of a multi-destination message, a non-delivery notification can refer to any or all of the recipient UAs to which the message could not be delivered. + +When a message is not delivered after distribution list expansion, then, depending on the policy of the distribution list, the notification can be sent to either the list owner, the message originator, or both. + +NOTE – Non-delivery notifications are generated automatically, and do not depend on a request by an originator. + +## **B.73 non-receipt notification request indication** + +**IPM PR** + +This element of service allows the originator to ask that he be notified should the IP-message be deemed unreceivable. In the case of a multi-recipient IP-message, the originator can request this element of service on a per-recipient basis. + +The originator's UA conveys his request to the recipient's UA. The recipient's UA automatically issues a non-receipt notification when any of the following events occur: + +- 1) The recipient's UA auto-forwards the IP-message to another user. +- 2) The recipient's UA discards the IP-message prior to receipt. +- 3) The recipient's subscription is terminated before he receives the IP-message. + +Since receipt can occur arbitrarily long after delivery, the recipient's failure to access the IP-message, even for a long period of time (for example, while on an extended business trip), does not constitute non-receipt and thus no notification is issued. However, an advice notification may be issued to advise the originator of the recipient's temporary absence. + +NOTE – No legal significance can be adduced from this element of service. + +## **B.74 non-repudiation of content received** + +**IPM PR** + +This element of service enables a recipient of an IP-message to provide an irrevocable proof that the original IP-message content was received by the recipient. This service provides irrevocable proof of the integrity of the content received and irrevocable proof of the authenticity of the recipient of the IP-message. This service fulfils the same function as the proof of content received element of service, but in a manner which cannot be repudiated. + +The corresponding irrevocable proof can be supplied in various ways depending on the security policy in force. The originator of the IP-notification always uses the non-repudiation of origin element of service when sending the IP-notification in response to the IP-message. + +One way of providing the irrevocable proof is to incorporate the following in the IP-notification: + +- A verified copy of the IP-message originator's non-repudiation of origin arguments (when present in the IP-message and verified by the recipient of the IP-message). +- A verified copy of the complete IP-message content, if the IP-message originator's "Non-repudiation of origin" arguments are not present in the IP-message. + +NOTE – As an alternative to invoking this element of service, equivalent security may be achieved by the use of a notarization mechanism, which requires bilateral agreement outside the scope of this Recommendation. + +The recipient is required to fulfil the request for this element of service only when the UA is subject to a security policy which mandates the support of this element of service. + +## **B.75 non-repudiation of delivery** + +**MT PR** + +This element of service allows the originator of a message to obtain from the recipient(s) of the message irrevocable proof that the message was delivered to the recipient(s). This will protect against any attempt by the recipient(s) to subsequently deny receiving the message or its content. Non-repudiation of delivery is provided to the originator of a message on a per-recipient basis using asymmetric encryption techniques. + +#### **B.76 non-repudiation of IP-notification** + +**IPM PR** + +This element of service provides the recipient of an IP-notification with irrevocable proof of the identity of the originator of the IP-notification and with proof that the corresponding IP-message was received by the recipient. + +This protects against any attempt by the recipient to deny subsequently that the IP-message was received or that the IP-notification was returned to the originator of the IP-message. This element of service fulfils the same service as proof of IP-notification but in a manner which cannot be repudiated. + +This element of service is used only in conjunction with non-repudiation of origin element of service applied to the IP-notification. + +The corresponding irrevocable proof can be supplied in various ways depending on the security policy in force. One way of providing the irrevocable proof is by means of the MTS-user to MTS-user data origin authentication security services defined in 10.2.1.1.1/X.402 | ISO/IEC 10021-2 applied to the IP-notification, when the security service has non-repudiation properties. + +The recipient is required to fulfil the request for this element of service only when the UA is subject to a security policy which mandates the support of this element of service. + +#### **B.77 non-repudiation of origin** + +**MT PR** + +This element of service allows the originator of a message to provide the recipient(s) of the message irrevocable proof of the origin of the message and the integrity of its content. This will protect against any attempt by the originator to subsequently revoke the message or its content. Non-repudiation of origin is provided to the recipient(s) of a message on a per-message basis using asymmetric encryption techniques. + +#### **B.78 non-repudiation of submission** + +**MT** + +This element of service allows the originator of a message to obtain irrevocable proof that a message was submitted to the MTS for delivery to the originally specified recipient(s). This will protect against any attempt by the MTS to subsequently deny that the message was submitted for delivery to the originally specified recipient(s). Non-repudiation of submission is provided to the originator of a message on a per-message basis, and uses an asymmetric encryption technique. + +#### **B.79 obsoleting indication** + +**IPM** + +This element of service allows the originator to indicate to the recipient that one or more IP-messages he sent previously are obsolete. The IP-message that carries this indication supersedes the obsolete IP-message. + +The action to be taken by the recipient or his IPM UA is a local matter. The intent, however, is to allow the IPM UA or the recipient to, for example, remove or file obsolete messages. + +#### **B.80 ordinary mail** + +**PD PR** + +This element of service enables the PDS to transport and deliver the letter produced from the MHS message in the mode available through the ordinary letter mail service in the country of destination. This is the default action for the transport and delivery of a physical message. + +#### **B.81 original encoded information types indication** + +**MT** + +This element of service enables an originating UA to specify to the MTS the encoded information types of a message being submitted. When the message is delivered, it also indicates to the recipient UA the encoded information types of the message specified by the originating UA. + +#### **B.82 originator indication** + +**IPM** + +This element of service allows the identity of the originator to be conveyed to the recipient. The intent of this IPM element of service is to identify the originator in a user-friendly way. In contrast, the MTS provides to the recipient the actual OR-address and directory name, if present, of the originator. DL names should not be used in originator indication. + +#### **B.83 originator reference indication** + +**IPM** + +This element of service enables the originator to indicate to a recipient a reference that is chosen by the originator. The originator reference may be used within the organization of the originator as an internal reference. Examples of possible originator references include: file number, claim number and legal case number. This information may be used by the recipient in later communications with the originator, possibly via non-MHS means, concerning a particular IP-message. + +#### **B.84 originator requested alternate recipient** + +**MT PR** + +This element of service enables an originating UA to specify, for each intended recipient, one alternate recipient to which the MTS can deliver the message, if delivery to the intended recipient is not possible. The alternate recipient can be a distribution list. For the purposes of determining success or failure (and hence delivery and non-delivery notifications), delivery to the originator requested alternate recipient is equivalent to delivery to the intended recipient. If the intended recipient has requested redirection of incoming messages, and if the originating UA has requested redirection allowed by the originator, the system first tries to redirect the message. If this fails, the system then attempts to deliver the message to the designated alternate recipient. + +#### **B.85 physical delivery notification by MHS** + +**PD PR** + +This element of service allows an originating user to request that an explicit notification, informing the originator of either successful or unsuccessful delivery of the physical message, be generated and returned by MHS. The notification provides information on delivery but no physical record is provided by the PDS. + +NOTE 1 – The notification includes the date and time of delivery based on the delivery confirmation given by the delivery person, the addressee or another authorized person. This is subject to national regulations in the destination country and is also dependent on the type of delivery requested (e.g., in the case of registered mail to addressee in person, the addressee would be the confirming person). + +NOTE 2 – This notification carries no implication that any action on the part of the recipient (such as examination of the message content) has taken place. + +NOTE 3 – When this element of service is requested, and the physical message is undeliverable, it is either returned or destroyed depending on national regulations in the destination country, which means that the default action of the element of service B.132 is overridden. + +#### **B.86 physical delivery notification by PDS** + +**PD PR** + +This element of service allows an originating user to request that an explicit notification, informing the originator of either successful or unsuccessful delivery of the physical message, be generated and returned by the PDS. The notification serves as a record of delivery for the originating user to retain for reference. + +NOTE 1 – The notification includes the date and time, and, in the case of successful delivery, the signature of the person confirming the delivery. The confirming person can be the delivery person, the addressee or another authorized person. This is subject to national regulations in the destination country and is also dependent on the type of delivery requested (e.g., in the case of registered mail to addressee in person, the addressee would be the confirming person). + +NOTE 2 – This notification carries no implication that any action on the part of the recipient (such as examination of the message content) has taken place. + +NOTE 3 – When this element of service is requested, and the physical message is undeliverable, it is either returned or destroyed depending on national regulations in the destination country, which means that the default action of the element of service B.132 is overridden. + +#### **B.87 physical forwarding allowed** + +**PD PR** + +This element of service enables the PDS to forward the physical message to a forwarding address if the recipient has changed his address and indicated this to the PDS. This is the default action taken by the PDS. + +#### **B.88 physical forwarding prohibited** + +**PD PR** + +This element of service allows an originating user to instruct the PDS not to forward the physical message to a forwarding address. + +## **B.89 precedence indication** + +**IPM PR** + +This element of service enables an originator to indicate a precedence value for each recipient of the message. The per-recipient precedence value provides an indication of the perceived importance of the IP-message or its expected relevance to the recipients. The aggregate of precedence values assigned to all the recipients of a message shall also affect the grade of delivery selection (see B.52). Whether the UA performs multiple submissions to the MTS using different grade of delivery selection values is a local matter. + +The semantics of values used by this service shall be determined by local policy. The following is an example of possible precedence values: + +- *deferred*: The message is of relatively low importance to the recipient, and should be handled after all other messages; this value corresponds to a grade of delivery selection of *non-urgent*. +- *normal*: The message is a routine communication of no distinguishing importance to the recipient, and should be handled at the convenience of the recipient; this value corresponds to a grade of delivery selection of *normal*. +- *action-today*: The message is of high importance to the recipient, and should be acted upon by the recipient before the close of the current business day; this value corresponds to a grade of delivery selection of *urgent*. +- *immediate*: The message is for the immediate attention of the recipient, and should be acted upon by the recipient with all possible speed; this value corresponds to a grade of delivery selection of *urgent*. + +## **B.90 prevention of non-delivery notification** + +**MT PR** + +This element of service enables an originating UA to instruct the MTS not to return a non-delivery notification to the originating UA in the event that the message being submitted is judged undeliverable. In the case of a multi-destination message, the originating UA can request this element of service on a per-recipient basis. + +## **B.91 primary and copy recipients indication** + +**IPM** + +This element of service allows the originator to provide the names of zero or more users, or DLs, who are the intended primary recipients of the IP-message, and the names of zero or more users, or DLs, who are the intended copy recipients of the IP-message. It is intended to enable a recipient to determine the category in which each of the specified recipients (including the recipient himself) was placed. The exact distinction between these two categories of recipients is unspecified. However, the primary recipients, for example, might be expected to act upon the IP-message, while the copy recipients might be sent the IP-message for information only. + +The names of users and DLs can be expressed in several ways, some of which are intended for human use only and not intended to be names by which the MHS could make deliveries. This element of service allows originators to distinguish between the names which the originating UA will add to the multi-destination delivery (see B.70) and the names to which the MHS will not be requested to make deliveries. This may result in none of the primary recipients or copy recipients contributing to the Multi-destination Delivery. + +NOTE – As an example of this element of service in a typical memorandum, the primary recipients are normally designated by the directive "to:" while "cc:" identifies the copy recipients. + +## **B.92 probe** + +**MT** + +This element of service enables a UA to establish before submission whether a particular message could be delivered. The MTS provides the submission information and generates delivery and/or non-delivery notifications indicating whether a message with the same submission information could be delivered to the specified recipient UAs. + +The probe element of service includes the capability of checking whether the content size, content type, and/or encoded information types would render it undeliverable. The significance of the result of a probe depends upon the recipient UA(s) having registered with the MTS the encoded information types, content type and maximum message size that it can accept. This element of service is subject to the same delivery time targets as for the urgent class. In the case of DLs, a Probe indicates nothing about the likelihood of successful delivery to the DL members, but only whether the originator has the right to submit to the DL. + +## **B.93 probe origin authentication** + +**MT** + +This element of service allows the originator of a probe to provide to any MTA through which the probe is transferred a means to authenticate the origin of the probe (i.e. a signature). Probe origin authentication is on a per-probe basis, and uses an asymmetric encryption technique. + +## **B.94 proof of content received** + +**IPM PR** + +This element of service enables a recipient of an IP-message to provide proof that the original IP-message content was received by the recipient. This service provides proof of the integrity of the content received and proof of the authenticity of the recipient of the IP-message. + +This element of service is used only in conjunction with content integrity or message origin authentication elements of service applied to the subject IP-notification. + +The corresponding proof can be supplied in various ways depending on the security policy in force. The originator of the IP-notification always uses the content integrity or message origin authentication elements of service when sending the receipt IP-notification in response to the IP-message. + +One way of providing the proof is to incorporate the following in the IP-notification: + +- A verified copy of the IP-message originator's content integrity or message origin authentication arguments (when present in the IP-message and verified by the recipient of the IP-message). +- A verified copy of the complete original IP-message content, if the IP-message originator's content integrity or message origin authentication arguments are not present in the IP-message. + +The recipient is required to fulfil the request for this element of service only when the UA is subject to a security policy which mandates the support of this element of service. + +NOTE 1 – The message origin authentication element of service may be provided on a per-message basis using the message-origin-authentication-check or on a per-recipient basis using the message-token as defined in ITU-T Rec. X.411 | ISO/IEC 10021-4. + +NOTE 2 – The content integrity element of service may be conveyed in several places on the message envelope. The content-integrity-check can be stand-alone security argument in the message envelope and/or attributes of the message-token as defined in ITU-T Rec. X.411 | ISO/IEC 10021-4. + +## **B.95 proof of delivery** + +**MT PR** + +This element of service allows the originator of a message to obtain from the recipient(s) of the message the means to authenticate the identity of the recipient(s) and the delivered message and content. Message recipient authentication is provided to the originator of a message on a per-recipient basis using either symmetric or asymmetric encryption techniques. + +## **B.96 proof of IP-notification** + +**IPM PR** + +This element of service provides the originator of an IP-message with proof that the IP-message was received by its recipient, and that the recipient was the originator of the received IP-notification. + +This protects against any attempt by the recipient IPM UA to deny subsequently that the IP-message was received and that the IP-notification was returned to the originator. + +This element of service is used only in conjunction with content integrity and/or the message origin authentication elements of service applied to the IP-notification. + +The corresponding proof can be supplied in various ways depending on the security policy in force. One way of providing the proof is by means of the MTS-user to MTS-user data origin authentication security services, defined in 10.2.1.1.1/X.402 | ISO/IEC 10021-2, applied to the IP-notification. + +The recipient is required to fulfil the request for this element of service only when the UA is subject to a security policy which mandates the support of this element of service. + +## **B.97 proof of submission** + +**MT** + +This element of service allows the originator of a message to obtain from the MTS the means to authenticate that the message was submitted for delivery to the originally intended recipient. Message submission authentication is provided on a per-message basis, and can use symmetric or asymmetric encryption techniques. + +## **B.98 receipt notification request indication** + +**IPM PR** + +This element of service allows the originator to ask that he be notified when the IP-message being sent is received. In the case of a multi-recipient message, the originator can request this element of service on a per-recipient basis. This element of service also implicitly requests non-receipt notification request indication. + +The originator's UA conveys his request to the recipient's UA. The recipient can instruct his UA to honour such requests, either automatically (for example, when it first renders the IP-message on the recipient's terminal) or upon his explicit command. The recipient can also instruct his UA, either in blanket fashion or case-by-case, to ignore such requests. + +#### **B.99 redirection disallowed by originator** + +**MT** + +This element of service enables an originating UA to instruct the MTS, if the recipient has requested the redirection of incoming messages element of service, that redirection should not be applied to a particular submitted message. + +#### **B.100 redirection of incoming messages** + +**MT** + +This element of service enables a recipient UA, through registration, to instruct the MTS to redirect incoming messages addressed to it, to another UA or to a DL. This instruction remains in effect for a specified period of time, or until revoked. + +NOTE 1 – This is an MT element of service that does not require delivery to the intended recipient to occur before redirection can take place. It is therefore distinct from the auto-forwarding of IP-messages element of service. + +NOTE 2 – Different incoming messages, on the basis of their content-types, security labels, and other criteria, may be redirected to different alternate recipients or not redirected at all. + +#### **B.101 registered mail** + +**PD PR** + +This element of service allows an originating user to instruct the PDS to handle the physical message as registered mail. + +#### **B.102 registered mail to addressee in person** + +**PD PR** + +This element of service allows an originating user to instruct the PDS to handle the physical message as registered mail and to deliver it to the addressee only. + +#### **B.103 reply request indication** + +**IPM PR** + +This element of service allows the originator to request that a recipient send an IP-message in reply to the IP-message that carries the request. The originator can also specify the date by which any reply should be sent, and the one or more users and DLs to whom the originator requests (but does not demand) be among the intended recipients of any reply. The recipient is informed of the date and names but it is up to the recipient to decide whether or not, and if so, to whom to reply. + +NOTE – A blind copy recipient should consider carefully to whom he sends a reply, in order that the meaning of the blind copy recipient indication element of service is preserved. + +#### **B.104 replying IP-message indication** + +**IPM** + +This element of service allows the originator of an IP-message to indicate to the recipient(s) that this IP-message is being sent in reply to another IP-message. A reply can, depending on the wishes of the originator of the replied-to message, and the final decision of the originator of the reply, be sent to: + +- 1) The recipients specified in the reply request indication of the replied-to message. +- 2) The originator of the replied-to message. +- 3) The originator and other recipients. +- 4) A distribution list, in which the originator of the replied-to message can be a receiving member. +- 5) Other recipients as chosen by the originator of the reply. + +The recipients of the reply receive it as a regular IP-message, together with an indication of which IP-message it is a reply to. + +#### **B.105 report origin authentication** + +**MT** + +This element of service allows the originator of a message (or probe) to authenticate the origin of a report on the delivery or non-delivery of the subject message (or probe), (a signature). Report origin authentication is on a per-report basis, and uses an asymmetric encryption technique. + +**B.106 request for forwarding address****PD PR** + +This element of service allows an originating user to instruct the PDS to provide the forwarding address if the recipient has changed his address and indicated this to the PDS. + +This element of service can be used with either physical forwarding allowed or prohibited. The provision of the forwarding address by the PDS to an originating user is subject to national regulations in the destination country. The default action is no provision of the forwarding address. + +**B.107 request for non-repudiation of content received****IPM PR** + +This element of service enables the originator of an IP-message to request the recipient of the IP-message to provide an irrevocable proof of the received IP-message content by means of an IP-notification. + +This element of service may be subscribed to only if the receipt notification request indication element of service is subscribed to. + +If this element of service is requested, the request for proof of content received element of service shall not be requested. + +This element of service provides only an indication of the originator's request. Fulfilment of the request requires support of the non-repudiation of content received element of service. + +**B.108 request for non-repudiation of IP-notification****IPM PR** + +This element of service enables the originator of an IP-message to request the recipient of the IP-message to provide irrevocable proof of the origin of an IP-notification generated in response to the IP-message. + +This element of service may be subscribed to only if the receipt notification request indication element of service is subscribed to. + +If this element of service is requested, the request for proof of IP-notification element of service shall not be requested. + +This element of service provides only an indication of the originator's request. Fulfilment of the request requires support of the non-repudiation of IP-notification element of service. + +**B.109 request for proof of content received****IPM PR** + +This element of service enables the originator of the IP-message to request the recipient of the IP-message to provide proof of the received IP-message content by means of an IP-notification. + +This element of service may be subscribed to only if the receipt notification request indication element of service is subscribed to. + +This element of service provides only an indication of the originator's request. Fulfilment of the request requires support of the proof of content received element of service. + +**B.110 request for proof of IP-notification****IPM PR** + +This element of service enables the originator of the IP-message to request the recipient of the IP-message to provide proof of the origin of an IP-notification generated in response to the IP-message. + +This element of service may be subscribed to only if the receipt notification request indication element of service is subscribed to. + +This element of service provides only an indication of the originator's request. Fulfilment of the request requires support of the proof of IP-notification element of service. + +**B.111 requested preferred delivery method****MT PR** + +This element of service allows a user to request, on a per-recipient basis, the preference of method or methods of message delivery (such as through an access unit). + +NOTE – This assumes availability of a Directory and specification of a directory name by the originator together with this element of service. It may not be possible to match the request with the OR-addresses available in the Directory. Non-delivery may occur if no feasible match can be found. + +#### **B.112 restricted delivery** + +**MT** + +This element of service enables a recipient UA to indicate to the MTS, through registration, that it is not prepared to accept delivery of messages which originate from, or are redirected by, or are DL-expanded by certain MTS-users. + +The recipient UA may register several sets of criteria to restrict the delivery of different categories of message. + +#### **B.113 return of content** + +**MT** + +This element of service enables an originating UA to request that the content of a submitted message be returned with any non-delivery notification. This will not be done, however, if any encoded information type conversion has been performed on the message's content. + +#### **B.114 secure access management** + +**MT** + +This element of service enables an MTS-user to establish an association with the MTS, or the MTS to establish an association with an MTS-user, or an MTA to establish an association with another MTA. It also establishes the strong credentials of the objects to interact, and the context and security-context of the association. Secure Access Management can use either an asymmetric or a symmetric encryption technique. When access security is achieved through strong credentials, they can be periodically updated. + +#### **B.115 sensitivity indication** + +**IPM** + +This element of service allows the originator of an IP-message to specify guidelines for the relative sensitivity of the message upon its receipt. It is the intent that the sensitivity indication should control such items as: + +- 1) Whether the recipient should have to prove his identity to receive the IP-message. +- 2) Whether the IP-message should be allowed to be printed on a shared printer. +- 3) Whether an IPM UA should allow the recipient to forward the received IP-message. +- 4) Whether the IP-message should be allowed to be auto-forwarded. + +The sensitivity indication can be indicated to the recipient or interpreted directly by the recipient's IPM UA. + +If no sensitivity level is indicated, it should be assumed that the IP-message's originator has advised no restriction on the recipient's further disposition of the IP-message. The recipient is free to forward, print, or otherwise do as he chooses with the IP-message. + +Three specific levels of sensitivity above the default are defined: + +*Personal:* The IP-message is sent to the recipient as an individual, rather than to him in his role. There is no implication that the IP-message is private, however. + +*Private:* The IP-message contains information that should be seen (or heard) only by the recipient, and not by anyone else. The recipient's IPM UA can provide services to enforce this intent on behalf of the IP-message's originator. + +*Company-confidential:* The IP-message contains information that should be treated according to company-specific procedures. + +#### **B.116 special delivery** + +**PD PR** + +This element of service allows an originating user to instruct the PDS to transport the letter produced from the MHS message through the ordinary letter mail circulation system and to deliver it by special messenger delivery. + +#### **B.117 storage of draft messages** + +**MS-94** + +This element of service enables an MS-user to store draft messages in the MS. The user may obtain summaries of draft messages and may access a draft message by means of the stored message listing and stored message fetching elements of service. + +**B.118 storage on submission****MS-94** + +This element of service enables an MS-user to instruct the MS to store a copy of a message upon its submission, either by the MS-user or as a result of the performance of an auto-action. Storage of a submitted message is conditional upon the success of the submission. The user may instruct the MS to store all submitted messages, or may control storage on a per-message basis. + +**B.119 storage period assignment****MS-94** + +This element of service enables an MS-user to assign a storage period to a stored message. The storage period indicates the period of time for which the user anticipates the message should be retained in the MS; this may be expressed as a period of time (from the start of storage), or as an absolute date and time. This element of service must be subscribed to if the auto-deletion after storage period or auto-assignment of storage period elements of service are subscribed to. + +**B.120 stored message alert****MS** + +This element of service allows a user of an MS to register relevant sets of criteria that can cause an alert to be generated to the user when a message arrives at the MS satisfying the selected criteria. The generation of the alert can occur as follows: + +- 1) If the UA is connected and on-line to the MS, the alert message will be sent to the UA as soon as a message arrives at the MS that satisfies the registered criteria for generating alerts. If the UA is off-line, then the next time the UA connects to his MS after a message arrives at the MS satisfying the registered criteria, the user will be informed that one or more alert cases have occurred, the details of which can be determined by performing a stored message summary. +- 2) In addition to, or as an alternative to 1 above, the MS can use other mechanisms to inform the user. + +**B.121 stored message annotation****MS-94** + +This element of service enables an MS-user to attach one or more textual annotations to a stored message. Annotations apply to the complete message and may not be applied selectively to different parts of the message. Annotations are local to the MS and MS-user and are not transmitted through the MTS in any message. The "cover note" described in B.19 is not related to message annotations. + +**B.122 stored message deletion****MS** + +This element of service enables a recipient UA to delete certain of its messages from the MS. Subject to subscription, deletion may be restricted to messages meeting certain criteria, e.g., messages stored for longer than an agreed period of time. Messages cannot be deleted if they have not been previously listed. + +**B.123 stored message fetching****MS** + +This element of service enables a recipient UA to fetch from the MS a message, or portions of a message. The UA can fetch a message (or message portion) based on the same search criteria that can be used for stored message listing. + +**B.124 stored message grouping****MS-94** + +This element of service enables an MS-user to attach group names to messages stored in the MS. A message can have zero, one, or more group names associated with it that can subsequently be used for selection purposes. Each message group name comprises a sequence of components which may be regarded as modelling a storage hierarchy. The setting, changing, or deletion of the group names attached to a message can be performed by the MS-user. + +The UA indicates to the MS, through registration, the name of each distinct group which the UA will employ to label each group of related messages. Each group name may be assigned a descriptive text registered together with the group name. The MS will verify that the group names subsequently employed by the user belong to the registered set of group names, and will prevent the user from deregistering group names which are currently attached to stored messages, or which are registered for use by the auto-assignment of group names element of service. A group name remains valid until it is deregistered. The MS will prohibit an attempt to register the same group name twice. + +**B.125 stored message listing****MS** + +This element of service provides a recipient UA with a list of information about certain of its messages stored in the MS. The information comprises selected attributes from a message's envelope and content and others added by the MS. The UA can limit the number of messages that will be listed. + +**B.126 stored message summary****MS** + +This element of service provides a recipient UA with a count of the number of messages satisfying a specified criteria based on one or more attributes of the message stored in the MS. + +**B.127 subject indication****IPM** + +This element of service allows the originator to indicate to the recipient(s) the subject of an IP-message being sent. The subject information is to be made available to the recipient. + +**B.128 submission log****MS-94** + +This element of service enables an MS-user to access a log that records details of the messages submitted from the MS to the MTS. These records are generated regardless of whether a copy of the submitted message is stored by means of the storage on submission element of service. Even where a copy is stored, the corresponding submission log entry may persist after the message has been deleted. Both successful and unsuccessful submissions are recorded. A submission log entry contains a subset of the information that may be stored for a submitted message. The quantity of information stored in the submission log for each message is specified at subscription time. The MS-user is able to determine whether the submitted message corresponding to a submission log entry has been deleted. The MS-user is able to retrieve information from the submission log by means of the stored message listing, stored message fetching and stored message summary elements of service. The ability to delete submission log entries is subject to subscription, and may be restricted to messages meeting certain criteria, e.g., messages stored longer than an agreed period of time. + +**B.129 submission of IP-messages incorporating stored messages****IPM MS-94** + +This element of service enables an MS-user to instruct the MS to incorporate parts of one or more stored messages as body parts of a submitted IP-message. The submitted IP-message may also contain body parts supplied in the submission from the MS-user. + +The stored message which is the source of a body part may be a delivered, submitted or draft message. Individual body parts or the whole content of a stored IP-message may be incorporated. When the content is incorporated, it will form a Forwarded IP-message. Delivery, information may also be incorporated from delivered messages when the content is incorporated. + +The MS may optionally support the forwarding of body parts from messages which are not IP-messages. In this case, only body parts whose definition is compatible with IPM (or for which rules of conversion into IPM body parts are defined) may be forwarded. The complete content of a message cannot be forwarded if the message is not an IP-message. + +The message submitted to the MTS, incorporating the stored messages or body parts, may be stored in the MS if the user subscribes to the storage on submission element of service. An extract of the message will also be stored in the submission log if this element of service is subscribed to. + +**B.130 submission time stamp indication****MT** + +This element of service enables the MTS to indicate to the originating UA and each recipient UA the date and time at which a message was submitted to the MTS. In the case of physical delivery, this element of service also enables the PDAU to indicate the date and time of submission on the physical message. + +**B.131 typed body****IPM** + +This element of service permits the nature and attributes of the body of the IP-message to be conveyed along with the body. Because the body can undergo conversion, the body type can change over time. + +NOTE 1 – One example is the use of a file transfer body part. This provides for conveying the contents of a stored file and other information associated with the file from originator to recipient. The other information includes: + +- file attributes, which are typically stored along with the file contents; +- information on the environment from which the transfer originated; + +- references to existing stored files or earlier messages. + +NOTE 2 – Another example is the use of a voice body part. + +**B.132 undeliverable mail with return of physical message PD PR** + +This element of service enables the PDS to return the physical message without delay, with reason indicated to the originator, if it cannot be delivered to the addressee. This is the default action to be taken by the PDS. + +NOTE – In the case of "poste restante", the return of the physical message will take place after some period of time. + +**B.133 use of distribution list MT PR** + +This element of service enables an originating UA to specify a distribution list in place of all the individual recipients (users or nested DLs) mentioned therein. The MTS will add the members of the list to the recipients of the message and send it to those members. Distribution lists can be members of distribution lists, in which case the list of recipients can be successively expanded at several places in the MTS. + +**B.134 user/UA capabilities registration MT** + +This element of service enables a UA to indicate to its MTA, through registration, the categories of message it is capable of handling, and which the MTA may deliver to it. A message category is defined as a combination of various properties: + +- 1) the content-type(s) of messages which may be delivered; +- 2) the encoded information type(s) of messages which may or may not be delivered; +- 3) additional properties, including the maximum message length, and the security labels present. + +NOTE – It is possible to register certain encoded information types such that they cause a message to be delivered regardless of the other encoded information types present. A user may declare certain encoded information types undeliverable to cause the MTS to perform implicit conversion. + +The UA may specify different sets of registration information to control the delivery of different categories of message. + +The MTA will not deliver to a UA a message that does not match, or exceeds, the capabilities registered. + +## Annex C + +### Elements of service changes from 1992 + +(This annex does not form an integral part of this Recommendation) + +#### C.1 New elements of service in 1996 version + +| Elements of service | MT | IPM | PD | MS | Annex B reference (Note) | +|---------------------------------------------------------|----|-----|----|----|--------------------------| +| Auto-acknowledgement of IP-messages | X | X | | | B.7 | +| Auto-action log | X | | | | B.8 | +| Auto-assignment of annotations | | X | | | B.10 | +| Auto-assignment of group names | | X | | | B.11 | +| Auto-assignment of storage period | | X | | | B.12 | +| Auto-correlation of IP-messages | X | X | | | B.13 | +| Auto-correlation of IP-notifications | X | X | | | B.14 | +| Auto-correlation of reports | | X | | | B.15 | +| Auto-deletion after storage period | | X | | | B.16 | +| Auto-discarding of IP-messages | X | X | | | B.17 | +| Copy precedence | | X | | | B.XXX | +| Delivery log | X | | | | B.38 | +| Distribution codes indication | | X | | | B.44 | +| Exempted address | | X | | | B.XXX | +| Extended authorization information | | X | | | B.XXX | +| IP-message action status | X | X | | | B.58 | +| Message instructions | | X | | | B.XXX | +| Message type | | X | | | B.XXX | +| Originator reference indication | | X | | | B.83 | +| Other recipients indicator | | X | | | B.XXX | +| Primary precedence | | X | | | B.XXX | +| Storage of draft messages | | X | | | B.117 | +| Storage on submission | | X | | | B.118 | +| Storage period assignment | | X | | | B.119 | +| Stored message annotation | | X | | | B.121 | +| Stored message grouping | | X | | | B.124 | +| Submission log | X | | | | B.128 | +| Submission of IP-messages incorporating stored messages | X | X | | | B.129 | + +NOTE – This indication is a reference to clause numbering and tile of Annex B in the 1999 version. + +B.XXX Reference removed from the 1996 version + +## **C.2 Classification of new elements of service** + +The new elements of service that were added to the 1992 X.400-series Recommendations to create the 1996 F.400-series and X.400-series Recommendations are all classified as additional optional user facilities. + +## **Annex D** + +### **Differences between ITU-T Recommendation F.400/X.400 and ISO/IEC 10021-1** + +(This annex does not form an integral part of this Recommendation) + +#### **D.1 Differences** + +This annex points out the major differences between this Recommendation and the corresponding ISO/IEC International Standard. Because the differences in many cases involve the inclusion or exclusion of a word, a phrase, or a sentence, and these occur in many places throughout the text, this annex does not specifically point to these instances. Rather it summarizes the intent of these differences. + +The following are the major differences: + +- 1) the ITU-T text makes references throughout to ITU-T services and their relationship to MHS; +- 2) the ITU-T text refers to the use of MHS in provision of public services (clause 17); +- 3) the ITU-T text provides description for multinational PRMD (Annex E). + +## Annex E + +### Multinational PRMD + +(This annex does not form an integral part of this Recommendation) + +See also Annex G/X.402 | ISO/IEC 10021-2. + +Many large organizations and service providers will have the requirement for message handling management domains covering more than one country. The naming conventions of X.400-series and F.400-series Recommendations and the idea of global uniqueness of names have to be applied where ADMDS are involved in international messaging. While direct PRMD to PRMD messaging is mentioned in these Recommendations, it is not viewed as the choice for global applicability. Each PRMD is free to use its own conventions but may have problems entering public transfer if the standard concept of naming is not observed. + +Starting from the assumption that in every country the responsible naming authorities are in place, names of PRMDs can be registered by the responsible naming authority at the country level. Furthermore, an organization may wish to use the same PRMD name with various ADMDS in a country. The PRMD name together with the ADMMD name concerned and the country name establishes global uniqueness. A multinational PRMD can easily manage its structure to meet these requirements by subdividing its global domain into sub-domains, each sub-domain registering its sub-domain name separately in the country concerned. Unambiguous names in the country and a unique name globally are the result of these provisions. Relaying of messages being transferred via ADMDS through PRMDs may be restricted by ADMMD practices. See Figure E.1. + +![Diagram illustrating the model for a multinational PRMD. A large oval labeled 'Multinational PRMD x' contains four sub-domains: PRMD x1, PRMD x2, PRMD x3, and PRMD x4. These sub-domains are connected to ADMDS: ADMMD A (Country A), ADMMD B (Country B), ADMMD C1 (Country C), and ADMMD C2 (Country C). Dashed lines represent connections between PRMD sub-domains and ADMDS across different countries. Solid lines represent connections between PRMD sub-domains and ADMDS within the same country. The diagram shows that PRMD x1 is connected to ADMMD A, PRMD x2 to ADMMD B, PRMD x3 to both ADMMD C1 and ADMMD C2, and PRMD x4 to both ADMMD C1 and ADMMD C2. Dashed lines also show connections between PRMD x1 and ADMMD B, PRMD x2 and ADMMD A, PRMD x3 and ADMMD A, and PRMD x4 and ADMMD B.](97b3ed12d7f6c65cf3f7187326728409_img.jpg) + +Diagram illustrating the model for a multinational PRMD. A large oval labeled 'Multinational PRMD x' contains four sub-domains: PRMD x1, PRMD x2, PRMD x3, and PRMD x4. These sub-domains are connected to ADMDS: ADMMD A (Country A), ADMMD B (Country B), ADMMD C1 (Country C), and ADMMD C2 (Country C). Dashed lines represent connections between PRMD sub-domains and ADMDS across different countries. Solid lines represent connections between PRMD sub-domains and ADMDS within the same country. The diagram shows that PRMD x1 is connected to ADMMD A, PRMD x2 to ADMMD B, PRMD x3 to both ADMMD C1 and ADMMD C2, and PRMD x4 to both ADMMD C1 and ADMMD C2. Dashed lines also show connections between PRMD x1 and ADMMD B, PRMD x2 and ADMMD A, PRMD x3 and ADMMD A, and PRMD x4 and ADMMD B. + +**Examples** + +| | | | | | | +|----------------------|----------------------|----------------------|----------------------|----------------------|----------------------| +| Country: A | Country: B | Country: C | Country: C | Country: C | Country: C | +| ADMD: A | ADMD: B | ADMD: C 1 | ADMD: C 2 | ADMD: C 1 | ADMD: C 2 | +| PRMD: x 1 | PRMD: x 2 | PRMD: x 3 | PRMD: x 3 | PRMD: x 4 | PRMD: x 4 | + +Name of the multinational PRMD: x + +Sub-domain names: x1, x2, x3, x4 + +NOTE 1 – Numbers used for PRMD names are for illustration only. + +NOTE 2 – PRMD names can be the same in various countries if legal requirements are met. + +**Figure E.1/F.400/X.400 – Model for a multinational PRMD** + +ITU-T X-SERIES RECOMMENDATIONS +**DATA NETWORKS AND OPEN SYSTEM COMMUNICATIONS** + +| | | +|-----------------------------------------------|--------------------| +| PUBLIC DATA NETWORKS | | +| Services and facilities | X.1–X.19 | +| Interfaces | X.20–X.49 | +| Transmission, signalling and switching | X.50–X.89 | +| Network aspects | X.90–X.149 | +| Maintenance | X.150–X.179 | +| Administrative arrangements | X.180–X.199 | +| OPEN SYSTEMS INTERCONNECTION | | +| Model and notation | X.200–X.209 | +| Service definitions | X.210–X.219 | +| Connection-mode protocol specifications | X.220–X.229 | +| Connectionless-mode protocol specifications | X.230–X.239 | +| PICS proformas | X.240–X.259 | +| Protocol Identification | X.260–X.269 | +| Security Protocols | X.270–X.279 | +| Layer Managed Objects | X.280–X.289 | +| Conformance testing | X.290–X.299 | +| INTERWORKING BETWEEN NETWORKS | | +| General | X.300–X.349 | +| Satellite data transmission systems | X.350–X.369 | +| IP-based networks | X.370–X.399 | +| MESSAGE HANDLING SYSTEMS | X.400–X.499 | +| DIRECTORY | X.500–X.599 | +| OSI NETWORKING AND SYSTEM ASPECTS | | +| Networking | X.600–X.629 | +| Efficiency | X.630–X.639 | +| Quality of service | X.640–X.649 | +| Naming, Addressing and Registration | X.650–X.679 | +| Abstract Syntax Notation One (ASN.1) | X.680–X.699 | +| OSI MANAGEMENT | | +| Systems Management framework and architecture | X.700–X.709 | +| Management Communication Service and Protocol | X.710–X.719 | +| Structure of Management Information | X.720–X.729 | +| Management functions and ODMA functions | X.730–X.799 | +| SECURITY | X.800–X.849 | +| OSI APPLICATIONS | | +| Commitment, Concurrency and Recovery | X.850–X.859 | +| Transaction processing | X.860–X.879 | +| Remote operations | X.880–X.899 | +| OPEN DISTRIBUTED PROCESSING | X.900–X.999 | +| TELECOMMUNICATION SECURITY | X.1000– | + +*For further details, please refer to the list of ITU-T Recommendations.* + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|--------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series B | Means of expression: definitions, symbols, classification | +| Series C | General telecommunication statistics | +| Series D | General tariff principles | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Construction, installation and protection of cables and other elements of outside plant | +| Series M | TMN and network maintenance: international transmission systems, telephone circuits, telegraphy, facsimile and leased circuits | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks and open system communications | +| Series Y | Global information infrastructure, Internet protocol aspects and Next Generation Networks | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.401-199208-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.401-199208-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..5140403e89f077eeeb9294fed5c5c46220b2513d --- /dev/null +++ b/marked/F/T-REC-F.401-199208-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:db3e11afc4d718c6b62ba3d91c8d8318612ee5e94fef06065ae0fa1602aaf0ff +size 8271 diff --git a/marked/F/T-REC-F.401-199208-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.401-199208-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..0a5e5e9f052bc24cff243556c788c192a8235fd9 --- /dev/null +++ b/marked/F/T-REC-F.401-199208-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:340c9aadbd3636a3ae143ef772f44603e808a82f76cde7dc6d713508ff94a31e +size 5956 diff --git a/marked/F/T-REC-F.410-199208-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.410-199208-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1452c78311ec43dd3ffbccc901b58475247d3c72 --- /dev/null +++ b/marked/F/T-REC-F.410-199208-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ce152ac77595bdee06c45615766f01584acdf20abd14a6da57885c0b74857a7b +size 7382 diff --git a/marked/F/T-REC-F.410-199208-I_PDF-E/3121ebddccf183ca63bb9781be440a7e_img.jpg b/marked/F/T-REC-F.410-199208-I_PDF-E/3121ebddccf183ca63bb9781be440a7e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..095a748ddaf53beebedd0ed48fa4ab51c9f8666b --- /dev/null +++ b/marked/F/T-REC-F.410-199208-I_PDF-E/3121ebddccf183ca63bb9781be440a7e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9c10275d37568a7ae7391a3b2200ade55d1e9c02a7a6fe0891cff3d32d20a657 +size 46693 diff --git a/marked/F/T-REC-F.410-199208-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg b/marked/F/T-REC-F.410-199208-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..2630c29d73ac072879991d7dcb2a973a203aafff --- /dev/null +++ b/marked/F/T-REC-F.410-199208-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3cdb6a531165755391eb316c3367ccc31c28927759890154b9fa54a10152e6d3 +size 5956 diff --git a/marked/F/T-REC-F.410-199208-I_PDF-E/raw.md b/marked/F/T-REC-F.410-199208-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..8f9a0675517067c4914884cd5fa578b694e50a58 --- /dev/null +++ b/marked/F/T-REC-F.410-199208-I_PDF-E/raw.md @@ -0,0 +1,352 @@ + + +![ITU logo: A globe with a lightning bolt and the letters ITU.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with a lightning bolt and the letters ITU. + +INTERNATIONAL TELECOMMUNICATION UNION + +**CCITT** + +**F.410** + +THE INTERNATIONAL +TELEGRAPH AND TELEPHONE +CONSULTATIVE COMMITTEE + +(08/92) + +**MESSAGE HANDLING SERVICE +OPERATIONS AND DEFINITION OF SERVICE** + +--- + +**MESSAGE HANDLING SERVICES: +THE PUBLIC MESSAGE TRANSFER SERVICE** + +![CCITT logo: A globe with the letters CCITT.](6ed175c791b5e156d9c98a8dbcc3318c_img.jpg) + +CCITT logo: A globe with the letters CCITT. + +**Recommendation F.410** + +--- + +# FOREWORD + +The CCITT (the International Telegraph and Telephone Consultative Committee) is a permanent organ of the International Telecommunication Union (ITU). CCITT is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The Plenary Assembly of CCITT which meets every four years, establishes the topics for study and approves Recommendations prepared by its Study Groups. The approval of Recommendations by the members of CCITT between Plenary Assemblies is covered by the procedure laid down in CCITT Resolution No. 2 (Melbourne, 1988). + +Recommendation F.410 was revised by Study Group I and was approved under the Resolution No. 2 procedure on the 4th of August 1992. + +# --- CCITT NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized private operating agency. + +# **MESSAGE HANDLING SERVICES: THE PUBLIC MESSAGE TRANSFER SERVICE** + +*(revised 1992)* + +The establishment in various countries of message handling services in association with public networks creates the need to produce Recommendations covering the aspects of public message handling services. + +# **CONTENTS** + +| | | +|---|--------------------------------------------------------------------------------| +| 1 | Purpose and scope | +| 2 | MT service | +| 3 | Operation of the service | +| 4 | Quality of service | +| 5 | Network requirements | +| 6 | Use of Message Transfer service within CCITT defined telematic services | + +*Annex A – Abbreviations* + +*Appendix I – Errata to Recommendation F.415* + +# **1 Purpose and scope** + +## **1.1 General** + +This Recommendation specifies the general, operational and quality of service aspects of the public international Message Transfer service. + +This type of Message Handling (MH) service is an international telecommunication service offered by Administrations, enabling subscribers' user agents to submit standardized classes of messages to Message Transfer agents for their transfer to another Message Transfer agent in the same Administration's domain, in another Administration's domain, or to private domains, via telecommunication networks using store and forward techniques. + +The Message Transfer service also may transfer messages submitted through a message store (MS), and delivered to a message store, and to and from access units (AUs) to other services. + +Locally provided functions, for which communication with other user agents or message transfer agents is not required, are not covered by CCITT Recommendations. + +The Message Transfer (MT) service enables subscribers to request a variety of features to be performed during the transfer of messages. + +Some features are inherent in the basic MT service. Other non-basic features may be selected by the subscriber, either on a per-message basis or for an agreed contractual period of time, if they are provided by Administrations. + +Elements of service belonging to the basic Message Transfer service and essential optional user facilities are to be made available internationally by Administrations. + +MT service may be provided using any physical network. MT service may be offered separately or in combination with various telematic or data communication services. It can be obtained by making appropriate arrangements. + +Technical specifications and protocols, to be used in the MT service are defined in the X.400-Series of Recommendations. + +The service definition is contained in § 2. Sections 3 and 4 describe the operation of the service and quality of service, and network requirements are given in § 5. + +## 1.2 *Message handling systems used in the provision of Message Transfer service* + +### 1.2.1 *1984 and 1992 implementations* + +This Recommendation assumes that the message handling systems implemented to provide the service outlined herein are based on the 1992 version of the X.400-Series of technical Recommendations. It is recognized however, that for some time after the publication of this Recommendation, several implementations of Message Transfer service will still be based on the 1984 X.400-Series of Recommendations. + +### 1.2.2 *1988 and 1992 implementations* + +Minor revisions between 1988 and 1992 MH service Recommendations are referenced in Annex C of F.400. All revisions introduced are classified as optional additional user facilities. The Quality of Service time targets should be supported as detailed in this Recommendation. + +Administrations are encouraged to adopt the latest CCITT Recommendations. + +### 1.2.3 *Interworking* + +In order to protect the investment of Administrations who have implemented previous systems for the provision of Message Transfer service, 1992 Administration management domain (ADMD) implementations shall be able to interwork to 1984 and 1988 ADMDs (see Note). Interworking from 1992 ADMDs to private management domain (PRMDs), which support previous versions is a local matter. + +*Note* – A future version of this Recommendation will make support for the interworking with 1984 ADMDs optional. + +# 2 **Message transfer service** + +## 2.1 *General service requirements* + +2.1.1 The fundamental ability of the MT service is to provide for the transfer of messages submitted by other services subscribing to the MT service. These other services may submit messages from their user agents, if they are services that follow the X.400-Series of Recommendations. Services may also access the MT service from standardized access units. Messages may also be transferred to and from message stores. The access units and message stores are not part of the MT service. Conversion of messages when different codings and other formats are used may be provided by the MT service. + +2.1.2 The public MT service will be provided by Administrations using systems that conform to the X.400-Series Recommendations. + +Management domains (MDs) are defined for the purpose of responsibilities boundaries. The MD managed by an Administration is called an Administration management domain (ADMD). The MD managed by an organization is called a private management domain (PRMD). + +2.1.3 International exchange of messages are performed between Administration management domains through CCITT standardized public data transmission services. Each Administration will designate one or more message transfer agent (MTAs) in its management domain as international access points to the MT service. + +2.1.4 Different classes of messages may be exchanged through this service. Some classes of messages may be standardized by CCITT Recommendations, such as Recommendation F.420. Other classes of messages may also be transferred, provided that the format adheres to the appropriate X.400-Series of Recommendations. + +2.1.5 An Administration may provide different methods of access to the MT service. The possible methods are: + +- 1) from a subscribing service's user agent, message store, or access unit; +- 2) from an MTA in a private management domain. + +2.1.6 Each Administration is responsible for the national access to its management domain. + +2.1.7 The characteristics of the direct interfaces to the MT service, or between a private domain and the MT service are a national matter, although they should generally conform to the X.400-Series Recommendations. Interworking with postal systems, or other physical delivery systems, should be in accordance with Recommendation F.415. + +2.1.8 The national implementation of the MT service may provide intercommunication of subscribing services with other telematic services such as telex, teletex, facsimile and videotex. When implemented, the interface between the MT service and the other services shall be according to relevant CCITT Recommendations. Intercommunication may also be provided to a physical delivery system (PDS). + +2.1.9 As the service is providing indirect communication, cases of non-delivery of the message to the intended recipient may occur. The MT service provides for non-delivery notification and, as an optional user facility, for delivery notification. + +2.1.10 Due to the intermediate storage of the message, the service may provide conversion optional user facilities: speed, access procedures, networks and coding of message contents. + +2.1.11 The message belongs to the originator until delivery has taken place. After delivery, the message belongs to the recipient. + +2.1.12 Where sender and recipient have different and conflicting requirements, the sender's requirements shall take precedence (e.g. content type conversion or redirection control). + +2.1.13 Management domains shall relay messages even if some additional optional user facilities are not supported by that domain. + +## 2.2 *Message Transfer service features* + +### 2.2.1 *Introduction* + +Recommendation F.400, § 19, defines elements of service which are available in the MT service and are classified as either belonging to the basic service or as MT optional user facilities. Elements of service comprising the basic MT service are inherently part of the service, and are always provided and available. The optional user facilities that are classified as essential are always provided and those classified as additional may be available nationally or internationally on the basis of bilateral agreement. + +In the MT service there is the following grouping of elements of service: + +- 1) basic service which corresponds with the basic elements of service listed in Table 4/F.400; +- 2) optional user facilities, which correspond to the MT optional user facilities listed in Table 5/F.400. + +Basic features are inherent in the service. Optional user facilities may be selected on a per-message basis or for an agreed contractual period of time. + +### 2.2.2 *The basic Message Transfer service* + +The basic MT service shall be implemented according to the requirements of Recommendation X.411. The basic MT service enables user agents (UAs) to access and be accessed by the message transfer system (MTS) in order to exchange messages. Each message is assigned a unique message reference identification. If a message cannot be delivered, the originating UA is informed. To facilitate meaningful communication, a UA may specify the types of encoded information that can be contained in messages delivered to it. The content type, the original encoded information types, the time of submission and delivery and whether conversion occurred are indicated for each message. The elements of service comprising the basic MT service are listed in Recommendation F.400, Table 4/F.400. + +### 2.2.3 *Optional user facilities in the Message Transfer service* + +Two classes of optional user facilities are available in the MT services. The first class is selectable on a per-message basis. The second class may be provided to the subscribing service when agreed to over a contractual period of time. The classes are described and cited in § 19.3 of Recommendation F.400 and in Table 5/F.400, and are available in the service based on the MT service. + +### 2.2.4 *Naming and addressing* + +Naming and addressing as used in the MT service, are described in overview in Recommendation F.400, § 12. The rules for naming and addressing in an Administration management domain are given in Recommendation F.401. + +# 3 **Operation of the service** + +## 3.1 *General* + +3.1.1 The MT service provides that messages can be sent, transferred, delivered and received using fully automatic procedures. + +Manual delivery of messages can be provided in the case of interworking with postal systems, and is described in Recommendation F.415. + +3.1.2 Messages are prepared by subscribers services user agents/access units or by user agents/access units in other management domains. + +3.1.3 Each Administration providing the MT service should validate its subscribers identities, at the time of access. It should also validate the identity of other management domains at their points of access. + +3.1.4 Connectivity of the MT service to message transfer in private management domains, which will allow users of these systems to exchange messages, is desirable. This is recognized to be a national matter. If these interconnections are provided, they should take place between management domains in accordance with CCITT Recommendations. + +3.1.5 When implicit conversion is provided by the Administration via the message transfer service, the message will be converted if necessary, unless prohibited by the originator. The conversion will be in accordance to the rules specified in Recommendation X.408. + +## 3.2 *Message transfer* + +Message transfer is initiated when a message is received from a user agent/message store or access unit. Delivery is attempted to the address of the message. The body part of the message will be transferred in the form in which it was received, unless conversion has been performed. + +The results of the transfer attempt may be conveyed by two notifications: + +- non-delivery notification; +- delivery notification. + +Delivery notification may be given to the originating domain by the destination domain to indicate successful delivery. This delivery notification should be provided if requested. + +Non-delivery notification is automatically originated by the MTS, while delivery notification will be generated by the recipients MTA on request of the originator. If non-delivery notification is prevented, and delivery notification is not requested, no notification is possible. In the case of a message to a teletex terminal (auto) receipt notification may be returned by the teletex access unit (TTXAU). + +# 4 Quality of Service + +## 4.1 *Message status* + +The unique identification of messages conforming to the requirements of X.400-Series Recommendations enables the system to provide information about, e.g. the status of an interpersonal (IP) message or other class of message. + +In the event of system failure, all accepted and non-delivered messages should be traceable. If messages cannot be delivered, the originator must be informed by a non-delivery notification. + +## 4.2 *Responsibility for messages* + +The subscribers to the service using the MTS are responsible for the messages in their user agents/message stores. The service using the MT service is responsible for the transfer between the UAs/MSs in that service and the MT service. + +The Administration providing the MT service is responsible for the message transfer and the optional user facilities performed within its management domain and for messages coming from or directed to private management domains connected to its management domain, unless other national regulations apply. In international interconnection of ADMDS, the responsibility to deliver passes from managements domains with the message. + +Administrations should provide assistance to their subscribers, with regard to status and tracing of non-delivered messages. + +*Note* – The international implications of this are for further study. + +## 4.3 *Model of delivery and notification times* + +See Figure 1/F.410. + +## 4.4 *Delivery time targets* + +The recipient ADMD should force non-delivery notification if it has not been able to deliver the message to the receiving UA after submission to the originating MTA according to the maximum delivery times given in Table 1/F.410 (or after date and time indicated for deferred delivery). The maximum delivery times are dependent on the grade of delivery requested by the originator as shown in Table 1/F.410. + +To be able to meet these time targets, a message has to transit a transitting ADMD according to the transit time targets given in Table 2/F.410. These time targets are dependent on the grade of delivery requested by the originator. + +![Diagram of the delivery and notification time model. It shows a sequence of components: UA/MS, MTA, MTA, and MS/UA. Arrows indicate the flow: Creation of message, Submission of message, Transfer of message, Delivery of message, Notification generation, and Notification delivery. Two vertical double-headed arrows on the left indicate time intervals: T1 (see Note 1) from Submission of message to Delivery of message, and T2 (see Note 2) from Delivery of message to Notification delivery. The text T0102720-92 is at the bottom right.](3121ebddccf183ca63bb9781be440a7e_img.jpg) + +Diagram of the delivery and notification time model. It shows a sequence of components: UA/MS, MTA, MTA, and MS/UA. Arrows indicate the flow: Creation of message, Submission of message, Transfer of message, Delivery of message, Notification generation, and Notification delivery. Two vertical double-headed arrows on the left indicate time intervals: T1 (see Note 1) from Submission of message to Delivery of message, and T2 (see Note 2) from Delivery of message to Notification delivery. The text T0102720-92 is at the bottom right. + +T1 Delivery time +T2 Notification time + +*Note 1* – Starting time of T1 corresponds to the submission time stamp indication. Ending time of T1 corresponds to the delivery time stamp indication. + +*Note 2* – Starting time of T2 corresponds to the delivery time stamp indication. Ending time of T2 is the time that the delivery notification is made available to the user through the UA or MS. + +FIGURE 1/F.410 +**Delivery and notification time model** + +TABLE 1/F.410 +**Delivery time targets** + +| Grade of delivery | 95% delivered within | Maximum delivery time | +|-------------------|----------------------|-----------------------| +| Urgent | 0.25 hours | 2 hours | +| Normal | 1.0 hours | 6 hours | +| Non-urgent | 4.0 hours | 12 hours | + +*Note* – It is expected that PRMDs will abide by these delivery time targets. If for any reason (including hold for delivery) the message cannot be delivered to the UA/AU/MS, a non-delivery notification shall be returned. + +TABLE 2/F.410 +**Transit time targets** + +| Grade of delivery | 95% transmitted within | +|-------------------|------------------------| +| Urgent | 10 minutes | +| Normal | 35 minutes | +| Non-urgent | 2.4 hours | + +## 4.5 *Delivery notification time targets* + +Non-delivery notifications or requested delivery notifications should be returned on a per-recipient basis, in order not to delay notifications for those messages in a multi-addressed message which have already been delivered, to enable the originating management domain either to return per-recipient notifications or to batch notifications to its subscribers (see Table 3/F.410). + +TABLE 3/F.410 +**Notification time targets** + +| Type | 95% returned within | +|-----------------|---------------------| +| ND-notification | 0.25 hours | +| D-notification | 0.25 hours | + +*Note* – It is expected that PRMDs will abide by these notification time targets. + +## 4.6 *Error protection* + +Error protection on transmission is provided by the message handling system (MHS) and underlying protocols used in the provision of the MT service. + +## 4.7 *Availability of service* + +In principle the MT service should be available continuously. User agents or message stores connected to the MT service should be available for submission or delivery continuously (unless hold for delivery is invoked). + +## 4.8 *Minimum storage capacity* + +The storage capacity of the message transfer agent shall be sufficient to provide a high grade of service. + +*Note* – This is for further study. + +# 5 Network requirements + +## 5.1 General + +The MT service is network independent, that is, the basic service and the essential optional user facilities are provided independently of the type of network used for service access. Additional optional user facilities chosen by an Administration to offer may vary. + +## 5.2 Network requirements for international interconnection + +For interconnection of the public international message transfer service between Administrations, public packet switching connections shall be used. This does not preclude Administrations from using different means for this interconnection on a bilateral basis. + +## 5.3 Network requirements for service access + +Access to the public message transfer service is a local matter. + +# 6 Use of the message transfer service within CCITT defined telematic services + +See relevant F-Series Recommendations. + +# ANNEX A (to Recommendation F.410) + +## Abbreviations + +The following abbreviations are used in this Recommendation: + +| | | +|-------|-------------------------------------| +| A | Additional (optional user facility) | +| AU | Access unit | +| ADMD | Administration management domain | +| E | Essential (optional user facility) | +| IP | Interpersonal | +| MD | Management domain | +| MH | Message handling | +| MHS | Message handling system | +| MS | Message store | +| MT | Message transfer | +| MTA | Message transfer agent | +| MTS | Message transfer system | +| PDS | Physical delivery system | +| PRMD | Private management domain | +| TTXAU | Teletex access unit | +| UA | User agent | + +*Note 1* – For a glossary of terms see Annex A of Recommendation F.400. + +*Note 2* – For references see Recommendations F.400 and F.401. + +# APPENDIX I + +(to Recommendation F.410) + +## **Errata to Recommendation F.415 in Fascicle II.6 CCITT Blue Book 1989** + +### **I.1 Errata to Recommendation F.415** + +(Page 99, section 1, fourth paragraph). "(PM)" should be corrected to "(IPM)". + +(Page 104, subsection 6.6, third paragraph). "MS" should be corrected to "MH". + +(Page 104, section 7, sixth paragraph). "apply" should be corrected to "supply". + +(Page 106, subsection B.2, first line). "printablea" should be corrected to "printable". + +(Page 110, subsection B.7, first paragraph). "Annex A" should be corrected to "Annex A/X.408". \ No newline at end of file diff --git a/marked/F/T-REC-F.421-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.421-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..fd6d37344e1dcb73e876a09f0e28fcc3cc12e0bf --- /dev/null +++ b/marked/F/T-REC-F.421-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:473a66f8b6837745a7f32ac1741f274c5a10ec4eec34cbcf59ae9cab1ea41190 +size 7270 diff --git a/marked/F/T-REC-F.421-198811-I_PDF-E/8e85c350ee10ead46f7834b148b8b7d3_img.jpg b/marked/F/T-REC-F.421-198811-I_PDF-E/8e85c350ee10ead46f7834b148b8b7d3_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..b5e632ee7b7524212cdefd82eabc0608e77260de --- /dev/null +++ b/marked/F/T-REC-F.421-198811-I_PDF-E/8e85c350ee10ead46f7834b148b8b7d3_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c3e95d20d64a5c1f94413b825acd0fd0daf06584b2423bdea288b0c5d808428c +size 26996 diff --git a/marked/F/T-REC-F.421-198811-I_PDF-E/a26e142d3df5bef41a84a9dd099d7825_img.jpg b/marked/F/T-REC-F.421-198811-I_PDF-E/a26e142d3df5bef41a84a9dd099d7825_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..97ce44827da163b9c94aee7b4fb09122c566c2bf --- /dev/null +++ b/marked/F/T-REC-F.421-198811-I_PDF-E/a26e142d3df5bef41a84a9dd099d7825_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9dc793957b1d9212192ced2c2e5a6e10ce4777e736a354614c50b86a1a45ac7c +size 91051 diff --git a/marked/F/T-REC-F.421-198811-I_PDF-E/cab0834804fb031b43865554cc8d06ab_img.jpg b/marked/F/T-REC-F.421-198811-I_PDF-E/cab0834804fb031b43865554cc8d06ab_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..fcde165060202ac798c539baa4aa3eaa22005c55 --- /dev/null +++ b/marked/F/T-REC-F.421-198811-I_PDF-E/cab0834804fb031b43865554cc8d06ab_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1c2e39a146bfaacc4688c66cc2928df593a4e0174eee697662bac223c2488574 +size 52068 diff --git a/marked/F/T-REC-F.423-199208-I_PDF-E/raw.md b/marked/F/T-REC-F.423-199208-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..d77b8a51dc1240ca0a756be21a245151b248d61a --- /dev/null +++ b/marked/F/T-REC-F.423-199208-I_PDF-E/raw.md @@ -0,0 +1,243 @@ + + +![ITU logo: A globe with a lightning bolt and the letters ITU.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with a lightning bolt and the letters ITU. + +INTERNATIONAL TELECOMMUNICATION UNION + +**CCITT** + +THE INTERNATIONAL +TELEGRAPH AND TELEPHONE +CONSULTATIVE COMMITTEE + +**F.423** + +(08/92) + +**MESSAGE HANDLING SERVICE** + +**OPERATIONS AND DEFINITION OF SERVICE** + +--- + +**MESSAGE HANDLING SERVICES: +INTERCOMMUNICATION BETWEEN THE +INTERPERSONAL MESSAGING SERVICE +AND THE TELEFAX SERVICE** + +![CCITT logo: A globe with the letters CCITT.](84a1d09fb489061482111515543b60dc_img.jpg) + +CCITT logo: A globe with the letters CCITT. + +**Recommendation F.423** + +--- + +# FOREWORD + +The CCITT (the International Telegraph and Telephone Consultative Committee) is a permanent organ of the International Telecommunication Union (ITU). CCITT is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The Plenary Assembly of CCITT which meets every four years, establishes the topics for study and approves Recommendations prepared by its Study Groups. The approval of Recommendations by the members of CCITT between Plenary Assemblies is covered by the procedure laid down in CCITT Resolution No. 2 (Melbourne, 1988). + +Recommendation F.423 was prepared by Study Group I and was approved under the Resolution No. 2 procedure on the 4th of August 1992. + +# --- CCITT NOTES + +- 1) In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized private operating agency. +- 2) A list of abbreviations used in this Recommendation can be found in Annex B. + +# **MESSAGE HANDLING SERVICES: INTERCOMMUNICATION BETWEEN THE INTERPERSONAL MESSAGING SERVICE AND THE TELEFAX SERVICE** + +(1992) + +The establishment in various countries of message handling services in association with public networks, creates the need to produce Recommendations covering the aspects of public message handling services. + +# **TABLE OF CONTENTS** + +| | | +|---|-------------------------------| +| 1 | Scope | +| 2 | Functional model | +| 3 | Service outline | +| 4 | Operational procedures | +| 5 | Quality of Service | + +*Annex A – Action to be taken by the PFAXAU* + +## **1 Scope** + +1.1 This Recommendation describes the service definition and general operational procedures for intercommunication between the public (Interpersonal Messaging) (IPM) service and the telefax services. + +1.2 The intercommunication is based on store-and-forward principles and allows the users of one service to exchange messages with the users of the other service. + +1.3 For facsimile users who are also registered as users of the IPM service, see Recommendation F.420. + +## **2 Functional model** + +2.1 The IPM service is a messaging service which may be provided on a variety of networks and allows several forms of address, whereas the telefax services provide direct connection between user terminals in the public switched networks. + +2.2 Therefore, to match the dissimilar characteristic of the services, it is necessary to provide intercommunication via a public facsimile access unit (PFAXAU). In both the IPM service to the telefax service, and the telefax service to the IPM service directions, the complete message is deposited in the PFAXAU for onward transmission. + +2.3 The functional model for intercommunication described in this Recommendation is shown in Figure 1/F.423. + +![Functional model diagram showing the interaction between an IPM user, IPM-UA, MT service, PFAXAU, IPM service, Telefax service, and Facsimile user. The IPM service is a large oval containing the IPM-UA, MT service, and PFAXAU. The IPM user is connected to the IPM-UA. The PFAXAU is connected to the Telefax service, which is in turn connected to the Facsimile user.](2763901b7a1fd1b5d704cdc450d12ed0_img.jpg) + +Functional model diagram showing the interaction between an IPM user, IPM-UA, MT service, PFAXAU, IPM service, Telefax service, and Facsimile user. The IPM service is a large oval containing the IPM-UA, MT service, and PFAXAU. The IPM user is connected to the IPM-UA. The PFAXAU is connected to the Telefax service, which is in turn connected to the Facsimile user. + +T0101630-90 + +UA User agent + +FIGURE 1/F.423 +Functional model + +# 3 Service outline + +- 3.1 Delivery of message to telefax service users from the IPM service and vice versa is provided by means of a PFAXAU. +- 3.2 The PFAXAU belongs to the message handling system (MHS) domain. +- 3.3 In the IPM service to telefax service direction, the IPM service PFAXAU is responsible for the message until delivery to the telefax service user terminal has been completed. +- 3.4 In the telefax service to IPM service direction, the IPM service PFAXAU is responsible for the delivery of the message to the IPM service user terminal, once the input is completed under normal conditions. +- 3.5 The PFAXAU may be located in the country of origin or country of destination as shown in Figure 2/F.423. Where two Administrations offer an IPM service, the international exchange of messages should be via the IPM service. + +![Location of PFAXAU diagram showing two scenarios. The left scenario shows Country A with a Fax terminal and Country B with an IPM-UA, MTS, and PFAXAU. The right scenario shows Country A with a Fax terminal and PFAXAU, and Country B with an IPM-UA and MTS.](12a6537c92844d5b393104c02e8dfc2f_img.jpg) + +Location of PFAXAU diagram showing two scenarios. The left scenario shows Country A with a Fax terminal and Country B with an IPM-UA, MTS, and PFAXAU. The right scenario shows Country A with a Fax terminal and PFAXAU, and Country B with an IPM-UA and MTS. + +MTS Message transfer system + +T0101640-90 + +FIGURE 2/F.423 +Location of PFAXAU + +3.6 Where an Administration does not provide an IPM service, then connections between the telefax service user and the PFAXAU should be via the international public switched networks. + +3.7 The elements of service applicable to this intercommunication should follow Recommendation F.420. + +## 4 Operational procedures + +4.1 IPM service to telefax service direction + +4.1.1 Messages from an IPM service user to a telefax user are sent as normal interpersonal (IP)-messages with the appropriate IPM elements of service, in accordance with Recommendation F.420. + +4.1.2 The terminal originator/recipient (O/R) address of the telefax service user, as defined in Recommendation F.401, should be used to route a message to the telefax service user terminal. + +4.1.3 When a message is received by the PFAXAU, the IP-message will be converted into the group 3 format defined for the telefax service. The delivery of IP-message to group 4 terminals is for further study. + +*Note* – The conversion may take place in the message transfer agent (MTA) associated with the PFAXAU. + +4.1.4 The PFAXAU shall be responsible for the action to be taken for IPM elements of service received in accordance with Recommendation F.420. Annex A shows the examples of the elements of service and the proposed action to be taken by the PFAXAU. + +4.1.5 Call establishment to the telefax user and delivery of the facsimile message should be in accordance with normal telefax service procedures (see Recommendation T.30). + +4.1.6 The IP-message sent to the telefax user terminal may be preceded by a PFAXAU identification. The content of this identification is a local matter. + +4.1.7 A general layout of an IP-message delivered to a telefax user is shown in Figure 3/F.423. + +![Diagram illustrating the conversion of an IP-message structure to a telefax format. The diagram shows two columns: 'IPM service' on the left and 'Telefax service' on the right. On the left, three stacked boxes represent 'Envelope', 'Heading', and 'Body'. On the right, two stacked boxes represent 'Header' and 'Text'. Arrows show the mapping: 'Envelope' and 'Heading' both map to 'Header', and 'Body' maps to 'Text'. On the far left, vertical double-headed arrows label the top level as 'Message', the middle level as 'Content', and the bottom level as 'Body'.](84a01685710d24f113b18758ed3c6fcb_img.jpg) + +Diagram illustrating the conversion of an IP-message structure to a telefax format. The diagram shows two columns: 'IPM service' on the left and 'Telefax service' on the right. On the left, three stacked boxes represent 'Envelope', 'Heading', and 'Body'. On the right, two stacked boxes represent 'Header' and 'Text'. Arrows show the mapping: 'Envelope' and 'Heading' both map to 'Header', and 'Body' maps to 'Text'. On the far left, vertical double-headed arrows label the top level as 'Message', the middle level as 'Content', and the bottom level as 'Body'. + +T0101650-90 + +FIGURE 3/F.423 +IP-message to a telefax format + +4.1.8 The elements of service related to the IP-message envelope and heading shall be converted into readable header. The PFAXAU shall deliver the header to a telefax user. The format and the language of this header is a local matter. The PFAXAU shall transmit the originator O/R address in a form suitable for recall. + +4.1.9 Upon delivery of the message to the telefax user terminal, a delivery notification should be sent back to the originating IPM service user if requested. In the event of non-delivery of the message to the telefax user terminal, a non-delivery notification should be sent back to the IPM service user. + +4.1.10 If a notification cannot be delivered to the originating IPM service user terminal, it is the responsibility of the operator of the MHS domain providing the PFAXAU to advise the originator via other suitable means. + +### 4.2 *Telefax to IPM service direction* + +4.2.1 The general selection procedures for telefax service users is for further study. + +### 4.3 *Construction of IPM message* + +For further study. + +# 5 **Quality of Service** + +5.1 Provision of intercommunication between the IPM service and the telefax services should maintain, as far as possible, the Quality of Service objectives of each individual service. + +5.2 The PFAXAU shall be capable of supporting the basic requirements of the facsimile terminal equipment as defined in the T-Series Recommendations. The support of standardized optional user facilities is for further study. + +5.3 The PFAXAU design, dimensioning and retry algorithm is a local matter. + +5.4 The method of recovery when the transmission of a message from the PFAXAU to the destination telefax user terminal is interrupted, should be in accordance with store and forward facsimile service (see Recommendation F.162). + +## ANNEX A + +(to Recommendation F.423) + +## **Actions to be taken by the PFAXAU** + +Basic IPM elements of service and essential optional IPM user facilities which have to be processed by the PFAXAU in the case where a message is sent from the IPM service to the fax service direction (see Table A-1/F.423). + +TABLE A-1/F.423 + +| References
Rec. F.400
Annex B | Elements of service | Action to be taken by the PFAXAU | +|-------------------------------------|---------------------------------|-----------------------------------------------------------------------------------------------------------------------------| +| B.5 | Authorizing users indication | Display in header | +| B.6 | Auto-forwarded indication | Ignore | +| B.8 | Blind copy recipient indication | Display the O/R descriptor information to the blind copy recipient(s) but not disclose to either primary or copy recipients | +| B.9 | Body part encryption indication | The PFAXAU shall send a non-delivery notification to the originator | +| B.12 | Content type indication | National matter for content types different than P2 and P22 | +| B.13 | Conversion prohibition | If G3 fax body part, ignore; otherwise the PFAXAU generate a non-delivery notification | +| B.15 | Converted indication | Local matter | +| B.18 | Cross referencing indication | Display in header | +| B.21 | Delivery notification | The PFAXAU shall send a delivery notification to the originator | +| B.22 | Delivery time stamp indication | Local matter | +| B.26 | DL expansion history indication | Ignore | +| B.29 | Expiry date indication | Display in header | +| B.31 | Forwarded IP-message indication | The PFAXAU shall display all headers | +| B.32 | Grade of delivery selection | Local matter | +| B.35 | Importance indication | Display in header | +| B.37 | IP-message identification | Display in header | +| B.38 | Language indication | Ignore | +| B.39 | Latest delivery designation | PFAXAU should honour | +| B.41 | Message identification | Local matter | +| B.45 | Multi-destination delivery | Deliver the message to all recipients | +| B.46 | Multi-part body | Messages containing not supported body parts are not delivered. Send back a non-delivery notification to the originator | + +Table A-1/F.423 (cont.) + +| References
Rec. F.400
Annex B | Elements of service | Action to be taken by the PFAXAU | +|-------------------------------------|-----------------------------------------------|------------------------------------------------------------------------| +| B.47 | Non-delivery notification | The PFAXAU shall generate a delivery report | +| B.48 | Non-receipt notification request indication | Note 1 | +| B.52 | Obsoleting indication | Local matter | +| B.54 | Original encoded information types indication | Local matter | +| B.55 | Originator indication | Display in header | +| B.56 | Originator requested alternate recipient | Local matter | +| B.62 | Primary and copy recipients indication | Display the O/R descriptor information of the recipients in the header | +| B.63 | Probe | Local matter | +| B.67 | Receipt notification request indication | Note 2 | +| B.72 | Reply request indication | Display in header | +| B.73 | Replying IP-message indication | Local matter | +| B.80 | Sensitivity indication | Local matter | +| B.88 | Subject indication | Display in header | +| B.89 | Submission time stamp indication | Display in header | +| B.90 | Typed body | Local matter | + +*Note 1* – Non-delivery notification is returned when transmission to the destination facsimile terminal is not successful. + +*Note 2* – Receipt notification may be returned according to the request in the subject message when message transmission to the destination facsimile terminal is completed. In this context, the use of receipt notification has an equivalent meaning to positive delivery notification for PFAXAU as described in § 4.1.9 of this Recommendation. + +## ANNEX B (to Recommendation F.423) + +## **Alphabetical list of abbreviations used in this Recommendation** + +| | | +|--------|------------------------------| +| IP | Interpersonal | +| IPM | Interpersonal messaging | +| MHS | Message handling system | +| MT | Message transfer | +| MTA | Message transfer agent | +| MTS | Message transfer system | +| O/R | Originator/recipient | +| PFAXAU | Public facsimile access unit | \ No newline at end of file diff --git a/marked/F/T-REC-F.435-199906-I_PDF-E/11edb7fcedf09ac6a817f8d7b8c61eec_img.jpg b/marked/F/T-REC-F.435-199906-I_PDF-E/11edb7fcedf09ac6a817f8d7b8c61eec_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..be23e5edfd9f06cbc0adeaf204e041fd763dbf62 --- /dev/null +++ b/marked/F/T-REC-F.435-199906-I_PDF-E/11edb7fcedf09ac6a817f8d7b8c61eec_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0deb0ed10c70dc667a9a94f1beee95c80575c085614c8e638ed4b27f130fe8dc +size 68272 diff --git a/marked/F/T-REC-F.435-199906-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.435-199906-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..07eef93937cfd0b52f90e0b24b91be936b203f37 --- /dev/null +++ b/marked/F/T-REC-F.435-199906-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5f034ae8781ccbcd245452d6237b72d7e69ad25899a48cd43e812b879469772d +size 8226 diff --git a/marked/F/T-REC-F.435-199906-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg b/marked/F/T-REC-F.435-199906-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..be955f2d3b1eb8f56f8f3c5352a80983ac4beedd --- /dev/null +++ b/marked/F/T-REC-F.435-199906-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:81e7a7c39a4ede3b83510beefca72c6d6bbcdc79a9df384476ecc382bb53dfbc +size 111493 diff --git a/marked/F/T-REC-F.435-199906-I_PDF-E/40f30e4d577a17052f8b1e6dc802a0d8_img.jpg b/marked/F/T-REC-F.435-199906-I_PDF-E/40f30e4d577a17052f8b1e6dc802a0d8_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ca342b88bcffbe5e3ae2fdfb9e242d83fbd01e14 --- /dev/null +++ b/marked/F/T-REC-F.435-199906-I_PDF-E/40f30e4d577a17052f8b1e6dc802a0d8_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f3e794ce87602f67350333759598bd15fd025e3d1390597ce4798a6ca706c2c0 +size 81891 diff --git a/marked/F/T-REC-F.435-199906-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg b/marked/F/T-REC-F.435-199906-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..2d20292ccf154e0d69c8f1931e24c3821d31c1f5 --- /dev/null +++ b/marked/F/T-REC-F.435-199906-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b2b9750bd94c2c292786a0c37f8656831c8f89e1b0d3f9f3eed30fdc52502ac9 +size 109715 diff --git a/marked/F/T-REC-F.435-199906-I_PDF-E/4e4be0bd8b235167902f2c03e41da651_img.jpg b/marked/F/T-REC-F.435-199906-I_PDF-E/4e4be0bd8b235167902f2c03e41da651_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..4372944af8886d08c9ca54c22736f2033ec65239 --- /dev/null +++ b/marked/F/T-REC-F.435-199906-I_PDF-E/4e4be0bd8b235167902f2c03e41da651_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e393da770d4a683781bfef8de3f1ef9b5688840e03375d455fa93419f5a292eb +size 37040 diff --git a/marked/F/T-REC-F.435-199906-I_PDF-E/5ee1bbbf85b473f78af9ec8368a4159a_img.jpg b/marked/F/T-REC-F.435-199906-I_PDF-E/5ee1bbbf85b473f78af9ec8368a4159a_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9b07deae3010a87305eb3222cca74e882e3a81d2 --- /dev/null +++ b/marked/F/T-REC-F.435-199906-I_PDF-E/5ee1bbbf85b473f78af9ec8368a4159a_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ac1eef13d0cceeebc399c94b813538e86786cdb3caca712499d1ca1d24886a21 +size 57937 diff --git a/marked/F/T-REC-F.435-199906-I_PDF-E/7133ccf78043568ca62ecbcd43628a4a_img.jpg b/marked/F/T-REC-F.435-199906-I_PDF-E/7133ccf78043568ca62ecbcd43628a4a_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3e6ad0d264c23b58f48c548546239704ee291e91 --- /dev/null +++ b/marked/F/T-REC-F.435-199906-I_PDF-E/7133ccf78043568ca62ecbcd43628a4a_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9505f900d5115a06e73e3c9bf009d899e347a235741825f7484a5ffc6500da7f +size 119676 diff --git a/marked/F/T-REC-F.435-199906-I_PDF-E/78e846f2858d131dc5a4dcbabcca1e4d_img.jpg b/marked/F/T-REC-F.435-199906-I_PDF-E/78e846f2858d131dc5a4dcbabcca1e4d_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..cfeffe875069a43d691d06a91caaf174e8b6c0ef --- /dev/null +++ b/marked/F/T-REC-F.435-199906-I_PDF-E/78e846f2858d131dc5a4dcbabcca1e4d_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e4809cb92a97ab715814580e820d14062e877161d30b85a37b9522ffc04c1122 +size 66889 diff --git a/marked/F/T-REC-F.435-199906-I_PDF-E/7efae06af3af43ffe5d4b956a679cf54_img.jpg b/marked/F/T-REC-F.435-199906-I_PDF-E/7efae06af3af43ffe5d4b956a679cf54_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..07f147fa39e617d1639ebe4c70b815aecf88e55a --- /dev/null +++ b/marked/F/T-REC-F.435-199906-I_PDF-E/7efae06af3af43ffe5d4b956a679cf54_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:83ea31797c7b0773417b8ab07d248776b59e54e8529598012ee870789965d127 +size 37835 diff --git a/marked/F/T-REC-F.435-199906-I_PDF-E/7fe5741e83bc9702d1b1d7585ddf66bd_img.jpg b/marked/F/T-REC-F.435-199906-I_PDF-E/7fe5741e83bc9702d1b1d7585ddf66bd_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c5f728151b9e5767237636d6780dbff8c305c945 --- /dev/null +++ b/marked/F/T-REC-F.435-199906-I_PDF-E/7fe5741e83bc9702d1b1d7585ddf66bd_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fb79a286d7c40170e8af3423ca4a07be6b9eb7df77df50f3cfc0c4dcafff6898 +size 84125 diff --git a/marked/F/T-REC-F.435-199906-I_PDF-E/853f59c89931a666c07903b31d098277_img.jpg b/marked/F/T-REC-F.435-199906-I_PDF-E/853f59c89931a666c07903b31d098277_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..774763798a7fc8a7001b566566c1b9c160535cc1 --- /dev/null +++ b/marked/F/T-REC-F.435-199906-I_PDF-E/853f59c89931a666c07903b31d098277_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:55ed9dad0e11ce8d504d94d38e51fbb4c10c9406b84f0a28f7561f9a43907856 +size 17053 diff --git a/marked/F/T-REC-F.435-199906-I_PDF-E/8d66c9c295023a1380f9986d3663bb1e_img.jpg b/marked/F/T-REC-F.435-199906-I_PDF-E/8d66c9c295023a1380f9986d3663bb1e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..5bfd5729b6ffbfe25e66b075811e212c9114a363 --- /dev/null +++ b/marked/F/T-REC-F.435-199906-I_PDF-E/8d66c9c295023a1380f9986d3663bb1e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6241772ad52f3c4675a2516df6883a9a8ea42a5bb0dcb7fe9caad5df48a29c1b +size 50075 diff --git a/marked/F/T-REC-F.435-199906-I_PDF-E/a3472689858b068ef469213682965325_img.jpg b/marked/F/T-REC-F.435-199906-I_PDF-E/a3472689858b068ef469213682965325_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f9844e0689fb546fbe193b36d7c624e420ba70ac --- /dev/null +++ b/marked/F/T-REC-F.435-199906-I_PDF-E/a3472689858b068ef469213682965325_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:746f623a535b65f2428ef1fdb46a67069058ea72897bf09ac9447e55cfdd941d +size 76956 diff --git a/marked/F/T-REC-F.435-199906-I_PDF-E/cf8bd014a50b7c69435e804f67f9617f_img.jpg b/marked/F/T-REC-F.435-199906-I_PDF-E/cf8bd014a50b7c69435e804f67f9617f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..bcb981746900700b300124c5f56ae5d133c13652 --- /dev/null +++ b/marked/F/T-REC-F.435-199906-I_PDF-E/cf8bd014a50b7c69435e804f67f9617f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:643a4dd080401db3591c3efa323930f72e5702d1049eb5f02393c21074db93f9 +size 80440 diff --git a/marked/F/T-REC-F.435-199906-I_PDF-E/dfe556fea00682b09a59427aaf72051c_img.jpg b/marked/F/T-REC-F.435-199906-I_PDF-E/dfe556fea00682b09a59427aaf72051c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..23b6586e86ebdae4b1cbc83491bfcf193412e2d1 --- /dev/null +++ b/marked/F/T-REC-F.435-199906-I_PDF-E/dfe556fea00682b09a59427aaf72051c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0639954e51793d7b0d47e9d3c32274aed37f340e40e3681a885d3a6b8092a0e7 +size 78690 diff --git a/marked/F/T-REC-F.435-199906-I_PDF-E/e69b9188aa2c14ec6b21c83f711fef65_img.jpg b/marked/F/T-REC-F.435-199906-I_PDF-E/e69b9188aa2c14ec6b21c83f711fef65_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..cf69972360a2d82cfe3ddbb0ef9b262a368ed4db --- /dev/null +++ b/marked/F/T-REC-F.435-199906-I_PDF-E/e69b9188aa2c14ec6b21c83f711fef65_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:72785b6852d147bbe24c9c35c1bff648c5c32e478935822f5291d44f45b9c2b9 +size 68649 diff --git a/marked/F/T-REC-F.435-199906-I_PDF-E/fa859e4e468bfb2710a94527f2c504af_img.jpg b/marked/F/T-REC-F.435-199906-I_PDF-E/fa859e4e468bfb2710a94527f2c504af_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..80718ac5e19cc5a39055d343a9e146dc481fccb3 --- /dev/null +++ b/marked/F/T-REC-F.435-199906-I_PDF-E/fa859e4e468bfb2710a94527f2c504af_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e7077796e60f302771be1cbfca2ed7bd7a454673ddcbdb04e35fd7824676ad91 +size 47412 diff --git a/marked/F/T-REC-F.435-199906-I_PDF-E/raw.md b/marked/F/T-REC-F.435-199906-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..8ea566b26296b52d60e6e3f5968061d364855be4 --- /dev/null +++ b/marked/F/T-REC-F.435-199906-I_PDF-E/raw.md @@ -0,0 +1,2055 @@ + + +![ITU logo](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +The logo of the International Telecommunication Union (ITU) features the letters 'ITU' in a bold, sans-serif font, superimposed on a stylized globe with intersecting lines. + +ITU logo + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.435** + +(06/99) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Message handling services + +--- + +**Message handling services: Electronic Data +Interchange messaging service** + +ITU-T Recommendation F.435 + +(Previously CCITT Recommendation) + +--- + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +## **TELEGRAPH SERVICE** + +| | | +|-----------------------------------------------------------------|-------------| +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | + +## **MOBILE SERVICE** + +| | | +|--------------------------------------------------------|-------------| +| Mobile services and multdestination satellite services | F.110–F.159 | +|--------------------------------------------------------|-------------| + +## **TELEMATIC SERVICES** + +| | | +|-------------------------------------------|-------------| +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | + +## **MESSAGE HANDLING SERVICES** + +**F.400–F.499** + +## **DIRECTORY SERVICES** + +F.500–F.549 + +## **DOCUMENT COMMUNICATION** + +| | | +|--------------------------------------|-------------| +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | + +## **DATA TRANSMISSION SERVICES** + +F.600–F.699 + +## **AUDIOVISUAL SERVICES** + +F.700–F.799 + +## **ISDN SERVICES** + +F.800–F.849 + +## **UNIVERSAL PERSONAL TELECOMMUNICATION** + +F.850–F.899 + +## **HUMAN FACTORS** + +F.900–F.999 + +*For further details, please refer to ITU-T List of Recommendations.* + +# **MESSAGE HANDLING SERVICES: ELECTRONIC DATA INTERCHANGE MESSAGING SERVICE** + +## **Summary** + +This Recommendation is one of a set of Recommendations for message handling. The entire set provides a comprehensive specification for message handling comprising any number of cooperating open-systems. + +Message handling systems and services enable users to exchange messages on a store-and-forward basis. A message submitted by one user, the originator, is conveyed by the message transfer system (MTS), the principal component of a larger message handling system (MHS), and is subsequently delivered to one or more additional users, the message's recipients. + +An MHS comprises a variety of interconnected functional entities. Message transfer agents (MTAs) cooperate to perform the store-and-forward message transfer function. Message stores (MSs) provide storage for messages and enable their submission, retrieval and management. User agents (UAs) help users access MHS. Access units (AUs) provide links to other communication systems and services of various kinds (e.g. other telematic services, postal services). + +This Recommendation defines the overall system and service description of the message handling application called EDI messaging. + +This edition introduces new Elements of Services (EOS) to provide the EDI message store with capabilities similar to those of the IPM message store for message store correlation and services. + +## **Source** + +ITU-T Recommendation F.435 was revised by ITU-T Study Group 7 (1997-2000) and was approved under the WTSC Resolution No. 1 procedure on 18 June 1999. + +## FOREWORD + +ITU (International Telecommunication Union) is the United Nations Specialized Agency in the field of telecommunications. The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of the ITU. The ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Conference (WTSC), which meets every four years, establishes the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics. + +The approval of Recommendations by the Members of the ITU-T is covered by the procedure laid down in WTSC Resolution No. 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +## INTELLECTUAL PROPERTY RIGHTS + +The ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. The ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, the ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementors are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database. + +© ITU 2000 + +All rights reserved. No part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and microfilm, without permission in writing from the ITU. + +# CONTENTS + +*Page* + +| | | | +|------|----------------------------------------------------------------------------------|----| +| 1 | Scope | 1 | +| 2 | Normative references | 1 | +| 2.1 | Identical Recommendations International Standards | 1 | +| 2.2 | Paired Recommendations International Standards equivalent in technical content | 1 | +| 2.3 | Additional references | 2 | +| 3 | Definitions | 2 | +| 3.1 | EDI forwarding | 2 | +| 3.2 | EDI message | 2 | +| 3.3 | EDI messaging user | 2 | +| 3.4 | EDI notification | 2 | +| 3.5 | EDI message responsibility | 2 | +| 4 | Abbreviations | 2 | +| 5 | Conventions | 3 | +| 6 | EDI messaging service | 3 | +| 6.1 | Introduction | 3 | +| 6.2 | EDI messaging | 4 | +| 6.3 | EDI messaging environment | 4 | +| 6.4 | EDI messaging user | 4 | +| 7 | EDI messaging system | 4 | +| 7.1 | Introduction | 4 | +| 7.2 | Information flow in the EDIMS | 5 | +| 7.3 | EDI messaging service functional model | 5 | +| 7.4 | Structure of EDI messages | 6 | +| 7.5 | EDI notification | 8 | +| 8 | EDIM responsibility and forwarding | 9 | +| 8.1 | Introduction | 9 | +| 8.2 | Forwarding and secondary distribution | 9 | +| 8.3 | Case 1: No forwarding | 10 | +| 8.4 | Case 2: Content not changed and EDIM responsibility forwarded | 11 | +| 8.5 | Case 3: EDIM responsibility not forwarded | 13 | +| 9 | EDI naming, addressing and use of directory | 13 | +| 10 | EDI security | 14 | +| 11 | Intercommunication with physical delivery services | 15 | +| 11.1 | Introduction | 15 | +| 11.2 | Delivery and notifications | 15 | +| 11.3 | Transfer of EDIM responsibility | 16 | +| 11.4 | Physical rendition | 16 | +| 12 | Use of message store for EDI | 16 | +| 13 | Elements of service | 16 | +| 14 | Classification of elements of service | 16 | +| 14.1 | Basic EDI messaging service | 16 | +| 14.2 | EDI messaging service optional user facilities | 18 | + +| | Page | +|----------------------------------------------------------------|-------------| +| 15 Quality of service | 21 | +| 15.1 EDI message status | 21 | +| 15.2 Support by providers of EDI service | 21 | +| 15.3 Model of delivery and notification times | 21 | +| 15.4 EDI message delivery time targets | 21 | +| 15.5 EDI notification time targets | 22 | +| 15.6 Error protection | 23 | +| 15.7 Availability of service | 23 | +| Annex A – Glossary of Terms | 23 | +| A.1 EDI application | 23 | +| A.2 EDI interchange | 23 | +| A.3 EDI message (EDIM) | 23 | +| A.4 EDI message store (EDI-MS) | 23 | +| A.5 EDI messaging (EDIMG) | 24 | +| A.6 EDI messaging environment (EDIME) | 24 | +| A.7 EDI messaging service | 24 | +| A.8 EDI messaging system (EDIMS) | 24 | +| A.9 EDI messaging user (EDIMG user) | 24 | +| A.10 EDI notification (EDIN) | 24 | +| A.11 EDI message responsibility | 24 | +| A.12 EDI security | 25 | +| A.13 EDI user | 25 | +| A.14 EDI user agent (EDI-UA) | 25 | +| A.15 Electronic data interchange (EDI) | 25 | +| A.16 GS | 25 | +| A.17 IEA | 25 | +| A.18 Interchange | 25 | +| A.19 ISA | 25 | +| A.20 MHD | 25 | +| A.21 ST | 25 | +| A.22 STX | 26 | +| A.23 UNA | 26 | +| A.24 UNB | 26 | +| A.25 UNG | 26 | +| A.26 UNH | 26 | +| A.27 UNT | 26 | +| A.28 UNZ | 26 | +| Annex B – Definitions of Elements of Service | 26 | +| B.1 application security element [EDI.1] | 26 | +| B.2 auto-acknowledgement of EDI messages [EDI.31] | 27 | +| B.3 auto-correlation of EDI messages [EDI.32] | 27 | +| B.4 auto-correlation of EDI notifications [EDI.33] | 27 | +| B.5 character set [EDI.2] | 27 | +| B.6 cross reference information [EDI.3] | 27 | +| B.7 EDI forwarding [EDI.4] | 27 | +| B.8 EDI message type(s) [EDI.5] | 27 | +| B.9 EDI notification request [EDI.6] | 27 | +| B.10 EDI standard indication [EDI.7] | 28 | +| B.11 EDI message Identification [EDI.8] | 28 | +| B.12 EDIM responsibility forwarding allowed indication [EDI.9] | 28 | +| B.13 EDIN receiver [EDI.10] | 28 | +| B.14 expiry date/time indication [EDI.11] | 28 | + +| | Page | +|------------------------------------------------------------------------------|-------------| +| B.15 incomplete copy indication [EDI.12]..... | 29 | +| B.16 interchange header [EDI.13]..... | 29 | +| B.17 multi-part body [EDI.14] ..... | 29 | +| B.18 non-repudiation of content originated [EDI.15]..... | 29 | +| B.19 non-repudiation of content received [EDI.16] ..... | 29 | +| B.20 non-repudiation of content received request [EDI.17]..... | 29 | +| B.21 non-repudiation of EDI notification [EDI.18] ..... | 29 | +| B.22 non-repudiation of EDI notification request [EDI.19] ..... | 30 | +| B.23 obsoleting indication [EDI.20]..... | 30 | +| B.24 originator indication [EDI.21] ..... | 30 | +| B.25 proof of content received [EDI.22]..... | 30 | +| B.26 proof of content received request [EDI.23]..... | 30 | +| B.27 proof of EDI notification [EDI.24] ..... | 30 | +| B.28 proof of EDI notification request [EDI.25]..... | 30 | +| B.29 recipient indication [EDI.26] ..... | 31 | +| B.30 related message(s) [EDI.27]..... | 31 | +| B.31 services indication [EDI.28] ..... | 31 | +| B.32 stored EDI message auto-forward [EDI.29]..... | 31 | +| B.33 submission of EDI messages incorporating stored messages [EDI.34] ..... | 31 | +| B.34 typed body [EDI.30] ..... | 31 | +| Annex C – Security Overview ..... | 32 | +| C.1 Introduction..... | 32 | +| C.2 Vulnerabilities..... | 32 | +| C.3 Vulnerabilities countered ..... | 33 | +| C.4 Additional pervasive mechanisms ..... | 38 | +| Annex D – EDI naming, addressing, and use of Directory ..... | 39 | +| D.1 Introduction..... | 39 | +| D.2 EDI naming..... | 39 | +| D.3 Suggested DIT structure for EDI ..... | 39 | +| D.4 Name resolution..... | 40 | +| D.5 Authentication..... | 41 | +| D.6 Capabilities assessment..... | 42 | +| Annex E – Cross Referencing Overview ..... | 43 | + + + +# MESSAGE HANDLING SERVICES: ELECTRONIC DATA INTERCHANGE MESSAGING SERVICE1 + +(revised in 1999) + +# 1 Scope + +This Recommendation defines the overall system and service of EDI messaging. + +Other aspects of message handling systems and services are defined in other Recommendations. The layout of Recommendations defining the message handling system and services is shown in Table 1/F.400/X.400. The public services built on MHS, as well as access to and from the MHS for public services are defined in the F.400-series of Recommendations. + +The technical aspects of MHS are defined in the X.400-series of Recommendations. The overall system architecture of MHS is defined in ITU-T Rec. X.402 | ISO/IEC 10021-2. The technical aspects of EDI messaging are defined in ITU-T Rec. X.435 | ISO/IEC 10021-9. + +# 2 Normative references + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; all users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. + +## 2.1 Identical Recommendations | International Standards + +- ITU-T Recommendation X.402 (1999) | ISO/IEC 10021-2:1999, *Information technology – Message Handling Systems (MHS): Overall architecture.* +- ITU-T Recommendation X.413 (1999) | ISO/IEC 10021-5:1999, *Information technology – Message Handling Systems (MHS) – Message store: Abstract service definition.* +- ITU-T Recommendation X.435 (1999) | ISO/IEC 10021-9, *Information technology – Message handling systems: Electronic data interchange messaging system.* +- ITU-T Recommendation X.501 (1997) | ISO/IEC 9594-2:1998, *Information technology – Open Systems Interconnection – The Directory: Models.* +- ITU-T Recommendation X.509 (1997) | ISO/IEC 9594-8:1998, *Information technology – Open Systems Interconnection – The Directory: Authentication framework.* +- ITU-T Recommendation X.521 (1997) | ISO/IEC 9594-7:1998, *Information technology – Open Systems Interconnection – The Directory: Selected object classes.* + +## 2.2 Paired Recommendations | International Standards equivalent in technical content + +- ITU-T Recommendation F.400/X.400 (1999), *Message handling system and service overview.* +ISO/IEC 10021-1:....2, *Information technology – Message Handling System (MHS) – Part 1: System and Service Overview.* + +--- + +1 Recommendation F.435 and ISO/IEC 10021-8, *Information technology – Text Communication – Message Handling Systems: Electronic data interchange messaging system*, are technically aligned. + +2 To be published. + +## **2.3 Additional references** + +- CCITT Recommendation F.401 (1992), *Naming and addressing for public message handling services*. +- CCITT Recommendation F.415 (1988), *Intercommunication with public physical delivery services*. + +# **3 Definitions** + +For the purposes of this Recommendation, the following definitions, and those defined in Annex A apply. + +Definitions of the elements of service applicable to EDI messaging are contained in Annex B. The elements of service applicable to the Message Transfer Service, and used by EDI messaging, are called out in this Recommendation, however their definitions are contained in Annex B/F.400/X.400. + +## **3.1 EDI forwarding** + +Onward transfer of a received EDIM to one or more recipients determined by the forwarding EDI user agent/message store. + +EDI forwarding takes place when an EDI message having been delivered to an EDI user agent or EDI message store is forwarded onward to another EDI user agent or EDI message store. + +## **3.2 EDI message** + +Information in electronic form that is transferred between EDI messaging users. An EDI message is a member of the primary class of information objects conveyed between EDI messaging users. + +See also clause 5 of *ITU-T Rec. X.435 | ISO/IEC 10021-9*. + +## **3.3 EDI messaging user** + +User that engages in EDI messaging. An EDI messaging user originates, receives, or both originates and receives EDI messages. The EDI messaging environment contains any number of EDI messaging users. An EDI messaging user may be a person or a computer process. An EDI messaging user may access the EDI messaging system through an access unit. + +## **3.4 EDI notification** + +Member of the secondary class of information objects that indicates to the originator of an EDI message the disposition of EDIM responsibility for the EDI message. + +## **3.5 EDI message responsibility** + +EDI message responsibility indicates whether the subject EDI message has been made available to a specific user by its EDI user agent/message store. EDI message responsibility carries no legal significance within this Recommendation and *ITU-T Rec. X.435 | ISO/IEC 10021-9*. + +# **4 Abbreviations** + +This Recommendation uses the following abbreviations: + +| | | +|------|---------------------------------------| +| ANSI | American National Standards Institute | +| AU | Access Unit | +| DIT | Directory Information Tree | +| DL | Distribution List | +| DUA | Directory User Agent | +| EDI | Electronic Data Interchange | + +| | | +|---------|------------------------------------------------------------------------| +| EDIFACT | Electronic Data Interchange for Administration, Commerce and Transport | +| EDIM | EDI Message | +| EDIME | EDI Messaging Environment | +| EDIMG | EDI Messaging | +| EDIMS | EDI Messaging System | +| EDI-AU | EDI Access Unit | +| EDI-MS | EDI Message Store | +| EDI-UA | EDI User Agent | +| EDIN | EDI Notification | +| FN | Forwarded Notification | +| ID | Identifier | +| MD | Management Domain | +| MH | Message Handling | +| MHS | Message Handling System | +| MS | Message Store | +| MT | Message Transfer | +| MTA | Message Transfer Agent | +| MTS | Message Transfer System | +| NDN | Non-Delivery Notification | +| NN | Negative Notification | +| O/R | Originator/Recipient | +| PD | Physical Delivery | +| PDAU | Physical Delivery Access Unit | +| PDS | Physical Delivery System | +| PN | Positive Notification | +| PRMD | Private Management Domain | +| TLMA | Telematic Agent | +| UA | User Agent | +| UNTDI | United Nations, Trade Data Interchange | +| UTC | Coordinated Universal Time | + +# 5 Conventions + +In clause 2, ISO/IEC aligned standards are cited. + +Common language practices have been applied as far as possible in the use of capitalization of words. + +# 6 EDI messaging service + +## 6.1 Introduction + +The EDI messaging service provides an EDI messaging user with features to assist in communicating with other EDI messaging users. EDI messaging users are in many cases computer processes. The EDI messaging service uses the capabilities of the Message transfer service (see also Recommendation F.410) for sending and receiving EDI messages. The elements of service describing the features of the EDI messaging service are defined in Annex B, and classified in clause 14. + +EDI, electronic data interchange, can be described as computer to computer exchange of structured business data, such as invoices and purchase orders. In some cases, the EDI messaging service can be used to transmit an EDI interchange to a physical rendition system, such as a physical delivery system, or facsimile. + +The EDI messaging service is provided by EDI messaging. + +## **6.2 EDI messaging** + +EDI Messaging (EDIMG) consists of the exchange of EDI Messages (EDIMs), and EDI Notifications (EDINs), which are information objects specified in ITU-T Rec. X.435 | ISO/IEC 10021-9. + +## **6.3 EDI messaging environment** + +The environment in which EDI messaging takes place can be modelled as a functional object which is hereafter referred to as the EDI Messaging Environment (EDIME). When refined (i.e. functionally decomposed), the EDIME can be seen to comprise lesser objects referred to as the primary objects of EDI messaging. They include a single central object, the EDI Messaging System (EDIMS), and numerous peripheral objects called EDI Messaging users (EDIMG users). + +The structure of the EDIME is depicted in Figure 1. + +![Diagram of the EDI messaging environment (EDIME) showing a central EDIMS object connected to two EDIMG user objects.](853f59c89931a666c07903b31d098277_img.jpg) + +``` + +graph LR + subgraph EDIME [EDI messaging environment] + EDIMS((EDIMS)) + User1((EDIMG user)) + User2((EDIMG user)) + User1 --- EDIMS + User2 --- EDIMS + end + +``` + +The diagram illustrates the structure of the EDI messaging environment (EDIME). It consists of a large oval labeled "EDI messaging environment". Inside this oval, there is a central oval labeled "EDIMS". To the left of the "EDIMS" oval, there are two smaller ovals, each labeled "EDIMG user". Lines connect each "EDIMG user" oval to the central "EDIMS" oval, indicating communication or interaction. + +Diagram of the EDI messaging environment (EDIME) showing a central EDIMS object connected to two EDIMG user objects. + +T0101260-90/d01 + +**Figure 1/F.435 – EDI messaging environment** + +## **6.4 EDI messaging user** + +An EDI Messaging user (EDIMG user) is a user that engages in EDI messaging. An EDIMG user originates, receives, or both originates and receives EDIMs. The EDIME contains any number of EDIMG users. + +An EDIMG user may be a person or a computer process. An EDIMG user may access the EDIMS through an access unit. + +# **7 EDI messaging system** + +## **7.1 Introduction** + +The EDI Messaging System (EDIMS) is the functional object by means of which all EDIMG users communicate with one another in EDI messaging. + +The EDIMS can be modelled as comprising lesser functional objects which interact with one another. These lesser objects are referred to as the secondary objects of EDI messaging. They include a single, central object, the message transfer system (MTS), and numerous peripheral objects of three kinds: EDI User Agents (EDI-UAs), EDI Message Stores (EDI-MSs), and EDI Access Units (EDI-AUs). + +The structure of the EDIMS is depicted in Figure 2. As shown in Figure 2, EDI-UAs, EDI-MSs, and EDI-AUs are the objects by which the EDIMS provides service to EDIMG users. + +![Diagram of the EDI messaging system structure. A large oval labeled 'EDI messaging system' contains four components: 'EDI-UA' (top left), 'EDI-MS' (middle left), 'Message transfer system' (center), and 'EDI-AU' (right). The 'EDI-UA' components are connected to the 'Message transfer system'. Outside the oval, three 'EDIMG user' nodes are shown: one at the top left connected to the first 'EDI-UA', one at the bottom left connected to the second 'EDI-UA', and one at the bottom right connected to the 'EDI-AU'. A reference code 'T0101270-90/d02' is located at the bottom right of the diagram.](4e4be0bd8b235167902f2c03e41da651_img.jpg) + +Diagram of the EDI messaging system structure. A large oval labeled 'EDI messaging system' contains four components: 'EDI-UA' (top left), 'EDI-MS' (middle left), 'Message transfer system' (center), and 'EDI-AU' (right). The 'EDI-UA' components are connected to the 'Message transfer system'. Outside the oval, three 'EDIMG user' nodes are shown: one at the top left connected to the first 'EDI-UA', one at the bottom left connected to the second 'EDI-UA', and one at the bottom right connected to the 'EDI-AU'. A reference code 'T0101270-90/d02' is located at the bottom right of the diagram. + +Figure 2/F.435 – EDI messaging system + +### 7.1.1 EDI user agents + +An EDI User Agent (EDI-UA) is a user agent tailored so as to better assist a single EDIMG user to engage in EDI messaging. It helps that EDIMG user originate and receive messages containing EDIMS. The EDIMS contains any number of EDI-UAs. + +NOTE – An exact definition of the boundary between the EDI-UA and the EDIMG user is beyond the scope of this Recommendation. + +### 7.1.2 EDI message store + +An EDI Message Store (EDI-MS) is a message store tailored so as to better assist a single EDI-UA engage in EDI messaging. It helps that EDI-UA submit, take delivery of, store, and retrieve messages containing EDIMS. + +### 7.1.3 Message transfer system + +In the present context, the Message Transfer System (MTS) conveys EDIMS or EDI notifications (EDINs) between EDI-UAs, or between an EDI-UA and an access unit. The EDIMS contains a single MTS. + +### 7.1.4 EDI access units + +An EDIMG user may have access to/from the EDIMS through an access unit (AU). One type of access unit is the physical delivery access unit (PDAU). In EDIMG, the physical delivery access unit provides the ability to send messages to EDIMG recipients through a physical delivery system (PDS). Other types of EDI-AUs (e.g. facsimile access units) may be the subject of future standardization. + +## 7.2 Information flow in the EDIMS + +Figure 3 expands on Figure 2 and shows the principal information flows in EDI messaging. + +## 7.3 EDI messaging service functional model + +Figure 4 shows the functional model of the EDI messaging service. The UAs used in the EDI messaging service comprise a specific class of cooperating UAs. The optional PDAU allows EDIMG users to send messages to indirect users outside of the EDI messaging environment. The message stores used in the EDI messaging service have specific EDI related functions and can optionally be used by EDIMG users to take delivery of messages on their behalf. The Telematic Agent (TLMA) shown in Figure 4 will allow access to telematic services and may be the subject of future standardization. + +![Diagram of information flow in EDI messaging showing the flow from an EDIMG user through an EDI interchange, EDI messaging system (EDI-UA, MTA, MTS), and another EDI messaging system to a second EDIMG user.](4801720824e4b5e2361a5564f91cfb70_img.jpg) + +``` + +graph TD + User1[EDIMG user +EDI application] <--> Interchange + subgraph Interchange [EDI interchange including interchange control information] + direction LR + EDIFACT[EDIFACT +UNA...UNZ] --- or1((or)) + ANSI[ANSI X12 +ISA...IEA] --- or2((or)) + UNTDI[UNTDI +STX...END] + end + Interchange <--> UA1[EDI-UA] + subgraph MessagingSystem [EDI messaging system] + UA1 <--> MTS + subgraph MTS [MTS] + MTA1[MTA] <--> MTA2[MTA] + MTA1 <--> MTA3[MTA] + MTA2 <--> MTA3 + end + MTS <--> MS[EDI-MS] + MS <--> UA2[EDI-UA] + UA3[EDI-UA] <--> MTA1 + end + User2[EDIMG user] <--> UA3 + UA2 <--> User3[EDIMG user] + +``` + +The diagram illustrates the information flow in EDI messaging. At the top, an **EDIMG user (EDI application)** is shown. Below it, an **EDI interchange including interchange control information (i.e. receiver ID, etc.)** is shown, containing three options: **EDIFACT** (UNA, UNZ), **ANSI X12** (ISA, IEA), or **UNTDI** (STX, END). The **EDI messaging system** is shown in a large box, containing an **EDI-UA** and an **MTS** (Message Transfer System). The **EDI-UA** is connected to the EDI interchange and the **MTS**. The **MTS** contains three **MTA** (Message Transfer Agent) components. The flow is as follows: + +- Generates EDI heading and submit envelope from interchange control info plus other info in EDI interchange (receiver ID maps into F.400/X.400 O/R name) +- Places EDI heading + EDI interchange in submit envelope and submits to MTS +- On delivery from MTS extracts EDI interchange and passes to EDI application + +Diagram of information flow in EDI messaging showing the flow from an EDIMG user through an EDI interchange, EDI messaging system (EDI-UA, MTA, MTS), and another EDI messaging system to a second EDIMG user. + +NOTE – For abbreviations and acronyms see clause 4 and Annex A. + +T0101280-90/d03 + +**Figure 3/F.435 – Information flow in EDI messaging** + +## 7.4 Structure of EDI messages + +The EDI class of UAs create messages containing a content specific to the EDI messaging service. The specific content that is sent from one EDI-UA to another is a result of an originator, which is generally an application process, composing and sending a message, called an EDI Message (EDIM). The EDIM carries the EDI interchange and optionally other information associated with the EDI interchange. Only one EDI interchange shall be present in an EDIM. Every EDIM shall contain an EDI interchange body part on origination of the EDIM. Any of the body parts can subsequently be removed (wholly, not partially) when forwarding an EDIM, except a forwarded body part, which cannot be removed. Body parts that are removed when forwarding are replaced with place holders to indicate what type of body part was removed. The heading of an EDIM shall not be removed when forwarding an EDIM. The structure of an EDIM as it relates to the basic message structure of MHS is shown in Figure 5. The EDIM is conveyed with an envelope when being transferred through the MTS. + +![EDI messaging service functional model diagram showing the flow from EDIMG user through EDI messaging service (EDI-UA, EDI-MS, MT service, PDAU, TLMA) to Physical delivery services (PDS) and finally to the Recipient.](33ed1f9b27c7c21c797aa928b0f06851_img.jpg) + +The diagram illustrates the functional model of an EDI messaging service. At the top, an 'EDIMG user' is connected to an 'EDI-UA' node. This node is part of a larger 'EDI messaging service' block, which also contains 'EDI-MS', 'MT service', 'PDAU', and 'TLMA' nodes. The 'EDI-UA' node connects to another 'EDIMG user' on the left and to an 'EDI-MS' node on the right. The 'EDI-MS' node connects to another 'EDI-UA' node, which in turn connects to another 'EDIMG user' on the right. The 'MT service' node is connected to the 'PDAU' node. The 'PDAU' node connects to a 'Physical delivery services' block, which contains a 'PDS' node. The 'PDS' node connects to the final 'Recipient' node at the bottom. + +EDI messaging service functional model diagram showing the flow from EDIMG user through EDI messaging service (EDI-UA, EDI-MS, MT service, PDAU, TLMA) to Physical delivery services (PDS) and finally to the Recipient. + +T0101290-90/d04 + +Figure 4/F.435 – EDI messaging service functional model + +![EDI message structure diagram showing the hierarchy of an EDI message, including Envelope, Heading, Body (with Body parts), and Content.](7efae06af3af43ffe5d4b956a679cf54_img.jpg) + +The diagram shows the structure of an EDI message. It is divided into two main parts: the 'Basic message structure' and the 'EDI message' structure. The 'Basic message structure' consists of an 'Envelope' box followed by a chevron shape labeled 'Envelope'. The 'EDI message' structure is shown on the left, with a vertical double-headed arrow labeled 'EDI message'. It includes a 'Heading' box, a 'Body' box containing multiple 'Body part' boxes, and a 'Content' box. A line connects the 'Heading' box to the 'Content' box, and a plus sign is placed between the 'Body' box and the 'Content' box. + +EDI message structure diagram showing the hierarchy of an EDI message, including Envelope, Heading, Body (with Body parts), and Content. + +T0101300-90/d05 + +Figure 5/F.435 – EDI message structure + +![Diagram illustrating the mapping between an EDIFACT interchange and an EDI message structure. The EDIFACT interchange is shown on the left with segments: UNA, UNB, UNH, User data segments, UNT, and UNZ. The EDI message structure is shown on the right, divided into a Heading and a Body. The Heading contains Field 1, Field 2, Field 3, and Field n. The Body contains Primary body part, Body part n, and additional information (e.g. drawings). Dashed lines indicate the mapping: UNA, UNB, and UNH map to the Heading; User data segments, UNT, and UNZ map to the Primary body part; Additional information maps to the Body part n.](fa859e4e468bfb2710a94527f2c504af_img.jpg) + +T0101310-90/d06 + +Diagram illustrating the mapping between an EDIFACT interchange and an EDI message structure. The EDIFACT interchange is shown on the left with segments: UNA, UNB, UNH, User data segments, UNT, and UNZ. The EDI message structure is shown on the right, divided into a Heading and a Body. The Heading contains Field 1, Field 2, Field 3, and Field n. The Body contains Primary body part, Body part n, and additional information (e.g. drawings). Dashed lines indicate the mapping: UNA, UNB, and UNH map to the Heading; User data segments, UNT, and UNZ map to the Primary body part; Additional information maps to the Body part n. + +**Figure 6/F.435 – EDI message structure for a typical EDI transaction** + +Figure 6 shows a mapping between a typical EDI interchange, and the corresponding EDI message structure. The EDI interchange is mapped entirely within one body part, called the primary body part, and may be an EDIFACT, ANSI X12, UNEDIFACT or privately defined EDI interchange. Other body parts are available to convey information associated with the EDI interchange such as drawings, explanatory text, etc. The heading of the EDIM contains various fields of information, some of which are present in the EDIFACT interchange header segments (or corresponding ISA or STX segments for ANSI X12 and UNEDIFACT), and others containing service requests from the originator. The heading and body part(s) form the EDIM. + +## 7.5 EDI notification + +An EDIMG user can request that a recipient return an EDI Notification (EDIN) indicating the disposition of the EDI message received. This notification is requested by an originating EDI-UA, and is generated by a recipient EDI-UA, EDI-MS, or AU. There are 3 possible conditions that can be requested and reported on, resulting in either the generation of a Positive Notification (PN), a Negative Notification (NN), or a Forwarded Notification (FN). The implied meanings of the responses PN, NN, and FN are described in 8.1. It is possible to forward a received EDI message unchanged and forward the obligation to respond to the notification request to the recipient to whom the EDI message is forwarded, or intermediate recipients, who then shall respond to the original originator of the message. An originating EDI-UA may request to be notified if the obligation to respond to the notification request has been forwarded. In this case, the EDI-UA or EDI-MS that forwards the EDIM shall send to the originating EDI-UA an EDI Forwarded Notification (FN). + +In all cases, including notifications sent by EDI-UAs to whom the EDIM has been forwarded, the notifications shall contain the O/R name of the recipient that was specified by the original originator. + +The originating EDI-UA may request any combination of the several EDINs from any combination of the recipients to whom the EDIM is sent. If no notifications are requested by an originator, none shall be sent by the recipient(s). + +EDI notifications cannot be forwarded, and EDI notifications cannot be requested for EDINs. + +# 8 EDIM responsibility and forwarding + +## 8.1 Introduction + +The EDIMS includes a concept called EDIM responsibility. This concept is key to the description below of EDINs and forwarding. In order to simplify the descriptions in the text below, all forwarding is shown as performed by the EDI-UA. It should be noted that the descriptions apply equally to forwarding performed by the EDI-MS. + +The purpose for introducing the concept of EDIM responsibility is primarily to provide a method for confirming the passing of messages amongst EDI-UAs. EDIM responsibility may apply to access units in certain cases. The concept of EDIM responsibility is described as follows. + +EDIM responsibility indicates that the EDIM is made available to the EDIMG user by the receiving EDI-UA. EDIM responsibility shall always be accepted when the EDI-UA adds or removes body parts when forwarding. An EDIM cannot leave the EDIMS unless EDIM responsibility has been accepted (delivery to a PDAU is a special case as described in 11.3). If requested to do so by the originating EDI-UA, the recipient EDI-UA, and possibly intermediate EDI-UAs (if requested), shall send EDINs to the originating EDI-UA. + +When an EDI-UA receives an EDIM it shall, if requested to do so, inform the originating EDI-UA that the recipient EDI-UA has accepted or refused EDIM responsibility by sending an appropriate EDIN. Subclause 8.2 below contains a detailed description of the EDINs that are sent in various scenarios. + +If notifications are requested, then when an EDI-UA accepts, refuses, or forwards EDIM responsibility, it shall send an appropriate EDIN to the originator, and if forwarding, it shall create the appropriate heading fields in the forwarded EDIM. The details of these operations are described in ITU-T Rec. X.435 | ISO/IEC 10021-9. + +Body parts that are forwarded cannot be changed in any way. If EDIM responsibility is forwarded, the forwarded EDIM cannot be changed in any way. If EDIM responsibility is accepted, body parts may be removed from, or added to the original EDIM when creating the forwarded EDIM. Body parts that are removed when forwarding are replaced with place holders to indicate what type of body part was removed. EDIM responsibility forwarding is limited to only one recipient. + +EDIMG includes mechanisms to prevent looping when forwarding. + +## 8.2 Forwarding and secondary distribution + +In EDIMG it may be desirable to receive EDI messages at a central EDI user agent, with subsequent forwarding to the final EDI user agents. Such a practice would, for example, enable a large organization to perform centralized functions such as logging, auditing, etc. on all EDI message traffic entering that organization. After performance of these functions the traffic would be distributed to the EDI user agents serving the recipient EDI applications. Similarly, a value added network service provider might operate a similar intermediary stage on behalf of its customers. The following text describes the use of an EDI-UA as such an intermediary stage. + +Since an intermediate EDI-UA will generally not be the final EDI-UA, there is a need to provide end-to-end confirmation of EDIM responsibility acceptance for an EDIM within EDIMG. The element of service "EDI notification Request" allows an originator to request from each recipient, positive, negative and forwarded notifications. Together with protocol elements defined in ITU-T Rec. X.435 | ISO/IEC 10021-9, the "EDI notification request" allows intermediate EDI-UAs to indicate, in a forwarded message, whether or not EDIM responsibility has been accepted. These tools allow EDIM responsibility acceptance to be deferred until an EDIM reaches the final EDI-UA, and provide an indication to that EDI-UA that a notification is to be returned to the original originator. + +In order to illustrate the use of an EDI-UA as an intermediate stage, three cases are described below. In all cases, an EDIM originates in EDI-UA1 and terminates in EDI-UA3. EDI-UA2 is the intermediate EDI-UA. In cases 1 and 2 it is assumed that the EDIM is forwarded with content unchanged. In all three cases it is assumed that EDI-UA1 has requested notifications. + +NOTE – Events described in the following tables are not necessarily performed in the exact sequential order shown in the Table 1. + +## 8.3 Case 1: No forwarding + +The EDIM prepared by EDI-UA1 is addressed to EDI-UA3. The EDIM is submitted to MTA1, transferred to MTA3, delivered to EDI-UA3 and retrieved by EDIMG user 3. EDI-UA3 will respond with an appropriate EDIN, accepting EDIM responsibility (i.e. PN). [If EDI-UA3 had determined that EDIMG user 3 could not retrieve the message, EDI-UA3 would have responded with an EDIN refusing EDIM responsibility (i.e. NN)]. Figure 7 illustrates the flow of information. The sequence of EDIMs and EDINs is depicted in Table 1. + +![Diagram illustrating the flow of information for Case 1: No forwarding. The diagram shows nested ellipses representing the EDI messaging environment (EDIME), EDI messaging system (EDIMS), and Message transfer system (MTS). Components include EDIMG user 1, EDI-UA1, MTA1, MTA2, MTA3, EDI-UA3, and EDIMG user 3. Arrows indicate the direction of transfer for EDIM (solid) and EDIN (dashed).](dfe556fea00682b09a59427aaf72051c_img.jpg) + +The diagram illustrates the flow of information for Case 1: No forwarding. It shows three nested ellipses. The outermost is labeled "EDI messaging environment (EDIME)", the middle is "EDI messaging system (EDIMS)", and the innermost is "Message transfer system (MTS)". Inside the MTS are MTA1, MTA2, and MTA3. MTA1 and MTA3 are connected by a solid arrow (2) pointing right and a dashed arrow (5) pointing left. MTA2 is above them. Outside the MTS but inside the EDIMS are EDI-UA1 (left), EDI-UA2 (top), and EDI-UA3 (right). EDI-UA1 is connected to MTA1 by a solid arrow (1) pointing right and a dashed arrow (6) pointing left. EDI-UA3 is connected to MTA3 by a solid arrow (3) pointing right and a dashed arrow (4) pointing left. Outside the EDIMS but inside the EDIME are EDIMG user 1 (left), EDIMG user 2 (top), and EDIMG user 3 (right). EDIMG user 1 is connected to EDI-UA1, and EDIMG user 3 is connected to EDI-UA3. + +Legend: + +---> Direction of transfer EDIN +—> Direction of transfer EDIM + +T0101320-90/d07 + +Diagram illustrating the flow of information for Case 1: No forwarding. The diagram shows nested ellipses representing the EDI messaging environment (EDIME), EDI messaging system (EDIMS), and Message transfer system (MTS). Components include EDIMG user 1, EDI-UA1, MTA1, MTA2, MTA3, EDI-UA3, and EDIMG user 3. Arrows indicate the direction of transfer for EDIM (solid) and EDIN (dashed). + +**Figure 7/F.435 – Case 1: No forwarding** + +**Table 1/F.435 – Case 1: No forwarding** + +| Events | EDIM | EDIN | +|--------|-------------------------------|--------------------------------| +| 1 | EDI-UA1 submits EDIM to MTA1 | | +| 2 | MTA1 transfers EDIM to MTA3 | | +| 3 | MTA3 delivers EDIM to EDI-UA3 | | +| 4 | | EDI-UA3 submits PN/NN to MTA3 | +| 5 | | MTA3 transfers PN/NN to MTA1 | +| 6 | | MTA1 delivers PN/NN to EDI-UA1 | + +## 8.4 Case 2: Content not changed and EDIM responsibility forwarded + +In this case an intermediary EDI-UA forwards a message from EDI-UA1 to EDI-UA3. The final recipient is EDI-UA3, and EDI-UA2 performs a forward operation, forwarding EDIM responsibility to EDI-UA3. The EDIM prepared by EDI-UA1 is addressed to EDI-UA2. The EDIM is delivered to EDI-UA2, which forwards it unchanged to EDI-UA3, based on selection criteria known to EDI-UA2. + +EDIM responsibility is handled as follows: + +- 1) When EDI-UA2 forwards EDIM responsibility, it shall create the forwarded EDIM so that requested EDINs are received by EDI-UA1, (see ITU-T Rec. X.435 | ISO/IEC 10021-9 for details). The following EDINs may be sent. + - a) If EDI-UA1 requested notification of forwarding of EDIM responsibility, EDI-UA2 shall send forwarded notification - FN to EDI-UA1. This EDIN is sent when EDI-UA2 successfully submits the EDIM to MTA2. + - b) If EDI-UA2 receives a non-delivery notification from MTA3 (via MTA2) it may send negative notification – NN to EDI-UA1. Note that EDI-UA2 has the choice to send or not to send the EDIN in this case. + +No other EDINs may be requested or sent. For example, EDI-UA2 cannot request notifications from EDI-UA3, and EDI-UA3 cannot send EDINs to EDI-UA2. + +In the case of non-delivery, EDI-UA2 may attempt to resubmit the EDIM to the intended recipient. In this case, the NN to EDI-UA1 is sent only when EDI-UA2 determines that it shall no longer attempt to resubmit the EDIM to EDI-UA3. + +- c) If forwarding succeeds, EDI-UA3 shall send an appropriate EDIN to EDI-UA1, accepting or refusing EDIM responsibility. + +Figure 8 illustrates the information flow described above for case 2. The sequence of possible EDIMs and EDINs is explained in Table 2. Events (8, 11, 13, 15) and (10, 12, 14, 16) are mutually exclusive. + +![Figure 8: Information flow for Case 2: EDIM responsibility forwarded. The diagram shows nested circles representing EDIME, EDIMS, and MTS. Components include EDI-UA1, MTA1, MTA2, EDI-UA2, MTA3, and EDI-UA3. Arrows with numbers 1-16 show the flow of EDIM (solid) and EDIN/NDN (dashed) messages.](40f30e4d577a17052f8b1e6dc802a0d8_img.jpg) + +``` + +graph TD + subgraph EDIME + subgraph EDIMS + subgraph MTS + MTA1 + MTA2 + MTA3 + end + UA1[EDI-UA1] + UA2[EDI-UA2] + UA3[EDI-UA3] + end + User1[EDIMG user 1] + User2[EDIMG user 2] + User3[EDIMG user 3] + end + + User1 --- UA1 + User2 --- UA2 + User3 --- UA3 + + UA1 -- 1: EDIM --> MTA1 + MTA1 -- 2: EDIM --> MTA2 + MTA2 -- 3: EDIN --> UA2 + UA2 -- 4: EDIM --> MTA2 + UA2 -- 5: EDIN --> MTA2 + MTA2 -- 7: EDIM --> MTA3 + MTA3 -- 10: EDIM --> UA3 + UA3 -- 12: EDIN --> MTA3 + MTA3 -- 14: EDIN --> MTA1 + MTA2 -- "6, 13: EDIN" --> MTA1 + MTA1 -- "9, 15, 16: EDIN" --> UA1 + UA2 -- 8: EDIN --> UA1 + UA3 -- 11: EDIN --> UA1 + +``` + +The diagram shows the flow of messages between three users (EDIMG user 1, 2, and 3) through their respective User Agents (EDI-UA1, 2, and 3) and Message Transfer Agents (MTA1, 2, and 3). The system is organized into nested layers: the Message Transfer System (MTS) contains the MTAs; the EDI Messaging System (EDIMS) contains the MTS and the UAs; and the EDI Messaging Environment (EDIME) contains the EDIMS and the users. Solid arrows indicate the direction of EDIM transfer, while dashed arrows indicate the direction of EDIN or NDN transfer. Numbered steps detail the sequence of message exchanges. + +Legend: + +---> Direction of transfer EDIM + +- - -> Direction of transfer EDIN, NDN + +Figure 8: Information flow for Case 2: EDIM responsibility forwarded. The diagram shows nested circles representing EDIME, EDIMS, and MTS. Components include EDI-UA1, MTA1, MTA2, EDI-UA2, MTA3, and EDI-UA3. Arrows with numbers 1-16 show the flow of EDIM (solid) and EDIN/NDN (dashed) messages. + +T0101330-90/d08 + +**Figure 8/F.435 – Case 2: EDIM responsibility forwarded** + +**Table 2/F.435 –Case 2: EDIM responsibility forwarded** + +| Events | EDIM | EDIN | NDN | +|--------|----------------------------------------|------------------------------------------|-----------------------------| +| 1 | EDI-UA1 submits EDIM to MTA1 | | | +| 2 | MTA1 transfers EDIM to MTA2 | | | +| 3 | MTA2 delivers EDIM to EDI-UA2 | | | +| 4 | | If requested, EDI-UA2 submits FN to MTA2 | | +| 5 | EDI-UA2 submits forwarded EDIM to MTA2 | | | +| 6 | | MTA2 transfers FN to MTA1 | | +| 7 | MTA2 transfers EDIM to MTA3 | | | +| 8 | | | MTA2 sends NDN to EDI-UA2 | +| 9 | | MTA1 delivers FN to EDI-UA1 | | +| 10 | MTA3 delivers EDIM to EDI-UA3 | | | +| 11 | | | EDI-UA2 submits NN to MTA2 | +| 12 | | EDI-UA3 submits PN/NN to MTA3 | | +| 13 | | | MTA2 transfers NN to MTA1 | +| 14 | | MTA3 transfers PN/NN to MTA1 | | +| 15 | | | MTA1 delivers NN to EDI-UA1 | +| 16 | | MTA1 delivers PN/NN to EDI-UA1 | | + +The following should be noted: + +- 1) EDI-UA1 will usually receive several EDINs if it requests FN ( forwarded notification). +- 2) EDI-UA1 may receive EDINs in a sequence other than that in which they were created. +- 3) EDI-UA1 may receive no EDIN whatsoever even if it requested FN (for example, in the case of catastrophic failure of EDI-UA2 after MTA2 has delivered the EDIM to EDI-UA2). + +It is up to EDI-UA1 to correctly handle 1) through 3) above. Item 1 can be handled for example, by keeping track of: + +- a) the EDIM ID; +- b) the original recipient; +- c) the submission time, and +- d) the EDI notifications expected. + +Item 2 can be handled by using the UTC time included in the EDIN (EDIN creation time). Item 3 can be handled with a time-out mechanism in EDI-UA1. Mechanisms to handle 1 to 3 are local implementation issues, thus beyond the scope of this Recommendation. + +## 8.5 Case 3: EDIM responsibility not forwarded + +This scenario provides for the case where the EDIM prepared by EDI-UA1 is addressed to EDI-UA2, and EDI-UA2 accepts EDIM responsibility for the message prior to forwarding to EDI-UA3. This would occur, for example, if EDI-UA2 were to add or remove body parts when forwarding (changes of the content). When EDIM responsibility is accepted, EDI-UA2 sends an EDIN to the originator (i.e. PN). and creates the forwarded EDIM so that no further EDINs are received by EDI-UA1 (the originator) (see ITU-T Rec. X.435 | ISO/IEC 10021-9 for details). As in case 2, EDI-UA1 addresses the EDIM to EDI-UA2. As in both previous cases EDI-UA3 represents the final destination. + +Upon retrieval of the EDIM, EDI-UA2 returns an appropriate notification to EDI-UA1. The message is then forwarded to EDI-UA3. Since initial EDIM responsibility has now been accepted, EDI-UA2 is at liberty to request EDIM responsibility or not, as desired. If requested, the resulting EDIM responsibility relationship shall apply between EDI-UA3 and EDI-UA2, i.e. not end to end as in the previous cases. In the scenario described here EDIM responsibility is assumed to have been requested, with the result that EDI-UA3 responds to EDI-UA2 with an appropriate notification. + +Figures 9 and 10 illustrate the flow of information for case 3. The sequence of EDIMs and EDINs for case 3 are explained in Table 3. + +![Diagram of EDI messaging environment showing message flow between users and agents. Dashed arrow Solid arrow](a3472689858b068ef469213682965325_img.jpg) + +The diagram illustrates the flow of information for Case 3: EDIM responsibility not forwarded, Part 1. It consists of three nested elliptical regions: + +- EDI messaging environment (EDIME)**: The outermost region containing EDIMG user 1, EDIMG user 2, and EDIMG user 3. +- EDI messaging system (EDIMS)**: The middle region containing EDI-UA1, EDI-UA2, and EDI-UA3. +- Message transfer system (MTS)**: The innermost region containing MTA1, MTA2, and MTA3. + +The message flow is indicated by numbered arrows: + +- Solid arrow from EDIMG user 1 to EDI-UA1. +- Solid arrow from EDI-UA1 to MTA1. +- Solid arrow from MTA1 to MTA2. +- Solid arrow from MTA2 to EDI-UA2. +- Solid arrow from EDI-UA2 to MTA2. +- Dashed arrow from EDI-UA2 to MTA2 (labeled 6). +- Dashed arrow from MTA2 to MTA1. +- Dashed arrow from MTA1 to EDI-UA1 (labeled 8). + +Note: The diagram also shows EDIMG user 2 connected to EDI-UA2, and EDIMG user 3 connected to EDI-UA3. MTA3 is shown within the MTS but without specific numbered flow arrows in this 'Part 1' view. + +Legend: + + Direction of transfer EDIN + + Direction of transfer EDIM + +Diagram of EDI messaging environment showing message flow between users and agents. Dashed arrow Solid arrow + +T0101340-90/d09 + +**Figure 9/F.435 – Case 3: EDIM responsibility not forwarded, Part 1** + +# 9 EDI naming, addressing and use of directory + +The MHS use of Directory, as defined in clause 13/F.400/X.400, is used to provide the Directory services required for EDI messaging. + +Each management domain should provide directory services for its EDIMG users. + +EDI messaging, naming and addressing and the subsequent directory service requirements are outlined in Annex D. + +![Diagram of EDI messaging environment (EDIME) showing nested systems: Message transfer system (MTS), EDI messaging system (EDIMS), and EDIME. It includes components like EDIMG user 1, EDI-UA1, MTA1, EDI-UA2, EDIMG user 2, MTA2, MTA3, EDI-UA3, and EDIMG user 3. Arrows indicate transfer directions for EDIN (dashed) and EDIM (solid) with numbered steps 1-12.](e69b9188aa2c14ec6b21c83f711fef65_img.jpg) + +The diagram illustrates the EDI messaging environment (EDIME) as the outermost ellipse. Inside it is the EDI messaging system (EDIMS), which contains the Message transfer system (MTS). +- Within the MTS, there are three Message Transfer Agents: MTA1, MTA2, and MTA3. +- Outside the MTS but within the EDIMS are User Agents: EDI-UA1 (connected to MTA1), EDI-UA2 (connected to MTA2), and EDI-UA3 (connected to MTA3). +- Outside the EDIMS but within the EDIME are the users: EDIMG user 1, EDIMG user 2, and EDIMG user 3. +- Numbered arrows show the flow: + - Arrow 5 (solid): EDI-UA2 to MTA2. + - Arrow 7 (solid): MTA2 to MTA3. + - Arrow 9 (solid): MTA3 to EDI-UA3. + - Arrow 10 (dashed): EDI-UA3 to MTA3. + - Arrow 11 (dashed): MTA3 to MTA2. + - Arrow 12 (dashed): MTA2 to EDI-UA2. +- A legend indicates: + - Dashed arrow (--->): Direction of transfer EDIN + - Solid arrow (——>): Direction of transfer EDIM + +Diagram of EDI messaging environment (EDIME) showing nested systems: Message transfer system (MTS), EDI messaging system (EDIMS), and EDIME. It includes components like EDIMG user 1, EDI-UA1, MTA1, EDI-UA2, EDIMG user 2, MTA2, MTA3, EDI-UA3, and EDIMG user 3. Arrows indicate transfer directions for EDIN (dashed) and EDIM (solid) with numbered steps 1-12. + +T0101350-90/d10 + +---> Direction of transfer EDIN +——> Direction of transfer EDIM + +Figure 10/F.435 – Case 3: EDIM responsibility not forwarded, Part 2 + +**Table 3/F.435 – Case 3: responsibility not forwarded** + +| Events | Figure 9/F.435 | Figure 10/F.435 | +|--------|-------------------------------|----------------------------------------| +| 1 | EDI-UA1 submits EDIM to MTA1 | | +| 2 | MTA1 transfers EDIM to MTA2 | | +| 3 | MTA2 delivers EDIM to EDI-UA2 | | +| 4 | EDI-UA2 submits PN to MTA2 | | +| 5 | | EDI-UA2 submits forwarded EDIM to MTA2 | +| 6 | MTA2 transfers PN au MTA1 | | +| 7 | | MTA2 transfers EDIM to MTA3 | +| 8 | MTA1 delivers PN to EDI-UA1 | | +| 9 | | MTA3 delivers EDIM to EDI-UA3 | +| 10 | | EDI-UA3 submits PN/NN to MTA3 | +| 11 | | MTA3 transfers PN/NN to MTA2 | +| 12 | | MTA2 delivers PN/NN to EDI-UA2 | + +# 10 EDI security + +The MHS security capabilities are defined in clause 15/F.400/X.400, and are also applicable for EDI messaging. In addition, the following extensions are provided to 15.4/F.400/X.400. + +An overview of the extended security capabilities in EDIMG is as follows: + +*Proof of EDI notification:* Enables the recipient of an EDIM to create an EDIN which may be used by the recipient of the EDIN to authenticate the originator of the EDIN. + +*Non-repudiation of the EDI notification:* Provides the recipient of an EDIN with proof of the origin of the EDIN which will protect against any attempt by the originator of the EDIN from falsely denying sending the EDIN. + +*Proof of content received:* Enables the originator of an EDIM to verify that the message content received by the recipient was the same as the message content originated by the originator. + +*Non-repudiation of content originated:* Provides the recipient of the EDIM with proof that the message content received was the same as the message content originated. This protects against any attempt by the originator to falsely deny originating the message content. + +*Non-repudiation of content received:* Provides the originator of the EDIM with proof that the message content received was the same as the message content originated. This proof will protect against any attempt by the recipient to falsely deny the content of the EDIM received. + +**Table 4/F.435 – Provision and use of secure messaging elements of service by MHS components** + +| Elements of service | EDIM originator | MTS | EDIM recipient | +|--------------------------------------------------------|------------------------|------------|-----------------------| +| Proof of EDI notification | U | – | P | +| Non-repudiation of EDI notification | U | – | P | +| Proof of content received | U | – | P | +| Non-repudiation of content originated | P | – | U | +| Non-repudiation of content received | U | – | P | +| P A provider of the service
U A user of the service | | | | + +Annex C describes the EDIMS vulnerabilities and details how they are countered. Annex A ITU-T Rec. X.435 | ISO/IEC 10021-9 describes the enhancements required to the ITU-T Rec. X.402 | ISO/IEC 10021-2 security model for EDIMS (the enhanced security services). ITU-T Rec. X.435 | ISO/IEC 10021-9 describes operations and procedures for security services. + +# 11 Intercommunication with physical delivery services + +## 11.1 Introduction + +As defined in Recommendation F.415, MH/PD intercommunication is a generic capability of the Message Transfer service. To make use of this capability, the originator may use a postal O/R address on submission, or, if using a directory name on submission, select physical delivery as the "Requested delivery method" and choose any desired options from the MH/PD elements of service (Table 1/F.415). + +The originator provides the address of the recipient as defined in Recommendation F.401, postal O/R address. This may be done through the Directory. + +## 11.2 Delivery and notifications + +Delivery to the access unit occurs when the EDIM is passed from the final MTA to the PDAU (MTS to EDI-AU). + +Delivery notifications and EDI notifications relevant to physical delivery apply as defined in Recommendation F.415 with the addition of an EDIN, as depicted below in Figure 11. + +These notifications are generated by the MTA/PDAU system components, which are considered to be co-located. + +Definitions of "T" times are provided in Recommendation F.415; "Tedi" can be defined as: + +Tedi = Generation and delivery of the EDIN. + +NOTE 1 – Start time corresponds to the time at which the EDIN is generated. + +NOTE 2 – End time corresponds to the time that the EDIN is made available to the EDIMG user. + +## **11.3 Transfer of EDIM responsibility** + +While it is up to the PDAU to physically render and subsequently deliver an EDIM sent to it, a PDAU can never accept EDIM responsibility for an EDIM. If an "EDI notification request" is asked for, two possibilities exist for the PDAU. If it determines that it can render the EDIM for physical delivery, it shall return an FN to the originator of the EDIM. However, if it determines that it cannot render or deliver the EDIM, it shall return an NN to the originator of the EDIM. + +## **11.4 Physical rendition** + +The basic physical rendition details defined in Annex B/F.415, should be used as a base, primarily for the rendition of routing and delivery information such as the address blocks, position on the page relative to the window, etc. + +For hard copy physical rendition specific to EDI, three approaches are identified; + +- 1) Standardized rendition; +- 2) Privately defined rendition; +- 3) Accompanying information for rendition (may be the subject of future standardization). + +Alternatively, if rendition rules are not available, the EDIM could simply be printed "as is", if possible, assuming that the recipient is able to work with the information, possibly with guidelines provided through some other means or message. (Additional guidelines and rules for the physical rendition of EDIMs may be the subject of future standardization.) + +# **12 Use of message store for EDI** + +Message store features may be used for EDI messaging. The general MS elements of service, "stored message fetching", "stored message listing", "stored message summary", "stored message deletion", and "stored message alert" are applicable for EDI messaging. + +General MS attributes and auto-actions are described in ITU-T Rec. X.413 | ISO/IEC 10021-5. EDI specific MS attributes and auto actions are described in ITU-T Rec. X.435 | ISO/IEC 10021-9. + +An EDI specific MS element of service, "Stored EDI message auto-forward", provides suitable MS auto-forwarding capabilities for EDI messaging. + +Security policies may restrict the use of message store elements of service. + +# **13 Elements of service** + +Elements of service are particular features, functions, or capabilities of MHS. The elements of service applicable for EDI messaging are made up of MT elements of service and EDI messaging elements of service. The MT elements of service used in EDI messaging are called out in this Recommendation in Tables 5 to 7, however they are defined in Annex B/F.400/X.400. The definitions of elements of service specific to EDI messaging are also listed in Tables 5 to 7, and are defined in Annex B of this Recommendation. The realization of all the elements of service applicable to EDI messaging is described in other Recommendations in the X.400-series. + +# **14 Classification of elements of service** + +## **14.1 Basic EDI messaging service** + +The basic EDI messaging service, which makes use of the MT service, enables an EDIMG user to send and receive EDI messages. An EDIMG user prepares EDI messages with the assistance of an EDI user agent (EDI-UA). EDI-UAs cooperate with each other to facilitate communication between their respective EDIMG users. To send an EDI message, the originating EDIMG user submits the message to his EDI-UA specifying the O/R name of the recipient who is to receive the EDI message. The EDI message, which has an identifier conveyed with it, is then sent by the originator's EDI-UA to the recipient's EDI-UA/MS via the Message Transfer service. + +![Sequence diagram showing MH/PD delivery and notification times model. It includes entities: EDIMG user, UA/MS, MTA, MTA/PDAU, PDS, and Recipient. The diagram is divided into three time intervals: T1, T2, and T3. T1 covers 'Creation of message' to 'Submission of message'. T2 covers 'Transfer of message' to 'Notification delivery'. T3 covers 'EDI notification generation' to 'Notification by MHS generation'. Vertical time markers Ta, Tb, and Tc are shown on the right side. A legend at the bottom indicates that grey chevrons represent 'Electronic event' and white envelopes represent 'Physical event'.](7133ccf78043568ca62ecbcd43628a4a_img.jpg) + +The diagram illustrates the flow of messages and notifications between an EDIMG user and a Recipient through various system components: UA/MS, MTA, MTA/PDAU, and PDS. The process is segmented into three time intervals: T1, T2, and T3. + +- T1:** From 'Creation of message' (Electronic event) to 'Submission of message' (Electronic event). +- T2:** From 'Transfer of message' (Electronic event) to 'Notification delivery' (Electronic event). This interval includes 'Delivery of message' (Electronic event), 'Notification generation' (Electronic event), 'Physical rendition' (Physical event), 'Physical transport' (Physical event), and 'Physical delivery' (Physical event). Vertical markers Ta and Tb are shown here. +- T3:** From 'EDI notification generation' (Electronic event) to 'Notification by MHS generation' (Electronic event). This interval includes 'EDI notification delivery' (Electronic event) and 'Notification by PDS delivery' (Physical event). Vertical marker Tc is shown here. + +Legend: + +- Electronic event (Grey chevron) +- Physical event (White envelope) + +Sequence diagram showing MH/PD delivery and notification times model. It includes entities: EDIMG user, UA/MS, MTA, MTA/PDAU, PDS, and Recipient. The diagram is divided into three time intervals: T1, T2, and T3. T1 covers 'Creation of message' to 'Submission of message'. T2 covers 'Transfer of message' to 'Notification delivery'. T3 covers 'EDI notification generation' to 'Notification by MHS generation'. Vertical time markers Ta, Tb, and Tc are shown on the right side. A legend at the bottom indicates that grey chevrons represent 'Electronic event' and white envelopes represent 'Physical event'. + +T0101360-90/d11 + +Figure 11/F.435 – MH/PD delivery and notification times model + +Following a successful delivery to the recipient's EDI-UA/MS, the EDI message is available for the recipient. To facilitate meaningful communication, a recipient can specify the encoded information type(s) contained in EDI messages that he will allow to be delivered to his EDI-UA, as well as the maximum length of an EDI message that he is willing to accept. The original encoded information type(s) and an indication if any conversions that have been performed and the resulting encoded information type(s) are supplied with each delivered EDI message. In addition, the submission time and delivery time are supplied with each EDI message. Non-delivery notification is provided with the basic MT service. The elements of service belonging to the basic EDI messaging service are listed in Table 5. + +**Table 5/F.435 –Elements of service belonging to +the basic EDI messaging service** + +| Elements of service | References | +|-----------------------------------------------------------------------------------------------------------|-------------------| +| Access management | B.1 | +| Content type indication | B.12 | +| Converted indication | B.15 | +| Delivery time stamp indication | B.22 | +| EDI message identification | EDI.8 | +| Message identification | B.41 | +| Non-delivery notification | B.47 | +| Original encoded information types indication | B.54 | +| Submission time stamp indication | B.89 | +| Type body | EDI.30 | +| User/UA capabilities registration | B.93 | +| NOTE – B references are to Annex B/F.400/X.400, and EDI references are to Annex B in this Recommendation. | | + +## **14.2 EDI messaging service optional user facilities** + +A set of the elements of service of the EDI messaging service are optional user facilities. The optional user facilities of the EDI messaging service, which may be selected on a per-message basis or for an agreed contractual period of time, are listed in Table 6 and Table 7, respectively. + +The optional user facilities of the EDI messaging service that are selected on a per-message basis are classified for both origination and reception by EDI-UAs. If a management domain offers these optional user facilities for origination by EDI-UAs, then an EDIMG user is able to create and send EDI messages according to the procedures defined for the associated element of service. If a management domain offers these optional user facilities for reception by EDI-UAs/MSs/AUs, then the receiving EDI-UA/MS/AU shall be able to receive and recognize the indication associated with the corresponding element of service and to inform the EDIMG user of the requested optional user facility. Each optional user facility is classified as additional (A) or essential (E) for EDI-UAs/MSs/AUs from these two perspectives. + +**Table 6/F.435 – EDI messaging optional user facilities selectable on a per-message basis** + +| Elements of service | Origination | Reception | References | +|-------------------------------------------------------|-------------|-----------|------------| +| Additional physical rendition | A | A | B.2 | +| Alternate recipient allowed | E | E | B.3 | +| Application security element | A | A | EDI.1 | +| Basic physical rendition | A | E* | B.7 | +| Character set | E | E | EDI.2 | +| Content confidentiality | A | A | B.10 | +| Content integrity | A | A | B.11 | +| Conversion prohibition | E | E | B.13 | +| Conversion prohibition in case of loss of information | A | A | B.14 | +| Counter collection | A | E* | B.16 | +| Counter collection with advice | A | A | B.17 | +| Cross reference information | A | E | EDI.3 | +| Deferred delivery | E | N/A | B.19 | +| Deferred delivery cancellation | E | N/A | B.20 | +| Delivery notification | E | N/A | B.21 | +| Delivery via bureaufax service | A | A | B.23 | +| Designation of recipient by directory name | A | N/A | B.24 | +| Disclosure of other recipients | E | E | B.25 | +| DL expansion history indication | N/A | E | B.26 | +| DL expansion prohibited | A | N/A | B.27 | +| EDI forwarding | A | N/A | EDI.4 | +| EDI message type(s) | E | E | EDI.5 | +| EDI notification request | E | E | EDI.6 | +| EDI standard indication | E | E | EDI.7 | +| EDIM responsibility forwarding allowed indication | E | E | EDI.9 | +| EDIN receiver | A | E | EDI.10 | +| EMS (Express Mail Service) a | A | E* | B.28 | +| Expiry date time indication | A | E | EDI.11 | +| Explicit conversion | A | N/A | B.30 | +| Grade of delivery selection | E | E | B.32 | +| Incomplete copy indication | A | E | EDI.12 | +| Interchange header | E | E | EDI.13 | +| Latest delivery designation | A | N/A | B.39 | +| Message flow confidentiality | A | N/A | B.40 | +| Message origin authentication | A | A | B.42 | +| Message security labelling | A | A | B.43 | +| Message sequence integrity | A | A | B.44 | +| Multi-destination delivery | E | N/A | B.45 | +| Multi-part body | A | E | EDI.14 | +| Non-repudiation of content originated | A | A | EDI.15 | +| Non-repudiation of content received | A | A | EDI.16 | +| Non-repudiation of content received request | A | A | EDI.17 | +| Non-repudiation of delivery | A | A | B.49 | +| Non-repudiation of EDI notification | A | A | EDI.18 | + +**Table 6/F.435 – EDI messaging optional user facilities selectable on a per-message basis (concluded)** + +| Elements of service | Origination | Reception | References | +|----------------------------------------------------|-------------|-----------|------------| +| Non-repudiation of EDI notification request | A | A | EDI.19 | +| Non-repudiation of origin | A | A | B.50 | +| Non-repudiation of submission | A | A | B.51 | +| Obsoleting indication | A | E | EDI.20 | +| Ordinary mail | A | E* | B.53 | +| Originator indication | E | E | EDI.21 | +| Originator requested alternate recipient | A | N/A | B.56 | +| Physical delivery notification by MHS | A | A | B.57 | +| Physical delivery notification by PDS | A | E* | B.58 | +| Physical forwarding allowed | A | E* | B.59 | +| Physical forwarding prohibited | A | E* | B.60 | +| Prevention of non-delivery notification | A | N/A | B.61 | +| Probe | A | N/A | B.63 | +| Probe origin authentication | A | N/A | B.64 | +| Proof of content received | A | A | EDI.22 | +| Proof of content received request | A | A | EDI.23 | +| Proof of delivery | A | A | B.65 | +| Proof of EDI notification | A | A | EDI.24 | +| Proof of EDI notification request | A | A | EDI.25 | +| Proof of submission | A | N/A | B.66 | +| Recipient indication | E | E | EDI.26 | +| Redirection disallowed by originator | A | N/A | B.68 | +| Registered mail | A | A | B.70 | +| Registered mail to addressee in person | A | A | B.71 | +| Related message(s) | A | E | EDI.27 | +| Report origin authentication | A | A | B.74 | +| Request for forwarding address | A | A | B.75 | +| Requested delivery method | E | N/A | B.76 | +| Services indication | A | A | EDI.28 | +| Special delivery a) | A | E* | B.81 | +| Stored message deletion | N/A | E** | B.84 | +| Stored message fetching | N/A | E** | B.85 | +| Stored message listing | N/A | E** | B.86 | +| Stored message summary | N/A | E** | B.87 | +| Undeliverable mail with return of physical message | A | E* | B.91 | +| Use of distribution list | A | N/A | B.92 | + +E    Essential optional user facility shall be provided +E\*    Essential optional user facility only applying to PDAUs +E\*\*   Essential optional user facility only applying to MSs +A    Additional optional user facility may be provided +N/A   Not applicable +a)   At least EMS or "Special delivery" shall be supported by the PDAU and associated PDS. +NOTE 1 – Bilateral agreement may be necessary in cases of reception by EDI-UA of elements of service classified as "A". +NOTE 2 – B references are to Annex B/F.400/X.400, and EDI references are to Annex B to this Recommendation. + +**Table 7/F.435 –EDI messaging service optional user facilities agreed +for a contractual period of time** + +| Elements of service | Classification | References | +|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-----------------------|--------------------| +| Auto-acknowledgement of EDI messages | A | EDI.31 | +| Auto-correlation of EDI messages | A | EDI.32 | +| Auto-correlation of EDI notifications | A | EDI.33 | +| Alternate recipient assignment | A | B.4 | +| Hold for delivery | A | B.33 | +| Implicit conversion | A | B.34 | +| MS register | A | B.nn a) | +| Redirection of incoming messages | A | B.69 | +| Restricted delivery | A | B.77 | +| Secure access management | A | B.79 | +| Stored EDI message auto-forward | A | EDI.29 | +| Stored message alert | A | B.82 | +| Stored message auto-forward | A | B.83 b) | +| Submission of EDI messages incorporating stored messages | A | EDI.34 | +|

a) This element of service shall be defined and assigned a "B" number in the next publication of Recommendation F.400/X.400. It describes a capability that is supported in Recommendation X.413, but not described in Recommendation F.400/X.400.

b) The use of this element of service, which is a general MS capability, is discouraged for EDI messaging. "Stored EDI message auto-forward", which is an EDI specific MS capability, provides a suitable alternative.

NOTE – B references are to Annex B/F.400/X.400, and EDI references are to Annex B to this Recommendation.

| | | + +# **15 Quality of service** + +## **15.1 EDI message status** + +The unique identification of each EDI message enables the system to provide information about e.g. the status of an EDI message. + +In the event of system failure all accepted and non-delivered EDI messages should be traceable. If EDI messages cannot be delivered, the originator shall be informed by a "Non-delivery notification" unless the originator invoked "Prevention of non-delivery notification". + +## **15.2 Support by providers of EDI service** + +Service providers should provide assistance to their subscribers, with regard to non-delivery notifications not being received in due time, as far as system components are concerned. Additional provision of support for status and tracing of messages may be provided under national responsibility. + +## **15.3 Model of delivery and notification times** + +See Figure 12. + +## **15.4 EDI message delivery time targets** + +The delivery time targets (including transfer times) are dependent on the message transfer system, on the number of transiting domains, and on message sizes. Values significantly less than those currently specified for the IPM service are aimed at. + +The management domain of the recipient EDI-UA should force non-delivery notification if the EDI message has not been delivered within *x* hours after submission (or after date and time indicated for deferred delivery), the value of *x* being dependent on the grade of delivery requested by the originator. + +The specification of these values for the EDI service may be the subject of future standardization. + +![Figure 12/F.435 – Notification time model. A sequence diagram showing the flow of an EDIM from an EDIMG user through EDI-UA/EDI-MS, two MTAs, and back to an EDIMG user. Time intervals T1, T2, and Tedi are marked on the left. T1 covers the submission and transfer of the EDIM. T2 covers the delivery of the EDIM and the generation of a delivery notification. Tedi covers the generation and delivery of an EDI notification.](11edb7fcedf09ac6a817f8d7b8c61eec_img.jpg) + +The diagram illustrates the notification time model for an EDIM. The sequence of events is as follows: + +- Creation of EDIM**: Initiated by the EDIMG user. +- Submission of EDIM**: The EDIM is submitted to the EDI-UA/EDI-MS. +- Transfer of EDIM**: The EDIM is transferred between MTAs. +- Delivery of EDIM**: The EDIM is delivered to the receiving EDI-UA/EDI-MS. +- Delivery notification generation**: A notification is generated by the receiving EDI-UA/EDI-MS. +- Delivery notification delivery**: The notification is delivered to the EDIMG user. +- EDI notification generation**: An EDI notification is generated by the receiving EDI-UA/EDI-MS. +- EDI notification delivery**: The EDI notification is delivered to the EDIMG user. + +Time intervals are defined on the left side of the diagram: + +- T1**: Delivery time, from the submission of the EDIM to the delivery of the EDIM. +- T2**: Delivery notification, from the delivery of the EDIM to the delivery of the delivery notification. +- Tedi**: Generation and delivery of EDI notification, from the generation of the EDI notification to the delivery of the EDI notification. + +T0101370-90/d12 + +Figure 12/F.435 – Notification time model. A sequence diagram showing the flow of an EDIM from an EDIMG user through EDI-UA/EDI-MS, two MTAs, and back to an EDIMG user. Time intervals T1, T2, and Tedi are marked on the left. T1 covers the submission and transfer of the EDIM. T2 covers the delivery of the EDIM and the generation of a delivery notification. Tedi covers the generation and delivery of an EDI notification. + +### **T1** *Delivery time* + +NOTE 1 – Start time of T1 corresponds to the submission time stamp indication. + +NOTE 2 – End time of T1 corresponds to the delivery time stamp indication. + +### **T2** *Delivery notification* + +NOTE 3 – Start time of T2 corresponds to the delivery time stamp indication. + +NOTE 4 – End time of T2 is the time the delivery notification is made available to the EDIMG user through the EDI-UA or EDI-MS. + +### **Tedi** *Generation and delivery of EDI notification* + +NOTE 5 – Start time corresponds to the time at which the EDIN is generated. + +NOTE 6 – End time corresponds to the time that the EDIN is made available to the EDIMG user. + +**Figure 12/F.435 – Notification time model** + +## **15.5 EDI notification time targets** + +Delivery time targets for EDI notifications depend on local arrangements. When EDINs are initiated by the receiving EDI-UA they have the same time targets as the EDI messages that caused them to occur (see Table 8). + +**Table 8/F.435 – EDIN time targets** + +| Grade of delivery | 95% delivered before | +|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-----------------------------| +| Urgent | 15 minutes | +| Normal | 60 minutes | +| Non-urgent | 4 hours | +| NOTE 1 – Intercommunication with PRMDs is not included in the calculation of the time targets.
NOTE 2 – The values are provisional and due to revision after proven experience.
NOTE 3 – For quality of service for physical delivery see clause 11. | | + +## **15.6 Error protection** + +Error protection on transmission is provided by the MHS and underlying protocols used in the provision of the EDI service. + +## **15.7 Availability of service** + +In principle the EDI service should be available continuously. The EDI-UA or the EDI-MS should be available for submission or delivery continuously (unless hold for delivery is invoked). In cases where the EDI-UA is not available for delivery continuously, an EDI-MS should be used. + +# **Annex A** + +## **Glossary of Terms** + +NOTE – The explanations given below are not necessarily definitions in the strict sense. See also the definitions in Annex B and the Glossary in Recommendation F.400/X.400 and terms provided in the other X.400-series Recommendations (especially ITU-T Rec. X.435 | ISO/IEC 10021-9). The terms have, depending on the source, varying levels of abstraction. + +## **A.1 EDI application** + +A computer process that creates and/or processes EDI messages. + +See also A.13. + +## **A.2 EDI interchange** + +"Communication between partners in the form of a structured set of messages and service segments starting with an interchange control header and ending with an interchange control trailer" (see ISO 9735). + +In the context of EDI messaging, the contents of the primary body part of an EDI message. + +## **A.3 EDI message (EDIM)** + +See definition in clause 3. + +## **A.4 EDI message store (EDI-MS)** + +See definition in 3.5 of ITU-T Rec. X.435 | ISO/IEC 10021-9. + +## **A.5 EDI messaging (EDIMG)** + +EDI messaging consists of the exchange and associated procedures of EDI messages and EDI notifications, which are information objects specified in ITU-T Rec. X.435 | ISO/IEC 10021-9. + +## **A.6 EDI messaging environment (EDIME)** + +The environment in which EDI messaging takes place can be modelled as a functional object which is referred to as the EDI messaging environment. When refined (i.e. functionally decomposed), the EDI messaging environment can be seen to be comprised of lesser objects referred to as the primary objects of EDI messaging. They include a single central object, the EDI messaging system, and numerous peripheral objects called EDI messaging users. + +## **A.7 EDI messaging service** + +A service that provides an EDI messaging user with features to assist in communicating with other EDI messaging users. EDI messaging users are in many cases computer processes. The EDI messaging service uses the capabilities of the message transfer service for sending and receiving EDI messages. Certain elements of service describing the features of the EDI messaging service are defined in Annex B, and classified in clause 14. + +## **A.8 EDI messaging system (EDIMS)** + +The EDI messaging system is the functional object by means of which users communicate with one another in EDI messaging. + +The EDI messaging system can be modelled as comprising lesser functional objects which interact with one another. These lesser objects are referred to as the secondary objects of EDI messaging. They include a single, central object, the message transfer system, and numerous peripheral objects of three kinds: EDI user agents, EDI message stores, and EDI access units. + +## **A.9 EDI messaging user (EDIMG user)** + +See definition in 3.3. + +NOTE – In the context of ITU-T Rec. X.435 | ISO/IEC 10021-9, for conciseness the term "user" is used with the meaning "EDIMG user". + +See also A.13 below. + +## **A.10 EDI notification (EDIN)** + +See definition in 3.4. + +In EDI messaging an EDIMG user can request that a recipient return an EDI notification indicating the disposition of the EDI message it received. This notification is requested by an originating EDI user agent, and is generated by a recipient EDI user agent/message store, or access unit. There are 3 possible conditions that can be requested and reported on, resulting in either the generation of a Positive Notification (PN), a Negative Notification (NN), or a Forwarded Notification (FN). The one notification serves to carry either the positive notification, the negative notification, or the forwarding notification. It is possible to forward a received EDI message unchanged and forward the obligation to respond to the notification request to the forwarded recipient, or intermediate recipients, who then shall respond to the original originator of the message. An originating UA may request to be notified if the obligation to respond to the notification request has been forwarded. In this case, the UA or MS that forwards the EDI message will send to the originating UA an EDI forwarded notification. + +In all cases, including notifications sent by UAs to whom the EDI message has been forwarded, the notifications shall contain the O/R name of the recipient that was specified by the original originator. + +The originating UA may request any combination of the several EDI notifications from any combination of the recipients to whom the EDI message is sent. If no notifications are requested by an originator, none shall be sent by the recipient(s). + +EDI notifications cannot be forwarded, and EDI notifications cannot be requested for EDI notifications. + +## **A.11 EDI message responsibility** + +See definition in 3.5. + +NOTE – EDIM responsibility is a trace-keeping means for confirming and tracking the passage of EDI messages among EDI user agents and EDI message stores. + +## **A.12 EDI security** + +The MHS security capabilities as defined in clause 15/F.400/X.400 and in clause 10 of ITU-T Rec. X.402 | ISO/IEC 10021-2, are used for EDI to provide the security features for the EDI messaging system. EDI messaging system vulnerabilities and how they are countered are outlined in Annex C. + +## **A.13 EDI user** + +See ITU-T Rec. X.435 | ISO/IEC 10021-9. + +The EDI user is an object not necessarily belonging to the EDI messaging environment. In the context of message handling, largely identical with an EDI messaging user. + +See also A.1 and A.9 above, and the Note to A.14. + +## **A.14 EDI user agent (EDI-UA)** + +See definition in 3.5 of ITU-T Rec. X.435 | ISO/IEC 10021-9. + +NOTE – An exact definition of the boundary between the EDI-UA and the EDI messaging user is beyond the scope of this Recommendation. + +## **A.15 Electronic data interchange (EDI)** + +EDI can be defined as computer to computer exchange of structured business data, such as invoices and purchase orders. In the context of the F.400-series Recommendations, it refers to the standardized way of performing the interchange in using the protocol means of ITU-T Rec. X.435 | ISO/IEC 10021-9 and the service outlined in this Recommendation. + +## **A.16 GS** + +Functional group header. + +Segment name in ANSI X12. + +## **A.17 IEA** + +Interchange trailer. + +Segment name in ANSI X12. + +## **A.18 Interchange** + +See EDI interchange in A.2. + +## **A.19 ISA** + +Interchange header. + +Segment name in ANSI X12. + +## **A.20 MHD** + +Message header. + +Segment name in UNEDIFACT. + +## **A.21 ST** + +Transaction set header. + +Segment name in ANSI X12. + +## **A.22 STX** + +Start of transmission. + +Defined in UNTDI. + +## **A.23 UNA** + +Service string advice. + +Defined in EDIFACT. + +## **A.24 UNB** + +Interchange header. + +Segment name in EDIFACT. + +## **A.25 UNG** + +Functional group header. + +Segment name in EDIFACT. + +## **A.26 UNH** + +Message header. + +Segment name in EDIFACT. + +## **A.27 UNT** + +Message trailer. + +Segment name in EDIFACT. + +## **A.28 UNZ** + +Interchange trailer. + +Segment name in EDIFACT. + +# **Annex B** + +## **Definitions of Elements of Service** + +This annex contains definitions for the elements of service unique to EDI messaging. It does not contain definitions for those elements of service of the MT service applicable to EDI messaging. Those are contained in Annex B/F.400/X.400. The abbreviation PR in the title line means that this element of service is available to be used on a per-recipient basis. The numbering for EDI elements of service uses, for ease of reference, and to distinguish them from message transfer and IPM elements of service, the abbreviation "EDI.n". + +## **B.1 application security element [EDI.1]** + +Element of service that allows the originator and the recipient to indicate in the heading of the EDI message application security information in order to support end-to-end security services. + +## **B.2 auto-acknowledgement of EDI messages [EDI.31]** + +Element of service that enables an EDI-MS-user to instruct the EDI-MS to generate a positive notification automatically for each EDIM requesting a PN which is delivered to the EDI-MS. The PN is sent when the complete EDIM has been retrieved by the EDI-MS-user or when the user indicates to the EDI-MS that he regards the message as having been retrieved. Use of this EoS implies that responsibility will be accepted whenever auto-acknowledgement occurs. + +## **B.3 auto-correlation of EDI messages [EDI.32]** + +Element of service that enables the MS-user to retrieve information, automatically generated by the MS, concerning the correlation between various related EDI messages. The following types of message may be correlated: + +- 1) the EDI messages which forwarded (or auto-forwarded) an EDI message; +- 2) the received or submitted EDI messages that obsolete an EDI message; +- 3) the received or submitted EDI messages that cross-reference an EDI message; +- 4) the received or submitted EDI messages that are related to an EDI message. + +## **B.4 auto-correlation of EDI notifications [EDI.33]** + +Element of service that enables the MS-user to retrieve information, automatically generated by the MS, concerning the EDI notifications that have been received in response to a previously submitted EDI message. Information may also be retrieved concerning EDI notifications sent by the MS-user or the MS in response to a delivered EDI message. The MS identifies each EDI notification (positive, negative, or forwarded) related to a given submitted or delivered EDI message, and for submitted messages, it also provides a summary of received EDI notifications. + +## **B.5 character set [EDI.2]** + +Element of service that allows the originator to indicate in the heading of an EDI message, the character set used in the EDI body of the message. + +## **B.6 cross reference information [EDI.3]** + +Element of service that allows the originator to indicate in the heading of an EDI message, information that can be used for cross referencing between application specified reference IDs within an EDI interchange and body parts of this or other EDI messages. + +## **B.7 EDI forwarding [EDI.4]** + +Element of service that enables an EDI-UA to forward with or without changes, and an EDI-MS to forward without changes, a received EDIM. Support of the element of service "EDIN receiver" is also required when forwarding. + +## **B.8 EDI message type(s) [EDI.5]** + +Element of service that allows the originator to indicate in the heading of an EDI message the type(s) of EDI messages contained in the EDI interchange (e.g. invoices, purchase orders, etc.). + +## **B.9 EDI notification request [EDI.6]** + +PR + +Element of service that allows the originating EDI-UA to request that it be notified of a recipient's acceptance, refusal or forwarding of EDIM responsibility, in any combination, for the message carrying this request. The originating EDI-UA conveys this request to the recipient EDI-UA/MS/AU. + +If the recipient EDI-UA/MS accepts EDIM responsibility for the message it issues a positive notification (PN) back to the originator of the message and no further notifications are issued back to this originator for this message. + +In the case where the recipient EDI-UA/MS does not accept EDIM responsibility and successfully forwards the message with content unchanged, the forwarded recipient UA/MS, or optionally any intermediate UAs/MSs, has the same obligations as the first recipient UA/MS with respect to responding to this request, and the response is due to the original originator of the message. A forwarding notification (FN) is sent back to the originator. + +If the recipient EDI-UA/MS/AU refuses EDIM responsibility for the message, or is unable to successfully forward the message, it issues a negative notification (NN) back to the originator of the message, with a reason indicated. Reasons for refusing EDIM responsibility for the message are as follows: + +- 1) the EDI interchange could not be passed over to the EDIMG user; +- 2) the EDI interchange could not be passed over to the EDIMG user within a specified time limit; +- 3) the message was discarded before processing; +- 4) the recipient's subscription was terminated after delivery but before responding; +- 5) EDI forwarding and forwarding of EDIM responsibility was attempted, but failed; +- 6) PDAU could not render the message; +- 7) security error; +- 8) unspecified local reasons; + +In the case of physical delivery access units, a PN is not meaningful, so a forwarded notification (FN) is returned to the originator instead of a PN. + +A negative notification indicates that this message shall not be made available to the EDIMG user and implies that the EDIM shall not be processed by an EDI application. + +Subject to the security policy, the capabilities of the message store may be restricted, e.g., when a secure notification is requested, the message store shall not be allowed to generate a PN. + +## **B.10 EDI standard indication [EDI.7]** + +Element of service that enables the originating EDI-UA to indicate in the heading of an EDI message the type of EDI standard that is being used in this EDI message (e.g. EDIFACT, etc). + +## **B.11 EDI message Identification [EDI.8]** + +Element of service that enables cooperating EDI-UAs to convey a globally unique identifier for each EDI message sent or received. The EDI message identifier is composed of an O/R name of the originator and an identifier that is unique with respect to that name. EDI-UAs and EDIMG users use this identifier to refer to a previously sent or received EDI message (for example, in EDI notifications). + +## **B.12 EDIM responsibility forwarding allowed indication [EDI.9] PR** + +Element of service that allows an originating EDI-UA to indicate that the EDIM responsibility for this EDI message may be forwarded on by the recipient EDI-UA. + +## **B.13 EDIN receiver [EDI.10]** + +Element of service that allows the originator, or a forwarding EDI-UA/MS, to indicate to a recipient the O/R address that requested notifications should be returned to. + +## **B.14 expiry date/time indication [EDI.11]** + +Element of service that allows the originator to indicate to the recipient the date and time after which the originator considers the EDI message to be invalid. The intent of this element of service is to state the originator's assessment of the current applicability of an EDI message. The particular action by the recipient, or by the recipient's EDI-UA, is unspecified. Possible actions might be to file or delete the EDI message after the expiry date has passed. + +## **B.15 incomplete copy indication [EDI.12]** + +Element of service that allows a forwarding EDI-UA to indicate that the forwarded EDI message is an incomplete copy of an EDI message with the same EDI message identification in that one or more body parts of the original EDI message are absent. + +## **B.16 interchange header [EDI.13]** + +Element of service that enables the originating EDI-UA to place data elements of the EDI interchange headers in corresponding fields in the EDIM. + +## **B.17 multi-part body [EDI.14]** + +Element of service that allows an originator to send to a recipient an EDI message with a body that is comprised of several parts. The nature and attributes, or type, of each body part are conveyed along with the body part. + +## **B.18 non-repudiation of content originated [EDI.15]** + +Element of service that enables an originating EDI-UA to provide a recipient EDI-UA with an irrevocable proof as to the authenticity and integrity of the content of the message as it was submitted into the MH environment. + +The corresponding proof data can be supplied in two ways depending on the security policy in force: + +- 1) Using the Non-repudiation of Origin Security service applied to the original message or, +- 2) By means of a notarization mechanism. + +NOTE – Use of a notarization mechanism is not reflected in protocol elements, but is subject to bilateral agreement. + +## **B.19 non-repudiation of content received [EDI.16]** + +PR + +Element of service that enables an originating EDI-UA to get from a recipient EDI-UA an irrevocable proof that the original subject message content was received by the recipient EDI-UA and EDIM responsibility was accepted, forwarded or refused. This service provides irrevocable proof as to the integrity of the content received and irrevocable proof as to the authenticity of the recipient of the message. It will protect against any attempt by the recipient(s) to subsequently deny having received the message content. This service is stronger than the "Proof of Content Received" service. + +The corresponding proof data can be supplied in two ways depending on the security policy in force: + +- 1) By returning a "non-repudiation of origin" of the "EDI notification" which incorporates the following: + - the originator's "non-repudiation of origin" arguments (if present), + - the complete original message content, if the originator's "non-repudiation of origin" arguments are not present. +- 2) By means of a notarization mechanism. + +NOTE – Use of a notarization mechanism is not reflected in protocol elements, but is subject to bilateral agreement. + +## **B.20 non-repudiation of content received request [EDI.17]** + +PR + +Element of service that enables the originating EDI-UA to request the recipient EDI-UA to provide it with an irrevocable proof of the received message content by means of an EDI notification. + +NOTE – This element of service requires the "EDI notification request" also to be present. + +## **B.21 non-repudiation of EDI notification [EDI.18]** + +PR + +Element of service that provides the originator of a message with irrevocable proof that the subject message was received by the recipient EDI-UA and EDIM responsibility was accepted, forwarded or refused. + +This shall protect against any attempt by the recipient EDI-UA to deny subsequently that the message was received and that the EDIM responsibility for the message has been accepted as indicated. This element of service provides the originator with irrevocable proof of the "proof of EDI notification". + +Such a proof may be provided by means of the "Non-repudiation of Origin" Security service, currently defined in 10.2.5.1 of ITU-T Rec. X.402 | ISO/IEC 10021-2, applied to the notification. + +This service is stronger than the "Proof of EDI Notification" service. + +## **B.22 non-repudiation of EDI notification request [EDI.19]** PR + +Element of service, used in conjunction with "EDI notification request", that enables the originating EDI-UA to request the responding EDI-UA to provide it with irrevocable proof of the origin of the notification. + +NOTE – This element of service supersedes the "Proof of EDI notification request" and assumes that "EDI Notification Request" is already present. + +## **B.23 obsoleting indication [EDI.20]** + +Element of service that allows the originator to indicate to the recipient that one or more EDI messages previously sent by the originator are obsolete. The EDI message that carries this indication supersedes the obsolete EDI message(s). + +The action to be taken by the recipient or the recipient's EDI-UA is a local matter. The intent, however, is to allow the EDI-UA or the recipient to, for example, remove or file an obsolete message(s). + +## **B.24 originator indication [EDI.21]** + +Element of service that allows the identity of the originator to be conveyed to the recipient. + +## **B.25 proof of content received [EDI.22]** PR + +Element of service that allows an originating EDI-UA to get from a recipient EDI-UA proof that the original subject message content was received by the recipient EDI-UA and EDIM responsibility was accepted, forwarded or refused. + +The corresponding proof is obtained by returning a proof of origin of the EDI notification which incorporates the originator's message origin authentication and/or content integrity arguments, if present, or the complete original message content otherwise. + +## **B.26 proof of content received request [EDI.23]** PR + +Element of service that enables the originating EDI-UA to request the recipient EDI-UA to provide it with proof of the received message content by means of an EDI notification. + +NOTE – This element of service requires the "EDI notification request" to also be present. + +## **B.27 proof of EDI notification [EDI.24]** PR + +Element of service that allows the originator of a message to obtain the means to corroborate that the subject message was received by the recipient EDI-UA and EDIM responsibility was accepted, forwarded or refused. Such a corroboration is provided by means of the MTS user-to-MTS user "Message Origin Authentication" Security service, currently defined in 10.2.1.1.1 of ITU-T Rec. X.402 | ISO/IEC 10021-2, applied to the EDI notification. + +## **B.28 proof of EDI notification request [EDI.25]** PR + +Element of service, used in conjunction with "EDI notification request", that enables the originating EDI-UA to request the responding EDI-UA to provide it with a corroboration of the source of the EDI notification. + +NOTE – This element of service assumes that "EDI notification request" is already present. + +## **B.29 recipient indication [EDI.26]** + +PR + +Element of service that allows the originator to provide the names of one or more EDIMG users, or DLs, who are intended recipients of the EDI message. In addition it is possible to specify an action request qualifier for each recipient, such as; + +- 1) for action; +- 2) copy; +- 3) other, as defined bilaterally. + +NOTE – The qualifier represents intent on the part of the originator with respect to the EDIM, however the recipient is not necessarily bound by this intent. + +## **B.30 related message(s) [EDI.27]** + +This element of service that allows the originator to associate with the EDI message being sent, the globally unique identifiers of one or more other messages which share the same identification space (e.g. IP-messages). This enables the recipient's EDI-UA, for example, to retrieve from storage a copy of the referenced messages. + +## **B.31 services indication [EDI.28]** + +Element of service that allows the originator to indicate in the heading of the EDI message various service requests to service suppliers that have bilateral meaning outside this Recommendation. + +## **B.32 stored EDI message auto-forward [EDI.29]** + +Element of service that allows a user of an EDI-MS to have the message store automatically perform EDI forwarding, with or without accepting EDIM responsibility. The user of the EDI-MS may establish criteria for selecting EDIMs through use of the element of service "MS register". The complete EDIM, as received from the originator, is forwarded unchanged, and if requested, an appropriate EDIN is generated by the EDI-MS. EDIM responsibility forwarding is limited to only one recipient. Support of the element of service "EDIN receiver" is also required when forwarding. + +Subject to the requirements of the security policy in force, the capabilities of the message store may be restricted, e.g., when a secure notification is requested, the message store shall not be allowed to generate a PN. + +## **B.33 submission of EDI messages incorporating stored messages [EDI.34]** + +Element of service that enables the MS-user to instruct the EDI-MS to incorporate parts of one or more stored messages as body parts of a submitted EDI message. The submitted EDI message may also contain body parts supplied in the submission from the EDI-MS-user. The primary body part shall contain either an EDI interchange or a forwarded EDIM. + +The stored message which is the source of a body part may be a delivered, submitted, or draft message. Individual body parts or the whole content of a stored message may be incorporated. The EDI-MS may optionally support the forwarding of body parts from messages which are not EDI messages. + +## **B.34 typed body [EDI.30]** + +Element of service that permits the nature and characteristics of the body of an EDI message to be conveyed along with the body. Permissible body part types are EDI body, forwarded EDIM body, and externally defined body parts. + +# Annex C + +## Security Overview + +## C.1 Introduction + +This annex details the vulnerabilities identified within an EDIME and the resulting security services required to counter those vulnerabilities. + +This annex is based on the assumption that an EDIME may use the secure messaging services as defined in Recommendation F.400/X.400. However, where vulnerabilities are not adequately covered by the existing MHS security services, provision has been made in ITU-T Rec. X.435 | ISO/IEC 10021-9 for additional security services in the EDIME. + +Some of the security services defined for the EDIME are of a generic message handling nature, others are specific to the EDIME. The security services defined for the EDIME are specific to EDIMG and are therefore fully defined in ITU-T Rec. X.435 | ISO/IEC 10021-9. + +## C.2 Vulnerabilities + +In most of the areas identified below, there will also be further vulnerabilities and corresponding service considerations at the level of the EDI applications (i.e. EDIMG users). The security model reflected in this paper assumes that such considerations are outside the scope of this Recommendation. The EDIMG security model assumes that the EDIMG user provides adequate security and trusted functionality in the operation of EDI applications sufficient to meet the user's security policy. + +NOTE – In practice this may necessitate co-location of the EDI application and the EDI-UA unless a suitably secure environment is established which includes both components. + +The following description of vulnerabilities is based on the threat definitions in Annex D of ITU-T Rec. X.402 | ISO/IEC 10021-2. In addition, it has been considered necessary to examine message loss independently of message sequencing and modification of information, and to take account of further vulnerabilities for EDIMG which are not currently identified in ITU-T Rec. X.402 | ISO/IEC 10021-2. + +An important aspect of the EDI environment which is not recognised within the ITU-T Rec. X.402 | ISO/IEC 10021-2 security model is the concept of EDIM responsibility for messages at each stage of the message path through the MHS environment. + +In an EDI context, the increased possibility of a number of service providers offering commercial services may require that the forwarding of EDIM responsibility be clearly identified and assured to provide further protection, not only to end users but also to such service providers. + +It is therefore necessary to establish the concept of EDIM responsibility domains, which may involve additional consideration of legal issues. One possible division of the EDIME into EDIM responsibility domains is as follows: + +- EDIMG user environment plus the EDI-UA; +- MTS management domain; +- EDI message store (if not co-located with either of the above). + +### C.2.1 Masquerade + +As defined in Annex D of ITU-T Rec. X.402 | ISO/IEC 10021-2. + +### C.2.2 Message sequencing + +As defined in Annex D of ITU-T Rec. X.402 | ISO/IEC 10021-2. + +Users should not assume that EDIMs shall be delivered in correct sequence. EDI applications should be able to recover from duplication and out-of-sequence messages, provided that MHS offers protection against the modification of information while messages are within the MHS environment. + +### **C.2.3 Message loss** + +Vulnerability to message loss is considered critical in the EDIMG environment. + +Two types of message loss are distinguished: + +- catastrophic failure of an EDI-UA, EDI-MS or MTA, +- loss of individual message(s). + +EDIME users and service providers may need to consider more carefully issues concerning transfer of messages between EDIM responsibility domains: + +- from the originating EDI-UA user domain; +- between relaying domains; +- to the recipient EDI-UA user domain. + +### **C.2.4 Modification of information** + +As defined in Annex D of ITU-T Rec. X.402 | ISO/IEC 10021-2. + +### **C.2.5 Denial of service** + +As defined in Annex D of ITU-T Rec. X.402 | ISO/IEC 10021-2. + +### **C.2.6 Repudiation** + +As defined in Annex D of ITU-T Rec. X.402 | ISO/IEC 10021-2. + +Furthermore repudiation vulnerability in the EDIME environment is considered to be critical. Such vulnerability may be increased by use of certain MHS services (e.g., auto-forwarding, redirection). + +### **C.2.7 Leakage of information** + +As defined in Annex D of ITU-T Rec. X.402 | ISO/IEC 10021-2. + +### **C.2.8 Manipulation of information by EDIMG user** + +The EDI community has additionally identified a further vulnerability where the integrity of a message content is altered subsequent to EDI interchange (i.e. by either or both of the originating EDI-UA and recipient EDI-UA). This vulnerability includes manipulation of message content in the originator's local store after non-repudiation of submission and/or manipulation of message content in the recipient's store after non-repudiation of delivery. + +### **C.2.9 Other vulnerabilities** + +Other vulnerabilities as defined in ITU-T Rec. X.402 | ISO/IEC 10021-2 are considered important such as: + +- misrouting; +- misdelivery (especially important in the context of redirection); +- insider threats; +- receipt of data that the EDI application is not prepared to accept. + +## **C.3 Vulnerabilities countered** + +Clause 10 of ITU-T Rec. X.402 | ISO/IEC 10021-2 provides an abstract security model for Message Transfer. The security model provides a framework for describing security services that counter potential vulnerabilities within the MTS and between MTS-User to MTS-User. EDIMG vulnerabilities may also be countered by security services which are outside the existing model in ITU-T Rec. X.402 | ISO/IEC 10021-2. The following text describes how the EDIM vulnerabilities are countered using ITU-T Rec. X.402 | ISO/IEC 10021-2 security services, enhanced security services defined in ITU-T Rec. X.435 | ISO/IEC 10021-9 and pervasive mechanisms defined in this Recommendation. + +### **C.3.1 Masquerade** + +The existing MHS security services which counter this vulnerability are: + +- message origin authentication; +- secure access management; +- security labelling; +- proof of delivery; +- proof of submission. + +Since an EDI-UA/MS is deemed in the MHS architecture as belonging to one user, it is not considered appropriate to provide selective access control for the various operations that may be performed on a EDI-MS. However, there is a requirement for security audit trail to record the actions of the EDIMG user. + +In this Recommendation such security audit trails are expected to be implemented as pervasive mechanisms (the term pervasive mechanism is defined in ISO 7498-2). Protocols to support audit capability may be the subject of future standardization. + +### **C.3.2 Message sequencing** + +The existing MHS Security service which counters this vulnerability is: + +- message sequence integrity. + +This security service has limited effect as it is based on the provision of an integer by the originating EDI-UA with no assurance as to uniqueness or consecutiveness. + +It is considered that the MHS environment should not be required to ensure message sequence integrity, but should support detection of sequence integrity failure (by additional provision of audit/logging facilities and/or the provision of third party notar services). In this Recommendation it is considered the responsibility of the EDIMG user to recover from sequence errors and message duplication. + +### **C.3.3 Message loss** + +Message loss could occur potentially over any peer-to-peer communications link (e.g. by deliberate malicious act), or by the failure or incorrect behaviour (whether by malicious intent or otherwise) of any MHS component (EDI-UA, EDI-MS, MTA). The following categories of message loss are distinguished: + +- catastrophic message loss (i.e. failure of a component); +- loss of individual messages in the EDI-MS – whether malicious or accidental; +- MTS message loss. + +#### **C.3.3.1 Catastrophic failure** + +Failure of the EDI-UA is outside the scope of this Recommendation. + +Failure of the EDI-MS is potentially catastrophic and desirably needs some protection, at least in terms of detection. This should be provided by an offline archive to hold all submitted and delivered messages. In this Recommendation detection and recovery from message loss using such archive mechanisms is a local matter. + +Failure of any component in the MTS may similarly be catastrophic and can again be protected by offline archive of messages. As for the message store, detection and recovery from message loss using such archive mechanisms in the MTS is a local matter and outside the scope of this Recommendation. + +#### **C.3.3.2 EDI-MS specific message loss** + +Loss of individual messages in the message store, whether malicious or accidental, shall require the provision of a secure audit trail to enable detection of such loss. Such a service may need to be provided to the EDIMG user and to EDI-MS management. In this Recommendation, secure EDI-MS audit trail could be realized as a pervasive mechanism and is a local issue. Protocol to support an audit trail may be the subject of future standardization. + +#### **C.3.3.3 MTS specific message loss** + +Loss of individual messages in the MTS (whether malicious or accidental) shall also require the provision of a secure audit trail to enable detection of such loss. Such a mechanism would need to be provided on a per-MTA and a per-MD basis depending on security policy in force. A secure MTA/MTS audit trail could be realized as a pervasive mechanism and is a local issue. The protocol to support an audit trail may be the subject of future standardization. + +#### **C.3.3.4 End-to-end message loss** + +The following description assumes that the functionality of the EDI-UA (including any associated components to meet such functionality e.g. encryption devices) is trusted. + +The existing "Message Sequence Integrity" service does not guarantee detection of message loss, since it relies on the provision of an integer value by the originating EDI-UA. In practice, effective operation of this service may be achieved with a common code of practice between EDIMG users which is outside the scope of this Recommendation. + +As a result, MHS services which may provide an indication of message loss are confined to services offered to the originating EDIMG user. Whereas, the existing "Proof of Submission and Delivery" services provide some degree of confidence that the message has not been lost they do not operate end-to-end. In particular they do not take account of the scenario where the recipient EDI-UA and EDI-MS are not co-located. There is therefore a requirement for a Proof of Receipt (i.e. by the recipient EDI-UA) service. This capability is realized by the user requesting an EDI notification which may be secured. The EDI notification indicating the status of EDIM responsibility as accepted, forwarded or refused includes elements which associates the notification with the subject message. + +In an EDIMG environment proof of receipt may therefore be provided by signing the EDI Notification service using the existing MTS security elements. In particular the EDI-UA to EDI-UA Security service of "message origin authentication" may be used to sign the EDI notification on submission of the EDI notification to the MTS. In this Recommendation the requirement for proof of receipt may be implemented by a trusted form of EDI notification in the EDIMG environment. + +NOTE – This service is called "proof of EDI notification" and/or "non-repudiation of EDI notification" in EDIMG depending on the strength of the mechanism provided. + +The MTS mechanism used on message submission to provide this service is defined as the MTS submission abstract operation in 8.2.1.1.1.28 of ITU-T Rec. X.411 | ISO/IEC 10021-4 "Content-integrity-check". In this instance the message content is the EDI notification. Proof of association between the subject message and replying EDI notification is provided by subject message EDI identifier and if included in the subject message the message content-integrity-check argument. + +### **C.3.4 Modification of information** + +The existing MHS security services which counter this vulnerability are: + +- connection integrity; +- content integrity. + +These security services provide sufficient protection against modification of message content. It is also noted that use of double enveloping (i.e. with encrypted checksum on outer envelope) may provide additional protection. + +NOTE – EDI-UAs are trusted entities in terms of content integrity. + +### **C.3.5 Denial of service** + +This is a very important vulnerability for EDIMG users, but is outside the scope of this Recommendation. + +### **C.3.6 Repudiation** + +Services which offer protection against repudiation in the EDIMG environment are fundamentally concerned with formalizing the forwarding of EDIM responsibility. + +The security services as defined in ITU-T Rec. X.402 | ISO/IEC 10021-2 are: + +- non-repudiation of origin; +- non-repudiation of submission ; +- non-repudiation of delivery. + +These security services only cover some areas of transfer between EDIM responsibility domains, which may be of significance in an EDIMG environment (as illustrated in Figure C.1). Areas which are not covered by security services provided in 1988 for message handling include: + +- between EDIMG user domains (i.e. end-to-end); +- between MTS management domains; +- between an EDI message store and a recipient EDI-UA. + +Therefore services and/or pervasive mechanisms defined in this Recommendation cover the above deficiencies: + +- non-repudiation/proof of transfer; +- non-repudiation/proof of retrieval; +- non-repudiation/proof of EDI notification; +- non-repudiation/proof of content. + +![Diagram illustrating EDIM responsibility transfer between an originator and a recipient across multiple domains (A, B, C).](cf8bd014a50b7c69435e804f67f9617f_img.jpg) + +The diagram illustrates the flow of security proofs in an EDIMG environment across three domains: Domain A, Domain B, and Domain C. At the top, two large boxes represent the 'EDIM originator Trusted functionality' and the 'EDIM recipient Trusted functionality'. Each contains an 'EDIMG user' and an 'EDI-UA' component. Below these, each side has an 'EDI-MS' (EDI Message Store) and an 'MTA' (Message Transfer Agent). Domain A contains the originator's MTA, Domain B is a central MTA, and Domain C contains the recipient's MTA. Arrows indicate various security proofs: 'Proof of submission' (upward on the left), 'Proof of transfer' (between MTAs in Domain A, B, and C), 'Proof of delivery' (from Domain C's MTA to the originator's EDI-UA), 'Proof of EDI notification' (from the recipient's EDI-UA to the originator's EDI-UA), 'Proof of content' (between EDIMG users), and 'Proof of retrieval' (downward on the right). A legend at the bottom defines the line styles: a dashed line for 'EDIM responsibility transfer boundary', a solid line with an arrowhead for 'Security service', and a dotted line with an arrowhead for 'Provided by pervasive mechanisms as a local issue'. The code 'T0101380-90/d13' is in the bottom right. + +Diagram illustrating EDIM responsibility transfer between an originator and a recipient across multiple domains (A, B, C). + +**Figure C.1/F.435 – EDIM responsibility transfer** + +"Non-repudiation/proof of transfer" counters the vulnerability of repudiation of responsibility between MTA and/or management domains. EDIMG environments may provide such a service using additional pervasive mechanisms, such as security logs and archives within MTA and/or MTS boundaries. Such pervasive mechanisms provide a "secure MT audit trail" to record the message details and trace information. + +"Non-repudiation/proof of retrieval" counters the vulnerability of repudiation of responsibility of a message between a UA and an MS. EDIMG environments may provide such a service using additional pervasive mechanisms, such as security logs and archives within EDI-MSs. Such pervasive mechanisms provide a "secure EDI-MS audit trail" to record EDIMG user actions in the EDI message store. + +"Non-repudiation/proof of EDI notification" counters the vulnerability of repudiation of an EDI notification EDI-UA to EDI-UA. This service is specific to EDIMG and a complete solution is included in this Recommendation. This vulnerability may be especially relevant in the case of EDI forwarding, redirection, etc, in addition to the scenario of delivery to an untrusted EDI message store. + +Two mechanisms have been defined for non-repudiation of EDI notifications, the first uses the trusted EDI notification as described above, the second using an external notary systems. Only the trusted EDI notification was fully defined in this Recommendation. External notary systems may be the subject of future standardization. + +"Non-repudiation/proof of content" counters the vulnerability of manipulation of information by the EDIMG user after the message has been received by the EDI-UA. Although such vulnerability is outside the MHS environment, the MHS environment may provide assistance in terms of trusted return of content and notarization services. There are several ways this requirement may be supported, using the secure messaging environment based on the security services provided in 1988. + +Firstly non-repudiation of content by the originating EDI-UA may be provided by the existing "Non-repudiation of Origin" Security service. + +Secondly non-repudiation of content by the recipient EDI-UA may be provided by returning the subject content within the EDI notification and submitting the EDI notification to the MTS using the "Non-repudiation of Origin" Security services. + +Thirdly by notarization services, such services may be achieved by forwarding messages via a mutually trusted third-party notary (i.e. using existing MHS security services). + +All three approaches would thus require no modification to the secure messaging environment based on the existing MHS Recommendations. + +NOTE – Non-repudiation services (which may imply the involvement of a third party) are considered stronger than "proof-of" services. + +### **C.3.7 Leakage of information** + +The existing MHS security services which counter this vulnerability are: + +- connection confidentiality; +- content confidentiality; +- secure access management; +- message flow confidentiality. + +These security services provide sufficient protection against leakage of message content. It is also noted that use of double enveloping could provide some protection against traffic analysis. Traffic padding is outside the scope of this area of work. + +NOTE – UAs are trusted entities in terms of content and message flow confidentiality. + +### **C.3.8 Manipulation of information by EDIMG user** + +Manipulation of information by the EDIMG user may be countered by use of the "Non-repudiation of Content" Security service. + +### **C.3.9 Other vulnerabilities** + +The use of "security access management" and "security labelling" to counter all other vulnerabilities is also applicable in the EDIMG environment. In addition, there is a requirement for auditing and accountability which is likely to require at least a "secure audit trail", this may be provided by a pervasive mechanism as a local matter. + +### **C.3.10 Other EDI application vulnerabilities** + +Within the EDIMG environment the EDI application itself may be vulnerable to security threats. To counter these vulnerabilities the EDI application may wish to generate its own security services and mechanisms (such as, signatures from EDI application to EDI application). These EDI application security services are conveyed in EDIMS security fields as purely information conveying elements of services within the EDIMG environment and may consequently be used for several end to end services including message recovery and non-repudiation. It is a local issue to determine how the EDI application security services are used. + +### **C.3.11 Summary** + +This annex identifies EDIMG vulnerabilities and the security services necessary to counter those vulnerabilities using MHS specification of 1988, then specifies the corresponding security elements required. + +EDIMG may provide additional pervasive mechanisms as follows: + +- secure EDI-MS audit trail, +- secure MT audit trail; +- EDI-MS archive; +- MD archive; +- security of MTA management and routing information. + +This Recommendation currently allows the use of both standard symmetric and standard asymmetric tokens. The use of trusted notary systems may be the subject of future standardization. + +## **C.4 Additional pervasive mechanisms** + +### **C.4.1 Secure EDI-MS audit trail** + +This facility would monitor and record EDI-UA actions on the message store. It would also provide support for "proof of retrieval". + +It is strongly recommended that "secure EDI-MS audit trail" should be controlled via a secure link or other secure local means to protect against masquerade. In this Recommendation "secure EDI-MS audit trail" may only be realized as a pervasive mechanism. The pervasive mechanisms mentioned may be the subject of future standardization. + +### **C.4.2 Secure MT audit trail** + +This facility would monitor and record all MTA actions. It would also provide additional support for: "proof of submission", "proof of transfer", "proof of delivery", security of the administration of the MTA. + +In this Recommendation secure MT audit trail may be realized as a pervasive mechanism. + +### **C.4.3 EDI-MS archive** + +This mechanism is potentially useful for providing recovery from MS failure i.e. by providing a secure offline archive of all submitted and delivered messages. Detection and recovery from message loss using such archive mechanism is a local matter. + +### **C.4.4 MT Archive** + +This mechanism is potentially useful for providing recovery from MTA failure i.e. by providing a secure offline archive of all messages. Detection and recovery from message loss using such archive mechanism is a local matter. + +# Annex D + +## EDI naming, addressing, and use of Directory + +This annex describes the use of the Directory by the EDI messaging service. While the Directory may be used by any EDI user, this annex is limited to the use of the Directory by an EDIMG user. + +## D.1 Introduction + +This annex describes the functions that an EDIMG user may obtain from the Directory, if the Directory is available to the EDIMG user. If the Directory is not available, the functions described in this annex may be performed as a local matter. + +This annex covers the following topics: + +- a) EDI naming in D.2; +- b) suggested DIT structure for EDI in D.3; +- c) name resolution in D.4; +- d) authentication in D.5; +- e) capabilities assessment in D.6. + +## D.2 EDI naming + +EDI users (trading partners) identify each other by a "name" which is essentially an arbitrary alphanumeric string. In this annex, an EDI name is defined to be such an alphanumeric string. EDI standards authorities (for example, EDIFACT and ANSI X12) define specific instances of EDI names. The EDI name is normally unique within a particular EDI community, but may not be globally. + +The EDI communities may be organized: + +- a) by industry group (e.g. CEFIC, EDIFICE); +- b) as a private trading group of a large corporation; +- c) around a third-party EDI service provider. + +The EDI names used by any of the above community types are one of the following forms: + +- d) a formalized name issued by an internationally recognized naming authority (e.g. DUNS, EAN, SIRET) which is globally unique; +- e) a formalized name issued by a multi-national company; the name is unique within the company's trading community and the multi-national company acts as a naming authority within this community; +- f) a free form name assigned by the trading partners themselves, subject only to a uniqueness check by the organizer or operator of the community, acting in the role of a naming authority. + +NOTE – All these name forms exist today, and it will take some time for EDIMG users to migrate to globally unique naming. + +EDI standards allow the use of a qualifier together with the EDI name. The qualifier identifies the naming authority that assigned or endorsed the alphanumeric string. Globally unique EDI naming is achieved by using EDI names with the appropriate qualifier code. + +There is no geographic element in an EDI name, such as country of operation. + +The EDIMG users send EDI interchanges with as little addressing information as possible. The EDI name is a static entity which exists for a long period, unlike an address which may change from time to time. + +## D.3 Suggested DIT structure for EDI + +Annex B of ITU-T Rec. X.521 | ISO/IEC 9594-7 suggests some common naming practices and DIT structures in which locality, country and root can be immediately superior to entries of the object class organization. + +When organization is immediately subordinate to root it denotes an international organization. The community types identified in D.2 operate internationally, and therefore, the majority of these community types may be classified as international organizations. + +A Directory structure is suggested in which each community of EDI names is grouped under the organization that serves as naming authority for that community (company, industry group, service provider). In this case the entry associated with each EDI name is an alias entry; actual entry for the EDIMG user is elsewhere in the DIT, as described in D.4. + +Figure D.1 illustrates a DIT structure which accommodates the requirements of the EDI community. A new generic object class, EDI user, is created. The attributes in its entry identify the name of the EDIMG user, and to the extent that they are present, the capabilities of the EDIMG user and the attributes of a message handling user as defined in ITU-T Rec. X.402 | ISO/IEC 10021-2. + +NOTE – Figure D.1 illustrates the directory information tree (DIT) that is implied by the object class EDI user as defined in Annex J of ITU-T Rec. X.435 | ISO/IEC 10021-9. + +![Figure D.1/F.435 – DIT structure for EDI requirements. The diagram shows a hierarchical tree structure starting from 'Root' at the top. 'Root' branches to 'Locality', 'Country', and 'Organization'. 'Locality' branches to 'Residential person' and 'Group of names'. 'Country' branches to 'Organization'. 'Organization' branches to 'Device', 'Organizational unit', 'EDI user', 'Organizational role', and 'Organizational person'. 'Group of names' branches to 'Application process'. 'Application process' branches to 'Application entity'. A legend box at the bottom right shows an arrow pointing from 'From immediate superior' to 'Object class' and another arrow pointing from 'Object class' to 'To immediate subordinate'.](7fe5741e83bc9702d1b1d7585ddf66bd_img.jpg) + +``` + +graph TD + Root((Root)) --> Locality((Locality)) + Root --> Country((Country)) + Root --> Organization((Organization)) + Locality --> ResidentialPerson((Residential person)) + Locality --> GroupOfNames((Group of names)) + Country --> Organization + Organization --> Device((Device)) + Organization --> OrganizationalUnit((Organizational unit)) + Organization --> EDIUser((EDI user)) + Organization --> OrganizationalRole((Organizational role)) + Organization --> OrganizationalPerson((Organizational person)) + GroupOfNames --> ApplicationProcess((Application process)) + ApplicationProcess --> ApplicationEntity((Application entity)) + +``` + +Figure D.1/F.435 – DIT structure for EDI requirements. The diagram shows a hierarchical tree structure starting from 'Root' at the top. 'Root' branches to 'Locality', 'Country', and 'Organization'. 'Locality' branches to 'Residential person' and 'Group of names'. 'Country' branches to 'Organization'. 'Organization' branches to 'Device', 'Organizational unit', 'EDI user', 'Organizational role', and 'Organizational person'. 'Group of names' branches to 'Application process'. 'Application process' branches to 'Application entity'. A legend box at the bottom right shows an arrow pointing from 'From immediate superior' to 'Object class' and another arrow pointing from 'Object class' to 'To immediate subordinate'. + +T0101390-90/d14 + +Figure D.1/F.435 – DIT structure for EDI requirements + +## D.4 Name resolution + +An EDIMG user may use the Directory to obtain the message handling O/R address of the EDI-UA corresponding to another EDIMG user. This process is defined as "name resolution" in clause 22 of ITU-T Rec. X.402 | ISO/IEC 10021-2. + +To obtain the message handling O/R address of an EDI-UA that corresponds to an EDIMG user whose Directory name it possesses, an EDIMG user presents the Directory name to the Directory, and obtains from the Directory the attribute message handling O/R address. + +To do this successfully, the EDIMG user shall authenticate itself to the Directory and have access rights to the information requested. + +The Directory name may contain a relative distinguished name that is an EDI name. The EDI name can be considered a "user friendly name", as defined in E.1 of ITU-T Rec. X.501 | ISO/IEC 9594-2. + +Figure D.2, which is similar to Figure E.1 of ITU-T Rec. X.501 | ISO/IEC 9594-2, illustrates an example of an EDI name. Whenever an EDI-UA requires access to an EDIMG user Directory entry, using the services of a DUA, it shall construct the distinguished name of the entry. The distinguished name shall contain the EDI name, organization name of the organization or naming authority that issued the EDI name, and if required, the country of that organization. How the EDIMG user obtains the EDI name is a local matter. The EDIMG user shall pass the EDI name to the EDI-UA, who shall pass it to the DUA. When the EDI name is globally unique, the organization name, and if required, country, may be derived from the qualifier code with the EDI name. When the EDI name is not globally unique, the EDIMG user or EDI-UA shall obtain the organization name and shall obtain the country, if required, via other means. + +![Figure D.2/F.435 – An aliasing example. A hierarchical tree diagram showing two paths to the same directory entry. The root branches into 'O = Multinational' and 'C = US'. 'O = Multinational' branches into 'EU = Invoicing'. 'C = US' branches into 'O = Telecom EDI', which branches into 'EU = Invoicing'. Both 'EU = Invoicing' nodes point to the same entry box containing 'EDI capabilities' and 'MHS O/R address' with attributes 'C US', 'A Some ADMD', 'P Telco', 'O Telecom EDI', and 'CN Invoices'. A legend on the left defines the attribute types: A ADMD, C Country, CN Common name, EU EDI user, O Organization, P PRMD. A code 'T0101400-90/d15' is at the bottom right.](8d66c9c295023a1380f9986d3663bb1e_img.jpg) + +The diagram illustrates an aliasing example in a directory tree. At the top is a 'Root' node. Two lines descend from 'Root': one to a box labeled 'O = Multinational' and another to a box labeled 'C = US'. From 'O = Multinational', a line descends to a box labeled 'EU = Invoicing'. From 'C = US', a line descends to a box labeled 'O = Telecom EDI'. From 'O = Telecom EDI', a line descends to a box labeled 'EU = Invoicing'. Both 'EU = Invoicing' boxes have arrows pointing to the same entry box at the bottom. This entry box contains two bullet points: 'EDI capabilities' and 'MHS O/R address'. Below these, the attributes are listed: 'C US', 'A Some ADMD', 'P Telco', 'O Telecom EDI', and 'CN Invoices'. To the left of the entry box is a legend defining the attribute types: 'A ADMD', 'C Country', 'CN Common name', 'EU EDI user', 'O Organization', and 'P PRMD'. At the bottom right of the diagram is the code 'T0101400-90/d15'. + +Figure D.2/F.435 – An aliasing example. A hierarchical tree diagram showing two paths to the same directory entry. The root branches into 'O = Multinational' and 'C = US'. 'O = Multinational' branches into 'EU = Invoicing'. 'C = US' branches into 'O = Telecom EDI', which branches into 'EU = Invoicing'. Both 'EU = Invoicing' nodes point to the same entry box containing 'EDI capabilities' and 'MHS O/R address' with attributes 'C US', 'A Some ADMD', 'P Telco', 'O Telecom EDI', and 'CN Invoices'. A legend on the left defines the attribute types: A ADMD, C Country, CN Common name, EU EDI user, O Organization, P PRMD. A code 'T0101400-90/d15' is at the bottom right. + +Figure D.2/F.435 – An aliasing example + +An alias name may be used to direct the search for a particular entry, for example, to enter the Directory with an organization name and the EDI name in order to extract a message handling O/R address. Figure D.2 shows an EDI-UA identified with the name (O = Multinational, EU = Invoicing). It is also identified by (C = US, O = Telecom EDI, EU = Invoicing). Both EDIMG user names resolve to the same message handling O/R address (C = US, A = Some ADMD, P = Telco, O = Telecom EDI, CN = Invoices). + +Figure D.3 illustrates that if the organization is not an international organization then the EDIMG user can still be accessed using country as a component of its name. + +## D.5 Authentication + +An EDIMG user may accomplish authentication using information stored in the Directory. This usage is as defined in Recommendations F.400/X.400 and ITU-T Rec. X.509 | ISO/IEC 9594-8. + +![A hierarchical tree diagram illustrating a country-oriented aliasing example. The root node is labeled 'Root'. It branches into a node labeled 'C = US'. This node branches into two nodes: 'O = Duns' on the left and 'O = Telecom EDI' on the right. The 'O = Duns' node branches into a node labeled 'EU = 1234599'. The 'O = Telecom EDI' node branches into two nodes: 'OU = Sales' on the left and 'EU = Invoicing' on the right. Below the 'EU = 1234599' node is a box containing the following attributes: A ADMD, C Country, CN Common name, EU EDI user, O Organization, OU Organizational unit, P PRMD. Below the 'EU = Invoicing' node is a box containing the following attributes: EDI capabilities, MHS O/R address, C US, A Some ADMD, P Telco, O Telecom EDI, CN Invoices. A curved arrow points from the 'EU = 1234599' node to the 'EU = Invoicing' node. The text 'T0101410-90/d16' is located at the bottom right of the diagram.](5ee1bbbf85b473f78af9ec8368a4159a_img.jpg) + +A hierarchical tree diagram illustrating a country-oriented aliasing example. The root node is labeled 'Root'. It branches into a node labeled 'C = US'. This node branches into two nodes: 'O = Duns' on the left and 'O = Telecom EDI' on the right. The 'O = Duns' node branches into a node labeled 'EU = 1234599'. The 'O = Telecom EDI' node branches into two nodes: 'OU = Sales' on the left and 'EU = Invoicing' on the right. Below the 'EU = 1234599' node is a box containing the following attributes: A ADMD, C Country, CN Common name, EU EDI user, O Organization, OU Organizational unit, P PRMD. Below the 'EU = Invoicing' node is a box containing the following attributes: EDI capabilities, MHS O/R address, C US, A Some ADMD, P Telco, O Telecom EDI, CN Invoices. A curved arrow points from the 'EU = 1234599' node to the 'EU = Invoicing' node. The text 'T0101410-90/d16' is located at the bottom right of the diagram. + +Figure D.3/F.435 – A country oriented aliasing example + +## D.6 Capabilities assessment + +An EDIMG user may assess the capabilities of another EDIMG user via the Directory. Capability assessment allows the EDIMG user to determine, for example, whether the other EDIMG user can process a specific version or release of an EDI document. + +The following Directory attributes represent EDI capabilities in the EDI messaging service: + +- a) standard; +- b) standard version; +- c) standard syntax identifier; +- d) document type; +- e) document version; +- f) document release; +- g) controlling agency; +- h) association assigned code; +- i) EDI character set; + +To assess a particular capability of an EDIMG user whose Directory name it possesses, the EDIMG user shall present that name to the Directory and request from the Directory the attribute EDI capabilities. + +To do this successfully, the EDIMG user shall first authenticate itself to the Directory, and have access rights to the information requested. + +# Annex E + +## Cross Referencing Overview + +EDIMG users have a need to reference other body parts from within the EDI interchange. For example, an EDI purchase order may refer to a drawing contained in another body part, either within that EDIM or within another message. The element of service "cross reference information" can be used to satisfy this requirement. This element of service corresponds to an EDIM heading field specifically designed to contain cross reference information. The EDI user arbitrarily chooses an application cross reference ID. The EDI-UA then uses information supplied by the EDIMG user to pair this cross reference ID with a globally unique body part ID and stores both in the cross reference heading field. + +The identifier for the body part can take different forms. It may be the sequence number of the body part within the EDIM, if the body part is contained within the EDIM. If the body part is contained in some other EDIM or IP-message, it shall be a globally unique identifier formed by concatenating the EDI message ID or IP-message ID and the body part number. + +An EDIMG user wishing to correlate a body part with an application cross reference ID found within an EDI interchange uses the application cross reference ID to perform a lookup in the cross reference information. It finds the corresponding body part ID in the data, and this can then be used to locate and extract that body part. + +The EDIMG user shall supply to the EDI-UA the information required to create the cross reference information when it requests the EDI-UA to create the EDIM. Similarly, the EDIMG user can use the cross reference data when processing a received EDIM. + +For informative purposes only, Figure E.1 illustrates the proposed mechanism. In this example, Body part 2, a drawing, is referenced from within the EDI interchange in the Primary body part, thus enabling a correlation between a particular line item in the purchase order and an engineering drawing. The application reference ID is "12345". + +If the EDIM is forwarded with no changes or body parts added, then all cross reference information is valid and the ultimate recipient EDI-UA can take the appropriate actions. + +![Diagram illustrating cross referencing in EDI messaging. It shows an EDIFACT interchange structure with a heading and user data segments. A callout shows a detailed view of the heading fields (Field 1, Field 2, Field 3, etc.) and a 'Cross reference data field' containing '12345 2'. Another callout shows the 'Primary body part containing interchange' which includes UNB, UNH, LIN, RFF+DW+12345, UNT, and UNZ segments. A third callout shows 'Body part 2' which contains an 'Engineering drawing #12345'. Dashed lines connect the segments to their respective callouts.](78e846f2858d131dc5a4dcbabcca1e4d_img.jpg) + +The diagram illustrates the structure of an EDIFACT interchange and its components for cross-referencing. On the left, the 'EDIFACT interchange' is shown as a vertical stack of elements: UNB, UNH, User data segments (containing LIN ... and RFF+DW+12345), UNT, and UNZ. A dashed line from the 'User data segments' points to a callout box on the right. This callout shows a detailed view of the 'Heading' section, which includes 'Field 1', 'Field 2', 'Field 3', and a 'Cross reference data field' containing the value '12345 2'. Another dashed line from the 'UNB' and 'UNH' segments points to a callout box labeled 'Primary body part containing interchange'. This box contains the segments UNB, UNH, LIN, RFF+DW+12345, UNT, and UNZ. A third dashed line from the 'RFF+DW+12345' segment points to a callout box labeled 'Body part 2', which contains an 'Engineering drawing #12345'. + +Diagram illustrating cross referencing in EDI messaging. It shows an EDIFACT interchange structure with a heading and user data segments. A callout shows a detailed view of the heading fields (Field 1, Field 2, Field 3, etc.) and a 'Cross reference data field' containing '12345 2'. Another callout shows the 'Primary body part containing interchange' which includes UNB, UNH, LIN, RFF+DW+12345, UNT, and UNZ segments. A third callout shows 'Body part 2' which contains an 'Engineering drawing #12345'. Dashed lines connect the segments to their respective callouts. + +Figure E.1/F.435 – Cross referencing in EDI messaging + + + + + +# ITU-T RECOMMENDATIONS SERIES + +| | | +|-----------------|--------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of the ITU-T | +| Series B | Means of expression: definitions, symbols, classification | +| Series C | General telecommunication statistics | +| Series D | General tariff principles | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Construction, installation and protection of cables and other elements of outside plant | +| Series M | TMN and network maintenance: international transmission systems, telephone circuits, telegraphy, facsimile and leased circuits | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks and open system communications | +| Series Y | Global information infrastructure | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.440-199208-I_PDF-E/raw.md b/marked/F/T-REC-F.440-199208-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..91f2d2f7ed802df71326ca61ad0a037504f0bbd1 --- /dev/null +++ b/marked/F/T-REC-F.440-199208-I_PDF-E/raw.md @@ -0,0 +1,1255 @@ + + +![ITU logo: A globe with a lightning bolt and the letters ITU.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with a lightning bolt and the letters ITU. + +INTERNATIONAL TELECOMMUNICATION UNION + +**CCITT** + +**F.440** + +THE INTERNATIONAL +TELEGRAPH AND TELEPHONE +CONSULTATIVE COMMITTEE + +(08/92) + +**MESSAGE HANDLING SERVICE +OPERATIONS AND DEFINITION OF SERVICE** + +--- + +**MESSAGE HANDLING SERVICES: +THE VOICE MESSAGING SERVICE** + +![CCITT logo: A globe with the letters CCITT.](6ed175c791b5e156d9c98a8dbcc3318c_img.jpg) + +CCITT logo: A globe with the letters CCITT. + +**Recommendation F.440** + +--- + +# FOREWORD + +The CCITT (the International Telegraph and Telephone Consultative Committee) is a permanent organ of the International Telecommunication Union (ITU). CCITT is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The Plenary Assembly of CCITT which meets every four years, establishes the topics for study and approves Recommendations prepared by its Study Groups. The approval of Recommendations by the members of CCITT between Plenary Assemblies is covered by the procedure laid down in CCITT Resolution No. 2 (Melbourne, 1988). + +Recommendation F.440 was prepared by Study Group I and was approved under the Resolution No. 2 procedure on the 4th of August 1992. + +# --- CCITT NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication Administration and a recognized private operating agency. + +# **MESSAGE HANDLING SERVICES: THE VOICE MESSAGING SERVICE** + +(1992) + +# **CONTENTS** + +- 1 *Purpose and scope* + - 1.1 General + - 1.2 Message handling systems used in the provision of VM-service +- 2 *VM-service* + - 2.1 General service requirements + - 2.2 VM-service features + - 2.3 Responsibility boundaries + - 2.4 Message store + - 2.5 Telephone service access unit + - 2.6 Use of directory + - 2.7 Security + - 2.8 Distribution lists +- 3 *Types of body parts* + - 3.1 Applicable body part types + - 3.2 Structure of voice messages + - 3.3 VM forwarding +- 4 *Conversion between different encoded information types* +- 5 *Naming and addressing in general* + - 5.1 Directory names + - 5.2 O/R names + - 5.3 O/R addresses +- 6 *Operation of the service* + - 6.1 General + - 6.2 Message handling phases +- 7 *Quality of service* + - 7.1 Message status + - 7.2 Support by Administrations + - 7.3 Model of delivery and notification times + - 7.4 Message delivery time targets + +- 7.5    Delivery notification time targets +- 7.6    Receipt notifications and non-receipt notifications +- 7.7    Error protection +- 7.8    Availability of service +- 7.9    Minimum storage capacity +- 8       *Tariff and accounting principles* +- 9       *Network requirements* +- 10      *User information and support* +- 11      *Use of the VM-service within CCITT defined telematic services* +- Annex A – Abbreviations* +- Annex B – Subscriber access and terminal requirements* +- Annex C – VM elements of service for 1984 systems* +- Annex D – Classification of elements of service for Voice Messaging* +- Annex E – Definitions of elements of service specific to Voice Messaging* +- Annex F – Classification of elements of service specific to the telephone service access unit* +- Annex G – Secure voice messaging elements of service* +- Annex H – Voice messaging security overview* + +# 1 Purpose and scope + +## 1.1 General + +This Recommendation specifies the general, operational and quality of service aspects of the public international Voice Messaging service. Voice Messaging services provided by Administrations belong to the group of Telematic Services defined in the F-Series Recommendations. + +This type of Message Handling (MH) service is an international telecommunication service offered by Administrations, enabling subscribers to send a message to one or more recipients and to receive messages via telecommunication networks using a combination of store and forward, and store and retrieve techniques. + +Locally provided functions, for which communication with other subscribers is not required, are not covered by CCITT Recommendations. + +The Voice Messaging (VM) service enables subscribers to request a variety of features to be performed during the handling and exchange of voice encoded messages. + +Some features are inherent in the basic VM-service. Other non-basic features may be selected by the subscriber, either on a per-message basis or for an agreed contractual period of time, if they are provided by Administrations. + +Intercommunication with the Interpersonal Messaging (IPM) service, may be provided as an option in the VM-service. + +Basic features have to be made available internationally by Administrations. Non-basic features, visible to the subscriber, are classified as either essential or additional. + +Essential optional features must be made available internationally by Administrations. Additional optional features may be made available by some Administrations for national use and internationally on the basis of bilateral agreement. Non-basic features are called optional user facilities. + +VM-service may be provided using any communications network. VM-service may be offered separately or in combination with various telematic or data communication services. It can be obtained by making appropriate arrangements. + +Technical specifications and protocols, to be used in the VM-service are defined in the X.400-Series Recommendations. + +The service definition is contained in § 2. Requirements for intercommunication between subscribers are described in §§ 3 and 4. Naming and addressing are described in § 5, while §§ 6, 7 and 8 describe the operation of the service, quality of service, tariff and accounting principles. Network requirements are given in § 9. The provision of subscriber information is in §§ 10, and 11 contains information on the use of the VM-service within CCITT defined telematic services. + +## 1.2 Message handling systems used in the provision of VM-service + +### 1.2.1 1984 implementations + +This Recommendation assumes that the message handling systems implemented to provide the service outlined herein are based on the 1988 version of the X.400-Series Recommendations. It is recognized however, that for some time after the publication of this Recommendation, the majority of implementations of the VM-service will be based on the 1984 X.400-Series Recommendations. Administrations are encouraged to adopt the latest CCITT Recommendations; however, in the interim, they may make use of this Recommendation with 1984 implementations as outlined below. + +### 1.2.2 *Elements of service* + +Elements of service available for message handling services are listed and classified in Recommendation F.400. Annex C of Recommendation F.440 provides a list of all the elements of service (called service elements in 1984) for VM from Recommendation X.400 (version 1984) which are based on the IPM elements of service. Annex C of Recommendation F.400 (version 1988) lists both the elements of service in 1988 as well as changes in classification to any 1984 elements of service. Annex D of Recommendation F.400 lists the elements of service specific to the VM-service. Annex E of Recommendation F.400 provides the definitions for the elements of service for VM additional to those described in Recommendation F.400 (1988). In all cases, 1984 elements of service may be used for the provision of the VM-service as described in this Recommendation, for a grace period ending in 1996. + +Administrations are urged to upgrade their implementations in this respect to the 1988 Recommendations. + +### 1.2.3 *Name forms* + +The name forms to be used for the VM-service are consistent with those specified in the F.400 (1988) and X.400 (1988) Recommendations. + +### 1.2.4 *Interworking* + +In order to protect the investment of Administrations who have implemented 1984 systems for the provision of VM-service, 1988, administration management domain (ADMD) implementations shall be able to interwork to 1984 ADMDs as outlined in Recommendation X.419, Annex B. + +Interworking from 1988 ADMDs to 1984 private management domains (PRMDs) is a national matter. + +# 2 **VM-service** + +## 2.1 *General service requirements* + +2.1.1 The fundamental ability of the VM-service is to provide a public interface between originators and recipients of voice communications to enhance their means of communication, especially where there is no immediate or convenient direct telecommunication service available between subscribers' equipment, or the telecommunication services available are incompatible. This service may also provide features available for the preparation and the presentation of the messages. + +2.1.2 The VM-service will be provided by Administrations using the Message Transfer service defined in Recommendation F.410, and by systems that conform to the X.400-Series Recommendations. + +Management domains (MDs) are defined for the purpose of responsibility boundaries. The MD managed by an Administration is called an administration management domain. The MD managed by an organization is called a private management domain (PRMD). + +2.1.3 International exchange of messages are performed between administration management domains through CCITT standardized public data transmission services. + +2.1.4 Different body part types of messages may be exchanged through this service. The various body part types are listed in § 3. + +2.1.5 An Administration may provide subscribers with different methods of access to the VM-service. The possible methods are: + +- 1) directly from the user's terminal, e.g. telephone set; +- 2) via a private message handling system. + +*Note* – A voice messaging management domain may be resident in a private automatic branch exchange. + +- 2.1.6 Each Administration is responsible for the national access to its management domain. +- 2.1.7 The characteristics of the interfaces and access methods used between terminals and the VM-service are a national matter, although they may follow provisions of the CCITT voice telephone service. However, the VM-service optional user facilities offered are defined and are independent of the access method and user's terminal. +- 2.1.8 The national implementation of the VM-service may provide intercommunication with existing services such as the IPM-service, teletex, facsimile and videotex. When implemented, the interfaces between the VM and the other services shall be according to relevant CCITT Recommendations. +- 2.1.9 As the service is providing indirect communication, cases of non-delivery of the message to the intended recipient may occur. The VM-service provides for non-delivery notification and, as optional user facilities, for delivery, receipt and non-receipt notifications. +- 2.1.10 Due to the intermediate storage of the message, the service may provide conversion optional user facilities: speed, access procedures, networks, and coding of message contents. +- 2.1.11 The message belongs to the originator until delivery has taken place. After delivery, the message belongs to the recipient. +- 2.1.12 Where sender and recipient have different and conflicting requirements, the sender's requirements shall take precedence (e.g. body type conversion or redirection control). + +## 2.2 *VM-service features* + +### 2.2.1 *Introduction* + +Recommendation F.400, § 19, defines elements of service which are available in the VM-service and are classified as either belonging to the basic service or as VM-optional user facilities. Elements of service comprising the basic VM-service are inherently part of the service and are always provided and available. The optional user facilities that are classified as essential are always provided and those classified as additional, may be available nationally, or internationally on the basis of bilateral agreement. + +### 2.2.2 *Basic VM-service* + +A set of elements of services comprises the basic VM-service. This set is defined in Recommendation F.400, and listed in Table 10/F.400. The basic VM-service, which is built upon the Message Transfer (MT) service, enables a user to send and receive voice messages. + +A user prepares V-messages with the assistance of his user agent (UA). User agents, which are a set of computer application processes, cooperate with each other to facilitate communication between their respective users. To send an V-message, the originating users make a request of their UA, specifying the name or address of the recipient who is to receive the V-message. The V-message, which has an identifier conveyed with it, is then sent by the originator's UA to the recipient's UA via the Message Transfer service. + +Following a successful delivery to the recipient's UA, the V-message can be received by the recipient. To facilitate meaningful communication, a receiving user may specify the encoded information type(s) that can be contained in V-messages delivered to the user, as well as the maximum length of a message that may be delivered. The original encoded information type(s) and an indication of any conversions that may have been performed and the resulting encoded information type(s) are supplied with each delivered V-message. In addition, the submission time, delivery time and other capabilities are supplied with each V-message. Non-delivery notification is provided with the basic service. + +### 2.2.3 *VM optional user facilities* + +A set of the elements of services of the VM-service are optional user facilities. The optional user facilities of the VM-service, which may be selected on a per-message basis or for an agreed contractual period of time, are listed in Tables 11/F.400 and 12/F.400, respectively. Local user facilities may be usefully provided in conjunction with some of these user facilities. + +The optional user facilities of the VM-service that are selected on a per-message basis are classified for both origination and reception by UAs. If an Administration provides the VM-service and offers these optional user facilities for origination by UAs, then a user is able to create and send V-messages according to the procedures defined for the associated element of service. If an Administration provides the VM-service and offers these optional user facilities for reception by UAs, then the receiving UA will be able to receive and recognize the indication associated with the corresponding element of service and to inform the user of the requested optional user facility. Each optional user facility is classified as additional or essential for UAs from these two perspectives. + +### 2.2.4 *Local functions* + +The message handling system (MHS) may perform many local functions for its subscribers in addition to providing VM features. For example, to assist subscribers in preparing and editing voice messages, MHS may provide an editing capability. The MHS could alert the subscribers when new messages have arrived (for example, by setting a message light on their telephone, or by displaying on their desktop terminals the originator's name and subject of all unread messages or by computer initiated voice indication). + +If alerting is not available, a subscriber may have to access MHS frequently to determine if new messages have arrived. + +The MHS may provide local database controls to help the subscriber find previously received and filed voice messages (for example, to find the message from Ms. Smith delivered sometime in August). A subscriber on vacation may request the MHS to automatically forward all voice messages to a delegate, or define rules for which voice messages should not be auto-forwarded (for example, personal messages). + +Local services such as those above, while perhaps utilizing some of the VM features, do not require coordination or cooperation with other subscribers. Thus, they do not impact the communication protocols associated with MHS. Therefore, local functions that may be provided by Administrations are outside the scope of CCITT. + +## 2.3 *Responsibility boundaries* + +The purpose of the MHS is to allow messages to be submitted for transfer to the destination and to be delivered to a UA/MS whose address is specified by the originator. + +A user interacts, using an access terminal, with a UA on the sending and on the receiving side. On request, a message is submitted to the message transfer system (MTS). It is also able to retrieve a received message from a UA or message store (MS). + +The responsibility for the message rests in the MHS when the originating user gives the command to send the message. The responsibility for a message is turned over to the receiving UA/MS after successful delivery. If the UA or MS is provided by an Administration, the responsibility for the message is taken over by the user when the message is rendered. As a basic feature, a non-delivery notification is created by the MHS when delivery to the receiving UA/MS is not possible. + +The conditions applied to this criterion may also depend on optional user facilities, e.g. conversion prohibition. An originating user may for a particular message specifically request a delivery notification, and/or a receipt notification, and/or a non-receipt notification. In the case of telematic addresses delivery takes place automatically when the message is transmitted to the telematic service. After delivery to a document store, or a message store, responsibility turns over to the user after having read the message once. When leaving the message in the store, the responsibility will be defined by the service provider. + +Loss of information may occur through the process of conversion as long as the conversion is not explicitly prohibited by the originating user. + +The responsibility of messages transferred through MDs starts at the moment of entering the domain and ends when leaving the domain, however, a later audit must be possible. + +When an ADMD interacts with a PRMD, the ADMD takes responsibility for the actions of the PRMD which are related to the interaction. In addition to ensuring that the PRMD properly provides the MT-service, and ADMD is responsible for ensuring that the accounting, logging, quality of service, and other related operations of the PRMD are correctly performed. An ADMD acts as the naming authority for the associated PRMDs, in accordance with national guidelines. + +## 2.4 *Message store* + +Administrations may optionally provide a message store (MS) to permit delivery of messages so that the recipient's UA does not have to be on line continuously. This is described in Recommendation F.400, § 7.4. A message delivered to an MS is deemed delivered by MHS. Messages delivered to an MS can be retrieved by the recipient as convenient and various optional user facilities can be provided to allow for retrieval for listing, fetching, and deletion of messages. When subscribing to an MS, all messages destined to the UA are delivered to the MS, an alert will be sent to the UA (from the MS) to inform the user of the fact that a message just arrived. + +## 2.5 *Telephone service access unit* + +The telephone service access unit (TSAU) is an access unit which provides for interworking between a Voice or Interpersonal Messaging service (VM-service or IPM-service) user and users of the telephone service. Users of the VM-service may request delivery of a voice message to any user of the telephone service who can be addressed by means of a telephone network address (telephone number). Users of the telephone service can cause a voice message to be recorded and delivered to users of the VM-service and IPM-service by means of the TSAU. + +### 2.5.1 *Description of TSAU functions* + +In an MHS environment, the IPM or VM-service user agent (UA) can access the telephone network through the TSAU. The MHS user composes the voice message (V-message) while interacting with the UA and designates it for delivery to a user of the telephone network. The MHS user also specifies the recipient's telephone number using a numeric originator/recipient (O/R) address. The MHS voice body part is sent in an F.400 envelope to the TSAU, which extracts the voice body part from the MHS envelope and calls the recipient on the telephone network to effect delivery of the V-message to the recipient. When the V-message is passed to the TSAU, a delivery notification is sent to the originator. Algorithms which control the calling behavior of the TSAU are a local matter. + +A user of the telephone network can call the TSAU, record a message, provide a delivery address (restricted to numeric characters provided on the telephone keypad), and request delivery of the message to a VM-service or IPM-service user. + +### 2.5.2 *Attendant-assisted delivery* + +The TSAU will normally deliver the message to anyone answering the telephone at the recipient's address. Attendant-assisted delivery allows a human attendant (telephone operator) to assist in the delivery of the message. When it is selected the VM-service user can specify that the message may only be delivered to a designated recipient. + +The attendant will ascertain if the recipient is available before playing the voice message and only deliver it if the designated recipient is available to receive the message. The attendant may use the recipient indicator for querying whether the recipient is present. + +### 2.5.3 *TSAU notifications* + +The TSAU servicing users of the telephone service uses MTS delivery reports to indicate transfer of the message from the MTS to the TSAU, and voice notifications (receipt and non-receipt) to indicate successful reception at the telephone terminal. + +## 2.6 *Use of directory* + +By making use of directory systems, VM-service users will be able to address recipients by using directory names or distribution list names, which are more user friendly than O/R addresses. The MHS will be able to access a directory systems and find out the O/R address(es) corresponding to a given directory name or distribution list name, for delivery of a message. An international telephone number has all the properties of a directory name. This capability is described in Recommendation F.400, § 14. + +## 2.7 *Security* + +Administrations may optional provide security mechanisms as outlined in Recommendation F.400, § 15, to counter the various security threats mentioned. This capability may use a directory system for storing certified copies of public keys for MHS-users. + +## 2.8 *Distribution lists* + +A group whose membership is stored in the directory can be used as a distribution list (DL). The originator simply supplies the name of the list on submission of a message, and the MHS can obtain the directory names (and then the O/R addresses) of the individual recipients, by consulting the directory. Upon receipt of a message addressed to a distribution list, the recipient can determine through which DL the message arrived. An originator can prohibit the expansion of the distribution if one of the recipients specified refers to a distribution list. Recommendation F.400, § 14, outlines the full capabilities available to DL-users. + +If a user unknowingly sends a message to a DL, charges may be incurred for multiple deliveries that were not expected. Because of this, names of distribution lists should be indicative of the fact that what is being named is a DL. Owners of DLs should also ensure that they respect a potential member's wishes about being a member and the rules of the country of the member that may prohibit inclusion without prior agreement. + +# 3 Types of body parts + +Messages sent and received in the VM-service can be composed of one or more body parts. + +## 3.1 *Applicable body part types* + +Applicable body part types are defined in Recommendation X.440 and comprise the following: + +- voice; +- message (e.g. for a forwarded message); +- notifications; +- externally defined. + +Initially, the voice encoding supported will be 32 kbit/s ADPCM as specified in Recommendation G.721(1988). Encoding schemes at lower bit rates may be added in the future. + +The voice messaging class of UAs (VM-UA) create messages containing a content specific to the Voice Messaging service. The specific content that is sent from one VM-UA to another is a result of an originator, which is generally a human interacting with a telephone, composing and sending a message, called a voice message. The voice message shall carry the voice-encoded object and optionally other information associated with the voice-encoded object. Only one voice-encoded object shall be present in a voice message, and every voice message shall contain a voice-encoded body part on origination of the voice message. The structure of a voice message as it relates to the basic message structure of MHS is shown in Figure 1/F.440. The voice message is conveyed with an electronic envelope when being transferred through the MTS. + +![Diagram illustrating the Voice message structure. On the left, a 'Heading' box is shown above a 'Body' box. The 'Body' box contains two 'Body part' boxes, with vertical ellipsis between them. On the right, an 'Envelope' box is shown above a '+' sign, which is above a 'Content' box. A dashed line connects the 'Heading' box to the 'Envelope' box, and another dashed line connects the 'Body' box to the 'Content' box.](e9314c83043183351ed74908e9bf2f90_img.jpg) + +The diagram illustrates the structure of a voice message. It is divided into two main columns. The left column shows the internal structure of the message: a 'Heading' box at the top, and a 'Body' box below it. The 'Body' box contains two 'Body part' boxes, with vertical ellipsis between them. The right column shows the message as it is transmitted: an 'Envelope' box at the top, a '+' sign in the middle, and a 'Content' box at the bottom. A dashed line connects the 'Heading' box to the 'Envelope' box, and another dashed line connects the 'Body' box to the 'Content' box. The text 'T0102940-93' is located at the bottom right of the diagram. + +Diagram illustrating the Voice message structure. On the left, a 'Heading' box is shown above a 'Body' box. The 'Body' box contains two 'Body part' boxes, with vertical ellipsis between them. On the right, an 'Envelope' box is shown above a '+' sign, which is above a 'Content' box. A dashed line connects the 'Heading' box to the 'Envelope' box, and another dashed line connects the 'Body' box to the 'Content' box. + +FIGURE 1/F.440 +**Voice message structure** + +Figure 2/F.440 shows a mapping between a typical voice-encoded object and the corresponding voice message structure. The voice-encoded object is mapped entirely within one body part, called the primary body part, and may be a G.721 or privately encoded voice object or a forwarded V-message. Unless specified otherwise, this voice object will be assumed to be encoded per G.721 32 kbit/s ADPCM. Other body parts are available to convey information associated with the voice-encoded object such as drawings, additional text information, etc. The heading of the VM contains various fields of information that convey service requests from the originator. This is illustrated in Figure 2/F.440 as V-E-Field *n* for voice encoded information and Field *n* for non-voice encoded information. + +The heading and body part(s) form the VM. A VM may contain only one voice-encoded object, which is totally contained in one body part. + +![Diagram illustrating the voice message structure for a typical voice messaging transaction. The diagram shows a 'Header' and a 'Body' on the left. The 'Header' is connected by dashed lines to a stack of fields on the right: 'V-E Field 1', 'V-E Field 2', a vertical ellipsis, 'V-E Field n', and 'Field n'. The 'Body' is connected by dashed lines to a stack of body parts on the right: 'Primary body part', a vertical ellipsis, and 'Body part n'. A small code 'T0102950-93' is at the bottom right of the diagram.](d4af765160d04ecef538e5066006dc77_img.jpg) + +Diagram illustrating the voice message structure for a typical voice messaging transaction. The diagram shows a 'Header' and a 'Body' on the left. The 'Header' is connected by dashed lines to a stack of fields on the right: 'V-E Field 1', 'V-E Field 2', a vertical ellipsis, 'V-E Field n', and 'Field n'. The 'Body' is connected by dashed lines to a stack of body parts on the right: 'Primary body part', a vertical ellipsis, and 'Body part n'. A small code 'T0102950-93' is at the bottom right of the diagram. + +FIGURE 2/F.440 +**Voice message structure for a typical voice messaging transaction** + +## 3.3 *VM forwarding* + +A user on the VM-service may cause received messages to be forwarded with or without acceptance. Requests for notifications may also be automatically forwarded along with the forwarded message. Three cases can be distinguished: + +- a) forwarding of the accepted V-message with requested notifications; +- b) forwarding of the V-message and notifications without acceptance; and +- c) forwarding of the V-message when notifications were not requested. + +The third case is effectively identical to the second case. A user may also request automatic forwarding to be set. + +If a V-message is forwarded after acceptance, the primary body part of the forwarded message is the content of the received message with or without changes. Additional VM body parts may be included or removed, but there may not be more than one additional voice encoded body part per instance of forwarding. The forwarded body part may not be removed. Receipt notifications will be issued as requested. In forwarding, a non-receipt notification (NRN) will not be sent back to the originator of the message. VM forwarding may be to one or more recipients, and notification requests may be set for any of these recipients. + +Body parts may not be added or deleted if a V-message is forwarded without acceptance. Forwarding may be to more than one recipient. The notification request should be the same as for the original message, but may only be sent to one of the recipients of the forwarded message. This will insure that the original originator only receives the notification that was requested. + +# 4 Conversion between different encoded information types + +The MTS provides conversion functions to allow VM-users to input messages in one encoded format, called encoded information type (EIT), and have them delivered in another EIT to cater to users with different terminal types. This capability is inherent in the VM-service, and increases the possibility of delivery by tailoring the message to the recipient's terminal capabilities. The EITs supported for the VM-service are defined in Recommendation X.440. + +The MTS may provide automatic conversion between the various standardized encoding schemes when required for the delivery of the message. + +The general aspects of conversion and the specific conversion rules for conversion between different EITs in the VM-service are for further study. + +# 5 Naming and addressing in general + +In MHS, the principal entity that requires naming is the user (the originator and recipient of messages). In addition distribution lists (DLs) have names for use in MHS. Users of MHS and DLs are identified by O/R names. O/R names are comprised of directory names and/or O/R addresses, all of which are described in this section. Recommendation F.401 provides more detail on naming and addressing for public message handling services, including naming restrictions and responsibilities of Administrations. + +## 5.1 *Directory names* + +Users of MHS service, and DLs, may be identified by a name, called a directory name. A directory name was to be looked up in a directory to find out the corresponding O/R address for a particular service. The structure and components of directory names are described in the X.500-Series Recommendations. + +A user may access a system directly to find out the O/R address of a user, or O/R addresses of the members of a DL (both of which are outside the scope of these Recommendations). + +As an alternative, a user may use the directory name and have the MHS access the directory to resolve the corresponding O/R address or addresses automatically. Every MHS-user or DL will not necessarily have a directory name, unless they are registered in a directory. As directories become more prevalent, it is expected that directory names will be the preferred method of identifying MHS-users to each other. + +## 5.2 *O/R names* + +Every MHS-user or DL will have an O/R name. An O/R name comprises a directory name, an O/R address, or both. The directory name unambiguously identifies an MHS-user but not necessarily uniquely. The O/R address uniquely identifies an MHS-user. + +Either or both components of an O/R name may be used on submission of a message. If only the directory name is present, the MHS will access a directory to attempt to determine the O/R address, which it will then use to route and deliver the message. If the directory name is absent, it will use the O/R address as given. When both are given on submission, the MHS will use the O/R address, but will carry the directory name and present both to the recipient. If the O/R address is incorrect, it will then attempt to use the directory name as above. + +## 5.3 *O/R addresses* + +An O/R address contains information that enables the MHS to uniquely identify a user to deliver a message or return a notification. (The prefix "O/R" recognizes the fact that the user can be acting as either the originator or recipient of the message or notification in question). + +Various forms of O/R addresses are currently defined, each serving its own purpose. These forms and their purpose are as follows: + +- *Numeric O/R address*: Provides a means of identifying users addressable by numeric keypads. +- *Terminal O/R address*: Provides a means of identifying users with terminals belonging to various networks. +- *Mnemonic O/R address*: Provides a user friendly means of identifying users in the absence of a directory. It is also used for identifying a distribution list. + +An O/R address is made up of a collection of information called attributes. These attributes as used in each of the O/R address forms above are detailed in Recommendation F.401. + +Management domains shall allow their users to originate messages using any of the above forms. The form in which names are input by or presented to the subscriber is a national matter (as for example the use of distribution lists or of friendly ways of identifying user agents). It is anticipated that the use of the numeric O/R address will be the most prevalent in the VM-service. + +Each Administration is responsible for the unique identification of each user agent in its management domain. + +# 6 Operation of the service + +## 6.1 General + +6.1.1 The VM-service provides that messages can be sent, transferred, delivered and received using fully automatic procedures. + +6.1.2 Messages are prepared in, sent from, and delivered to a memory. These memories are part of the UA/MS functionality and are under control of the subscriber. + +6.1.3 The transfer of messages between management domains will be in accordance to the Message Transfer service as described in Recommendation F.410. + +6.1.4 Each Administration providing the VM-service should validate the subscribers identities, at the time of access. + +6.1.5 It is a national matter whether to allow private messaging systems to connect to the public VM-service, in order to allow users of these systems to exchange messages. If these interconnections are provided, they should take place between administration management domains in accordance with CCITT Recommendations. + +6.1.6 When implicit conversion is provided by the Administration via the Message Transfer service, the message will be converted if necessary, unless prohibited by the originator. The conversion will be in accordance to the rules specified in Recommendation X.408. See also § 4 of this Recommendation. + +6.1.7 Deferred delivery shall be provided by the management domain of the originator, which is responsible for the storage of the message until the date and time specified for intended delivery. Therefore the element of service, deferred delivery, should not be used across international links. + +## 6.2 Message handling phases + +### 6.2.1 General + +The VM-service has different message handling phases visible to the user. + +### 6.2.2 Preparation phase + +In this phase messages are prepared by making use of the user agent functionality (e.g. capture, editing and filing). The way in which these functions are performed is outside the scope of the Recommendation. + +### 6.2.3 *Sending phase* + +In this phase the originator may request the user agent or message store to send a prepared message to one or more recipients and to request certain optional user facilities. + +### 6.2.4 *Receipt phase* + +In this phase the subscriber can receive delivered messages and notifications from the user agent or message store. The receipt phase can be initiated by the service (auto-receipt) or by the subscriber for message reception. The operation of the user agent receiving messages is specified in Recommendation X.440. + +Subscribers using terminals without user agent functionality may register for a contractual period of time during which they will receive delivered messages automatically from their user agent to a terminal, if the Administration provides for this alternative. Normally the user agent is called to receive incoming messages. + +In the case of auto-receipt, the MHS will initiate a call to the subscriber's terminal. In the other case, the subscriber shall initiate a call to the MHS at a time convenient to the subscriber. + +The body parts of the message will be received by the subscriber in the form in which the originator has sent it, unless conversion has been performed. + +The indication of the optional user facilities requested by the originator are presented by the user agent to the recipient in a form convenient to the user. + +*Notifications:* Four notifications can be received: + +- non-delivery notification; +- delivery notification; +- receipt notification; +- non-receipt notification. + +Non-delivery notification is automatically originated by the MTS, while delivery notification, receipt and non-receipt notifications depend on the action of the recipient. + +### 6.2.5 *Service notifications* + +Service notifications are sent only upon the originator's request. They are sent shortly after the subject message leaves the MTS for the recipient's UA or MS. Service notifications may be generated by a UA or MS or TSAU when requested in the per recipient voice notification (VN) request indicator. Service notification reasons are specified in Annex B for the element of service "VM notification request." + +# 7 **Quality of service** + +## 7.1 *Message status* + +The unique identification of each V-message enables the system to provide information about, e.g. the status of an V-message. + +In the event of system failure all accepted and non-delivered messages should be traceable. If messages cannot be delivered, the originator must be informed by a non-delivery notification. + +## 7.2 *Support by Administrations* + +Administrations should provide assistance to their subscribers, with regard to non-delivery notifications not being received in due time, as far as public system components are concerned. Additional provision on support of status and tracing of messages may be provided under national responsibility. + +When the user agent is provided by an Administration, additional functionality should be provided in order to minimize cases of not reading messages within a certain period of time (the definition of this period is for further study). This functionality could be for example, alert messages sent to an automatic reception terminal. + +See Figure 3/F.440. + +![Sequence diagram showing the notification time model. It illustrates the flow of a message from creation to playback, with associated notification times (T1, T2, T3) and VM notifications. The diagram includes entities: UA/MS, MTA, MTA, MS/UA. Arrows show message creation, submission, transfer, delivery, notification generation, notification delivery, playback, and VM notifications. Vertical dashed lines mark time points T1, T2, and T3.](0f985b39edc1d52ba3600c438bc8f0a5_img.jpg) + +The diagram illustrates the notification time model for a message delivery process. It shows the sequence of events and the corresponding notification times: + +- Creation of message:** The initial event, marked by a hatched arrow. +- Submission of message:** The message is submitted to the first UA/MS entity. +- Transfer of message:** The message is transferred between the two MTA entities. +- Delivery of message:** The message is delivered to the MS/UA entity. +- Notification generation:** A notification is generated at the MS/UA entity. +- Notification delivery:** The notification is delivered back to the first UA/MS entity. +- Playback of message:** The message is played back to the user at the MS/UA entity. +- VM notifications:** VM notifications are generated and delivered at various points in the process. + +Vertical dashed lines indicate key time points: + +- T1:** Delivery time (at the MS/UA entity). +- T2:** Delivery notification time (at the MS/UA entity). +- T3:** VM notifications time (starting at the MS/UA entity and ending at the first UA/MS entity). + +A bracket groups T2 and T3 with the text: "For more details see Recommendation F.410". + +Sequence diagram showing the notification time model. It illustrates the flow of a message from creation to playback, with associated notification times (T1, T2, T3) and VM notifications. The diagram includes entities: UA/MS, MTA, MTA, MS/UA. Arrows show message creation, submission, transfer, delivery, notification generation, notification delivery, playback, and VM notifications. Vertical dashed lines mark time points T1, T2, and T3. + +T0102960-93 + +- T1    Delivery time +- T2    Delivery notification time    } For more details see Recommendation F.410 +- T3    VM notifications time + +*Note 1* – Starting time of T3 corresponds to the time the message is played to the user and VM notification is actioned by the user. + +*Note 2* – Ending time of T3 is the time that the VM notification is made available to the user through the UA or MS. + +*Note 3* – Similar considerations apply to non-receipt notifications. + +FIGURE 3/F.440 +**Notification time model** + +## 7.4 *Message delivery time targets* + +The management domain of the recipient UA should force non-delivery notification if the message has not been delivered before $x$ hours after submission (or after date and time indicated for deferred delivery), the value of $x$ being dependent on the grade of delivery requested by the originator (see Recommendation F.410, § 4.4). + +## 7.5 *Delivery notification time targets* + +Non-delivery notifications or requested delivery notifications should be returned on a per-recipient basis, in order not to delay notifications for those messages in a multi-address message which have already been delivered, to enable the originating management domain either to return per-recipient notifications or to batch notifications to its subscribers (see Recommendation F.410, § 4.5). + +## 7.6 *Receipt notifications and non-receipt notifications* + +Delivery times for receipt, non-receipt, and service element availability notifications in the first place depend on local arrangements. When they are initiated by the receiving UA/user they have the same time targets as the messages that cause them to occur. + +TABLE 1/F.440 + +### **Delivery times for notifications** + +| Grade of delivery
of the referred message | 95% delivered before | Non-delivery forced after | +|----------------------------------------------|----------------------|---------------------------| +| Urgent | 0.25 hours | 2 hours | +| Normal | 1.0 hours | 6 hours | +| Non-urgent | 4.0 hours | 12 hours | + +*Note* – Intercommunication with PRMDs is not included in the calculation of the time targets. + +## 7.7 *Error protection* + +Error protection on transmission is provided by the MHS and underling protocols used in the provision of the VM-service. + +## 7.8 *Availability of service* + +In principle the VM-service should be available continuously. The user agent should be available for submission or delivery continuously (unless hold for delivery is invoked). In cases where the UA is not available for delivery continuously, a message store should be used. + +## 7.9 *Minimum storage capacity* + +The storage capacity of a user agent and message store shall be sufficient to provide a high grade of service. + +*Note* – This is for further study. + +# **8 Tariff and accounting principles** + +See the D-Series Recommendations. + +# **9 Network requirements** + +The VM-service is network independent, that is, the basic service and the essential optional user facilities are provided independently of the type of network used for service access. Additional optional user facilities chosen by an Administration to offer may vary. + +# **10 User information and support** + +A directory shall be provided by each Administration for its domain. The directory can be hard copy or preferably electronic form. + +The directory shall at least contain the following: + +- a) how to use the directory and the service; +- b) list of O/R addresses of subscribers belonging to the Administrations domain; +- c) list of standardized abbreviations for O/R address attributes; +- d) list of country and administration management domain names reachable by the public VM-service. + +# **11 Use of the VM-service within CCITT defined telematic services** + +In the telematic services, intercommunication between the VM-service and the IPM-service is desirable. Intercommunication with other Telematic Services is for further study. + +Intercommunication with the telephone service is provided for by means of the optional telephone service access unit. + +# **ANNEX A (to Recommendation F.440)** + +## **Abbreviations** + +The following abbreviations are used in this Recommendation: + +| | | +|-------|---------------------------------------------| +| ADMD | Administration management domain | +| ADPCM | Adaptive differential pulse code modulation | +| AU | Access unit | +| DL | Distribution list | +| EIT | Encoded information type | +| IP | Interpersonal | +| IPM | Interpersonal messaging | +| MD | Management domain | +| MH | Message handling | +| MHS | Message handling system | +| MS | Message store | +| MT | Message transfer | + +| | | +|-------|-------------------------------| +| MTA | Message transfer agent | +| MTS | Message transfer system | +| N/A | Not applicable | +| NRN | Non-receipt notification | +| O/R | Originator/recipient | +| PDN | Public data network | +| PRMD | Private management domain | +| TS | Telephone service | +| TSAU | Telephone service access unit | +| UA | User agent | +| VM | Voice messaging | +| VM-MS | Voice messaging store | +| VM-UA | Voice messaging user agent | +| VN | Voice notification | + +*Note 1* – For a glossary of terms, see Annex A of Recommendation F.400. + +*Note 2* – For references, see Recommendations F.400 and F.401. + +# ANNEX B + +(to Recommendation F.440) + +## **Subscriber access and terminal requirements** + +### **B.1      *General*** + +Various types of terminals may be used for accessing the service. These terminals are functionally divided into two categories; those without user agent functionality, and those with user agent functionality. The telematic terminals assume a special user agent. + +## **B.2      *Terminals without UA functionality*** + +Terminals in this category require additional functions to be provided by MHS to enable their participation in the VM-service. + +In this category are voice terminals or telephone subscriber sets as commonly used in telephony. + +## **B.3      *Terminals with UA functionality*** + +These terminals shall, as a minimum, have the capabilities to: + +- 1) provide the capabilities to subscribers of the basic features defined in § 2; +- 2) have speech input and output capability; +- 3) make use of the VM protocol specified in Recommendation X.440; +- 4) use the submission and delivery protocol specified in Recommendation X.419; +- 5) use the remote operation procedures specified in Recommendation X.419. + +This type of terminal may be configured to access both the IPM- and VM-services. + +# ANNEX C (to Recommendation F.440) + +## **VM elements of service for 1984 systems** + +TABLE C-1/F.440 + +| Elements of service | Classification | | | | +|-----------------------------------------------|----------------|-------------|-----------|-------------| +| | Basic | Optional | | | +| | | Origination | Reception | Contractual | +| Access management | X | | | | +| Alternate recipient allowed | | A | A | | +| Alternate recipient assignment | | | | A | +| Authorizing users indication | | A | E | | +| Auto-forwarded indication | | A | E | | +| Blind copy recipient indication | | A | E | | +| Body part encryption indication | | A | E | | +| Content type indication | X | | | | +| Conversion prohibition | | E | E | | +| Converted indication | X | | | | +| Cross referencing indication | | A | E | | +| Deferred delivery | | E | N/A | | +| Deferred delivery cancellation | | A | N/A | | +| Delivery notification | | E | N/A | | +| Delivery time stamp indication | X | | | | +| Disclosure of other recipients | | A | E | | +| Expiry date indication | | A | E | | +| Explicit conversion | | A | N/A | | +| Forwarded IP-message indication | | A | E | | +| Grade of delivery selection | | E | E | | +| Hold for delivery | | | | A | +| Implicit conversion | | | | A | +| Importance indication | | A | E | | +| IP-message identification | X | | | | +| Message identification | X | | | | +| Multi-destination delivery | | E | N/A | | +| Multi-part body | | A | E | | +| Non-delivery notification | X | | | | +| Non-receipt notification | | A | A | | +| Obsoleting indication | | A | E | | +| Original encoded information types indication | X | | | | +| Originator indication | | E | E | | +| Prevention of non-delivery notification | | A | N/A | | +| Primary and copy recipients indication | | E | E | | +| Probe | | A | N/A | | +| Receipt notification | | A | A | | +| Registered encoded information types | X | | | | +| Reply request indication | | A | E | | +| Replying IP-message indication | | E | E | | +| Return of contents | | A | N/A | | +| Sensitivity indication | | A | E | | +| Subject indication | | E | E | | +| Submission time stamp indication | X | | | | +| Typed body | X | | | | + +# ANNEX D + +(to Recommendation F.440) + +## **Classification of elements of service for Voice Messaging** + +TABLE D-1/F.440 + +| Elements of service | Origination | Reception | Reference (Note) | +|---------------------------------------------|-------------|-----------------|------------------| +| Access management | B | B | B.1 | +| Attendant-assisted delivery | A | A | E.1 | +| Auto-forwarding indication | A | A | B.6 | +| Body part encryption indication | A | A | B.9 | +| Content type indication | B | B | B.12 | +| Conversion prohibition | E | E | B.13 | +| Converted indication | B | B | B.15 | +| Deferred delivery | E | N/A | B.19 | +| Deferred delivery cancellation | A | N/A | B.20 | +| Delegation of recipient by directory name | A | N/A | B.24 | +| Delivery notification | E | N/A | B.21 | +| Delivery time stamp indication | B | B | B.22 | +| Disclosure of other recipients | A | E | B.25 | +| DL-Expansion prohibited | A | N/A | B.27 | +| Expiry date (+time) indication | A | A | B.29 | +| Forwarding V-message indication | E | E | E.2 | +| Grade of delivery selection | E | E | B.32 | +| Hold for delivery | A | C | B.33 | +| Implicit conversion | A | C | B.34 | +| Importance indication | E | E | B.35 | +| Language indication | A | A | B.38 | +| Latest delivery designation | A | N/A | B.39 | +| Message identification | B | B | B.41 | +| MS register | A | A | B.95 | +| Multi-destination delivery | E | N/A | B.45 | +| Multi-part body | A | A | B.46 | +| Non receipt notification request indication | E | E | B.48 | +| Non-delivery notification | B | B | B.47 | +| Obsoleting indication | A | A | B.52 | +| Prevention of non-delivery notification | E | N/A | B.61 | +| Receipt notification request indication | E | E | B.67 | +| Redirection disallowed by originator | A | N/A | B.68 | +| Redirection of incoming message | A | C | B.69 | +| Replying V-message indicator | A | E | E.3 | +| Restricted delivery | A | C | B.77 | +| Sensitivity indication | E | E | B.80 | +| Stored message alert | A | C | B.82 | +| Stored message deletion | N/A | E a) | B.84 | +| Stored message fetching | N/A | E a) | B.85 | +| Stored message listing | N/A | E a) | B.86 | +| Stored message summary | N/A | E a) | B.87 | +| Stored V-message auto-forward | A | A | E.4 | +| Submission time stamp indication | B | B | B.89 | +| TS-recipient spoken name indicator | A | A | E.5 | +| TS AU-recipient | E | E | E.6 | +| Typed body | B | B | B.90 | +| Use of distribution list | E | N/A | B.92 | + +TABLE D-1/Rec. F.440 (Cont.) + +| Elements of service | Origination | Reception | Reference (Note) | +|----------------------------------------|-------------|-----------|------------------| +| User/UA capability registration | B | B | B.93 | +| V-message creation time | B | B | E.7 | +| V-message duration indicator | B | B | E.8 | +| V-message identification | B | B | E.9 | +| VM-encoding algorithm indicator | B | B | E.10 | +| VM-forwarding | A | A | E.11 | +| VM-multi-part body | A | E | E.12 | +| VM-receiver | A | A | E.13 | +| VM-recipient indicator | E | E | E.14 | +| VM-service status request notification | E | E | E.15 | +| VM-spoken name indication | E | E | E.16 | +| VM-subject indication | E | E | E.17 | + +a) This classification applies if the message store is provided. + +- A Additional +- B Basic +- C Contractual +- E Essential +- N/A Not Applicable + +*Note* – References to B-sections are found in Annex B of Recommendation F.400. References to E-sections are found in Annex E of this Recommendation. + +# ANNEX E + +(to Recommendation F.440) + +## Definitions of elements of service specific to Voice Messaging + +### E.1 **attendant-assisted delivery** + +This element of service allows the voice messaging user agent to indicate that a human operator should be used in the delivery of the message by means of the telephone service access unit. + +### E.2 **forwarding voice message indicator** + +This element of service allows a forwarded voice message or a forwarded voice message plus its “delivery information” to be sent as the body (or as one of the body parts) of a voice message. An indication that the body part is forwarded is conveyed along with the body part. In a multi-part body, forwarded body parts can be included along with the body parts of other types. “Delivery information” is information which is conveyed from the message transfer system when a voice message is delivered (for example, time stamps and indication of conversion.) However, inclusion of this delivery information along with a forwarded voice message, in no way guarantees that this delivery information is validated by the message transfer system. + +*Note* – In the context of voice, the additional body types that may be added during forwarding is still for further study. + +### E.3 **replying voice message indicator** + +This element of service allows the originator of a voice message to indicate to the recipient(s) that this voice message is being sent in reply to another voice message. A reply can, depending on the wishes of the originator of the replied-to message, and on the final decision of the originator of the reply, be sent to: + +- a) the recipients specified in the reply request indication of the replied-to message; +- b) the originator of the replied-to message; +- c) the originator and other recipients; + +- d) a distribution list in which the originator of the replied-to message can be a receiving member; +- e) other recipients as chosen by the originator of the reply. + +### E.4 **stored VM-auto-forward** + +This element of service allows the user of a VM-MS to have the message store automatically perform voice message forwarding, with or without accepting the message. The user of the VM-MS may establish criteria for selecting voice messages through use of the element of service “MS-register”. The complete voice message, as received from the originator, is forwarded unchanged, and, if requested, an appropriate voice message notice is generated by the VM-MS. Forwarding is limited to one recipient. + +### E.5 **TS-recipient spoken name indicator** + +This element of service is used to allow the attendant in attendant-assisted delivery to ascertain delivery of the message to the intended recipient. + +### E.6 **TSAU-recipient** + +This element of service specifies the recipient on the telephone service for which the message is intended. + +### E.7 **voice message creation time** + +This element of service enables an originating user agent to convey the message creation time to the recipients user agent. This message creation time can be used for correlation of returning notifications. + +### E.8 **voice message duration indicator** + +This element of service enables an originator’s user agent to indicate, to the recipient(s), the duration of a message. + +### E.9 **voice message identification** + +This element of service enables cooperating voice messaging user agents to convey a globally unique identifier to each voice message sent or received. The voice message identifier is composed of an originator/recipient name of the originator and an identifier that is unique with respect to that name. A voice messaging user agent and user use this identifier, optionally together with message creation time, to refer to a previously sent or received voice message (for example, in receipt notifications.) + +### E.10 **VM-encoding algorithm indicator** + +This element of service enables the originating user agent to indicate to the recipient user agents, the voice encoding algorithm used to construct the message. + +*Note* – 32 kbit/s ADPCM defined by Recommendation G.721 is the default encoding algorithm. + +### E.11 **VM-forwarding** + +This element of service enables an voice messaging user agent to forward with or without changes, and a voice messaging store to forward without changes, a received voice message. Support of the element of service voice messaging receiver is also required when forwarding. + +### E.12 **VM-multi-part body** + +This element of service allows an originator to send to a recipient a voice message with a body that is partitioned in several parts. The nature and attributes, or type, of each part are conveyed along with the body part as well as its positional index into the body. + +### E.13 **VN-receiver** + +This element of service allows the originator, or a forwarding voice messaging user agent/message store, to indicate to a recipient the originator/recipient address that requested notifications should be returned to. Notifications should always take the same return path as was used for delivery of the voice message. + +### E.14 **VM-recipient indicator** + +This element of service allows the originator to provide the name of zero, or more users, or distribution list who are the intended recipients of voice message. In addition, it is possible to specify a qualifier for each recipient. + +### E.15 **VM-service status request notification** + +This element of service allows the originator to request notification when any requested service can not be performed by the recipient's user agent. + +*Note* – Qualifier values are for further study. + +### E.16 **VM-spoken name indication** + +This element of service allows the identity of the originator of a voice message to be conveyed in a voice form to the recipient. + +### E.17 **VM-subject indication** + +This element of service allows an originator to indicate in a voice form, the subject of a voice message to the recipient(s). The recipients of the reply receive it as a regular voice message, together with an indication of which voice message it is in reply to. + +# ANNEX F + +(to Recommendation F.440) + +## **Classification of elements of service specific to the telephone service access unit** + +The classification of the MH elements of service defined in Annex B of Recommendation F.400 that apply to the TSAU is shown in Table F-1/F.440. In addition, since the TSAU is provided as part of the VM service, it may use all elements of service that apply to the VM service listed in the Annex D of this Recommendation. + +TABLE F-1/F.440 + +### **Elements of service applying when using the telephone service access unit** + +| Elements of service | Origination | Reception | Reference (Note) | +|--------------------------------|-------------|-----------|------------------| +| Attendant-assisted delivery | A | A | E.1 | +| Conversion prohibition | E | E | B.13 | +| Converted indication | B | B | B.15 | +| Deferred delivery | E | N/A | B.19 | +| Deferred delivery cancellation | A | N/A | B.20 | +| Delivery notification | E | N/A | B.21 | +| DL expansion prohibited | A | A | B.27 | +| Grade of delivery selection | E | E | B.32 | + +TABLE F-1/F.440 (Cont.) + +| Elements of service | Origination | Reception | Reference (Note) | +|-----------------------------------------------|-------------|-----------|------------------| +| Hold for delivery | A | C | B.33 | +| Implicit conversion | A | C | B.34 | +| Importance indication | A | E | B.35 | +| Language indication | A | A | B.38 | +| Message identification | B | B | B.41 | +| Multi-destination delivery | E | N/A | B.45 | +| Non-delivery notification | B | B | B.47 | +| Non-receipt notification | A | E | B.48 | +| Original encoded information types indication | B | B | B.54 | +| Prevention of non-delivery notification | E | N/A | B.61 | +| Receipt notification request indication | A | E | B.67 | +| Redirection disallowed by originator | A | N/A | B.68 | +| Replying message indicator | A | A | B.73 | +| Submission time stamp indication | B | B | B.89 | +| TS-recipient spoken name indicator | A | A | E.5 | +| TSAU-recipient | E | E | E.6 | +| Typed body | B | B | B.90 | +| Use of distribution list | E | N/A | B.92 | +| User/UA capability registration | B | B | B.93 | +| V-message identification | B | B | B.37 | + +A Additional +B Basic +C Contractual +E Essential +N/A Not Applicable + +*Note* – References to B-sections are found in Annex B of Recommendation F.400. References to E-sections are found in Annex E of this Recommendation. + +# ANNEX G + +(to Recommendation F.440) + +## Secure voice messaging elements of service + +(This annex forms an integral part of this Recommendation) + +This Annex defines the secure voice messaging elements of service. + +### G.1 non-repudiation of content originated + +This element of service allows an originating VM-UA to provide a recipient VM-UA an irrevocable proof as to the authenticity and integrity of the content of the message as it was submitted into the MHS environment. + +The corresponding proof data can be supplied in two ways depending on the security policy in force: + +- 1) by using the non-repudiation of origin security applied to the original message; or +- 2) by using a notarization mechanism. + +*Note* – At this time, use of a notarization mechanism requires bilateral agreements, protocol is not provided. + +### G.2 **non-repudiation of content received** + +This element of service allows an originating VM-UA to get from a recipient VM-UA an irrevocable proof that the original subject message content was received by the recipient VM-UA and the subject voice message was accepted, forwarded or refused. This service provides irrevocable proof as to the authenticity of the recipient of the message and irrevocable proof as to the integrity of the content of the message. It will protect against any attempt by the recipient(s) to subsequently deny having received the message content. + +*Note 1* – This service is stronger than the Proof of Content Received service. + +The corresponding proof data can be supplied in two ways depending on the security policy in force: + +- 1) by returning a Non-repudiation of Origin of the voice notification which incorporates the following: + - the originator's Non-repudiation of Origin security arguments (if present); or + - the complete original message content, if the originators Non-repudiation of Origin arguments are not present; +- 2) by using a notarization mechanism. + +*Note 2* – At this time, use of a notarization mechanism requires bilateral agreements, protocol is not provided. + +### G.3 **non-repudiation of content received request** + +This element of service enables an originating VM-UA to request a recipient VM-UA to provide it with an irrevocable proof that the original subject message content was received by means of a receipt or non-receipt notification. + +*Note* – This element of service requires the “receipt notification request indication or non-receipt notification request indication” to also be requested of this recipient. + +### G.4 **non-repudiation of voice notification** + +This element of service provides the originator of a message with irrevocable proof that the subject message was received by the VM-UA and the subject voice message was accepted, forwarded, refused or that certain requested elements of service were not available to the recipient even though the message was accepted. + +This shall protect against any attempt by the recipient VM-UA to deny subsequently that the message was received and that the subject voice message was not accepted as indicated. This element of service provides the originator with irrevocable proof of the proof-of-voice message-notification. Such proof may be provided by means of the Non-repudiation of Origin security service, defined in Recommendation X.402 (1988), § 10.2.5.1, being applied to the notification. + +*Note* – This service is stronger than the Proof of Voice Notification service. + +### G.5 **non-repudiation of voice notification request** + +This element of service, used in conjunction with non-receipt notification request indication or receipt notification request indication or voice messaging-service status request notification, enables an originating VM-UA to request the responding VM-UA to provide it with irrevocable proof of origin of the Voice notification. + +*Note* – This element of service supersedes the proof of voice notification request and assumes that a request for at least one of the three voice messaging notifications is always present. + +### G.6 proof of content received + +This element of service allows an originating VM-UA to get from a recipient VM-UA proof that the original subject message content was received by the recipient VM-UA and that the subject voice message was accepted, forwarded or refused. + +The corresponding proof is obtained by returning a proof of origin of the voice message notification which incorporates the originator's message origin authentication and/or content integrity arguments, if present, or the complete original message content otherwise. + +### G.7 proof of content received request + +This element of service allows an originating VM-UA to request the recipient VM-UA to provide it with proof that the original subject message content was received by use of voice messaging receipt or non-receipt Notification. + +*Note* – This element of service requires the “receipt notification request indication or non-receipt notification request indication” to also be requested of this recipient. + +### G.8 proof of voice notification + +This element of service allows an originator of a message to obtain the means to corroborate that the subject message was received by the recipient VM-UA and that the voice message was accepted, forwarded or refused. Such corroboration is provided by means of the MTA user-to-message transfer system-user “Message Origin Authentication” security service defined in Recommendation X.402 (1988), § 10.2.1.1.1, being applied to the voice messaging notification. + +### G.9 proof of voice notification request + +This element of service, used in conjunction with non-receipt notification request indication or receipt notification request indication or voice messaging service status request notification, enables an originating VM-UA to request the responding VM-UA to provide it with a corroboration of the source voice notification. + +*Note* – This element of service requires the “voice messaging notification request” to also be present. + +### G.10 New voice messaging security – optional per recipient user facilities elements of service + +Table G-1/F.440 lists the new secure voice messaging elements of service. + +TABLE G-1/F.440 + +| Elements of service | Originator | Recipient | Section reference | +|-----------------------------------------------|------------|-----------|-------------------| +| Non-repudiation of content originated | A | A | G.1 | +| Non-repudiation of content received | A | A | G.2 | +| Non-repudiation of content received request | A | A | G.3 | +| Non-repudiation of voice notification | A | A | G.4 | +| Non-repudiation of voice notification request | A | A | G.5 | +| Proof of content received | A | A | G.6 | +| Proof of content received request | A | A | G.7 | +| Proof of voice notification | A | A | G.8 | +| Proof of voice notification request | A | A | G.9 | + +### G.11 *Voice messaging security – optional per message/recipient user facilities elements of service* + +These elements of service originally defined for IPM in Recommendation F.400 (1988), are applicable to voice messaging on a per message level, see Table G-2/F.440. + +TABLE G-2/F.440 + +| Element of service | Originator | Recipient | Section reference to F.400 | +|-------------------------------|------------|-----------|----------------------------| +| Non-repudiation of delivery | A | A | B.49 | +| Non-repudiation of origin | A | A | B.50 | +| Non-repudiation of submission | A | A | B.51 | +| Proof of delivery | A | A | B.65 | +| Proof of submission | A | A | B.66 | +| Report origin authentication | A | A | B.74 | + +# ANNEX H + +(to Recommendation F.440) + +## **Voice messaging security overview** + +(This annex does not form an integral part of this Recommendation) + +### H.1 *Introduction* + +This annex details the vulnerabilities identified within the Voice Messaging-service environment and the resulting security services required to counter those vulnerabilities. + +This annex is based on the assumption that a VM-service environment may use the secure messaging services as defined in Recommendation F.400. However, where vulnerabilities are not adequately covered by the existing MHS security services, provision has been made in Recommendation X.440 for additional security services in the VM-service environment. + +Some of the security services defined for the VM-service environment are of a generic message handling nature, others are specific to the VME. The security services defined for the VM-service environment are specific to VM and are therefore fully defined in Recommendation X.400. + +### H.2 *Vulnerabilities* + +In most of the areas identified below, there will also be further vulnerabilities and corresponding service considerations at the level of the voice messaging users (VM-users). The security model reflected in this annex assumes that such considerations are outside the scope of this Recommendation. The voice messaging (VM) security model assumes that the VM-user provides adequate security and trusted functionality in the operation of VM sufficient to meet the user's security policy. + +*Note* – In practice this may necessitate co-location of the VM-user and the VM-UA unless a suitably secure environment is established which includes both components. + +The following description of vulnerabilities is based on the threat definitions in Annex D of Recommendation X.402. In addition, it has been considered necessary to examine message loss independently of message sequencing and modification of information, and to take account of further vulnerabilities for VM which are not currently identified in Recommendation X.402. + +An important aspect of the VM environment which is not recognized within the Recommendation X.402 security model is the concept of VM acceptance for messages at each stage of the message path through the MHS environment. + +It is therefore necessary to establish the concept of VM-acceptance domains, which may involve additional consideration of legal issues. One possible division of the VM-service environment into VM acceptance domains is as follows: + +- 1) VM-user environment plus the VM-UA; +- 2) MTS management domain; +- 3) VM-message store (if not co-located with either of the above). + +#### H.2.1 *Masquerade* + +As defined in Recommendation X.402, Annex D. + +#### H.2.2 *Message sequencing* + +As defined in Recommendation X.402, Annex D. + +Users should not assume the voice message shall be delivered in correct sequence. Voice messaging users should be able to recover from duplication and out-of-sequence messages, provided the MHS offers protection against the modification of information while messages are within the MHS environment. + +#### H.2.3 *Message loss* + +Vulnerability to message loss is considered critical in the VMG environment. Two types of message loss are distinguished: + +- a) catastrophic failure of a VM-UA, VM-MS or MTA; +- b) loss of individual message(s). + +VM-users and service providers may need to consider more carefully issues concerning transfer of messages between VM-acceptance domains: + +- a) from the originating VM-UA user domain; +- b) between relaying domains; +- c) to the recipient VM-UA user domain. + +#### H.2.4 *Modification of information* + +As defined in Recommendation X.402, Annex D. + +#### H.2.5 *Denial of service* + +As defined in Recommendation X.402, Annex D. + +#### H.2.6 *Repudiation* + +As defined in Recommendation X.402, Annex D. + +Furthermore repudiation vulnerability in the VM-service environment is considered to be critical. Such vulnerability may be increased by use of certain MHS services (e.g. auto-forwarding, redirection). + +#### H.2.7 *Leakage of information* + +As defined in Recommendation X.402, Annex D. + +#### H.2.8 *Manipulation of information by VM-User* + +The VM community has additionally identified a further vulnerability where the integrity of a message content is altered subsequent to creation of the spoken message (i.e., by either or both of the originating VM-UA and recipient VM-UA). This vulnerability includes manipulation of message content in the originator's local store after non-repudiation of submission and/or manipulation of message content in the recipient's store after non-repudiation of delivery. + +#### H.2.9 *Other vulnerabilities* + +Other vulnerabilities as defined in Recommendation X.402 are considered important such as: + +- 1) misrouting; +- 2) misdelivery (especially important in the context of redirection); +- 3) insider threats; +- 4) receipt of data that the VM-user is not prepared to accept or process. + +### H.3 *Vulnerabilities countered* + +Recommendation X.402, § 10, provides an abstract security model for message transfer. The security model provides a framework for describing security services that counter potential vulnerabilities within the MTS and between MTS-user to MTS-user. VM vulnerabilities may also be countered by security services which are outside the existing model defined in Recommendation X.402. The following text describes how the VM vulnerabilities are countered using Recommendation X.402 security services, enhanced security services defined in Recommendation X.440 and pervasive mechanisms defined in this Recommendation. + +#### H.3.1 *Masquerade* + +The existing MHS-security services which counter this vulnerability are: + +- a) Message Origin Authentication; +- b) Secure Access Management; +- c) Security Labeling; +- d) Proof of Delivery; +- e) Proof of Submission. + +Since voice messaging UA/MS is deemed in the MHS architecture as belonging to one user, it is not considered appropriate to provide selective access control for the various operations that may be performed on a VM-MS. However, there is a requirement for security audit trail to record the actions of the VM-user. + +In this Recommendation, such security audit trails are expected to be implemented as pervasive mechanisms (the term pervasive mechanism is defined in ISO 7498-2). Protocols to support audit capability may be the subject of future standardization. + +#### H.3.2 *Message sequencing* + +The existing MHS-security service which counters this vulnerability is: Message Sequence Integrity. This security service has limited effect as it is based on the provision of an integer by the originating VM-UA with no assurance as to uniqueness or consecutiveness. It is considered that the MHS environment should not be required to ensure message sequence integrity, but should support detection of sequence integrity failure (by additional provision of audit/logging facilities and/or the provision of third party notary services). In this Recommendation it is considered the responsibility of the VM-user to recover from sequence errors and message duplication. + +#### H.3.3 *Message loss* + +Message loss could occur potentially over any peer-to-peer communications link (e.g., by deliberate malicious act), or by the failure or incorrect behavior (whether by malicious intent or otherwise) of any MHS component (VM-UA, VM-MS, MTA). The following categories of message loss are distinguished: + +- 1) catastrophic message loss (i.e., failure of a component); +- 2) malicious loss of individual messages in the VM-MS; +- 3) accidental loss of individual messages in the VM-MS; +- 4) MTS-message loss. + +##### H.3.3.1 *Catastrophic failure* + +Failure of the VM-UA is outside the scope of this Recommendation. + +Failure of the VM-MS is potentially catastrophic and desirably needs some protection, at least in terms of detection. This should be provided by an off-line archive to hold all submitted and delivered messages. In this Recommendation detection and recovery from message loss using such archive mechanisms is a local matter. + +Failure of any component in the MTS may similarly be catastrophic and can again be protected by off-line archive of messages. As for the message store, detection and recovery from message loss using such archive mechanisms in the MTS is a local matter and outside the scope of this Recommendation. + +##### H.3.3.2 *VM-MS specific message loss* + +Loss of individual messages in the message store – whether malicious or accidental – shall require the provision of a secure audit trail to enable detection of such loss. Such a service may need to be provided to the VM-user and to VM-MS management. In this Recommendation, secure VM-MS audit trail could be realized as a pervasive mechanism and is a local issue. Protocol to support an audit trail may be the subject of future standardization. + +##### H.3.3.3 *MTS specific message loss* + +Loss of individual messages in the MTS (whether malicious or accidental) shall also require the provision of a secure audit trail to enable detection of such loss. Such a mechanism would need to be provided on a per-MTA and a per-MD basis depending on security policy in force. A secure MTA/MTS audit trail could be realized as a pervasive mechanism and is a local issue. The protocol to support an audit trail may be the subject of future standardization. + +##### H.3.3.4 *End-to-end message loss* + +The following description assumes that the functionality of the VM-UA (including any associated components to meet such functionality – e.g., encryption devices), is trusted. + +The existing “Message Sequence Integrity” service does not guarantee detection of message loss, since it relies on the provision of an integer value by the originating VM-UA. In practice, effective operation of this service may be achieved with a common code of practice between VM users which is outside the scope of this Recommendation. + +As a result, MHS services which may provide an indication of message loss are confined to services offered to the originating VM-user. Whereas the existing Proof of Submission and deliver services provide some degree of confidence that the message has not been lost, they do not operate end-to-end. In particular they do not take account of the scenario where the recipient VM-UA and VM-MS are not co-located. + +There is, therefore, a requirement for a proof of receipt (i.e., by the recipient VM-UA) service. This capability is realized by the user requesting a voice messaging notification (VN) which may be secured. The VN indicating the status of VM as accepted, forwarded or refused includes elements which associates the notification with the subject message. + +In a VM environment, proof of receipt may therefore be provided by signing the VN-service using the existing MTS security elements. In particular, the VM-UA to VM-UA security service of Message Origin Authentication may be used to sign the VN on submission of the VN to the MTS. In this Recommendation the requirement for proof of receipt may be implemented by a trusted form of VN in the VM environment. + +*Note* – This service is called Proof of VM Notification and/or Non-repudiation of VM Notification in VM, depending on the strength of the mechanism provided. + +The MTS mechanism used on message submission to provide this service is defined as the MTS submission abstract operation in Recommendation X.411, § 8.2.1.1.1.28 *Content-integrity-check*. In this instance the message content is the VN. Proof of association between the subject message and replying VN is provided by subject message VM identifier and if included in the subject message the message content-integrity-check argument. + +#### H.3.4 *Modification of information* + +The existing MHS security services which counter this vulnerability are: + +- Connection Integrity; +- Content Integrity. + +These security services provide sufficient protection against modification of message content. It is also noted that use of double enveloping (i.e. with encrypted checksum on outer envelope) may provide additional protection. + +*Note* – VM-UAs are trusted entities in terms of content integrity. + +#### H.3.5 *Denial of service* + +This is a very important vulnerability for VM-users, but is outside the scope of this Recommendation. + +#### H.3.6 *Repudiation* + +Services which offer protection against repudiation in the VM environment are fundamentally concerned with formalizing the forwarding of a VM. The security services as defined in Recommendation X.402 are: + +- a) Non-repudiation of Origin; +- b) Non-repudiation of Submission; +- c) Non-repudiation of Delivery. + +These security services only cover some areas of transfer between VM-Acceptance domains, which may be of significance in a VM environment. Areas which are not covered by security services provided in 1988 for message handling include: + +- between VM-user domains (i.e., end-to-end); +- between MTS management domains; +- between a VM-MS and a recipient VM-UA. + +Therefore, services and/or pervasive mechanisms defined in this Recommendation covering the above deficiencies are: + +- non-repudiation/Proof of Transfer; +- non-repudiation/Proof of Retrieval; +- non-repudiation/Proof of VM Notification; +- non-repudiation/Proof of Content. + +“Non-repudiation/Proof of Transfer” counters the vulnerability of repudiation of responsibility between MTA and/or management domains. VM environments may provide such a service using an additional pervasive mechanisms, such as security logs and archives within MTA and/or MTS boundaries. Such pervasive mechanisms provide a “secure MT audit trail” to record the message details and trace information. + +Non-repudiation/Proof of Retrieval counters the vulnerability of repudiation of responsibility of a message between a UA and an MS. VM environments may provide such a service using an additional pervasive mechanisms, such as security logs and archives within VM-MSs. Such pervasive mechanisms provide a “secure VM-MS audit trail” to record VM-user actions in the VM-message store. + +Non-repudiation/Proof of VM Notification counters the vulnerability of repudiation of a voice messaging notification for VM-UA to VM-UA. This service is specific to VM and a complete solution is included in this Recommendation. This vulnerability may be especially relevant in the case of VM forwarding, redirection, etc., in addition to the scenario of delivery to an untrusted VM-message store. + +Two mechanisms have been defined for non-repudiation of VM notifications: the first uses the trusted VN as described above, the second uses an external notary systems. Only the trusted VN was fully defined in this Recommendation. External notary systems may be the subject of future standardization. + +Non-repudiation/Proof of Content counters the vulnerability of manipulation of information by the VM-user after the message has been received by the VM-UA. Although such vulnerability is outside the MHS environment, the MHS environment may provide assistance in terms of trusted return of content and notarization services. There are several ways this requirement may be supported, using the secure messaging environment based on the security services provided in 1988. + +Firstly, non-repudiation of content by the originating VM-UA may be provided by the existing Non-repudiation of Origin security service. + +Secondly, non-repudiation of content by the recipient VM-UA may be provided by returning the subject content within the VM notification and submitting the voice messaging notification to the MTS using the Non-repudiation of Origin security service. + +Thirdly, by notarization services, such services may be achieved by forwarding messages via a mutually trusted third-party notary (i.e., using existing MHS-security services). + +All three approaches would thus require no modification to the secure messaging environment based on the existing MHS Recommendations. + +*Note* – Non-repudiation services (which may imply the involvement of a third party) are considered stronger than “proof-of” services. + +#### H.3.7 *Leakage of information* + +The existing MHS-security services which counter this vulnerability are: + +- Connection Confidentiality; +- Content Confidentiality; +- Secure Access Management; +- Message Flow Confidentiality. + +These security services provide sufficient protection against leakage of message content. It is also noted that use of double enveloping could provide some protection against traffic analysis. Traffic padding is outside the scope of this area of work. + +*Note* – UAs are trusted entities in terms of content and message flow confidentiality. + +#### H.3.8 *Manipulation of information by VM-user* + +Manipulation of information by the VM-user may be countered by use of the “Non-repudiation of content” security service. + +#### H.3.9 *Other vulnerabilities* + +The use of “security access management” and “security labeling” to counter all other vulnerabilities is also applicable in the VM environment. In addition, there is a requirement for auditing and accountability which is likely to require at least a “secure audit trail”; this may be provided by a pervasive mechanism as a local matter. + +#### H.3.10 *Other VM-user vulnerabilities* + +Within the VM environment the VM-user may be vulnerable to security threats. To counter these vulnerabilities the VM-user may wish to generate its own security services and mechanisms (such as, signatures from one VM-user to another). These VM-user security services are conveyed in VM security fields as purely information conveying elements of services within the VM environment and may consequently be used for several end-to-end services including message recovery and non-repudiation. It is a local issue to determine how the VM-user security services are used. + +#### H.3.11 *Summary* + +This annex identifies VM vulnerabilities and the security services necessary to counter those vulnerabilities using the MHS specification of 1988, then specifies the corresponding security elements required. + +VM may provide additional pervasive mechanisms as follows: + +- secure VM-MS audit trail; +- secure MT-audit trail; +- VM-MS archive; +- MD archive; +- security of MTA management and routing information. + +This Recommendation currently allows the use of both standard symmetric and standard asymmetric tokens. The use of trusted notary systems may be the subject of future standardization. + +### H.4 *Additional pervasive mechanisms* + +#### H.4.1 *Secure VM-MS audit trail* + +This facility would monitor and record VM-UA actions on the message store. It would also provide support for “proof of retrieval”. + +It is strongly recommended that “secure VM-MS audit trail” be controlled via a secure link or other secure local means to protect against masquerade. In this Recommendation “secure VM-MS audit trail” may only be realized as a pervasive mechanism. The pervasive mechanisms mentioned may be the subject of future standardization. + +#### H.4.2 *Secure MT-audit trail* + +This facility would monitor and record all MTA actions. It would also provide additional support for: “proof of submission”, “proof of transfer”, “proof of delivery”, security of the administration of the MTA. In this Recommendation, secure MT-audit trail may be realized as a pervasive mechanism. + +#### H.4.3 *VM-MS archive* + +This mechanism is potentially useful for providing recovery from MS failure, i.e. by providing a secure off-line archive of all submitted and delivered messages. Detection and recovery from message loss using such archive mechanism is a local matter. + +#### H.4.4 *MT archive* + +This mechanism is potentially useful for providing recovery from MTA failure, i.e. by providing a secure off-line archive of all messages. Detection and recovery from message loss using such archive mechanism is a local matter. \ No newline at end of file diff --git a/marked/F/T-REC-F.472-199708-I_PDF-E/raw.md b/marked/F/T-REC-F.472-199708-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..425ef5fdf8585c80ad03c25414913206b67707ab --- /dev/null +++ b/marked/F/T-REC-F.472-199708-I_PDF-E/raw.md @@ -0,0 +1,619 @@ + + +![ITU logo](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +The logo of the International Telecommunication Union (ITU) features the letters 'ITU' in a bold, sans-serif font, superimposed on a stylized globe with intersecting lines. + +ITU logo + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.472** + +(08/97) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Message handling services + +--- + +**Service and operational requirements of the +voice-mail store-and-forward service** + +ITU-T Recommendation F.472 + +(Previously CCITT Recommendation) + +--- + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1–F.109 | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | F.110–F.159 | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | F.160–F.399 | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | F.550–F.599 | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| AUDIOVISUAL SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to ITU-T List of Recommendations.* + +# **ITU-T RECOMMENDATION F.472** + +# **SERVICE AND OPERATIONAL REQUIREMENTS OF THE VOICE-MAIL STORE-AND-FORWARD SERVICE** + +## **Summary** + +This Recommendation defines the general principles and basic operational requirements of the voice-mail store-and-forward service provided by service providers using computer-controlled voice-mail systems and Voice-mail Store-and-Forward Units (Voice SFU). Although voice-mail system access will be, predominantly, from within the same country in which the Voice SFU is situated, this Recommendation sets out the minimum service requirements that should be fulfilled to allow international extension of the service. International extension of this service is achieved by the interconnection with one or more other Voice SFUs, across national boundaries using automated transit facilities for the purpose of transfer and delivery of voice-mail messages. + +## **Source** + +ITU-T Recommendation F.472 was prepared by ITU-T Study Group 7 (1997-2000) and was approved under the WTSC Resolution No. 1 procedure on the 9th of August 1997. + +## FOREWORD + +ITU (International Telecommunication Union) is the United Nations Specialized Agency in the field of telecommunications. The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of the ITU. The ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Conference (WTSC), which meets every four years, establishes the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics. + +The approval of Recommendations by the Members of the ITU-T is covered by the procedure laid down in WTSC Resolution No. 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +## INTELLECTUAL PROPERTY RIGHTS + +The ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. The ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, the ITU had/had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementors are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database. + +© ITU 1997 + +All rights reserved. No part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and microfilm, without permission in writing from the ITU. + +## CONTENTS + +| | | Page | +|---------|--------------------------------------------------------------------------|-------------| +| 1 | Scope ..... | 1 | +| 2 | References ..... | 1 | +| 3 | Abbreviations ..... | 1 | +| 3.1 | Conventions ..... | 2 | +| 4 | Definitions ..... | 2 | +| 5 | Introduction ..... | 3 | +| 5.1 | Interaction between the VMS and its direct user ..... | 3 | +| 5.2 | Interaction between a Voice SFU and its direct user (i.e. VMS) ..... | 4 | +| 5.3 | Interaction between Voice SFUs ..... | 4 | +| 5.4 | Interaction between a Voice SFU and direct users of other services ..... | 4 | +| 6 | Service description ..... | 4 | +| 6.1 | Technical requirements ..... | 4 | +| 6.2 | Accounting and settlement ..... | 4 | +| 6.3 | Operational aspects ..... | 5 | +| 6.4 | Interworking ..... | 5 | +| 6.5 | Message submission ..... | 5 | +| 6.6 | Access to the service ..... | 5 | +| 6.7 | Service interconnection ..... | 5 | +| 7 | Service elements ..... | 5 | +| 7.1 | Voice message format service elements ..... | 5 | +| 7.1.1 | Message deferral services ..... | 5 | +| 7.1.2 | Voice message identification ..... | 6 | +| 7.2 | Classification of VM S&F service elements ..... | 6 | +| 7.2.1 | Access management ..... | 7 | +| 7.2.2 | Message recipient ..... | 7 | +| 7.2.3 | Classes of message delivery ..... | 7 | +| 7.2.4 | Voice-mail notifications ..... | 7 | +| 7.2.5 | Voice message encoding conversion ..... | 9 | +| 8 | Access procedures ..... | 9 | +| 9 | Message delivery procedures ..... | 9 | +| 9.1 | Automatic call retry ..... | 10 | +| 9.2 | Call records ..... | 10 | +| 9.3 | Recovery from interrupted message delivery ..... | 10 | +| 10 | Quality of service ..... | 10 | +| 10.1 | Message loss or failure ..... | 10 | +| 10.2 | Quality of service targets ..... | 10 | +| 10.3 | Storage capacity ..... | 11 | +| 10.4 | Public network circuits ..... | 11 | +| 10.4.1 | Provision ..... | 11 | +| 10.4.2 | Performance ..... | 11 | +| 10.5 | Duration of the service ..... | 11 | +| 10.6 | Inquiries and complaints ..... | 11 | +| Annex A | – Format element of service definitions ..... | 12 | + + + +# SERVICE AND OPERATIONAL REQUIREMENTS OF THE VOICE-MAIL STORE-AND-FORWARD SERVICE + +(Geneva, 1997) + +# 1 Scope + +This Recommendation defines the general principles and basic operational requirements of the voice-mail store-and-forward service provided by service providers using computer-controlled voice-mail systems and Voice-mail Store-and-Forward Units (Voice SFU). Although voice-mail system access will be, predominantly from within the same country in which the Voice SFU is situated, this Recommendation sets out the minimum service requirements that should be fulfilled to allow international extension of the service. International extension of this service is achieved by the interconnection with one or more other Voice SFUs, across national boundaries using automated transit facilities for the purpose of transfer and delivery of voice-mail messages. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; all users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. + +- ITU-T Recommendation F.471 (1997), *Operational requirements for the interconnection of voice-mail store-and-forward units*. +- ITU-T Recommendation F.400/X.400 (1996), *Message handling: System and service overview*. +- CCITT Recommendation F.440 (1992), *The voice messaging service*. +- CCITT Recommendation X.440 (1992), *Message handling systems: Voice messaging system*. +- CCITT Recommendation G.721 (1990) (See ITU-T Recommendation G.726.) +- CCITT Recommendation G.726 (1990) (Has replaced former CCITT Recs. G.721 and G.723), *40, 32, 24, 16 kbit/s Adaptive Differential Pulse Code Modulation (ADPCM)*. +- ITU-T Recommendation G.728 – Annex G (1994), *16 kbit/s fixed point specification*. + +## 3 Abbreviations + +This Recommendation uses the following abbreviations: + +| | | +|--------|---------------------------------------------| +| ADPCM | Adaptive Differential Pulse Code Modulation | +| DEST | Destination | +| EOS | Elements of Service | +| Rec. | Recommendation | +| SFU | Store-and-Forward Unit | +| VMS | Voice-mail System | +| VM S&F | Voice-mail Store-and-Forward | + +## 3.1 Conventions + +The classifications indicated in the tables use the following conventions: + +| | | | +|---|--------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| B | BASIC service | On origination, it is automatically provided by the service on behalf of the user, and on receipt it is always available for presentation to a recipient. | +| E | ESSENTIAL service | The user is always provided a mechanism to set a value (or override a default value) on origination and on receipt it is always available for presentation to a recipient. | +| A | ADDITIONAL service | The VMS may provide the user with a means to set a value (or override a default) on origination and on receipt the service may be available for presentation to a recipient. The choice of its availability is that of the service provider. | + +# 4 Definitions + +This Recommendation defines the terms listed below. + +NOTE – Use of the terms "delivery", "submission", "transfer", does not imply any specific protocol implementation. + +**4.1 message delivery:** Message delivery is defined as the transfer of voice-mail messages, including responsibility for the message from a Voice SFU to a voice-mail system for the recipient named on the subject message. + +NOTE – In the context of this service, the acceptance of responsibility for the message by the voice-mail system constitutes delivery. As a local matter, the voice-mail system may choose to accept responsibility for the message only if it has been deposited into the recipient's mailbox. + +**4.2 message submission:** Message submission is defined as the transfer of voice-mail messages and responsibility for the message from the VMS to its Voice SFU. + +**4.3 message transfer:** Message transfer is defined as the transfer of voice-mail messages and responsibility for the message from one Voice SFU to another Voice SFU via a transfer system. + +**4.4 recipient address:** A recipient address is defined as the ultimate destination of a message. A recipient address may designate a VMS user, a voice-mail terminal or a storage device empowered to receive messages for a VMS recipient. A recipient address may identify a single user or a distribution list. + +**4.5 transfer system:** A transfer system is defined as a system which accomplishes the transfer of voice-mail messages. A transfer system may or may not provide end user services. + +**4.6 transit facility:** A transit facility is either a Voice SFU or a transfer system. + +**4.7 voice-mail system:** The system used to collect/deliver voice messages from/to direct users and subscribers. The voice-mail system may be associated with a Voice SFU. This system consumes the services supplied by its Voice SFU and consumed by its users as described in this Recommendation. + +**4.8 voice-mail system interface to voice SFU:** The service interaction uses an analogue or digital interface between the VMS and its corresponding Voice SFU. + +NOTE – The realization of this interface is a local matter but may be the subject of future standardization. + +**4.9 voice SFU:** A Voice SFU identifies the boundary within which the VM S&F service is provided. An administration may control one or more voice-mail store-and-forward units. A Voice SFU is a main component of a VM S&F domain which interacts between a VMS and a transfer system. A Voice SFU must provide the functions of message submission, message delivery to its direct user (a VMS), and in the case of interconnection, consumes the services provided by a transfer service and its correspondent Voice SFU as identified in Recommendation F.471. + +**4.10 voice signature:** The voice signature (e.g. voice back) is a representation of recipient identifier in voice encoding form. It is of variable duration but may support ten seconds duration if required. Voice signature is a networking capability that allows the recipient's name to be spoken when sending or receiving networking messages. Its value is used by the receiving service for rendering by or to the intended recipient associated with the receiving voice-mail system. It is not intended for routing the message between Voice SFUs. It may be associated with a recipient address or an originator address, which constitutes the routeable information for intercommunication between Voice SFUs. The encoding of the information conveyed as a voice signature shall be that of the voice object contained in the subject message. If multiple objects are contained in a single voice-mail message, and these objects contain different encodings, then the encoding of the voice signature is conveyed in the message as part of the protocol that realizes this service. This voice signature element of service shall comply with the voice encodings identified in 7.3/F.471. + +NOTE – Signature in this context does not mean a security code. + +**4.11 messages:** Messages are the voice encoded objects exchanged between voice-mail systems. Voice encoded objects contain a digital representation of the originator's spoken, pre-recorded, or otherwise audio generated message. Examples of voice encoded objects may include, spoken language, high fidelity sound, etc. When the voice encoded object is delivered to the intended recipient, irrespective of any transcoding that might occur, it shall retain the same semantics and sound fidelity as the standard of the originator's recording device. + +**4.12 supporting data:** The information necessary for messages to be exchanged between users of voice-mail systems. The primary use of these data is to enable pre-processing of the subject voice encoded object(s) for proper rendering to the intended recipient. Supporting data may include encoding algorithm identifier and corresponding parameters needed for decoding, message duration, subject information, security parameters, etc. + +**4.13 voice-mail message:** The voice-mail message format comprises the message, supporting data, and any other voice-mail message pre-processing details. Abstractly, it consists of an envelope and the envelope contents which is characterized by the elements of service identified in this Recommendation. + +# 5 Introduction + +The development and growth of disparate voice store-and-forward equipment and the potential for the international interconnection of such equipment has created the need for a common set of service elements and operational procedures to meet the needs of users. The VM S&F service defined in this Recommendation is designed to meet these needs. As such, this Recommendation defines the VM S&F service to enable the international interconnection of disparate VMS equipment. To facilitate international interoperability, this Recommendation, along with Recommendation F.471, identifies or defines features and procedures for the service in order to promote and assist in the provisioning of such services in many countries. + +The VM S&F is comprised of one or more domains where a domain is a logical entity that incorporates at least a single Voice SFU and one or more VMSS, though not indicated in Figure 1. + +Figure 1 identifies the scope of both this Recommendation and Recommendation F.471. Furthermore, it identifies the following four (4) points of interaction. + +## 5.1 Interaction between the VMS and its direct user + +The interaction point between the VMS and its direct user is identified as "1" in Figure 1. The direct user consumes the services of the VMS identified in this Recommendation. The concrete realization of support for these services by a VMS is considered to be a local matter. + +![Diagram of VM S&F services architecture showing interactions between Users, VMS, Voice SFUs, and Other services.](a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg) + +The diagram illustrates the architecture of VM S&F services. At the top, five 'User' ovals are shown. The first three users are connected to two 'VMS' boxes, and the last two users are connected to a third 'VMS' box. These connections are labeled '1'. Below the VMS boxes is a large dashed rectangle labeled 'F.471' containing three 'Voice SFU' boxes. The top two Voice SFUs are connected to each other and to the third Voice SFU at the bottom, with these internal connections labeled '3'. Above the top two Voice SFUs, a dashed line labeled '4' connects them to an oval labeled 'Other services'. Below the 'F.471' box is another dashed rectangle labeled 'F.472', which contains a 'VMS' box. This 'VMS' box is connected to the bottom Voice SFU (labeled '2') and to a 'User' oval at the bottom (labeled '1'). The text 'T0727070-96/d01' is located to the right of the 'F.472' box. + +Diagram of VM S&F services architecture showing interactions between Users, VMS, Voice SFUs, and Other services. + +Figure 1/F.472 – VM S&F services + +## 5.2 Interaction between a Voice SFU and its direct user (i.e. VMS) + +The interaction point between a Voice SFU and its direct user is identified as "2" in Figure 1. The direct user consumes the services of a Voice SFU identified in this Recommendation. The concrete realization of support for these services by a Voice SFU is considered to be a local matter. + +## 5.3 Interaction between Voice SFUs + +The interaction point between Voice SFUs is identified as "3" in Figure 1. It is understood that the intercommunication of Voice SFUs employs a transfer mechanism, such as the message transfer service specified in Recommendation F.410. A Voice SFU consumes transfer services as well as the services provided by its correspondent Voice SFU as identified in Recommendation F.471. The concrete realization of support for these services by a Voice SFU is considered to be a local matter. + +## 5.4 Interaction between a Voice SFU and direct users of other services + +The interaction point between a Voice SFU and an indirect user who is a direct user of another service is identified as "4" in Figure 1. The indirect user through its association with other services may consume all or some of the services identified in this Recommendation for direct users. The concrete realization of support for these services may be the subject of future standardization. + +# 6 Service description + +## 6.1 Technical requirements + +Technical requirements of the service are covered in other publications, e.g. in the relevant G- and X-Series Recommendations. + +## 6.2 Accounting and settlement + +Accounting and settlement aspects are covered in D-Series Recommendations. Voice SFUs shall record and store sufficient information for charging and settlement of accounts, when applicable. + +## 6.3 Operational aspects + +The operational aspects of the interconnection of Voice SFUs are covered in Recommendation F.471. Quality of service aspects for VM S&F are covered in clause 10. + +## 6.4 Interworking + +The interworking between the VM S&F service and other ITU-T defined services is for further study. + +## 6.5 Message submission + +Users submit messages to a VMS which in turn submits messages to the Voice SFU for delivery to one or more recipients. The VM S&F service provides a range of service elements which may be used by the VMS and its users for message management purposes, for example: + +- Service elements such as class of message delivery and deferred delivery request allow the VMS to specify how messages are handled in their transmission from source to destination. +- Other service elements, such as voice-mail notifications, may provide the VMS with more extensive message statistics than are generally available in a real-time service, e.g. see clause 7/F.471. + +The service shall accept messages submitted in an agreed voice message format. Acceptance of messages in other formats is *optional* and subject to bilateral agreement. + +## 6.6 Access to the service + +The VM S&F service may be accessed by a variety of network access devices, such as telephone hand sets or computer oriented terminal devices over a variety of networking technologies. User access to a VM S&F service will be predominantly through a proprietary VMS connected to its associated national public networks. Access may also be provided by local mechanisms defined independent of public networks. + +## 6.7 Service interconnection + +Where the service is provided by means of the interconnection by two or more service providers interworking between Voice SFUs is as defined in Recommendation F.471. Interworking using the voice message encoding defined for 32 kbit/s in Recommendation G.726 is *mandatory*. Interworking using other encodings such as G.728, or proprietary encodings is *optional* and subject to bilateral agreements. + +# 7 Service elements + +## 7.1 Voice message format service elements + +The provision of service elements by a VMS is a local matter and may vary between service providers. However, service providers extending their service across international boundaries by interconnection with one or more Voice SFUs should regard support for the voice message format service elements defined in Annex A as indicated in Table 1. + +### 7.1.1 Message deferral services + +Message deferral relates to two elements of service, *deferred delivery request* and *cancel deferred delivery request*. How message deferral services are concretely realized is a local matter. + +Message submission may be deferred by two mechanisms in context of the *Deferred Delivery Request element of service*: + +- a) Deferred to a time specified by the originator. +- b) Deferred to a time determined by the service provider. + +NOTE 1 – In a) the voice-mail system may choose any of the classes of delivery supported to take effect at the specified time. + +NOTE 2 – Deferral b) is only provided when a voice-mail system has selected the non-urgent class of delivery. + +**Table 1/F.472 – Voice message format service elements** + +| Element of service | Originate | Receive | EOS | +|----------------------------------|-----------|---------|-----| +| Cancel deferred delivery request | A | N/A | A.1 | +| Class of message delivery | E | B | A.2 | +| Deferred delivery request | A | N/A | A.3 | +| Message identification | B | B | A.4 | +| Message recipient | E | B | A.5 | +| Originator identification | B | B | A.6 | + +### 7.1.2 Voice message identification + +This element of service enables cooperating voice messaging user agents to convey a globally unique identifier to each voice message sent or received. The voice message identifier is composed of an originator/recipient name of the originator and an identifier that is unique with respect to that name. A voice messaging user agent and user use this identifier, optionally together with message creation time, to refer to a previously sent or received voice message (for example, in receipt notifications.) + +## 7.2 Classification of VM S&F service elements + +Additionally, VM S&F service feature considerations for international communication are also characterized as service elements and are explained in 7.2 and Table 2. In this table the inference of B, E and A is slightly different as the classification implies both origination and receipt. + +**Table 2/F.472 – Classification of VM S&F service elements** + +| Service elements | Classification | Subclause reference | +|-------------------------------------------------------------------|----------------|---------------------| +| Access management | B | 7.2.1 | +| Message recipients: | E | 7.2.2 | +| • Single address message; | E | 7.2.2.1 | +| • Multiple address message; | A | 7.2.2.2 | +| • Distribution list identifier. | A | 7.2.2.3 | +| Class of message delivery | A | 7.2.3 | +| Voice-mail notifications: | A | 7.2.4 | +| • Successful message transfer notifications (positive Level 1); | B | 7.2.4.1 | +| • Unsuccessful message transfer notifications (negative Level 1); | B | 7.2.4.2 | +| • Successful message delivery notification (positive Level 2); | B | 7.2.4.3 | +| • Unsuccessful message delivery notification (negative Level 2). | B | 7.2.4.4 | +| Voice encoding conversion | A | 7.2.5 | + +### **7.2.1 Access management** + +Access management is the responsibility of the originating voice-mail system. It includes the identification and validation that the direct user of the voice-mail system is authorized to use its services. This may be achieved by passwords where pre-registration is involved. In cases where pre-registration is not employed, the identification of the originating voice-mail system may be achieved through the transfer of automatic number identification data from the access network to the Voice SFU. Where passwords are used, it should be possible for the voice-mail system to change the password periodically. The exact procedures for security of access should be a local matter. + +### **7.2.2 Message recipient** + +The message recipient element of service requires service providers to provide a mechanism for users to identify the intended recipient(s) for a message. The recipient identifier conveys the identity of the recipient in a routeable representation, i.e. the address, and optionally a voice signature. The method of specifying the recipient address is a local matter. The form of this address may include an E.164 number and/or a domain component. Voice-mail messages may be addressed to a single recipient, multiple recipient or distribution list. + +The Voice SFU supporting the originator may need to convert the canonical form of the address as created on the originating voice-mail system to one of a form used by the transit service for intercommunication between Voice SFUs. The receiving Voice SFU may need to convert the received address into the canonical form used by the voice-mail system. In such cases, the recipient address as created in the originating voice-mail system shall be preserved and returned in any notifications related to the subject message. + +##### **7.2.2.1 Single address message** + +A single address message is a message with one recipient address and optionally as voice signature. + +##### **7.2.2.2 Multiple address message** + +A multiple address message is a message which is addressed to more than one recipient. VM S&F service shall be able to accommodate a single message addressed to at least 256 recipients. + +##### **7.2.2.3 Distribution list identifier** + +A distribution list identifier is used to reference a pre-registered list of recipient addresses. A voice-mail system may use a distribution list identifier at the time of message submission. Distribution list may be supported by the originating or receiving voice-mail system. + +NOTE – This Recommendation does not assume that there is any distinguishable difference between the name of a single recipient address format and the name of a distribution list. + +#### **7.2.3 Classes of message delivery** + +Two classes of message delivery are defined for this service: + +- a) urgent; +- b) routine. + +The process by which the class of message delivery is selected should be a local matter. Quality of service objectives for the above classes of delivery are shown in 10.2, Table 3. + +#### **7.2.4 Voice-mail notifications** + +Message status information is exchanged in VM S&F Level 1 (message transfer notification) and Level 2 (message delivery notification) notifications whose EOSs are described in Recommendation F.471 and Annex A. The information provided by these two notifications are described in 7.2.4.1 through 7.2.4.4. Both Level 1 and 2 notifications shall be generated on a per recipient basis. + +The presentation of the content of Levels 1 and 2 notifications and the handling of multiple address reports is a local matter. The method of delivery and alert of an existence of a notification is a local matter, e.g. forced – voice message or queried – delivery. Message submission is employed for Level 2 notifications. + +##### 7.2.4.1 Successful message transfer notification (positive Level 1 notification) + +The **successful message transfer notification** reports to the originator the successful transfer and acceptance of responsibility of a message to the receiving Voice SFU associated with an intended recipient. Depending upon the message acceptance policy of the destination VMS: + +- the notification may be automatically generated immediately on *message delivery* by the accepting Voice SFU on behalf of its associated VMS based on some locally determined criteria; or +- the notification may be automatically generated after the VMS associated with the receiving Voice SFU has accepted responsibility for the subject message based on some locally determined criteria. + +In either case, it conveys the semantics that responsibility for the message has been transferred to the receiving Voice SFU. + +Level 1 successful transfer notifications shall provide at least the following information: + +- message reference; +- recipient address; +- time of delivery to recipient address. + +NOTE – Provision of time zone indication is a local matter although UTC is recommended. + +##### 7.2.4.2 Unsuccessful message transfer notification (negative Level 1 notification) + +The **unsuccessful message transfer notification** reports to the originator the failure to transfer a message to the receiving Voice SFU associated with an intended recipient. This notification is automatically generated by the service if the Voice SFU fails to accept responsibility for the subject message. Depending upon the message acceptance policy of the receiving VMS: + +- 1) the notification may be automatically generated immediately on *message delivery* by the rejecting Voice SFU (and its associated transfer system) on behalf of its associated VMS based on some locally determined criteria; or +- 2) the notification may be automatically generated after the VMS associated with the receiving Voice SFU signals non-acceptance of responsibility for the subject message based on some locally determined criteria. + +Level 1 unsuccessful delivery notifications shall provide at least the following information: + +- message identification; +- notification originator identification; +- original message recipient indication (as submitted by the originator); +- date and time of failure; +- failure reason and diagnostic information. + +NOTE 1 – Provision of time zone indication is a local matter, although UTC is recommended. + +NOTE 2 – Unsuccessful message transfer notifications do not need to be requested by the originator because the service always generates one on a per recipient basis if delivery to the destination Voice SFU fails. + +NOTE 3 – The service is provided by the transfer system associated with the destination Voice SFU. + +##### 7.2.4.3 Successful message delivery notification (positive Level 2 notification) + +A **successful message delivery notification** reports to the originator the service's success in delivering a message from the Voice SFU to the voice-mail system which serves the intended recipient as further qualified in 7.2.2/F.471. This notification is only generated if requested by the originator. The report to the service may provide **at least** the following information: + +- message identification; +- notification originator identification; +- original intended message recipient; +- receiving terminal identification; + +- message duration; +- delivery duration; +- date and time of responsibility acceptance. + +NOTE 1 – Provision of time zone indication is a local matter although UTC is recommended. + +NOTE 2 – Receiving terminal identification, message duration, and delivery duration are as defined in Recommendation F.440. + +##### 7.2.4.4 Unsuccessful message delivery notification (negative Level 2 notification) + +A **negative message acceptance notification** reports to the originator the failure to deliver a message from the Voice SFU to the voice-mail system which serves the intended recipient. The notification is always generated by the Voice SFU whenever the destination VMS fails to accept responsibility for the subject message. The local message acceptance policy is used to determine the basis for the rejection. For example, possible reasons for not accepting responsibility for the message is the message's address does not actually identify a user of that VMS, or the voice-mail system cannot place the message into the recipient's mailbox, e.g. box full or locked, or the message is forwarded to another user, etc. + +Level 2 notifications shall provide at least the following information: + +- message identification; +- notification originator identification; +- original intended message recipient (as submitted by the originator); +- date and time of failure; +- failure reason and diagnostic information; +- time last attempted; +- number of attempts; +- delivered message duration. + +NOTE 1 – Provision of time zone indication is a local matter, although UTC is recommended. + +NOTE 2 – Negative Level 2 notifications do not need to be requested by the originator because the service always generates one on a per recipient basis if delivery to the destination Voice SFU fails. + +NOTE 3 – *Time last attempted, number of attempts, and delivered message duration* are as defined in Recommendation F.440. + +### 7.2.5 Voice message encoding conversion + +The conversation service facility enables a message to be converted from one speech encoding to another, e.g. G.711 to G.726 conversion. If a Voice SFU supports speech encoding other than 32K ADPCM defined in Recommendation G.726, then it shall support conversion between Recommendation G.726 and that format. The conversion should be an automatic facility and should not require additional procedures for users. + +NOTE – Some conversions may result in a loss of information and degrade intelligibility of the message. Such conversions should be avoided. For example, Recommendation G.728 (16K LD-CELP) to G.726 (32K ADPCM) may result in as loss of information. + +# 8 Access procedures + +Access procedures are a local matter. However, the service provider should make provision for security of access, message recipient indication delivery information and message submission integrity. + +# 9 Message delivery procedures + +The following procedures are provided for provisioning the message delivery functionality. + +## 9.1 Automatic call retry + +When a Voice SFU attempts to deliver a message to the destination voice-mail system, procedures for automatic call retry on system or terminal busy, ringing no reply or incomplete calls (subject to national policy and interworking agreements) should be provided. Where the service is extended across national boundaries for the delivery of messages, then the target delivery times specified in clause 10 should be considered when setting the call retry periods. + +## 9.2 Call records + +A call record that captures the successful or unsuccessful acceptance of responsibility of the subject message by recipient should be maintained by the voice-mail system. The call record information should include, as a minimum, the relevant information provided by notifications. The storage and format of call records is a local matter. + +NOTE – The intended purpose for call records is accounting and settlement. + +## 9.3 Recovery from interrupted message delivery + +If the transfer of a voice-mail message or notification is interrupted during the delivery from the destination Voice SFU to the destination voice-mail system, subsequent attempt to deliver the message should commence with the start of the voice-mail message not successfully delivered. If the transfer of the message was interrupted during the transfer of the message from the voice-mail system to the recipient's terminal device, subsequent attempts to deliver the message shall commence with the start of the message not successfully delivered. Redelivery should be attempted as soon as possible to minimize user confusion and dissatisfaction. + +# 10 Quality of service + +## 10.1 Message loss or failure + +The introduction of a Voice SFU for communication between voice-mail systems should not increase, on a per address basis, the probability of message loss or corruption. + +In the event of system failure, all messages accepted by the originating Voice SFU should be traceable. Furthermore, if the following conditions occur, then procedures are to be invoked as indicated: + +- If messages with deferred delivery have to be cancelled by the originating Voice SFU because of system failure, an **unsuccessful message transfer notification (Level 1)** is to be generated to the originator of the cancelled message as part of system shutdown or system restart if the pending message cannot be recovered. +- If system failure occurs at the destination Voice SFU, an **unsuccessful message transfer notification (Level 1)** is to be generated to the originator indicating system failure. +- If failure occurs in the destination voice-mail system prior to the user accepting responsibility for the subject message, then: + - as part of system shutdown, then an **unsuccessful message delivery notification (Level 2)** is to be generated as part of the shutdown procedures; and + - subsequent to system restart, an **unsuccessful message delivery notification** is to be sent to the message originator indicating an appropriate reason. + +## 10.2 Quality of service targets + +When the service is extended across international boundaries for delivery purposes, it is recommended that the Voice S&F design be such that the end-to-end quality of service targets is that established in Table 3 as illustrated in Figure 2. + +The Voice S&F should be capable of meeting the quality of service targets at least 95% of messages received during any one hour period 95% of the time. + +While delivery is being attempted, the Voice S&F may generate interim notifications to the voice-mail system showing that delivery of the message is still outstanding. + +![Figure 2/F.472 – Illustration of quality of service times. A timeline diagram showing three vertical lines: 'End of submission', 'Delivery attempt', and 'Forced negative'. A horizontal double-headed arrow labeled T_A spans from 'End of submission' to 'Delivery attempt'. A longer horizontal double-headed arrow labeled T_B spans from 'End of submission' to 'Forced negative'. The text 'T0730510-97/d02' is at the bottom right.](8fbdfc3d17fb1dae7b2d8f5a287fa9fc_img.jpg) + +Figure 2/F.472 – Illustration of quality of service times. A timeline diagram showing three vertical lines: 'End of submission', 'Delivery attempt', and 'Forced negative'. A horizontal double-headed arrow labeled T\_A spans from 'End of submission' to 'Delivery attempt'. A longer horizontal double-headed arrow labeled T\_B spans from 'End of submission' to 'Forced negative'. The text 'T0730510-97/d02' is at the bottom right. + +$T_A$ Time from the end of submission of the message or the time specified for deferred delivery to the start of the first call attempt. + +$T_B$ Time from the end of submission of the message or the time specified for deferred delivery to the forced generation of a negative 2nd level notification and termination of message delivery attempt (unless transmission has already commenced). + +**Figure 2/F.472 – Illustration of quality of service times** + +**Table 3/F.472 – Quality of service targets** + +| Class of message delivery | $T_A$ | $T_B$ | +|---------------------------|-----------|---------| +| Urgent | 0.5 hours | 1 hour | +| Routine | 2 hours | 4 hours | + +## 10.3 Storage capacity + +Storage capacity will vary from one Voice SFU to another according to the volume of traffic. However, sufficient storage capacity should be provided to cater for an average message size of two minutes, and maximum message size of five minutes. + +## 10.4 Public network circuits + +### 10.4.1 Provision + +The provision of network circuits for the reception and delivery of traffic by the Voice SFU will vary between service providers. However, sufficient outgoing public network circuits should be provided to meet the quality of service targets of 10.2. + +### 10.4.2 Performance + +Service providers should make observations to evaluate the performance of the Voice SFU to ensure, in particular, that public network provisions are such that quality of service targets of 10.2 are being achieved. + +## 10.5 Duration of the service + +The VM S&F service should be continuously available. + +## 10.6 Inquiries and complaints + +User inquiry and complaint support services should be provided to VMS operators by service providers. VMS users should address any complaints to the OAM&P group providing them with the VM S&F service. + +# Annex A + +## Format element of service definitions + +**A.1 cancel deferred delivery request:** The *cancel deferred delivery request* element of service enables the user to indicate to the service the desire to cancel a message being held for deferred delivery. + +**A.2 class of message delivery:** The *class of message delivery* element of service enables the user to indicate to the service the desired grade of transfer and delivery for a submitted message. The underlying communications mechanisms interpret this information to provide delivery in accordance with the originator's requested transfer and delivery performance times indicated in Table 2. + +NOTE – Voice-mail notifications use the value of *class of message delivery* indicated in the original subject message as the value in the notification. + +**A.3 deferred delivery request:** This element of service allows the originator to request that the message be held locally for submission at a point in the future. The realization of this service is a local matter. + +Provisioning of this service requires the provisioning of element of service A.1 – cancel deferred delivery request. + +NOTE – It functions as the deferred delivery element of service of F.400 (B.19), but is not defined in terms of UA and MTA which are functional entities in the message handling service. + +**A.4 message identification:** A message identification element of service is provided by VM S&F to unambiguously identify a message related to the message originator on a global basis. + +NOTE – The value is returned in notifications to enable the user to correlate the received notification with the appropriate previously submitted message. + +**A.5 message recipient:** A message recipient element of service identifies the one or more intended recipients for this message. It may also identify a distribution list which is named like any other recipient. + +NOTE 1 – All recipients are considered to be primary recipients. + +NOTE 2 – The notion of blind copy or copy recipients are not featured in this service. + +**A.6 originator identification:** This element of service allows the identity of the originator to be conveyed to the recipient. The intent of this element of service is to identify the originator in a user-friendly way. It includes a routeable representation of the originator's name, and may include a voice signature as well. + +**A.7 notification details:** This element of service allows the originator to convey to the recipient the details associated with a Level 1 or Level 2 notification. The intent of this element of service is to identify the originator in a user-friendly way details associated with the successful acceptance or rejection of a voice-mail message. It includes date and time information, and an indication of acceptance or rejection (and *why* if rejected.) + +# ITU-T RECOMMENDATIONS SERIES + +- Series A Organization of the work of the ITU-T +- Series B Means of expression: definitions, symbols, classification +- Series C General telecommunication statistics +- Series D General tariff principles +- Series E Overall network operation, telephone service, service operation and human factors +- Series F Non-telephone telecommunication services** +- Series G Transmission systems and media, digital systems and networks +- Series H Audiovisual and multimedia systems +- Series I Integrated services digital network +- Series J Transmission of television, sound programme and other multimedia signals +- Series K Protection against interference +- Series L Construction, installation and protection of cables and other elements of outside plant +- Series M TMN and network maintenance: international transmission systems, telephone circuits, telegraphy, facsimile and leased circuits +- Series N Maintenance: international sound programme and television transmission circuits +- Series O Specifications of measuring equipment +- Series P Telephone transmission quality, telephone installations, local line networks +- Series Q Switching and signalling +- Series R Telegraph transmission +- Series S Telegraph services terminal equipment +- Series T Terminals for telematic services +- Series U Telegraph switching +- Series V Data communication over the telephone network +- Series X Data networks and open system communication +- Series Z Programming languages \ No newline at end of file diff --git a/marked/F/T-REC-F.500-199208-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.500-199208-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..16d91af14a99a0a23c7f6addf6154b8e10e625a6 --- /dev/null +++ b/marked/F/T-REC-F.500-199208-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8693efe5f59a9c0ad05b83c283a0d86b30b798232b0c81153138b1b76a4772b1 +size 8227 diff --git a/marked/F/T-REC-F.500-199208-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.500-199208-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..4fceed1f5a4ea91d0c3b9c48cdfe0fb72fa39ae6 --- /dev/null +++ b/marked/F/T-REC-F.500-199208-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:caba20f75e4127ce2e20906dc9a4911c639059c35825949b35d7d49d35d98507 +size 6021 diff --git a/marked/F/T-REC-F.500-199208-I_PDF-E/e6df2733626a85205c1db682e6259c46_img.jpg b/marked/F/T-REC-F.500-199208-I_PDF-E/e6df2733626a85205c1db682e6259c46_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1a2d8d9bafd92aed99752bd4615c6691d2b2d972 --- /dev/null +++ b/marked/F/T-REC-F.500-199208-I_PDF-E/e6df2733626a85205c1db682e6259c46_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fb3db3331e656d450bb14997783f6e322018ee42c04ea0b412a2ea51aed78319 +size 87688 diff --git a/marked/F/T-REC-F.500-199208-I_PDF-E/raw.md b/marked/F/T-REC-F.500-199208-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..c5ca2616be2ae13cbcbae8133c2161adb92f4878 --- /dev/null +++ b/marked/F/T-REC-F.500-199208-I_PDF-E/raw.md @@ -0,0 +1,1533 @@ + + +![ITU logo: A globe with a lightning bolt and the letters ITU.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with a lightning bolt and the letters ITU. + +INTERNATIONAL TELECOMMUNICATION UNION + +**CCITT** + +**F.500** + +THE INTERNATIONAL +TELEGRAPH AND TELEPHONE +CONSULTATIVE COMMITTEE + +(08/92) + +**DIRECTORY SERVICE** + +**OPERATIONS AND DEFINITION OF SERVICE** + +--- + +**INTERNATIONAL PUBLIC DIRECTORY +SERVICES** + +![CCITT logo: A globe with the letters CCITT.](84a1d09fb489061482111515543b60dc_img.jpg) + +CCITT logo: A globe with the letters CCITT. + +**Recommendation F.500** + +--- + +# **FOREWORD** + +The CCITT (the International Telegraph and Telephone Consultative Committee) is a permanent organ of the International Telecommunication Union (ITU). CCITT is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The Plenary Assembly of CCITT which meets every four years, establishes the topics for study and approves Recommendations prepared by its Study Groups. The approval of Recommendations by the members of CCITT between Plenary Assemblies is covered by the procedure laid down in CCITT Resolution No. 2 (Melbourne, 1988). + +Recommendation F.500 was revised by Study Group I and was approved under the Resolution No. 2 procedure on the 4th of August 1992. + +# --- **CCITT NOTE** + +In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized private operating agency. + +# **INTERNATIONAL PUBLIC DIRECTORY SERVICES** + +*(revised 1992)* + +Given the rapid multiplication and expansion of CCITT-defined telecommunication services, there is a growing need for subscribers to these telecommunication services to be able to communicate with each other. In order to facilitate such intercommunication for the subscribers of the various services, public directory services will be required. + +The CCITT, + +*considering* + +(a) that the CCITT-defined telecommunication services, including telegraphic, telematic and telephone services, have directory requirements; + +(b) that such requirements are being implemented as on-line electronic directories (in addition to traditional hard-copy versions); + +(c) that national initiatives are being taken to develop electronic integrated directories or service specific directories; + +(d) that the system definition is being undertaken by the CCITT in the field of electronic directories in the X.500-Series of Recommendations, + +*unanimously declares* + +that the specifications of this Recommendation should be applied to the provision of public directory services. + +# **CONTENTS** + +| | | +|----|-----------------------------------------------| +| 1 | Introduction | +| 2 | Purpose and scope | +| 3 | Organizational provisions | +| 4 | Public directory services | +| 5 | Names as the key to directory searches | +| 6 | Character repertoire and languages | +| 7 | Display of a response | +| 8 | Operational issues | +| 9 | Quality of Service aspects | +| 10 | References | + +*Annex A – Abbreviations* + +*Annex B – Service error messages* + +*Annex C – Selected object classes* + +*Annex D – Selected attribute types* + +*Annex E – MHS selected object classes* + +*Annex F* – MHS selected attribute types + +*Annex G* – User visibility of the search operation + +*Annex H* – Glossary of terms + +# **1 Introduction** + +International public directory services will enable subscribers to determine rapidly and easily what services are available and how to access and address their correspondents. Public directories may also be used internally by the various telecommunication services for the proper routing of calls or messages. However, this application of directory systems is not covered by this Recommendation. + +Service specific directories may be implemented as part of a global directory service. Consistent with the need to make directory information as widely available as possible, it is anticipated that Administrations will aim to provide global electronic directory services. + +In order to provide international public directory services, Administrations should mutually cooperate in handling inquiries for information across national boundaries. + +Public directory services should solve the primary problem of name to address association, e.g. obtaining a company's telex number by querying the directory with the name of the company. The reverse question, i.e., obtaining the name and other information from the address, may also be applicable in certain services and its provision is at the option of an Administration. + +Public directory services should include directory information concerning the provision of services, service descriptions, operational instructions, tariff conditions, etc. + +Public directory services should make provision for accessing information without knowing the name of the object sought, e.g., designating categories of goods, business areas or services. + +Advertising is included in the scope of public directory services, but is left to national implementations. + +Public directory services can be considered as supplementary to the services for which they provide information or by which they are accessed. + +Private directory services which are compliant with the public directory services defined in CCITT Recommendations may be permitted to intercommunicate with public directory services under national regulations. + +# **2 Purpose and scope** + +This Recommendation provides for the general framework for the provision of international public directory services in accordance with and based on the Recommendations of X.500-Series. It defines the requirements for and the service features associated with the provision of public directory services. It specifies naming aspects, describes operational issues to be taken into account in providing the public directory services as well as Quality of Service aspects. + +# **3 Organizational provisions** + +Provision of a public directory service will be done in accordance with the organizational model described in Recommendation X.501. An Administration directory management domain (ADDMD) is responsible for the application of the basic service features and the optional user facilities provided in that domain. Directory management domains shall intercommunicate with each other as far as the provision of the public directory services requires it. The protocols to be used for interworking as well as the directory's overall concept and behaviour, is described in the X.500-Series of Recommendations. + +*Note* – ADDMDs will typically have (international) connections. + +Private directory management domains (PRDMDs) may exist and intercommunicate with ADDMDs, following national regulations. + +A Directory management domain (DMD) consists of one or more directory system agents (DSAs) and zero or more directory user agents (DUAs). + +Each directory management domain may act as the naming authority for that domain. Names need to be unambiguous. + +*Note* – The system specification assumes only one occurrence of an organization name under the country name. Distributed entries are for further study. + +The intercommunication between PRDMDs is outside of the scope of this Recommendation. + +# 4 Public directory services + +## 4.1 Service requirements + +The fundamental ability of a public directory service is to provide a means by which subscribers or users of telecommunication services may, in a user-friendly manner, and from information they would normally possess, obtain information about a desired recipient, such as addresses or communication capabilities. + +This public directory service is provided in an on-line and interactive manner. It should be made available for subscribers or users at the discretion of the Administration offering the service. + +Each Administration is responsible for the access methods used. The characteristics of the access methods between terminals and the public directory service are a national matter. However, the directory service offered is independent of the access method, the terminal used and the location of the user. + +Public directories of Administrations should intercommunicate (or refer to each other) to fulfil requests made by users when the directory serving the user does not have available the information requested. + +### 4.1.1 Basic service requirements + +The following basic service requirements are fulfilled by the public directory services: + +- to provide subscribers with information, e.g. a telex number, needed for establishing communication with other subscribers or users of telecommunication services; +- to provide subscribers with information, e.g. service instructions, needed to use the telecommunication services and the directory itself; +- to assist subscribers in the formulation of queries to narrow the scope of the operation; +- to allow for flexibility in the formulation of a request, e.g. names should not artificially remove natural ambiguities; names should admit natural abbreviations and commonly used variations in spelling; +- to provide sufficient functionality of access control. + +### 4.1.2 Non-basic service requirements + +The following non-basic service requirements are fulfilled by features or the user facilities of the public directory services. + +- to provide subscribers with other information, e.g. advertising; +- to provide subscribers with “yellow page” information, e.g. categories of goods, business areas or services; +- to provide an inverted directory for specific services, e.g. for telex and teletex; +- to provide “wildcards” capability to ease, as far as possible, the input of the requests to the directory; + +- to provide means for the verification of credentials, under conditions specified by the provider of the directory service; +- to provide possibilities for the search of distribution lists; +- to provide means for the phonetic matches; +- to provide possibility for the geographic extension; +- to provide means for the shadowing (replication) of directory information. + +## 4.2 *Service features and optional user facilities* + +The service features and the optional user facilities of a public directory service will be provided in accordance with the X.500-Series of Recommendations. The terms used in the context of service features and optional user facilities discussed below are explained in Annex H. + +### 4.2.1 *Basic service features* + +Basic service features are inherent in directory services and are always available for use in directory service. They are provided by all service providers offering international public directory services or by private directories intercommunicating with public directory services. + +The basic features are: + +- read operation; +- search operation. + +Other basic features are for further study. + +### 4.2.2 *Optional user facilities* + +Optional user facilities may be selected by the user or subscriber at the time the service is being used. Each optional user facility visible to the user is classified as either essential or additional. Essential (E) optional user facilities are to be made available internationally by Administrations. Additional (A) optional user facilities may be made available by Administrations for national use and for international use on the basis of bilateral agreement. + +The major terms used in this Recommendation are contained in Annex H. + +The classification of optional user facilities is shown in Table 1/F.500. + +## 4.3 *Further features and facilities* + +Some of the following items are not yet specified as elements of service in the X.500-Series of Recommendations and will be studied further. Some others will need further study under service aspects. The following list may provisionally be considered as guidance for service providers to be taken into account for the provision of public directory services under national responsibility. The items may become basic features or optional user facilities in the future or/and will be included with descriptive text in future Recommendations. + +- Provision of additional information with or after the result of a query. +- Provision of query cost information. +- Provision of information about services, service instructions, tariffs, etc., in standardized formats taking into account additional attributes. +- The ability of the user to indicate the desire not to receive partial results when service control maximum parameters are exceeded. +- Provision of administrative procedures for authentication. +- Consequences of distributed directory services. + +TABLE 1/F.500 + +### **Classification of optional user facilities** + +| | Classification | +|------------------------------|----------------| +| Abandon | E (see Note 1) | +| Add | A | +| Additional service controls | A | +| Compare | A | +| Distribution lists | A | +| Inverted directory | A (see Note 2) | +| List | A | +| Management of access control | A (see Note 3) | +| Modify | A | +| Remove | A | +| Security capabilities | A | +| Time limit service control | E | + +*Note 1* – This abandon operation is not guaranteed outside of the local scope, i.e. the DSA or DMD to which the original request was made. + +*Note 2* – This facility may be provided only for the business use such as telex and teletex services. + +*Note 3* – The full functionality is presently not provided in the present system specification of the X.500-Series of Recommendations (see § 3 and Annex F of Recommendation X.501). This is for further study and referred to as being presently a national matter. Access control functions are for further study. + +Other optional user facilities are for further study. + +## 4.4 *Service controls and other parameters* + +Because of its generality and scope, the directory service can fulfil subscribers' requests that might require consumption of resources beyond a level desired by the subscriber or by the service provider. Service controls and other parameters help to prevent such situations by imposing limits on the resources that may be consumed in fulfilling a request for service. Service controls and other parameters not impacting the provision of international directory services are a local matter. The following service controls and a parameter are provided by the system application (see Recommendation X.511): + +### 4.4.1 *Prefer chaining* + +This service control indicates a preference for chaining rather than referral. For the international intercommunication of public directories, chaining is the preferred choice. + +The setting of this service control is for the service provider who may allow the user to invoke it. + +### 4.4.2 *Chaining prohibited* + +The scope of a search will then be limited to the local portion of the directory information base (DIB) by prohibiting chaining. + +The setting of this service control is for the service provider who may allow the user to invoke it. + +### 4.4.3 *Local scope* + +The scope of the operation will be limited to the local portion of the DIB. The determination of local is restricted to a single DSA or DMD in accordance with an Administration's policy. + +For the international intercommunication of public directories, generally no limitation to local scope is assumed. Public directories will aim to open their scope as much as possible. + +The setting of this service control is for the service provider who may allow the user to invoke it. + +### 4.4.4 *Do not use copy* + +This service control prevents a DMD from returning copied information. + +The setting of this service control is for the service provider who may allow the user to invoke it. + +### 4.4.5 *Do not dereference alias* + +This service control allows reference to an alias entry itself rather than to the aliased entry. + +The setting of this service control is for the service provider. + +### 4.4.6 *Priority: low, medium, high* + +The setting of this service control is for the service provider. + +The usefulness of this service control is for further study. + +### 4.4.7 *Time limit* + +The scope of this service control is to limit an operation in terms of total elapsed time such that if the limit is exceeded, then the operation will be terminated, and for search and list operations partial results should be returned, with the indication that results are incomplete due to the time limit. This service control shall be honoured by any DMD involved. + +The setting of this service control is for the service provider who may allow the user to invoke it. + +*Note* – This service control is an essential optional user facility. All service controls other than the time limit are a local matter and when implemented, need not be made available by the service provider to the user. + +### 4.4.8 *Attribute size limit* + +If an attribute included in returned entry information exceeds this limit, with the indication it is omitted from the returned entry information, the size of an attribute is taken to be its size in octets in the local concrete syntax of each service provider. + +The setting of this service control is for the service provider who may allow the user to invoke it. + +### 4.4.9 *Size limit (applicable to search or list operations)* + +If the list size specified is exceeded, any results equal in number to the size limit should be returned, with the indication that the results are incomplete due to the size limit. + +The setting is for the service provider who may allow the user to invoke it. + +### 4.4.10 *Scope of referrals* + +Indicates the scope to which a referral (or advice), if generated, is to be restricted to, i.e. limits the range of alternate access points at which the requestor (DUA or DSA) may alternately use to satisfy the request. The limitation can be restricted to a country or DMD. + +The setting of this service control is for the service provider who may allow the user to invoke it. + +### 4.4.11 *Paged result* + +The scope of this service control is to request from the DSA that the results of the list or search operation be returned "page-by-page". It requests the DSA to return only a subset of the results of the operation and query-reference which can be used to request the next set of results on a follow-up query. This subset is based on the requested number of entries to be returned, the ordering sequence and merging. + +The setting of this service control is for the service provider who may allow the user to invoke it. + +### 4.4.12 *Sub-entries* + +Indicates that the operation is to access sub-entries only. + +The setting of this service control is for the service provider who may allow the user to invoke it. + +### 4.4.13 *Copy shall do* + +Indicates that if the directory is able to partly but not fully satisfy a query of a copy of an entry, it need not chain the query to the DSA holding the master copy. It is meaningful only if "do not copy" is not set. If "copy shall do" is not set, the directory will use shadow data only, if it is sufficiently complete to allow the operation to be fully satisfied at the copy. + +The setting of this service control is for the service provider who may allow the user to invoke it. + +*Note* – Combination of some service controls may affect the quality of the results, e.g. combination of priority, time limit and size limit may conflict, or chaining cannot be both preferred and prohibited simultaneously. If no service controls are supplied with an operation, the following is assumed: referrals and/or chained operations may be used; no limit on the scope of the operation; locally held copies of information are permitted; no preference of priority for operation processing; there is no time or size constraint; referrals, if generated, are not restricted to a DMD or country; aliases are dereferenced; sub-entries are not accessible; and operations not fully satisfiable with shadowed data are subject to further chaining. + +# 5 Names as the key to directory searches + +## 5.1 *General* + +A name within the directory service is a label which is constructed to identify a particular object, that is, which singles out an object from the set of all objects. A name should not be ambiguous, that is, should not denote more than one object. However, there may be more than one name for an object. Thus, it is possible to call an object by the name International Widget Makers or IWM. In either case, one and only one object is identified. + +A more abstract definition of "name" can be found in Annex H. + +## 5.2 Entries + +The directory service will provide information about entries. The complete set of such information is called the directory information base (DIB). The information about entries is composed of attributes; attributes, in turn, are composed of an attribute type (one type of attribute could be a telex number) and one or more attribute values. (The actual telex number would be a value.) The entries are arranged in a tree, called the directory information tree (DIT). This is graphically illustrated in Figure 1/F.500. However, this does not preclude other directory information structures. + +Thus, an entry may be viewed as an entity which is named through one or a series of attributes. A company may be sufficiently named simply through the use of its actual legal name e.g. the Padraic Steel Co. A plumber in Secausus, N.J. can be named through the use of his common name, his postal address and his business category “plumber”. A human person may be named through the use of his or her common name and telephone number. + +![Diagram illustrating the structure of an entry in a directory. The top part shows a tree structure labeled 'DIT' with 'Alias entry' and 'Object entry' nodes. An arrow points from the 'Alias entry' to the 'Object entry'. Below this, a detailed view of an 'Entry' is shown, containing multiple 'Attribute' boxes. One 'Attribute' is expanded to show its internal structure: 'Attribute type' and 'Attribute value(s)'. The 'Attribute value(s)' are further expanded to show 'Distinguished attribute value', 'Attribute Value', and other 'Attribute value' boxes. The diagram is labeled T0102970-93.](e6df2733626a85205c1db682e6259c46_img.jpg) + +Diagram illustrating the structure of an entry in a directory. The top part shows a tree structure labeled 'DIT' with 'Alias entry' and 'Object entry' nodes. An arrow points from the 'Alias entry' to the 'Object entry'. Below this, a detailed view of an 'Entry' is shown, containing multiple 'Attribute' boxes. One 'Attribute' is expanded to show its internal structure: 'Attribute type' and 'Attribute value(s)'. The 'Attribute value(s)' are further expanded to show 'Distinguished attribute value', 'Attribute Value', and other 'Attribute value' boxes. The diagram is labeled T0102970-93. + +Note 1 – The alias entry has a pointer to the actual entry for which it is the alias and does not contain the actual object information. + +Note 2 – See also Recommendations of the X.500-Series. + +FIGURE 1/F.500 +**Structure of an entry in a directory** + +## 5.3 Distinguished names + +It should be noted that within the directory system Recommendations, the term “distinguished name” is used. This is the combination of the minimum attribute value assertions (AVAs) needed to denote an entry uniquely. This minimum will be established in accordance with the requirements of the naming authority and/or the directory + +management domain, and the preference of the owner of the entry named. Use of the distinguished name may be of assistance in performing the most effective search of the DIB. However, it should be recognized that in some instances, distinguished names may not be user friendly and may contain information, which, in fact, is the object of the directory search, i.e. a person's postal address. + +## 5.4 *Classification of requests* + +To satisfy the most common needs of directory users which are presently met by so-called "white pages" or "yellow pages" (classified directories) or organization directories, three classifications of requests to the directory service are provided. + +### 5.4.1 *Common name requests (type 1)* + +Information returned under this type of request includes information about one or more of the following entries. (Selected object classes can be found in Recommendation X.521; they are listed in Annex C.) + +- a) A person +*Example:* Bernadette L. Casey +- b) A residential person +*Example:* Cornelius Fecit +2 Humbug Road +Fun City, New York 11666 +USA +- c) An application entity +*Example:* Some logical name, usually a sequence of alpha and/or numeric characters identifying an application process; consequently not necessarily user friendly. +- d) A communication device +*Example:* The XYZ 9.6 modem (however, this information is normally associated with an organization and is thus generally of greatest utility to organizations). +- e) An alias +*Example:* Neil Fecit [an alias for the residential person in b)] +- f) An organizational role +*Example:* Director of regulatory affairs +- g) A group of names +*Example:* Members of Special Rapporteur's Group Question 14/1 + +### 5.4.2 *Business category requests (type 2)* + +Information returned under this type of request includes information about one or more of the following entries. (Selected object classes can be found in Recommendation X.521; they are listed in Annex C.) + +- a) A person +*Example:* John Smith +- b) A residential person +*Example:* John Smith, with the rest of the postal address +- c) An organization +*Example:* The Padraic Steel Company +- d) An organizational unit +*Example:* Regulatory Affairs Department +- e) A group of names +*Example:* The plumbers in Secausus + +### 5.4.3 *Organization requests (type 3)* + +Information returned under this type of request includes information about one or more of the following entries. (Selected object classes can be found in Recommendation X.521; they are listed in Annex C.) + +- a) An organization +*Example:* The Padraig Steel Company +- b) An organizational unit +*Example:* Regulatory Affairs Dept. +- c) An organizational person +*Example:* John Jones, Padraig Steel Company +- d) An organization role +*Example:* Chief Operating Office +- e) A group of names +*Example:* The President's Staff +- f) An application entity +*Example:* as above in § 5.4.1 c) +- g) A device +*Example:* An XYZ 9.6 modem +- h) An organizational unit alias +*Example:* the "bean counters" which is an alias for the "Controller's Dept." +- i) An organizational name alias +*Example:* GMC for "Good Modern Cooks Inc." + +### 5.4.4 *Use of attributes* + +Attribute types that are recommended to be included, whenever they exist (subject to the permission of the owner) in each entry of each group, either for query or retrieval, are listed in Table 2/F.500 (see also Annex D). + +## 5.5 *Naming of entries* + +To reach an entry, a user has to provide some information, a part of which is essential to the performance of the request (e.g. the provision of attributes CTN, ORG, CLASS, for an organizational object), as described in § 5.2. + +Depending on the knowledge the user has about the naming structure of the part of the directory information tree (DIT) to which the entry of the intended object belongs, the request information provided by this user to reach the intended entry is either the distinguished name of the entry (in which case the response is unique), or the value of some relevant search attributes (already known by the user) arranged in a logical pattern to act as a filter to reduce as far as possible the number of the directory responses. + +Since distinguished names have to be unambiguous, it is not expected that they will always be user-friendly. For instance, a name of a residential person may include the telephone number and thus be rather difficult to predict, especially if the telephone number is the information requested from the directory. It is recognized that the distinguished name (DN) of an object may not be commonly known, in which case the DN may be acquired by using a list operation and in some instances a search operation. + +To perform efficiently the search or list operation, it is recommended that one narrows as far as possible the scope of the search, either by giving a base object (from which the search starts in the DIT) near enough to the intended entry (in terms of DIT levels), or by obtaining and using the appropriate filtering. + +It should be possible to obtain from the directory which of the attributes (qualified with "Q" in Table 2/F.500) may be used as part of the search filter for a given object class starting from a given base object. However, it is recognized that the use of this feature across domain boundaries is subject to national restrictions and bilateral agreements. + +It is expected in most cases that a directory management domain will be able to provide from previous experience the useful search criteria of subordinate levels, whether or not they efficiently manage those levels, without exploring the DIT further for each request. Knowledge of the search criteria may also be acquired by DUAs from the directory by automatic means, e.g. by reading the "search guide attribute" if available. + +It is up to the directory management domain (DMD) managing a given entry to select from the attribute types specified in § 5.4 for use as search criteria. + +The use of wildcards to replace the value or part of the value of unknown recommended search criteria should be made possible. + +Phonetic or orthographic extensions, when requested, may be locally applied to the provided values for query operations. However, their actual provision depends on the capabilities of the DSA involved. The fall-back mode is phonetic or orthographic extensions not supported. + +TABLE 2/F.500 +Use of attributes for each type of request + +| Attribute type | Abbreviation | for type 1 | for type 2 | for type 3 | +|-----------------------------------------|--------------|------------|------------|------------| +| Business category | BCTG | – | M | R | +| Common name | COM | M | Q | Q | +| Country name | CTN | M | M | M | +| Description (free text) | DES | R | R | R | +| Destination indicator (public telegram) | DI | – | – | – | +| Facsimile telephone number | FAX | – | Q | Q | +| ISDN address | ISDN | – | Q | Q | +| Knowledge information | KI | – | – | – | +| Locality name | LOC | M | Q | Q | +| Member | MEM | R | R | R | +| Object class | CLASS | Q | Q | Q | +| O/R address (MHS) (see Note 1) | O/R | R | R | Q | +| Organization name | ORG | – | – | M | +| Organizational unit name | OUN | – | – | Q | +| Owner | OWN | – | – | – | +| Physical delivery office name | PDO | Q | Q | Q | +| Post office box | POB | Q | Q | Q | +| Postal address | PADD | Q | Q | Q | +| Postal code (see Note 2) | PCOD | Q | Q | Q | +| Preferred delivery method | DLM | R | R | R | +| Presentation address | PRADD | R | – | R | +| Registered address (public telegram) | RADD | – | R | R | +| Role occupant | RO | R | – | R | + +TABLE 2/F.500 (cont.) + +| € | Abbreviation | for type 1 | for type 2 | for type 3 | +|-----------------------------------|--------------|----------------|------------|------------| +| Search guide | SG | R | R | R | +| See also | SEE | R | R | R | +| Serial number | SN | – | – | – | +| State or province name | STN | M (see Note 3) | Q | Q | +| Street address | SADD | Q | Q | Q | +| Supported application context | SAC | Q | Q | Q | +| Surname | SUR | Q | Q | Q | +| Telephone number | TEL | Q | Q | Q | +| Teletex terminal identifier | TTX | R | Q | Q | +| Telex answerback (see Note 4) | A/B | R | R | R | +| Telex number | TLX | R | Q | Q | +| Title | TIT | – | – | Q | +| User certificate | UC | R | R | R | +| User password | UP | R | R | R | +| Videotex user number (see Note 4) | VTX | Q | Q | Q | +| X.121 address | X.121 | – | Q | Q | + +M Mandatory to reach an object of this type. + +Q May be used to reach an object of this type (within a distinguished name or as a search filter), but may also be part of the directory response. Additional attribute types may be used for selection criteria within national implementations. + +R Normally part of the directory response with regard to the request of the user. + +– This attribute type may either be part of a local sub-object class or used nationally. + +*Note 1* – This attribute type is defined in the X.400-Series of Recommendations. + +*Note 2* – The postal address will normally contain the postal code. Requirements may exist to justify the postal code as being a separate attribute type. Specific conditions are applied to a postal address for physical delivery (see Recommendation F.401). + +*Note 3* – Depending on the value of the attribute “CTN”. + +*Note 4* – This attribute type has not yet been defined in Recommendation X.520. + +Some terms used in Table 2/F.500 are explained in Annex H. Definitions of other terms can be found in the X.500-Series of Recommendations. + +## 5.6 Qualifications of attribute types + +Some criteria of the selected attribute types require qualification. + +“Mandatory” in Table 3/F.500 indicates that, if that attribute type exists in an entry of the directory, it shall be part of any response provided, when asked for by the user, and that no combination of access controls may be kept on attributes which would preclude provision of a meaningful directory service, subject to the owner’s approval. + +The “required length” of an attribute type in Table 3/F.500 designates the minimum number of character positions to be made available for the attribute type to be displayed on the terminal of a user, and can therefore assist Administrations in defining their attribute values with the assurance that the attribute value will not be truncated. (The X.500-Series Recommendations have system qualifications for the maximum length of attribute types.) + +The system specification does not provide multiple values for country name and preferred delivery method. All others may be recurring. For example, an organization may be “Padraic Steel” and “Padraic Steel Company”. At least one value needs to be displayed to the user. + +Table 3/F.500 contains a list of the user-visible selected attribute types to be used in the directory service. The figures shown may require revision in the light of experience. + +TABLE 3/F.500 + +### **Qualifications of attribute types** + +| € | Mandatory | Required length | +|-----------------------------------------|-----------|-----------------| +| Business category | Yes | 128 | +| Common name | Yes | 64 | +| Country name (see Note 1) | Yes | 30 | +| Description | Yes | 1024 | +| Destination indicator (public telegram) | Yes | 4 | +| Facsimile telephone number | No | 150 | +| ISDN addressee | No | 16 | +| Knowledge-information | No | – | +| Locality name | Yes | 64 | +| Member | No | – | +| Object class | No | – | +| O/R address MHS (see Note 2) | Yes | – | +| Organization name | Yes | 64 | +| Organizational unit name | Yes | 64 | +| Owner | No | – | +| Physical delivery office name | No | 64 | +| Post office box | No | 40 | +| Postal address | No | 180 | +| Postal code (see Note 2) | No | 20 | +| Preferred delivery method (see Note 3) | Yes | 15 | +| Presentation address | No | – | +| Registered address (public telegram) | Yes | 60 | +| Role occupant | No | – | +| Search guide | Yes | – | +| See also | Yes | – | +| Serial number | No | 64 | +| State or province | Yes | 64 | + +TABLE 3/F.500 (cont.) + +| € | Mandatory | Required length | +|-----------------------------------|-----------|-----------------| +| Street address | No | 64 | +| Supported application context | No | – | +| Surname | No | 64 | +| Telephone number | No | 16 | +| Teletex terminal identifier | No | 24 | +| Telex answerback (see Note 6) | No | 21 | +| Telex number (see Note 3) | No | 36 | +| Title | No | 64 | +| User password | No | – | +| User certificate | No | – | +| Videotex user number (see Note 6) | No | 17 | +| X.121 address | No | 15 | + +*Note 1* – The system specification provides only a 2-character length, to correspond to the ISO 3166 value. + +*Note 2* – The postal address will normally contain the postal code. Requirements may exist to justify the postal code as being a separate attribute type. Specific conditions are applied to a postal address for physical delivery (see Recommendation F.401). + +*Note 3* – The system specification provides a shorter field. + +*Note 4* – For some attribute types, values are stored in encoded/compressed format and will need to be displayed in a non-encoded format or human readable format. + +*Note 5* – See also Recommendation X.520, Annex C. + +*Note 6* – This attribute type has not yet been defined in Recommendation X.520. + +# 6 Character repertoire and languages + +## 6.1 Character repertoire + +Directory information will be entered and stored locally using a character repertoire suitable to the country where the directory is located. More than one character repertoire may be needed to cover different languages or to provide for access from different types of communication terminals. + +However, in order to provide international public service, the character repertoire to be used internationally should be limited to CCITT standardized sets, i.e. the IA5 and T.61 character repertoires. + +For the intercommunication between public directory services, the repertoires may be agreed to bilaterally. + +However, where no such agreement exists, the character repertoire to be used shall consist only of those characters defined as “printable string” in Recommendation X.208. Furthermore, those Administrations which use character repertoires other than this repertoire shall provide suitable conversion of the information into this character repertoire for directory requests from Administrations with which no bilateral agreement has been reached. + +Subscribers have to be instructed on the use of the appropriate character repertoires. + +## 6.2 *Language of requests to the directory and responses from the directory* + +Subject to the conditions in § 6.1, the results of requests to the directory should normally be provided in the language or languages of the DMD providing the information. However, the information is presented to the requestor is a national matter. + +# 7 **Display of a response** + +Attribute types and values will be displayed to the user, when required, by converting the values in accordance with Recommendation X.408. + +Though it is logical enough that the right response always be sought, in some cases where no such answer can be provided, and on explicit request of the requestor, the directory may also provide phonetic and orthographic extensions corresponding to the intended object. + +For displaying directory responses, the following order is recommended: + +- a) the right answer(s); +- b) the answer(s) approaching the right answer(s) using conjunctions, particles, articles, as well as extended or concatenated abbreviations; +- c) the phonetic and orthographic extensions (e.g. plural instead of singular denominations). It should be noted that such responses may be erroneous. + +Partial responses, including referrals, should be displayed to the requestor and properly identified as such. The cause for partial responses should also be displayed. + +# 8 **Operational issues** + +## 8.1 *Management* + +It is the responsibility of the directory management domains (DMDs) to exercise the management of information within their domains. Inter-domain management is for further study. + +## 8.2 *Authentication* + +Authentication in this context means that the identity of the subscriber or user is established. In some cases, the directory service has to ensure that directory information is released only to authorized requestor(s), and in some cases it has to ensure that data is modified only by an authorized originator (e.g. by employing techniques related to data origin authentication). + +Checking and keeping of credentials, when performed, are at the discretion of the DMD, taking into account the requirements of privacy of the owner of the information. The precise reason for credential failure will be masked from the user. The user will be advised that denial of the request was because an inappropriate authentication level was encountered. + +See also Recommendation X.509. + +Further study is required. + +## 8.3 *Access control* + +The access to the information in the directory is controlled to conform to the security policy of the domain in which the fragment of the DIB exists. When access control prohibits the return of the information requested, an appropriate error code will be returned. The directory will offer consistent access control even if the user has access to the replicated information (shadowed copy). + +The detail of the access control is described in § 3 of Recommendation X.502. + +## 8.4 *Operational actions* + +Actions performed within a directory can be categorized as: + +- 1) primary (subscriber/directory) action – always in direct support of a subscriber; +- 2) secondary action in support of a subscriber request, either serving the subscriber's DUA or an intermediary DSA. + +These actions are qualitatively different, and differ also in what they imply concerning the obligations of an ADDMD. + +Examples of such interactions can be found in Recommendation X.518. + +### 8.4.1 *Primary (subscriber/directory) action* + +The public directory service should provide three user-visible activities of support, as follows: + +#### a) *Request formation* + +In this activity, the subscriber composes a request to the directory. The way in which these functions are performed is a national matter. + +#### b) *Presentation of results* + +In this activity, the directory service presents to the subscriber the results of a previously entered request. The format, presentation medium and other aspects of result presentation are a national matter. + +#### c) *Subscriber assistance* + +In this activity, the directory service assists the subscriber by providing instructions on the use of the directory. The means through which the subscriber asks for such instruction, and the manner in which an instruction is delivered, are a national matter. + +### 8.4.2 *Secondary action for subscriber support* + +In order to provide the public directory service, DMDs shall cooperate. Such cooperation includes adherence to defined patterns of interaction, and also includes provision of requested directory information to one another, subject only to internationally agreed access controls (or bilateral arrangements). This technical cooperation among DMDs implies an equivalent level of cooperation in service terms, especially with regard to information sharing, among the DMDs. Examples of such interaction can be found in Recommendation X.518. + +## 8.5 *Maintenance of the directory information* + +The service provider has to ensure integrity of the information contained in the directory. Shadowing (controlled replication) of information in other DMDs is permitted by bilateral agreement. + +Creation and modification of directory information by the subscribers may be permitted by the DMDs concerned. + +## 8.6 *Error handling* + +Error conditions will be returned as a value of an error code for all standardized operations. The meaning will be displayed according to national implementations as service error messages to the user. + +See Annex B for guidance. + +## 8.7 *Operator assistance* + +For further study. + +# 9 Quality of Service aspects + +## 9.1 *Availability* + +In principle, a public directory service should be available to subscribers 24 hours a day, seven days a week. + +## 9.2 *Security of directory information* + +Information in public directories should be given the broadest dissemination. However, subscribers or users about whom information is available in a directory should be able to require the entity charged with the management of the directory to limit access to such information to ensure their own privacy. + +## 9.3 *Successful directory requests* + +Normally, a successful directory request will result in a report of all the requested information, unless it is denied because of authorization restrictions. + +Requests to the directory which do not provide sufficient information to execute a reasonable search will normally not lead to a successful result. + +## 9.4 *Access* + +Providers of a public directory service should ensure that an adequate number of access ports are available to accommodate subscribers' requests for information. In principle, this means that a requestor will receive a prompt within 15 seconds as a goal. + +## 9.5 *Response time* + +Recognizing that responses to requests will be controlled in part by the level of ambiguity tolerated in requests and the number of DMDs which shall be traversed to retrieve the information requested, a subscriber normally should expect an initial acknowledgement regarding his request within 5 seconds. The scope and priority of the request may have an impact on the response time. The requestor may terminate his request at any time. + +Responses will depend on the capabilities of the directories consulted. As a possible target for final response, 90% of inquiry should be answered within 10 seconds. A response indicating that no information or incomplete information is available (possibly with hints for further searches) should be given within one minute. + +*Note* – The figures for Quality of Service are provisional and may be revised in the future. + +# 10 References + +## 10.1 *Recommendations of the X.500-Series* – Data communication networks: directory + +X.500 The directory – Overview of concepts, models and services + +X.501 The directory – Models + +X.509 The directory – Authentication framework + +X.511 The directory – Abstract service definition + +X.518 The directory – Procedures for distributed operation + +X.519 The directory – Protocol specification + +X.520 The directory – Selected attribute types + +X.521 The directory – Selected object classes + +X.525 The directory – Replication + +- 10.2 *Recommendations of the X.200-Series* – Data communication networks: Open systems interconnection (OSI) +- 10.3 *Recommendations of the F.400-Series* – Message handling operations and definition of service +- 10.4 *Recommendations of the X.400-Series* – Data communication networks: Message handling systems + +# ANNEX A (to Recommendation F.500) + +### **Abbreviations** + +| | | +|-------|--------------------------------------------| +| A | Additional optional user facility | +| ADDMD | Administration directory management domain | +| AVA | Attribute value assertion | +| CA | Certification authority | +| DIB | Directory information base | +| DIT | Directory information tree | +| DMD | Directory management domain | +| DN | Distinguished name | +| DSA | Directory systems agent | +| DUA | Directory user agent | +| E | Essential optional user facility | +| ITU | International Telecommunication Union | +| MHS | Message handling system | +| O/R | Originator/recipient | +| PRDMD | Private directory management domain | +| RDN | Relative distinguished name | +| RPOA | Recognized private operating agency | + +## ANNEX B (to Recommendation F.500) + +### **Service error messages** + +Error codes produced while performing operations in directory systems are transformed by the local DUA into service error messages. The values of the error codes and the meaning are summarized in this annex. The presentation to the user is a local matter. + +See also Recommendation X.511. + +### **B.1      *Attribute error*** + +This error is displayed on a per-selection criteria basis (attribute type) and includes the attribute type, attribute value and problem reason value. The problem reason values are as follows (see Table B-1/F.500). + +TABLE B-1/F.500 + +| € | Meaning | +|---|-------------------------------------------------------------------------------------------------------------------------------------------------------| +| 1 | The requested information does not exist for the named entry. | +| 2 | The syntax of the value used for the distinguished name or the selection criteria is inappropriate. Contact support staff for assistance. | +| 3 | Attribute Type is not defined for this . | +| 4 | Inappropriate matching for the information type . | +| 5 | Attribute Type or Attribute Value is not within its constraints. | +| 6 | or already exists. | + +### **B.2      *Name error*** + +This will be displayed with one of the following reason values whenever a name provided by the user is detected to have a problem (see Table B-2/F.500). + +TABLE B-2/F.500 + +| € | Meaning | +|---|-------------------------------------------------------------------------------------------------------------| +| 1 | The name supplied, , cannot be found.
(Note – Alias names are resolved to the actual named entry.) | +| 2 | is an alias that can not be properly resolved. | +| 3 | Part, , of the name used is undefined. | +| 4 | An alias was encountered in an operation where it is not allowed. | + +### B.3 *Interconnect error* + +This error will be displayed whenever the operation cannot be carried further at this time. The possible access points for continuing the request are provided in the form: “name and access point”. + +### B.4 *Service error* + +This will be displayed with one of the following reason values whenever the operation requested has detected a problem that affects the user service (see Table B-3/F.500). + +TABLE B-3/F.500 + +| Reason value | Meaning | +|--------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| 1 | The directory system is too busy to perform the request, but may be able to do so after a short while. | +| 2 | The directory system is currently unavailable. | +| 3 | The directory system is not prepared to execute this request, e.g. because it would lead to excessive consumption of resources or violate the administrative authority involved. | +| 4 | The directory system is unable to accomplish the request other than by chaining, however chaining was prohibited by means of service control. | +| 5 | The directory system is unable to accomplish the request due to the lack of appropriate administrative authority for name resolution. | +| 6 | The directory system is unable to perform the request as directed via OperationProgress by the DUA. | +| 7 | The directory system has reached the limit of time set by the user in a service control. | +| 8 | The directory system has reached some limit set by an administrative authority. | +| 9 | The directory is unable to accomplish this request due to an internal loop. | +| 10 | The directory is unable to satisfy the request because one or more critical extensions are not available. | +| 11 | No referrals are available within the requested scope. | +| 12 | The directory is unable to accomplish the request due to a DIT consistency problem. | +| 13 | The parameters of the requested operation are invalid. This problem is reported if the query-reference in paged result request is invalid. | + +### B.5 *Update error* + +This will be displayed with one of the following reason values whenever the modify (Add, Change, or Delete) operation(s) requested has detected a problem (see Table B-4/F.500). + +TABLE B-4/F.500 + +| Reason value | Meaning | +|--------------|----------------------------------------------------------------------| +| 1 | The update violates directory naming rules. | +| 2 | The update violates the directory rules for that class of object. | +| 3 | Update not allowed because the object is not the leaf entry. | +| 4 | Update not allowed because this affects the RDN. | +| 5 | Entry already exists (relevant for add operation only). | +| 6 | Update denied, affects multiple directory systems. | +| 7 | Update not allowed because this modifies the object class attribute. | + +### B.6 *Security error* + +This will be displayed with one of the following reason values whenever the operation requested has detected a problem for security reasons (see Table B-5/F.500). + +TABLE B-5/F.500 + +| Reason value | Meaning | +|--------------|----------------------------------------------------------------------------------------| +| 1 | The security level of the requester is inconsistent with the required level. | +| 2 | The supplied credential is invalid. | +| 3 | The requester does not have the right to carry out the operation. | +| 4 | The signature of the requester is invalid. | +| 5 | The request is not carried out because the argument is not signed. | +| 6 | The request is not carried out by a security error without indicating any information. | + +### B.7 *Abandon error* + +This will be displayed with one of the following reason values whenever the abandon operation requested has detected a problem (see Table B.6/F.500). + +TABLE B-6/F.500 + +| Reason value | Meaning | +|--------------|------------------------------------------------------------------| +| 1 | The directory has no knowledge of the operation to be abandoned. | +| 2 | The operation has already responded. | +| 3 | The attempt is prohibited or cannot be performed. | + +### B.8 *Referral error* + +For further study. + +## ANNEX C (to Recommendation F.500) + +### **Selected object classes** + +See Recommendation X.521 (1988). + +Object identifiers are allocated to object classes. The concept makes use of the concept of subclasses (see Recommendation X.501) (1988). + +Selected object classes provided by the directory systems specifications depend on the scope of public directory service chosen by the service provider. It is assumed that the presently defined selected object classes will allow the provision of a useful directory service. + +- Top +- Alias +- Country +- Locality +- Organization +- Organizational unit +- Person +- Organizational person +- Organizational role +- Group of names +- Residential person +- Application entity +- Application process +- DSA +- Device +- Strong authentication user +- Certification authority + +*Note 1* – A certain object class is used as a classificatory attribute type. + +*Note 2* – The definition of additional selected object classes for public directory service is for further study. + +*Note 3* – Messaging handling, in X.400-Series of Recommendations (1988), defined additional object classes for message handling system (MHS) specific use (see Anne E). + +## ANNEX D (to Recommendation F.500) + +### **Selected attribute types** + +It is assumed that the presently defined selected attribute types will provide a useful directory service. The implementation of the attribute types used in the public directory service are left for the decision of the service provider. Selected attribute types provided by the directory system specification, Recommendation X.520 (1988), are: + +- a) *System attribute types* + - Aliased object name + - Knowledge information + - Object class +- b) *Labelling attribute types* + - Common name + - Serial number + - Surname +- c) *Geographical attribute types* + - Country name + - Locality name + - State or province name + - Street address +- d) *Organizational attribute types* + - Organization name + - Organizational unit name + - Title +- e) *Explanatory attribute types* + - Business category + - Description + - Search guide +- f) *Postal attributes* + - Physical delivery office name + - Post office box + - Postal address + - Postal code + - Registered address +- g) *Telecommunications addressing attribute types* + - Destination indicator + - Facsimile telephone number + - ISDN address + - Registered address + - Telephone number + - Teletex terminal identifier + - Telex number + - X.121 address + +- h) *Preferences attribute types* + - Preferred delivery method +- i) *OSI application attribute types* + - Presentation address + - Supported application context +- j) *Relational attribute types* + - Member + - Owner + - Role occupant + - See also +- k) *Security attribute types* + - User password + - User certificate + - Authority revocation list + - Certificate revocation list + - Certification authority (CA) certificate + +*Note 1* – Other attribute types may be defined for local scope or on bilateral agreement. + +*Note 2* – The definition of additional selected attribute types for public directory services is for further study. + +*Note 3* – Messaging handling, in Recommendation X.402, defined additional attribute types for MHS specific use (see Annex F). + +## ANNEX E + +(to Recommendation F.500) + +### **MHS selected object classes** + +See Recommendation X.402 (1988) for further details. + +Selected object classes provided by the directory systems for MHS depend on the scope of the public directory service chosen by the service provider. It is assumed that the presently defined selected MHS object classes will allow the provision of a useful directory service that intercommunicates well with MHS as defined in the X.400-Series of Recommendations (1988). + +### *MHS object classes* + +- MHS (Generic MHS user information) +- MHS organizational user +- MHS distribution list +- MHS message store +- MHS message transfer agent +- MHS user agent + +## ANNEX F (to Recommendation F.500) + +### **MHS selected attribute types** + +It is assumed that the presently defined attribute types defined in the X.400-Series of Recommendations (1988) will provide a useful directory service for message handling systems. The implementation of the attribute types used in the public directory service are left for the decision of the service provider. MHS selected attribute types provided by the X.400 system specification, Recommendation X.402 (1988), are: + +### *MHS attribute types* + +- MHS deliverable content length +- MHS deliverable content types +- MHS deliverable encoded information types +- MHS distribution list members +- MHS distribution list submit permissions +- MHS message store +- MHS originator/recipient (O/R) addresses +- MHS preferred delivery methods +- MHS supported automatic actions +- MHS supported content types +- MHS supported optional attributes + +## ANNEX G (to Recommendation F.500) + +### **User visibility of the search operation** + +Some examples of filters are shown for the practical use. + +### G.1 *Possible examples* + +ORG = Organization name + +OUN = Organizational unit name + +#### G.1.1 *Sales units of TTT or marketing units of TNT* + +[(ORG = "TTT"), AND, (OUN = "SALES")] OR [(ORG = "TNT") AND, (OUN = "MARKETING")] + +#### G.1.2 *Marketing or sales units of TTT* + +(ORG = "TTT"), AND, [(OUN = "MARKETING", OR OUN = "SALES")] + +#### G.1.3 *All departments of TTT except Marketing* + +[(ORG = "TTT"), AND, (OBJECT CLASS = OUN)], AND NOT, [(OUN = "MARKETING")] OR +[(OUN = MARK\*)] + +#### G.1.4 *All organizations in a country whose telex numbers are in the range of 5030 to 5067* + +(OBJECT CLASS = ORG), AND, [(TLX ≤ 5067), AND, (TLX > 5030)] + +### G.2 *Practical use and effect of filters* + +#### G.2.1 *Task* + +“Retrieve” in the United States, the location (state or province), the telefax number, and voice telephone number for the sales departments of TTT or the marketing departments of TNT. The total elapsed time for retrieving the information should not exceed 10 minutes (600 ns) and the maximum number of objects found should not exceed 20. + +#### G.2.2 *Solution/action* + +##### *Action* + +##### SEARCH + +Criteria: Base object: “CTN = US”. + +subset: “whole subtree” + +##### Filter + +[(ORG = “TTT”, AND, OUN = “SALES”) + +, OR, (ORG = "TNT", AND, OUN = "MARKETING")] + +Service controls: { + +time limit = 600, + +size limit = 20, + +priority = medium } + +Selection: { + +FAX, + +TEL, + +STN } + +###### *Result* + +The directory will return the requested information within the limits designated by the requestor. If the limits are exceeded, an error indicating the limit that was exceeded and arbitrary collection of partial results are displayed in this example. + +## ANNEX H + +(to Recommendation F.500) + +### **Glossary of terms1)** + +*Note* – Some of the terms included are quoted from X.500-Series of Recommendations and are only included to enhance understanding of system related descriptions. Some of the text provided are definitions and others are of explanatory nature. The *Blue Book* fascicle named “Terms and Definitions” may be used as a further source. + +### H.1 **abandon** + +A directory operation to terminate a request. This operation is not guaranteed outside of the local scope. + +*Note* – This directory system operation is considered to be an optional user facility in the service context. + +--- + +1) Additional terms may be added. + +- H.2 **access control** +Method of controlling access to information held in the directory either for retrieval, managing or updating purposes. +- H.3 **ADD** +A directory operation to add an object entry or an alias entry to the directory information tree. +*Note* – This directory system operation is considered to be an optional user facility in the service context. +- H.4 **additional service controls** +Function of a directory system to control certain additional performance criteria. +*Note* – These service controls are considered to belong to additional optional user facilities. +- H.5 **Administration** +Denotes a public telecommunications administration or recognized private operating agency. +- H.6 **Administration directory management domain** +A directory management domain which is managed by an Administration or by a recognized private operating agency. +- H.7 **alias (entry)** +An entry of the class “alias” containing information used to provide an alternate name for an object. It points to the entry that actually contains the information. +- H.8 **alias name** +A name for an object where at least one of whose relative distinguished names is that of an alias entry. +- H.9 **attribute** +The information of a particular type concerning an object and appearing in an entry describing that object in the directory information base. +*Note* – See X.500-Series of Recommendations for further details. +- H.10 **attribute type** +That component of an attribute which indicates the nature of information given by that attribute. +- H.11 **attribute value** +A particular instance of information indicated by an attribute type. +- H.12 **attribute value assertion** +A proposition, which may be true, false, or undefined, concerning the values (or perhaps only the distinguished values) of an entry. + +### H.13 **authentication** + +Method to establish security services by means of simple or strong authentication. There are two kinds of authentication: data origin authentication and peer entity authentication. + +*Note – See Recommendation X.509 for more information.* + +### H.14 **authentication mechanisms** + +Authentication mechanisms are used to provide for encryption, data integrity and digital integrity. + +### H.15 **business category** + +Attribute type which specifies the commercial activity of some common objects, e.g. people. + +### H.16 **chaining** + +A feature used by the directory system to communicate between directory system agents to satisfy the users request. To achieve this, multiple directory system agents must be able to intercommunicate as peers. This feature may be inhibited by the user or service provider through service control parameters that are supplied with the user's request. + +*Note – A set of agreements is required between the domains directory system agents wanting to interact based on this method.* + +### H.17 **classified information** + +In the context of the directory, directories presently known as “white pages”, “yellow pages”, etc. + +#### H.18 **common name** + +In the context of directory systems: + +An attribute type identifying an object that is named. It is the name by which the object is commonly named, and conforms to the naming conventions of the country or culture with which the object is associated. + +In the context of message handling systems: + +Standard attribute identifying a user or distribution list relative to the entity denoted by another attribute (e.g. an organization name). (See Recommendation X.402.) + +#### H.19 **compare** + +An operation of the directory system to compare a value (which is supplied as an argument of the request) with the value(s) of a particular attribute type in a particular object entry. + +*Note – This directory system operation is considered to be an optional user facility in the service context.* + +#### H.20 **copy information** + +Replicated information. + +#### H.21 **country name** + +An attribute type that identifies a country. A country name is a unique designation of a country. When used as a component of a directory name, it identifies the country in which the named object is physically located or with which it is associated in some other important way. In the context of directory systems a value from ISO 3166 (Alpha 2 country codes) is used. + +- H.22 **description** +An attribute type which describes the associated object, e.g. as “yellow pages” entries. +- H.23 **destination indicator (public telegram)** +An attribute type specifying the country and city associated with the object (the addressee) needed to provide the public telegram service. +*Note* – See Recommendations F.1 and F.31. +- H.24 **directory** +A collection of open systems cooperating to provide directory services. +- H.25 **directory entry** +A part of the directory information base which contains information about an object. +- H.26 **directory information base** +The complete set of information to which the directory provides access, and which includes all of the pieces of information which can be read or manipulated using the operations of the directory. +- H.27 **directory information tree** +The directory information base considered as a tree, whose vertices (other than the root) are the directory entries. +*Note* – The term directory information tree is used instead of directory information base only in contexts where the tree structure of the information is relevant. +- H.28 **directory interrogation** +Methods to get results from a request to a directory by read, compare, list, search or abandon operations. +- H.29 **directory management domain** +A domain responsible for managing the information contained in a directory and the operation on this information. +- H.30 **directory modification** +Methods to change information in a directory by add entry, remove entry, modify entry or modify relative distinguished name functions. +- H.31 **directory name** +A construct that singles out a particular object from all other objects. A directory name must be unambiguous (that is, denote just one object). However, it need not to be unique (that is, be the only name which unambiguously denotes the object). +See also *name*. +- H.32 **directory schema** +The set of definitions and constraints concerning directory information tree structure, object class definitions, attribute types and syntaxes which characterize the directory information base. + +### H.33 **directory system agent** + +An open systems interconnection application process which is part of the directory, and whose role is to provide access to the directory information base for directory user agents and/or other directory system agents. + +### H.34 **directory user agent** + +An open systems interconnection application process which represents a user in accessing the directory. Each directory user agent serves a single user so that the directory may control access to directory information on the basis of user's identity. Directory user agents may also provide a range of local facilities to assist users to compose requests (queries) and interpret the responses. + +### H.35 **directory management domain** + +A collection of one or more directory system agents and zero or more directory user agents which is managed by a single organization. Management of a directory user agent by a directory management domain implies an ongoing responsibility for service to that directory user agent, e.g. maintenance, or in some cases ownership, by the directory management domain. + +### H.36 **distinguished name** + +The sequence of relative distinguished names of the entry which represents the object and those of all its subordinate entries (in descending order). Because of the one to one correspondence between objects and object entries, the distinguished name of an object can be considered to also identify the object entry. + +### H.37 **distinguished value** + +An attribute value in an entry which has been designated to appear in the relative distinguished name of the entry. + +### H.38 **distribution list** + +List of O/R addresses for message handling services stored in the directory. + +*Note* – This feature is considered to be an optional user facility in the service context. + +### H.39 **directory information tree structure** + +The definition for an entry of an object class of the permissible object class or classes to which the immediate superior (or subordinate) may belong and its permissible relative distinguished name attribute types. + +### H.40 **do not dereference alias** + +A service control which allows to prohibit that any alias used to identify the entry effected by an operation is to be dereferenced. + +See also *alias*. + +### H.41 **do not use copy** + +A service control allowing for prohibition of copied information. + +### H.42 **entry (directory entry)** + +A part of the directory information base which describes a particular object, and which consists of information that the directory holds about that object. + +### H.43 **error code** + +Information provided from the directory system for the purpose of indicating to the requestor why a request could not be performed sufficiently. + +*Note* – A local directory domain may transfer the information to the requestor in a way appropriate to local requirements. Error codes may refer to service error, attribute error, update error, security error, referral error, abandon error or name error. They are transferred to service messages for the user. + +### H.44 **facsimile telephone number** + +An attribute type which specifies a telephone number for a facsimile terminal (and optionally its parameters) associated with an object. + +### H.45 **filter** + +A filter parameter applies a test to a particular entry and either is satisfied or not by the entry. The filter is expressed in terms of assertions about the presence or value of certain attributes of the entry, and is satisfied if and only if it evaluates to TRUE. + +### H.46 **intercommunication** + +In the context of directory services a relationship between services, where one of the services is a directory service, enabling the user of a service to communicate with the directory. + +*Note* – The term also applies for the relation between public and private directories, for the relation between directory services of different service providers and for the relation between directory management domains. + +### H.47 **ISDN address** + +An attribute type which specifies an ISDN address associated with an object. + +### H.48 **knowledge information** + +An attribute type which specifies a human-readable accumulated description of knowledge mastered by a specific directory system agent. + +### H.49 **locality name** + +An attribute type which specifies a locality. When used as a component of a directory name, it identifies a geographical area or locality in which the named object is physically located or with which it is associated in some other important way. + +### H.50 **list** + +An operation in the directory system to obtain a list of immediate subordinates of an explicitly identified entry. Under some circumstances, the list returned may be incomplete. + +*Note* – This directory system operation is considered to be an optional user facility in the service context. + +### H.51 **local scope** + +A service control which restricts the scope of directory operations. + +*Note* – The definition of local scope is itself a local matter, and may, for example, mean a limit within a single directory system agent or a single directory management domain. + +### H.52 **member** + +An attribute type which specifies a group of names associated with the object. + +### H.53 **modify** + +An operation in the directory system to perform a serie of one or more of the following modifications to a single entry: + +- add a new attribute; +- remove an attribute; +- add attribute values; +- remove attribute values; +- replace attribute values; +- modify the distinguished name of a leaf entry (automatically all subordinates are renamed); +- modify alias; +- modify entry. + +*Note* – This directory system operation is considered to be an optional user facility in the service context. + +### H.54 **modify operations** + +These are operations to alter the contents of the directory: add entry, remove entry, modify entry and modify relative distinguished name. + +### H.55 **multicasting** + +This is a special case of distributing simultaneously a request to more than one directory system agent. See Recommendation X.518. + +*Note* – A set of agreements is required between the domains wanting to interact based on this method. + +### H.56 **name** + +In the context of a directory, the designation of entries and parts thereof. A name must be unambiguous, that is, denote just one object. However, a name need not be unique, that is, not be the only name that unambiguously denotes the object. + +*Note* – See the X.500-Series of Recommendations for further study. + +### H.57 **naming authority** + +An authority responsible for the allocation of names. Each object whose object entry is located at a node in the directory information tree is, or is closely associated with, a naming authority. + +In the context of public directory services, the Administration directory management domain administers the part of the directory information tree covered by entries of that domain. It may act as naming authority for the distinguished names used in the scope of the domain. + +### H.58 **object (of interest)** + +Anything in some “world”, generally the world of telecommunications and information processing or some part thereof, which is identifiable (can be named), and which is of interest to hold information on the directory information base. + +### H.59 **object entry** + +An entry which is the primary collection of information in the directory information base about an object, and which can therefore be said to represent that object in the directory information base. + +- H.60 **object class** +An identified family of objects (or conceivable objects) which share certain characteristics. +*Note* – See the X.500-Series of Recommendations for further study. +- H.61 **O/R address; originator/recipient address** +Address of an originator/recipient of messages in the context of message handling. +- H.62 **organization name** +An attribute type which specifies an organization. When used as a component of a directory name it identifies an organization with which the named object is affiliated. +- H.63 **organization unit name** +An attribute type which specifies an organizational unit. When used as a component of a directory name it identifies an organizational unit with which the named object is affiliated. +- H.64 **owner** +In the context of a directory, that attribute type specifying the name of some object which has some responsibility for the associated object. +- H.65 **physical delivery office name** +An attribute type which specifies the name of the city, village, etc., where a physical delivery office is situated. +- H.66 **post office box** +An attribute type which specifies the post office box by which the object will receive physical delivery. If present, the attribute value is part of the object's postal address. +- H.67 **postal address** +An attribute type which specifies the address information required for the physical delivery of postal messages by the postal authority to the named object. Formatted and unformatted postal addresses exist. +*Note* – See also Recommendations F.401 and X.520. +- H.68 **postal code** +An attribute type which specifies the postal code of the named object. If this attribute value is present it will be part of the object's postal address. +- H.69 **preferred delivery method** +An attribute type which specifies the object's priority regarding the method to be used for communicating with it. +- H.70 **presentation address** +An attribute type which specifies a presentation address associated with an object representing an open systems interconnection application entry. +- H.71 **priority** +A service control which specifies the priority of a request (low, medium, high) for the service. This is not a guaranteed service in that the directory as a whole does not implement queuing. There is no relationship implied with the use of priorities in underlying layers. + +### H.72 **private directory management domain** + +A directory management domain managed by another organization than an Administration. + +### H.73 **public directory service** + +A service provided by Administrations to subscribers and users for the purpose of obtaining information on addresses for telecommunication services and other related information from an electronic directory. + +### H.74 **read operation** + +An operation of the directory system to extract an explicitly identified entry. It may also be used to verify a distinguished name. + +*Note* – This directory system operation is considered to be a basic service feature in the service context. + +### H.75 **referral** + +Request handling by the directory system agent in the case of failing to find the requested information in the first directory system agent. In this case the directory may return a referral, which suggests an alternative access point at which the directory user agent can make its request. + +*Note 1* – This is an alternative method to chaining or multicasting. The implementation is a local matter. + +*Note 2* – A set of agreements is required between the domains (directory system agents) wanting to interact on the basis of this method. Whether referrals are presented to the user or not is a local matter. It has to take into account whether the domain (directory system agent) being referred to will accept requests from these users. + +*Note 3* – Referrals to domains (directory system agents) without prior agreement (including accounting procedures) with them are undesired. + +### H.76 **registered address** + +An attribute type which specifies a mnemonic for an address associated with an object at a particular city location. The mnemonic is registered in the country in which the city is located and is used in the provision of the public telegram service. + +### H.77 **relative distinguished name** + +The unique name of an entry. It consists of a particular sequence of attribute value assertions, each of which is true, concerning the distinguished values of an entry. + +### H.78 **requestor** + +The subscriber, user or system entity making a particular request to the directory. + +### H.79 **role occupant** + +An attribute type which specifies the name of an object that fulfils an organizational role. An attribute value for role occupant is a distinguished name. + +### H.80 **search guide** + +An attribute type which specifies information of suggested search criteria which may be included in some entries expected to be a convenient base-object for the search operation, e.g. country or organization. + +### H.81 **search operation** + +An operation in the directory system to search a portion of the directory information tree for entries of interest, and to return selected information from those entries. + +*Note* – This directory system operation is considered to be a basic service feature in the service context. + +### H.82 **security capabilities** + +Capabilities of a directory system to provide protection against security threats. + +*Note 1* – These directory system capabilities are considered to be additional optional user facilities in the service context. + +*Note 2* – See Recommendation X.509 for explanation of security capabilities. + +### H.83 **see also** + +An attribute type which specifies names of other objects which may be other aspects (in some sense) of the same real-world object. + +### H.84 **serial number** + +**An attribute type which specifies an identifier, the serial number of a device.** + +### H.85 **service control** + +A function of a directory system to control certain performance criteria. A service control parameter contains the controls, if any, that are to direct the provision of the service. + +*Note* – One service control in the directory system (time limit) is an essential optional user facility. Other specific ones are additional optional user facilities in the service context, if the service provider offers them. See also § 4 of Recommendation F.500. + +### H.86 **shadowing** + +Replication between two directory system agents whereby shadowed information is copied and maintained using the directory information shadowing protocol. + +*Note* – Caches are not being used for public directory services. + +### H.87 **shadowing agreement** + +The terms conforming to an agreement between two directory system agent administrative authorities. + +*Note* – Appropriate procedures for accounting between public service providers will need to be developed. + +#### H.88 **state or province name** + +Identifies the geographical subdivision in which the named object is physically located or with which it is associated in some other important way. + +#### H.89 **street address** + +An attribute type which specifies a site for the local distribution and physical delivery in a postal address, i.e. the street name, place, avenue and the house number. When used as a component of a directory name, it identifies the street address at which the named object is located or with which it is associated in some other important way. + +### H.90 **subclass** + +Relative subordinate to a superclass, an object class derived from a superclass. The members of the subclass share all the characteristics of another object class (the superclass) and additional characteristics possessed by none of the members of that class (the superclass). + +### H.91 **subscriber** + +A user of a telecommunication service, normally based on a contract with the provider of a public service. + +### H.92 **superclass** + +Relative superior to a subclass, an object class from which a subclass is derived. + +### H.93 **supported application context** + +An attribute type which specifies the object identifier of an application context that the object (an Open Systems Interconnection application entity) supports. + +### H.94 **surname** + +An attribute type which specifies the linguistic construct which normally is inherited by an individual from the individual's parent or assumed by marriage, and by which the individual is commonly known. + +### H.95 **telephone number** + +An attribute type which specifies a telephone number associated with an object. + +*Note* – The format of internationally agreed telephone numbers follows Recommendation E.164. + +### H.96 **teletex terminal identifier** + +An attributed type which specifies the teletex terminal identifier for a teletex terminal associated with an object. + +*Note* – The format follows Recommendation F.200. + +### H.97 **telex answer-back** + +An attribute type which specifies the telex terminal identifier for a telex terminal associated with an object. + +*Note* – The format follows Recommendation F.60. + +### H.98 **telex number** + +An attribute type which specifies the telex number, country code, and answer-back code of a telex terminal. + +*Note* – The format follows Recommendation F.69. + +#### H.99 **time limit** + +A service control that indicates the maximum elapsed time, in seconds, within which the service should be provided. If the constraint cannot be met, an error is reported, unless it was a search or a list operation, in which case partial results should be returned to the directory user agent with the indication that a time limit problem has been encountered. If this component is omitted, no time limit is implied. + +*Note* – This service control is an essential optional user facility. + +### H.100 **title** + +An attribute type which specifies the designated position or function of the object within an organization. + +- H.101 **user** +In telecommunication service context: a human being using a service. +In a technical context: a human being, an entity or a process. +*Note* – A user will not necessarily be a subscriber of a telecommunication service. +- H.102 *user certificate* +See Recommendations X.520 and X.509. +- H.103 **wildcard** +In the context of directory services, a way to replace unknown parts of attributes for a request to the directory. +- H.104 **user password** +A sequence of characters to identify a user. +- H.105 **videotex user number** +An attribute type which specifies a videotex user number associated with an object. +- H.106 **white pages** +See under “classified information”. +- H.107 **X.121 address** +An attribute type which specifies a number from the X.121 numbering plan associated with an object. +- H.108 *yellow pages* +See under “classified information”. + diff --git a/marked/F/T-REC-F.510-200302-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.510-200302-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..368067765fe2f2723c2ae04f0707a1efc2dc204a --- /dev/null +++ b/marked/F/T-REC-F.510-200302-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3f859673df2ee3081065ccc834ef8c75f690bd9c8a4ae3cfc6cbf36212b86921 +size 8204 diff --git a/marked/F/T-REC-F.510-200302-I_PDF-E/2e399449b7f62d40167a65ca88888917_img.jpg b/marked/F/T-REC-F.510-200302-I_PDF-E/2e399449b7f62d40167a65ca88888917_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..0bf31643895444935fa47f7abe324cb138e6c863 --- /dev/null +++ b/marked/F/T-REC-F.510-200302-I_PDF-E/2e399449b7f62d40167a65ca88888917_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8c8c93be705ee4a4b809151b669192c4fa93735db3afa3507eaa6e08e7231c90 +size 28511 diff --git a/marked/F/T-REC-F.510-200302-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg b/marked/F/T-REC-F.510-200302-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..87edc6d196ce43fe4ccc958e5b0a1c0e2ebba9b6 --- /dev/null +++ b/marked/F/T-REC-F.510-200302-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:23885b0b22a8c3a5d587e3100de8e31f3b1189aedd454b8fff8126e932d160d0 +size 31451 diff --git a/marked/F/T-REC-F.510-200302-I_PDF-E/40a8c30f7ea5ecea4912e040c97c5b9c_img.jpg b/marked/F/T-REC-F.510-200302-I_PDF-E/40a8c30f7ea5ecea4912e040c97c5b9c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f68324f9fc4de063bc08dbf423d45f9500b453af --- /dev/null +++ b/marked/F/T-REC-F.510-200302-I_PDF-E/40a8c30f7ea5ecea4912e040c97c5b9c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:15fdc20e7523eee19cb25006cababb8aef4d03ee7266f6a0ececf90f76028a77 +size 88944 diff --git a/marked/F/T-REC-F.510-200302-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg b/marked/F/T-REC-F.510-200302-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..6d62c4d480187cfe4a59f997db9af5e8ad52efb2 --- /dev/null +++ b/marked/F/T-REC-F.510-200302-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1ba2632814591e80ccdcb6978ce5aeeee353b780e6b37086fcfc27eb9f3b4d4d +size 25159 diff --git a/marked/F/T-REC-F.510-200302-I_PDF-E/8fbdfc3d17fb1dae7b2d8f5a287fa9fc_img.jpg b/marked/F/T-REC-F.510-200302-I_PDF-E/8fbdfc3d17fb1dae7b2d8f5a287fa9fc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..370c91127d6611958a433da12ce20a9d6aeb26f4 --- /dev/null +++ b/marked/F/T-REC-F.510-200302-I_PDF-E/8fbdfc3d17fb1dae7b2d8f5a287fa9fc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:48c1fb0598b3793edd5b7f235de274ef06e7ea48deec9e043cbee090b59c470e +size 66238 diff --git a/marked/F/T-REC-F.510-200302-I_PDF-E/a0e8fe7862a6d7341faf5dac275277cc_img.jpg b/marked/F/T-REC-F.510-200302-I_PDF-E/a0e8fe7862a6d7341faf5dac275277cc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..8c1efbb673f096a3c0c228f2fac2dbfc3ee2a9ea --- /dev/null +++ b/marked/F/T-REC-F.510-200302-I_PDF-E/a0e8fe7862a6d7341faf5dac275277cc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9315092dba1da26b5513c9cb1be06f7bce63102930e6e3c7290e9399353668bc +size 69127 diff --git a/marked/F/T-REC-F.510-200302-I_PDF-E/af7916c89a458fdab6c3f443217388ae_img.jpg b/marked/F/T-REC-F.510-200302-I_PDF-E/af7916c89a458fdab6c3f443217388ae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..75f5673350a8d6afd47bfefe554d529cd682d273 --- /dev/null +++ b/marked/F/T-REC-F.510-200302-I_PDF-E/af7916c89a458fdab6c3f443217388ae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8d079c50e7257cf769545a518f397673958009aa311d05f53a810a856bd33582 +size 16687 diff --git a/marked/F/T-REC-F.510-200302-I_PDF-E/bd671b21db63e6fdb2196e9b18502aac_img.jpg b/marked/F/T-REC-F.510-200302-I_PDF-E/bd671b21db63e6fdb2196e9b18502aac_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..863e021399f2e971c1b5be46c58b85207fcdcb21 --- /dev/null +++ b/marked/F/T-REC-F.510-200302-I_PDF-E/bd671b21db63e6fdb2196e9b18502aac_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2651840a3d5a54b38bace1170c5f3aff71daf5a97e359495fec26cedd0240e49 +size 21562 diff --git a/marked/F/T-REC-F.510-200302-I_PDF-E/raw.md b/marked/F/T-REC-F.510-200302-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..d55a96e205bad8a9f46a5b695c29e5e3031a75f3 --- /dev/null +++ b/marked/F/T-REC-F.510-200302-I_PDF-E/raw.md @@ -0,0 +1,1854 @@ + + +![ITU logo](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +The logo of the International Telecommunication Union (ITU) features the letters 'ITU' in a bold, sans-serif font, superimposed on a stylized globe with intersecting lines. + +ITU logo + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.510** + +(02/2003) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Directory services + +--- + +**Automated directory assistance – White pages +service definition** + +ITU-T Recommendation F.510 + +--- + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | | +| | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | | +| | F.600–F.699 | +| AUDIOVISUAL SERVICES | | +| | F.700–F.799 | +| ISDN SERVICES | | +| | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | | +| | F.850–F.899 | +| HUMAN FACTORS | | +| | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# **ITU-T Recommendation F.510** + +# **Automated directory assistance – White pages service definition** + +## **Summary** + +Given the rapid multiplication and expansion of telecommunication services, there is a growing need for subscribers to those telecommunications services to be able to communicate with each other. In order to facilitate such intercommunications for the subscribers of the various services, operator assisted and public Directory Services will be required. The Service Definition in this Recommendation supports the provision of those Directory Services. + +## **Source** + +ITU-T Recommendation F.510 was revised by ITU-T Study Group 17 (2001-2004) and approved under the WTSA Resolution 1 procedure on 13 February 2003. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications. The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementors are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database. + +© ITU 2003 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## CONTENTS + +| | | Page | +|------|----------------------------------------------------|------| +| 1 | Scope..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 2 | +| 4 | Abbreviations..... | 4 | +| 5 | Conventions ..... | 4 | +| 6 | Directory Services model ..... | 4 | +| 6.1 | Directory Service use ..... | 4 | +| 6.2 | Directory Service Provision ..... | 6 | +| 6.3 | Directory Service Management..... | 6 | +| 7 | Directory information model ..... | 7 | +| 7.1 | Information structure ..... | 7 | +| 7.2 | Geography ..... | 9 | +| 7.3 | Business Classification..... | 11 | +| 7.4 | Organizational structure ..... | 11 | +| 7.5 | Subscriber Entry ..... | 11 | +| 8 | Service specification ..... | 15 | +| 9 | Common service specification elements ..... | 16 | +| 9.1 | Character set ..... | 17 | +| 9.2 | Matching rules ..... | 17 | +| 9.3 | Language differences..... | 19 | +| 9.4 | Ordering of entries within a response..... | 19 | +| 9.5 | Hierarchical Groups..... | 20 | +| 9.6 | Number of entries ..... | 21 | +| 9.7 | Entry count ..... | 21 | +| 9.8 | Alternate attribute values..... | 21 | +| 9.9 | Weighted attribute values ..... | 21 | +| 9.10 | Geographical extension ..... | 22 | +| 9.11 | Ignore if absent ..... | 22 | +| 9.12 | Paging ..... | 22 | +| 10 | White Pages DQ ..... | 22 | +| 10.1 | Service Data Model ..... | 22 | +| 10.2 | Search controls and response indicators..... | 25 | +| 10.3 | Predicates..... | 26 | +| 10.4 | Common query conditions ..... | 27 | +| 10.5 | Basic service offering..... | 28 | +| 10.6 | Enhanced service offering ..... | 32 | +| 11 | Directory Service Management Service Profile ..... | 37 | + +| | Page | +|----------------------------------------------------|-------------| +| 11.1 Introduction ..... | 37 | +| 11.2 Directory Service Managed Objects..... | 38 | +| 11.3 Directory Service Management Procedures ..... | 39 | +| 12 Operational issues ..... | 40 | +| 12.1 Fallback and User Support ..... | 40 | +| 12.2 Security ..... | 40 | +| 13 Charging and accounting ..... | 40 | +| 14 Quality of Service ..... | 40 | +| Annex A – Messages..... | 40 | +| Annex B – List of symbols ..... | 43 | +| Annex C – Service Data Model ..... | 44 | +| Annex D – Object Modelling Notation..... | 45 | + +# Introduction + +This Recommendation details a service definition for the provision of electronic access to an International number information service by cooperating ROAs. + +This service is suitable for use both by the operators of ROAs providing a DQ service to their customers, and by direct access of authorized customers. + +The service addresses not only the provision of fixed-line telephone numbers, but also the provision of the number information of other communications media – facsimile, mobile, videotelephony, etc. The service can also provide other communications address information such as electronic mail and application process addresses. + +The service definition meets the current and foreseeable future needs of the service and is designed for worldwide use if required. + +The service definition provides the basis for the progressive delivery of a constantly improving DQ service. The definition allows for the functional capabilities of systems (all or individual) to be expanded, and continually evolved, within but not limited to the constraints set by the definition. + +The service definition enables, and should encourage, the implementation of cooperative systems through which an efficient International DQ service is provided. + +The service includes, but is significantly different from that defined by ITU-T Rec. E.115, in that it defines services for both operators, end-users and processes, provides telecommunications addresses in addition to telephone numbers. + + + +# Automated directory assistance – White pages service definition + +# 1 Scope + +This service definition is the basis for the establishment of a truly International DQ service through the standardized interworking of numerous individual DQ information services. + +This service definition provides a general framework for the provision of electronic access to international Directory Services. It defines the requirements for, and the service features associated with the provision of such Directory Services. It specifies the requirements to support Directory Services that may be offered to both Directory Service operators and to end-users. It encompasses the operational aspects of the service and includes Quality of Service and Charging and Accounting. + +This service definition defines the White Pages Directory Service. However, the services and elements specified herein may be used to define other services, such as Yellow Pages. + +A DQ service might have more than one access point available for retrieving information from another country. At each of these access points information might be retrieved from one or more national service or information providers. There may be various types and levels of service quality provided. This service definition does not define any mechanism for selection of provider and of any explicit mechanism for selection of quality of service. + +This service definition assumes that information is retrieved from a single country at a time. The name of the country must be included in the request parameters. Removal of the single country restriction is for further study. + +This service definition does not specify the method of implementation which may use any appropriate protocols (or combination of protocols). The terms used in this Recommendation adopt their conventional meaning (dictionary definition), and the meaning of terms, such as "attribute", should not be assumed from other (standards) documents in which they may be defined. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +- ITU-T Recommendation D.37 (1996), *Accounting and settlement principles applicable to the provision of public directory services between interconnected Directory Management Domains*. +- ITU-T Recommendation E.104 (1995), *International telephone directory assistance service and public access*. +- ITU-T Recommendation E.115 (1995), *Computerized directory assistance*. +- ITU-T Recommendation E.123 (2001), *Notation for national and international telephone numbers, e-mail addresses and Web addresses*. +- ITU-T Recommendation E.160 (1993), *Definitions relating to national and international numbering plans*. + +- ITU-T Recommendation T.51 (1992), *Latin based coded character sets for telematic services*. +- ITU-T Recommendation T.51 (1992)/Amd.1 (1995), *Latin based coded character sets for telematic services – Amendment 1*. +- ITU-T Recommendation T.52 (1993), *Non-Latin coded character sets for telematic services*. +- ITU-T Recommendation X.509 (2000) | ISO/IEC 9594-8:2001, *Information technology – Open Systems Interconnection – The Directory: Public-key and attribute certificate frameworks*. +- ISO 3166-1:1997, *Codes for the representation of names of countries and their subdivisions – Part 1: Country codes*. + +## 3 Definitions + +3.1 This Recommendation defines the following terms. + +3.1.1 **directory information:** The information held within a directory associated with, and for the purpose of identifying, a communications address, as described in 7.5.2. + +3.1.2 **directory entry:** Information associated with a particular type of real world object. An object consists of one or more attributes. This Recommendation defines the objects and the information required to support the service. + +3.1.3 **attribute:** An attribute has a type and a value. A particular attribute type may be used in one or more directory entries. A new attribute may also be formed by the combination of (simpler) attribute types. + +3.1.4 **key attribute:** An attribute parameter in a search request that is required to match a real-world attribute, e.g., a locality, but not to be ambiguous to a level determined by local considerations. + +3.1.5 **service provider:** A Service Provider is the function, supported by an ROA, that supplies a Directory Service to one or more Query Service Agents. + +3.1.6 **query service agent:** The Query Service Agent issues physical search requests to and receives responses from one or more Service Providers. The Query Service Agent may perform Service Provider selection, query modification (or rejection) on the basis of knowledge of the Directory Services offered by Service Providers to which it has access. + +3.1.7 **directory service:** A Directory Service is a service which supports the following functionalities. + +- 1) Standardization of directory objects and attributes and their interpretation and utilization. +- 2) Defined Directory Information. +- 3) Unambiguous definition of the enquiries that have to be supported by the Service Providers. +- 4) Resilience to requests specifying search attributes and matching types which are not supported by the Service Provider. +- 5) Means of communicating to the Directory Service user what is required in order to fulfil request, i.e., definition of services on offer. +- 6) Defined matching rules and character sets, including its conversion and interpretation rules. +- 7) Defined handling of language translation. +- 8) Ordering of returned entries. +- 9) Definition of information returned for unsuccessful or failed searches. + +- 3.1.8 hierarchical group:** A set of directory entries which form the hierarchical external view of a larger body or organization. +- 3.1.9 named service:** A Named Service is a Directory Service which comprises a specific set of Directory functions and Directory data that provide a coherent service. The only Named Service that is defined to date is the White Pages Directory Service defined in this Recommendation. +- 3.1.10 service specification:** A service specification defines a specific Named Service. The specification includes the elements, which form part of all Directory Services, defined in clause 8 and the specification utilizes, where appropriate, common terminology and definitions which are defined in clause 9. +- 3.1.11 service data model:** The Service Data Model defines the specific directory information required to support a Named Service. The Service Data Model is defined using object modelling techniques (see Annex D) and defines the data object, attributes and relationships between object types. +- 3.2** The following definitions are used to define directory searches in clause 10: +- 3.2.1** Parameter classifications (used in clause 10): +- 3.2.1.1 mandatory (M):** A request parameter or response parameter that is mandatory shall be present. +- 3.2.1.2 optional (O):** A request parameter or response parameter that is optional may be present. +- 3.2.1.3 omitted (X):** A request parameter or response parameter that is omitted shall not be present. +- 3.2.1.4 not applicable (–):** The use of this request parameter or response parameter is not applicable for the search in question. +- 3.2.2 predicate:** A predicate (Pn) describes the conditions where a request parameter or response parameter may be mandatory, optional or omitted. +- 3.2.3 target object:** The primary object that contains the information to which the search request applies. +- 3.2.4 search request parameters:** The items of information including attributes that form the search request. +- 3.2.5 search response parameters:** The items of information including attributes for the entries returned in the search response. +- 3.2.6 search parameters:** The items of information including attributes that form both the search request and the search response. +- 3.2.7 SearchType:** A SearchType uniquely identifies each of the searches defined in 10.5 and 10.6. +- 3.2.8 message:** Information in the search response which is identified by a code and includes, when necessary, precise information, such as attribute types, attribute values or other identifiers, that relate to problems detected. Messages are listed in Annex A. +- 3.2.9 matching rule:** A matching rule defines the relation between a request and the entries returned in the corresponding response. A matching rule is applied to each request parameter by the Service Provider to select entries, that satisfy the matching rule, to be contained in the response. Only entries that satisfy the matching rules for all the parameters in the request shall be contained in the response. Matching rules are indicated as MR1, MR2, etc. and are defined in 9.2.2, 9.2.3 and 9.2.4. +- 3.2.10 essential:** The attributes that are required to support the minimum service (defined in clause 10) are identified as essential attributes. + +**3.2.11 additional:** The attributes that may be provided, in order to support higher levels of service (defined in clause 10), are identified as additional attributes. + +## **4 Abbreviations** + +This Recommendation uses the following abbreviations. + +| | | +|------|--------------------------------------------------------| +| DQ | Directory Enquiry | +| FTP | File Transfer Protocol | +| HTTP | Hypertext Transfer Protocol | +| ISDN | Integrated Services Digital Network | +| NACE | Nomenclature des Activités de la Communauté Européenne | +| NDC | National Destination Code | +| PSTN | Public Switched Telephone Network | +| ROA | Recognized Operating Agency | +| RWL | Real World Locality | +| SIC | Standard Industry Code | +| SMS | Short Message Service | +| SMTP | Simple Mail Transfer Protocol | + +## **5 Conventions** + +The symbols, such as MR1 to indicate WordExact matching, which are defined in clauses 7 and 9, and used in clause 10 are listed in Annex B. + +The Service Data Model described in clause 10 is formally defined in Annex C using the Object Modelling notation described in Annex D. + +# **6 Directory Services model** + +This clause describes the organization of components of the service for both the delivery of Directory Services to users and the management of those services by administrators. + +## **6.1 Directory Service use** + +The public Directory Service shall be accessible not only to professional Directory Service enquiry operators but also to end-users, which can be human users and application processes. + +The service must allow for different facilities to be made available to different classes of user, e.g., available to operators, but not end-users. This F.510 Service supports a number of facilities but does not define specific classes of user or the combination of facilities appropriate to, or the functionality offered to, any particular class of user. + +The F.510 Service supports a set of query functions that may be utilized by a local agent in order to offer appropriate facilities to different types of end-users, as shown in Figure 1. + +The delivery of the service is subject to legal and privacy rules which are outside the scope of this Recommendation. + +![Diagram of Figure 1/F.510 – F.510 Service Interface. It shows an 'End-user' (oval) connected to a 'Local Agent' (rectangle), which is connected to an 'F.510 Service' (rectangle) via a dashed vertical line labeled 'F.510 Service Interface'. The label 'F.510_F01' is at the bottom right.](af7916c89a458fdab6c3f443217388ae_img.jpg) + +``` + +graph LR + End-user((End-user)) --- Local-Agent[Local Agent] + Local-Agent --- F510-Service[F.510 Service] + subgraph Interface + direction TB + F510-Interface[F.510 Service Interface] + end + F510-Interface -.-> F510-Service + F510_F01[F.510_F01] + +``` + +Diagram of Figure 1/F.510 – F.510 Service Interface. It shows an 'End-user' (oval) connected to a 'Local Agent' (rectangle), which is connected to an 'F.510 Service' (rectangle) via a dashed vertical line labeled 'F.510 Service Interface'. The label 'F.510\_F01' is at the bottom right. + +**Figure 1/F.510 – F.510 Service Interface** + +### 6.1.1 Local Agent + +A Local Agent provides access to a Directory Service on behalf of an end-user. The Local Agent is a person, such as an enquiry operator using computerized facilities (i.e., PC or telematic terminal), or a function (or both) that utilizes the services at the F.510 Service Interface. + +A Local Agent acts on behalf of the end-user in the usage of one or more Directory Services. Typically, the Local Agent has detailed knowledge of specific Directory Services in terms of their definition, physical location and accessibility. The Local Agent provides enhanced service usability and physical data location transparency to Directory Users. + +The Local Agent may also provide and control access to a range of Directory Services that provide different data and quality of service. The choice of service may also be subject to regulatory, legal or commercial considerations. This service definition does not define any parameters relating to service selection and the choice of service is outside the scope of this service definition. + +### 6.1.2 End-user + +An end-user is the ultimate consumer of Directory Services and can be a human user or application process. The end-user utilizes the services of a Local Agent in order to interact with Directory Service Providers. + +### 6.1.3 Possible service scenarios + +Figures 2 and 3 show different possible arrangements for the provision of operated assisted and direct access services. Services and mechanisms, such as access control, user preferences and help (assistance) may be provided by a human operator, terminal application or access service function, possibly in combination. The definition of these services and mechanisms is outside the scope of this Recommendation. + +![Diagram of Figure 2/F.510 – Operator assisted (dial-up) Service. It shows an 'End-user' (oval) connected to a dashed box labeled 'Local Agent'. Inside the box are three rectangles: 'Operator', 'Operator terminal', and 'Access service'. The box is connected to an 'F.510 Service' (rectangle) via a dashed vertical line labeled 'F.510 Service Interface'. The label 'F.510_F02' is at the bottom right.](2e399449b7f62d40167a65ca88888917_img.jpg) + +``` + +graph LR + End-user((End-user)) --- Local-Agent + subgraph Local-Agent [Local Agent] + Operator[Operator] + Operator-terminal[Operator terminal] + Access-service[Access service] + end + Local-Agent --- F510-Service[F.510 Service] + subgraph Interface + direction TB + F510-Interface[F.510 Service Interface] + end + F510-Interface -.-> F510-Service + F510_F02[F.510_F02] + +``` + +Diagram of Figure 2/F.510 – Operator assisted (dial-up) Service. It shows an 'End-user' (oval) connected to a dashed box labeled 'Local Agent'. Inside the box are three rectangles: 'Operator', 'Operator terminal', and 'Access service'. The box is connected to an 'F.510 Service' (rectangle) via a dashed vertical line labeled 'F.510 Service Interface'. The label 'F.510\_F02' is at the bottom right. + +**Figure 2/F.510 – Operator assisted (dial-up) Service** + +![Diagram of Figure 3/F.510: Direct customer (computerized) access service. An oval labeled 'End-user' is connected to a dashed box labeled 'Local Agent'. Inside the 'Local Agent' box are two rectangles: 'End-user process e.g., terminal' and 'Access Service'. A solid line connects the 'Access Service' to a rectangle labeled 'F.510 Service'. A vertical dashed line labeled 'F.510 Service Interface' separates the 'Local Agent' from the 'F.510 Service'. The label 'F.510_F03' is at the bottom right.](4801720824e4b5e2361a5564f91cfb70_img.jpg) + +Diagram of Figure 3/F.510: Direct customer (computerized) access service. An oval labeled 'End-user' is connected to a dashed box labeled 'Local Agent'. Inside the 'Local Agent' box are two rectangles: 'End-user process e.g., terminal' and 'Access Service'. A solid line connects the 'Access Service' to a rectangle labeled 'F.510 Service'. A vertical dashed line labeled 'F.510 Service Interface' separates the 'Local Agent' from the 'F.510 Service'. The label 'F.510\_F03' is at the bottom right. + +**Figure 3/F.510 – Direct customer (computerized) access service** + +## 6.2 Directory Service Provision + +The F.510 Service is refined into two components, as in Figure 4: the Query Service Agent and the Service Provider which interact to provide directory query services that satisfy regulatory, geographical and language differences between Directory Service users and Directory Service Providers in different domains. + +![Diagram of Figure 4/F.510: F.510 Service Provision. A vertical dashed line labeled 'F.510 Service Interface' is on the left. To its right is a dashed box labeled 'F.510 Service' containing two rectangles: 'Query Service Agent' and 'Service Provider'. A solid line connects the 'Query Service Agent' to the 'Service Provider'. The label 'F.510_F04' is at the bottom right.](bd671b21db63e6fdb2196e9b18502aac_img.jpg) + +Diagram of Figure 4/F.510: F.510 Service Provision. A vertical dashed line labeled 'F.510 Service Interface' is on the left. To its right is a dashed box labeled 'F.510 Service' containing two rectangles: 'Query Service Agent' and 'Service Provider'. A solid line connects the 'Query Service Agent' to the 'Service Provider'. The label 'F.510\_F04' is at the bottom right. + +**Figure 4/F.510 – F.510 Service Provision** + +### 6.2.1 Service Provider + +A Service Provider is the function that supplies a Directory Service to one or more Query Service Agents. Additionally, the Directory Service Provider may regulate access to the service against a set of known legitimate Query Service Agents. + +A Service Provider may utilize the services of one or more other Service Providers in order to provide a Directory Service. + +### 6.2.2 Query Service Agent + +The Query Service Agent issues physical requests to and receives responses from one or more Service Providers. The Query Service Agent may perform Service Provider selection and query modification (or rejection) on the basis of knowledge of the Directory Services offered by Service Providers to which it has access. + +## 6.3 Directory Service Management + +Directory Service Management provides the mechanisms to supply information for management decisions and the procedures to execute those decisions. Management functions are associated with the Query Service Agent and Service Provider to implement the Directory Service offered by a Service Provider and consumed by a Query Service Agent. The information and procedures required to manage the Directory Service that operates between the Query Service Agent Manager and Service Provider Manager (see Figure 5) are specified in 11.3. + +![Diagram of F.510 Service Management showing the relationship between the F.510 Service Interface and the F.510 Service components.](33ed1f9b27c7c21c797aa928b0f06851_img.jpg) + +The diagram illustrates the F.510 Service Management structure. On the left, a vertical dashed line represents the 'F.510 Service Interface'. A horizontal line extends from this interface into a dashed rectangular box labeled 'F.510 Service' at the top. Inside the box, there are four components arranged in a 2x2 grid: 'Query Service Agent' (top left), 'Service Provider' (top right), 'Query Service Agent Manager' (bottom left), and 'Service Provider Manager' (bottom right). The 'Query Service Agent' is connected to the 'Service Provider' by a horizontal line. The 'Query Service Agent Manager' is connected to the 'Service Provider Manager' by a horizontal line. Vertical lines connect the 'Query Service Agent' to the 'Query Service Agent Manager' and the 'Service Provider' to the 'Service Provider Manager'. A label 'F.510\_F05' is located at the bottom right of the diagram. + +Diagram of F.510 Service Management showing the relationship between the F.510 Service Interface and the F.510 Service components. + +**Figure 5/F.510 – F.510 Service Management** + +### 6.3.1 Service Provider Manager + +A Directory Service Provider Manager performs a number of functions which cover management of Directory Services offered by a Directory Service Provider negotiated by Directory Service Administration. Typically these include operational aspects of the service provision, such as distribution of availability and physical access information. Service Provider Managers will operate in cooperation with Directory Query Service Agent Managers. + +The Directory Service Provider Manager will have the capability to create, and subsequently monitor and configure, Directory Services, disseminate knowledge of the existence of Directory Services to other Directory Service Managers (both Directory Query Service Agent Managers and peer Directory Service Provider Managers) and establish usage of Directory Services through the configuration of system and communications resources. + +### 6.3.2 Query Service Agent Manager + +A Query Service Agent Manager performs a number of functions which cover negotiations of usage of Directory Services offered by Directory Service Providers and performing the configuration of the Directory Query Service Agent needed to use offered Directory Services. Query Service Agent Managers will exist for Query Service Agents and will work in cooperation with Directory Service Provider Managers. + +# 7 Directory information model + +## 7.1 Information structure + +The purpose of this clause is to describe the inherent structure and relationships for the information contained in subscriber listings and to provide definitions of object types for that information. + +It is important that there is a common understanding and unambiguous definition of the information in order to develop specific services using that information. The information contained in subscriber listings **does not define** the services but does limit the overall scope and capability of any service. + +Many items of information **could** be used to identify a particular subscriber listing (and ultimately the communications addresses and associated information for the subscriber) such as subscriber name, house number and street name. + +Where information is available that is sufficiently distinguishing **and** the service provider supports a search function using that combination of attributes, then identification of the subscriber may be achieved directly. + +Where such information or search function is not available, then it is usually necessary to identify information that can be used reliably to select smaller sets of subscriber *listings*. The information that is of this type, which is generally useful (commonly known), is Geographic Location, Business Classification and Organization Structure. This type of information may be used to aid identification of the actual attribute values held in the directory from alternate information available for searching such as alias and synonym values. For example, with reference to Figure 6, a real world locality (a) Heathrow may be used to identify the actual locality (b) London Airport that is the value (c) held in the subscriber listings. + +NOTE – The terms Geography and Geographic Location Information are intended to be equivalent. + +Geography, which can include information of various scope (street, town, county, region, etc.), is generally known and useful for the selection of smaller sets of subscriber listings, but alternate information that identifies the same Geographic Location (place) typically exists. + +Business Classification is also generally known and useful for the selection of smaller sets of subscriber listings, but no common system for the classification of business function is generally known or used and alternate information that identifies similar business functions can exist. + +Organizational Structure is more complex because there are many ways and different types of information that may be used to describe an Organization. For example, a company may have a structure that is described by location, function, product or activity; an Administration may have a structure that is described by department, region and service; a family may be described by genealogy. For a particular Organizational Structure various relationships may be used to aid in searching. However, within this service definition use is only made of the hierarchical relationship that may exist between directory listings. + +The general model, that is appropriate to directory enquiry services, is illustrated in Figure 6. The broken lines represent the optional relationships, while the solid lines indicate mandatory relationships. Formal models, which are subsets of the general model, on which each specific service is based are specified in clause 10. + +In Figure 6, information about Geography, Organizational Structure or Business Classification may be used to identify the value (i.e., the actual attribute values for a listing held in the directory) for Subscriber Locality, Subscriber Entry (organizational) and Subscriber Business Category, respectively that may be used to select Subscriber Entries reliably. + +![Figure 6/F.510 – Directory information model diagram. The diagram shows three top-level boxes: 'Geography' (containing 'Real World Locality (a)' and 'Subscriber Locality (b)'), 'Organizational structure' (containing 'Subscriber Entry'), and 'Business Classification' (containing 'Real World Business Category' and 'Subscriber Business Category'). Dashed lines connect 'Subscriber Locality (b)', 'Subscriber Entry', and 'Subscriber Business Category' to a central 'Subscriber Entry' box. This central box contains a 'Subscriber' box with 'Subscriber Locality (c)' and 'Subscriber Business Category', and a 'Comms. Address' box. The label 'Subscriber Listings' is to the right of the central box. The code 'F.510_F06' is at the bottom right.](8fbdfc3d17fb1dae7b2d8f5a287fa9fc_img.jpg) + +Figure 6/F.510 – Directory information model diagram. The diagram shows three top-level boxes: 'Geography' (containing 'Real World Locality (a)' and 'Subscriber Locality (b)'), 'Organizational structure' (containing 'Subscriber Entry'), and 'Business Classification' (containing 'Real World Business Category' and 'Subscriber Business Category'). Dashed lines connect 'Subscriber Locality (b)', 'Subscriber Entry', and 'Subscriber Business Category' to a central 'Subscriber Entry' box. This central box contains a 'Subscriber' box with 'Subscriber Locality (c)' and 'Subscriber Business Category', and a 'Comms. Address' box. The label 'Subscriber Listings' is to the right of the central box. The code 'F.510\_F06' is at the bottom right. + +F.510\_F06 + +**Figure 6/F.510 – Directory information model** + +Subscriber entries may be associated with structured data relating to a Geographical Location or Business Classification. Within each of these structured data sets, the entry to which a subscriber is immediately related will be related to other "real world" entries within that set. To be most useful these sets should be well structured, well-defined and have widely, generally known values. + +Subscriber entries may also be related to each other through an organizational structure based on a combination of geographical, business or other data. + +## 7.2 Geography + +Structured information regarding "real world localities" enables queries to be translated or resolved into one or more "subscriber localities" to be searched. + +### 7.2.1 Real World Locality + +This term may be applied to any "named" geographic locality (or area), the name (or identifier) of which is likely to **be known and specified by** a user making an enquiry on the directory. Different types of RWL are likely to be appropriate within different countries, and while the names of such objects are likely to be known abroad, their type often will not. The most common localities will be those used as components in Postal Addresses. + +It includes both political and administrative localities (Country, County, State, Province, Department, Region, City, Town, Village, Rural District, etc.) and natural geographic areas (e.g., Dovedale, Peak District). These political, administrative or natural geographic localities (or areas) may be precise, with well-defined borders, or imprecise with somewhat fuzzy borders. They may be officially named places, or local familiarly named places. The exact interpretation of each of these terms is generally fully appreciated only within the countries where they apply. + +All such localities may be known by one or more alias names. + +Relationships between RWLs include: + +- a locality is part of another locality (strict hierarchy); +- a locality neighbours (is adjacent to) another locality; +- a locality is an alias for another locality; +- a locality roughly maps to a set of localities, e.g., Yorkshire (an area consisting of three counties), Smaaland in Sweden (consists also of three counties), Ameland (an island of three localities). + +Which of these relationships should be represented within the directory is a local consideration for the Service Provider. + +The attributes for Geographic Locality are given in Table 1. A Geographic Locality will be of a certain type, such as Country, StateOrProvince, Locality, Small Locality or Locality Street Address. Certain attributes may be complex, that is, may have more than one value, for example, LocalityName may be held in more than one language (with an indication of the language for each value). + +**Table 1/F.510 – Geographic Locality attributes** + +| Attribute | Description | +|-------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| LocalityName | The name by which the Geographic Locality is identified. | +| LocalityCode | A common code used to identify a Geographic Locality, e.g., Country ID, Post Code. | +| LocalityNameLanguage | Language for the LocalityName. | +| LocalityNDC | National Destination Code for Locality (see 7.2.3 below). | +| LocalityDescription | Description provides additional locality information. The information, which is not structured or standardized, is held on existing directory databases. This attribute is not a mechanism to introduce data or service enhancements. | +| NoSubscriberInformation | Indicator to show that Subscriber data is not held for the particular Geographic Locality. | + +### **7.2.2 Subscriber Locality** + +A Subscriber Locality represents a convenient grouping of geographically related subscribers used to aid searching. The areas represented by these may have a direct mapping to real world localities, or may be artificially created and named, e.g., to create manageable subdivisions. All geographically based Subscriber Entries are associated with a Subscriber Locality. + +The Subscriber Locality provides the means by which imprecise and unstructured RWL specified as search parameters can be translated into more well-defined search areas. + +Subscriber Locality attributes are as for Geographic Locality attributes. + +### **7.2.3 National Destination Code (NDC)** + +National Destination Code is defined by ITU-T Rec. E.160 and is summarized here. + +The National Destination Code is a field, within the E.164 numbering plan, which combined with the subscriber's number will constitute the national (significant) number of the international ISDN number. The NDC will have a network and/or trunk code selection function. + +NDC assignments are a national responsibility and therefore the NDC structure varies from one country to another. + +In general numbering areas do not line up with subscriber locality boundaries and not all the communications apparatus for a subscriber are necessarily routed through a single exchange. + +## 7.3 Business Classification + +Business Classification is the system used to classify Subscribers according to their "Business". + +### 7.3.1 Real world business category + +Real world business categorization is composed of formal categories (which make up a strict hierarchy), and informal categories which may be commonly known (and hence provided in requests). Informal categories will be mapped to the formal ones in order that information provided in a request can be translated to the category with which the required entry is likely to be associated. + +Standard classification of Business categories (e.g., NACE, SIC) may be adopted in order to circumvent language problems. + +### 7.3.2 Subscriber Business Category + +A Subscriber Business Category is the Business Category with which a Subscriber Entry is associated. The precise meaning is determined by the Service Provider. + +## 7.4 Organizational structure + +Subscriber Entries may form part of the external public view of a larger body or organization. The structure of an organization may be described by the hierarchical relationships between the Subscriber Entries that form the organization. The set of entries which form the (complete) external public view of a larger body or organization is referred to as the Hierarchical Group. + +## 7.5 Subscriber Entry + +A Subscriber Entry is associated with none, one or more communications addresses of a variety of possible types, e.g., telephone, facsimile, mobile, e-mail, EDI, videotelephone, ISDN, toll free voice telephony (also known as green numbers). + +The attributes for Subscriber are given in Table 2. Certain subscriber entry attributes may be complex, having more than one value, for example, StreetAddress may be held in more than one language (with an indication of the language for each value). + +Each Subscriber Entry will be optionally related to a Subscriber Locality, a Subscriber Business Category and (hierarchically) to a Subscriber Entry. + +**Table 2/F.510 – Subscriber attributes** + +| Attribute | Description | +|------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| SubscriberType | Categorization of Subscriber Entries into residential subscribers, organizational subscribers and governmental subscribers. | +| SubscriberName | Name by which the Subscriber chooses to be known in the directory. The SubscriberName is the surname for a residential subscriber and organizational name for organizational or governmental subscribers. | +| SecondFamilyName | The SecondFamilyName is the second surname for a residential subscriber and may be used in countries where the SubscriberName is not sufficiently unique. | +| GivenName | Forename or initials of residential subscriber. | +| Title | Honorific distinction of residential subscriber. | +| HouseIdentifier | Unit number or name and further refinements. For example, this may include building, apartment or suite number. | + +**Table 2/F.510 – Subscriber attributes** + +| Attribute | Description | +|-----------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| StreetAddress | Address street name (this may have more than one value). | +| StreetAddressLanguage | Language for each value of StreetAddress. See 9.3.2. | +| Zip/Post Code | Address zip or post code. | +| SubscriberAddress | The address provided in the subscriber entry; this may or may not be the correct postal address, and it will often be split into its components – house identifier, street address and zip or post code (above). | +| Profession | The Profession of a subscriber. The profession attribute describes the profession or job of an individual, business or department. | +| BusinessCategory | The BusinessCategory of an organization or Administration. It describes the activity domain in which the business or administrative authority is involved. The BusinessCategory may be a general activity, e.g., telecommunications, or a more specific activity, e.g., mobile telephony. | +| Language | Preferred language, which has been specified by the subscriber, in which their entry shall be displayed. In some countries there is a requirement that a subscriber should be able to specify the language in which their entry should be displayed or provided to the directory service user. | +| SubscriberDescription | Description provides additional subscriber information. The information, which is not structured or standardized, is held on existing directory databases. This attribute is not a mechanism to introduce data or service enhancements. | +| HierarchyBelow | The HierarchyBelow indicates if any Subscriber Entries exist (hierarchically) below the Subscriber Entry within the same Hierarchical Group. | +| HierarchyLevel | Indicates the relative position for a Subscriber Entry that is part of a Hierarchical Group. The HierarchyLevel for a Subscriber Entry is the level at which the entry occurs within the hierarchy, where the top, or root, of the hierarchy has a HierarchyLevel of zero. | +| EntryIdentifier | A directory entry identifier has been supplied in a response and that can be used, in a new request, to uniquely identify a directory entry.
NOTE – This identifier can have more general use but is only used within this service definition to identify Subscriber Entries. | +| LinkedEntries | A directory entry identifier that may be used, in a request, to uniquely identify a related entry, e.g., a "see also" reference to another entry. This attribute may have many values. | + +### 7.5.1 Subscriber Types + +Subscriber Types may optionally be used to classify different subscribers in the directory assistance service. The Subscriber Type differentiation can be utilized both in the Directory Information Model and Service Data Model. Each Subscriber Type has its own information content and naming structure. It is also possible to customize special services, which are allowed only for certain entry types, such as search for organizations in a country. Subscribers may be classified into residential, organization, government. + +In practice it is sometimes difficult for the requestor to know whether certain subscribers are classified within the directory as an organization or government; and not all countries support classification of entry types. It should be possible to retrieve both organization and government entries without the requestor knowing the exact subscriber type. + +#### 7.5.1.1 Residential + +The residential entry type normally, but not necessarily, describes a person rather than an organization. The subscriber entry type normally, but not necessarily, is not associated with an administration, business or commercial service. The precise definition may vary in each country. + +#### 7.5.1.2 Organization + +The organization entry type is used to describe organizations which are not part of the administrative machinery in a country. + +#### 7.5.1.3 Government + +The government entry type is used to describe the entities, which are part of the administrative machinery in a country. + +#### 7.5.1.4 Alternate name forms + +Subscriber listings commonly occur as alternate representations, such as acronyms and abbreviations, for the same item of information. For example, the same organization may be listed as HAL, H.A.L., H A L or Henley Ascot Limited, or the address Acacia Avenue may be represented by Acacia Av. or Acacia Ave. + +To remove the need for the end-user to have knowledge of how an item should be entered or how it is stored with the Directory database, the Service Provider is advised to standardize the way in which these alternate name forms are supported, either by holding a single form and converting request parameters to the same form or by matching the request parameter against all (possible) forms. + +### 7.5.2 Communications Address + +One or more Communications Addresses are associated with each Subscriber Entry, and each communications address provides the actual address, e.g., telephone number, e-mail address, etc., and it provides additional information about the address. A Subscriber entry may be associated with multiple "numbers" for a variety of different communications address types. + +Information that affects the usability or availability of a communications address may include: when the address is valid, within what (geographic) area the address may be used, the class of tariff associated with the address and access restriction to the address. The range of values for each classification may be extended and is not limited to those values given in Table 3. The addition of values is for further study. + +**Table 3/F.510 – Communications address attributes** + +| Attribute | Values | Description | +|-----------------------|--------|----------------------------------------------------------------------------------------------------| +| CommunicationsAddress | | The communications address is a common nomination for different types of communications addresses. | +| TelephoneNumber | Number | | +| ORAddress | String | X.400 address as per RFC 1474. | +| Mail | String | | +| URL | String | | + +**Table 3/F.510 – Communications address attributes** + +| Attribute | Values | Description | +|-----------------------|----------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| CommunicationsService | | Identifies the type of communications service. | +| | Voice | | +| | Facsimile | | +| | textPhone | | +| | Videotelephone | | +| | publicPhoneBox | | +| | switchBoard | | +| | Data | | +| | E-mail | | +| | Pager | | +| | Internet | | +| CommunicationsNetwork | | Communication address number type. The type may be generic or specific. | +| | PSTN | | +| | ISDN | | +| | GSM | | +| | UMTS | | +| | Pager | e.g., SMS. | +| | X.400 | | +| | Internet | e.g., SMTP, HTTP, FTP. | +| | | | +| AddressValidity | | Classification of validity. The value for AddressValidity may also include a date and a reference to a dependent listing. | +| | Old | | +| | Current | | +| | Future | | +| | Temporary | | +| AddressCoverage | | The GeographicLocality within which the CommunicationsAddress may be used. The GeographicLocality may be "any" for international use, the country providing the number for national use or may be some other locality such as a region or city. | +| AddressTariffClass | | Classification of service tariff. | +| | Normal | | +| | Premium | | +| | Toll Free | | + +**Table 3/F.510 – Communications address attributes** + +| Attribute | Values | Description | +|--------------------|---------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------| +| AddressRestriction | | Classification of access restriction – public, secret number, etc. | +| | Public | | +| | Secret | | +| | Call Screen | | +| | No Marketing | | +| | Complete Only | | +| AddressDescription | | Informal, unstructured information related to the communications address, for example, to indicate that the communications address is for use in an emergency. | + +# 8 Service specification + +The purpose of this clause is to describe general principles and features required of any Named Service and to describe the elements that are needed for a complete specification of each specific service. + +The primary function of the Directory Service is to provide users with communications addressing information (telephone, facsimile numbers, etc.) on Organizations and Individuals they wish to contact, and about whom they can supply commonly available data (name, address, etc.) which can be used in identifying the subscriber entry. + +As a general principle, the extent and functionality of the service offered by any Service Provider is limited only by what the Service Provider is prepared to and capable of providing, and not by unnecessary constraints imposed in the name of uniformity. The consequence of this is that the service definition must ensure compatibility between systems providing different levels of service. + +The aim is for minimal imposition on Service Providers, allowing individual Service Providers maximum freedom with regard to the service they are able to deliver. It is then critical that the service definition embodies sufficient flexibility to allow the most highly developed Service Providers to fully employ their capabilities, whilst not excluding or compromising interworking with ROAs with less sophisticated systems. + +In practice, there will need to be different levels of service. The minimum level of service, comprising basic functionality and essential directory information, should be met by all cooperating Service Providers. Other levels of services, which enhance the basic functionality and directory information, can also be defined for each Named Service. + +In terms of query capability, the goal for a "generic service" should be that users input whatever they know about the target object with minimal structural constraints, and that the systems apply the necessary "intelligence" to the supplied data in order to locate the target directory entries. + +In the generic service, any set of known criteria (object types, attribute types, values and matching specification) can be supplied, by the user, to the Service Provider which will then apply whatever intelligence is within its means in order to identify the subscriber entries (see Table 4). The Service Provider may provide one or more of the following: + +- search on locality in order to aid refinement of geographic data; +- search on subscriber category in order to aid refinement of function data; + +NOTE – Subscriber category is a generic term used to indicate that other services than a White Pages Service may involve additional categorization of subscribers, e.g., according to business category. + +- search on hierarchical group to aid refinement of external, public view of organization data; +- search for subscriber. + +**Table 4/F.510 – Generic service definition** + +| Supplied by user | Information required | | | | +|-----------------------|----------------------|---------------------|--------------------|-----------------| +| | Subscriber Locality | Subscriber Category | Hierarchical Group | Subscriber | +| Country b) | M | M | M | M | +| Locality | O a) | | | O a) | +| SubscriberName | | | | O a) | +| SubscriberType | | | | O a) | +| SubscriberAddress | | | | O a) | +| HierarchySelection | | | O a) | O a) | +| SubscriberCategory | | O a) | | O a) | +| CommsAddress | | | | O a) | + +a) Matching technique selected by the user. +b) Note that country is just a special case of locality. + +Search features should include: + +- alternative forms of "approximate matching" on any or all input attributes; +- all input fields potentially optional (the only restrictions should be those imposed by Service Provider); +- some agreed mechanism for dealing with language translation problems, e.g., character translation or use of codes; +- selection of the attributes; +- order of entries returned; +- provision of information to aid searching where searches are unsuccessful. + +A number of distinct services may be defined on the basis of the generic services described above, the data held within the directory, and the enquiry services supported. Clause 10 defines the White Pages. Other Named Services, such as Yellow Pages, may also be defined on the basis of the generic services and the common elements in clause 9. + +For each of these Named Services to be delivered a basic definition is specified that is mandatory; all cooperating Service Providers need to support its implementation. Any such definition ensures compatibility and compliance between its users. In addition to the basic service, additional or enhanced services are defined. + +# 9 Common service specification elements + +This clause describes aspects of service specifications that are common to all services. This clause provides terminology and definitions that may be used within the specification of any Named Service. + +## 9.1 Character set + +An 8-bit environment is used which contains the Latin based character set as described in Annex D of Amendment 1 to ITU-T Rec. T.51. + +NOTE – ITU-T Recs T.51 (and T.52) have replaced ITU-T Rec. T.61, which has been deleted. + +The character set consists of the Primary set (see Figure D.1/T.51), designated as the G0 set and invoked in columns 2 to 7, and the Supplementary set (see Figure D.2/T.51), designated as the G2 set and invoked in columns 10 to 15. + +## 9.2 Matching rules + +The relation between a request and the corresponding reply is defined by the "Matching rules". The reply should consist of those entries in the directory where the attributes values "match" the values specified in the request. Generally, different matching rules will be used for different attributes. In order to avoid ambiguity and differing interpretations of matching, it is important that the rules for each type of matching are precisely defined. + +Each character, word and string matching rule has a name, e.g., WordPhonetic, and an abbreviated reference, e.g., MR3. + +### 9.2.1 Definitions + +A **string** is a non-empty finite sequence of words. + +A **word** is a non-empty finite sequence of elements from the character set given in 0 but not including the space character. + +The words in a string are bounded by either space characters, the beginning of the string or the end of the string. + +A **weak word** is a word that is common, such as an article, which is normally ignored by the Service Provider when matching strings. Their presence or absence does not affect matching. The set of words that are defined as weak are language and culture-dependent and is determined by the Service Provider. + +A matching rule determines whether a value "matches" another value. The value contents of the directory is irrelevant for the definition of the matching rule. The matching rule will define for any pair of values whether one matches the other. + +### 9.2.2 Character matching rules + +#### 9.2.2.1 CharacterMap (CR1) + +The request character matches the directory character according to a mapping determined by the Service Provider. This mapping includes exact matching. + +Descriptions of mappings shall be made widely available and might be agreed in standards or other groups. + +A particularly important case for end-user friendliness is the matching of the basic characters A-Z with national characters listed in Figure A.2/T.51. Service Providers shall provide mappings such that it is possible to find every entry using only the characters A-Z and 0-9 in requests. + +Examples of character mapping might be the matching of basic letters of Figure A.2/T.51 with those with diacritical marks, etc. in the other columns. Other mappings might be two characters matching one character or vice versa, e.g., ae matching æ, ue matching ü or ü matching ue. + +#### **9.2.2.2 CharacterCaseIgnore (CR2)** + +The request character matches the directory character exactly, except for case. + +For example: Ü matches Ü and ü, but not U or u. + +#### **9.2.2.3 CharacterExact (CR3)** + +The request character matches the directory character exactly. + +For example: Ü matches Ü, but not ü or U. + +### **9.2.3 Word matching rules** + +Matching a word is matching in sequence characters of the word one by one, using character matching rule CR1 or CR2. + +The following rules can be applied to individual words. (Word 2 is a word from the database; Word 1 is a word in the request.) + +#### **9.2.3.1 WordExact (MR1)** + +Word 2 matches word 1 if and only if they are equal. + +#### **9.2.3.2 WordTruncated (MR2)** + +Word 2 matches word 1 if and only if word 1 results from word 2 by truncating word 2. (Truncating a word means replacing it with a non-empty initial substring.) The minimum size to which a word may be truncated shall be specified by the Service Provider. + +#### **9.2.3.3 WordPhonetic (MR3)** + +The word 2 matches word 1 if they are equal or word 1 matches word 2 according to a phonetic rule determined by the Service Provider. + +### **9.2.4 String matching rules** + +String 2, which is a string from the database, matches string 1, which is a string in the request, if they match according to one of the following rules: + +#### **9.2.4.1 StringExact (MR4)** + +String 2 matches string 1 if and only if string 1 results from string 2 without any deletion of words and without any change in their ordering. + +#### **9.2.4.2 WordDeletion (MR5)** + +String 2 matches string 1 if and only if string 1 results from string 2 by deletion of zero or more, but not all, words from string 2 and the words present are in the same order. + +#### **9.2.4.3 WordRestrictedDeletion (MR6)** + +String 2 matches string 1 if and only if string 1 results from string 2 by deletion of zero or more, but never the first, words from string 2 and the words present are in the same order. + +#### **9.2.4.4 WordPermutation (MR7)** + +String 2 matches string 1 if and only if string 1 results from string 2 by permuting zero or more of the words in string 2. + +#### **9.2.4.5 WordPermutationAndDeletion (MR8)** + +String 2 matches string 1 if and only if string 1 results from string 2 by deletion of zero or more, but not all, words from string 2, and permuting zero or more of the remaining words in string 2. + +#### 9.2.4.6 ProviderDefined (MR9) + +Local knowledge applied. This is an indication that the matching to be performed is left up to the system performing the matching, i.e., the "providing" system. ProviderDefined matching may include omission of common words and modification of words (by separation or combination) in the request. For example, modification of *Du Pont* into *Dupont*. + +### 9.2.5 Examples + +Table 5 gives examples of string values that are matched by the string value "TRUTH OR CONSEQUENCES". In this example "OR" has not been considered as a weak word. + +**Table 5/F.510 – Examples of string matching rules** + +| Matching rule | String values that are matched | +|----------------------------|--------------------------------------| +| StringExact | "TRUTH OR CONSEQUENCES" | +| WordDeletion | "TRUTH CONSEQUENCES", "CONSEQUENCES" | +| WordRestrictedDeletion | "TRUTH CONSEQUENCES", "TRUTH OR" | +| WordPermutation | "OR CONSEQUENCES TRUTH" | +| WordPermutationAndDeletion | "OR TRUTH" | + +## 9.3 Language differences + +### 9.3.1 General + +Many countries have more than one official language in use and directory entries may hold the value for an attribute in more than one language. + +Each Directory Service Provider shall declare the official languages that may be used in responses by the Directory Service. + +As a general principle the language of one type of data within a directory is independent of the language for other types of data. For example, a locality with a name in French may occur in a State or Province with a name in Dutch. + +It is assumed that the requirements for the support of language differences are generic and apply to all services. + +### 9.3.2 Locality names and street addresses + +When returning a locality name or street address, where different values for different languages are available, then all the values shall be returned with an indication of the language. + +*For example:* + +*Response: Locality = Brussel (NL), Bruxelles (FR)* + +*Street Address = Vossenplein (NL), Place de jeu de balles (FR)* + +The language codes defined by ISO 639 shall be used. How the language code is provided in the response is not specified by this Recommendation. + +## 9.4 Ordering of entries within a response + +The ordering of entries within a response is for further study. + +## 9.5 Hierarchical Groups + +When searching for subscriber entries, the requester need not be aware of the hierarchical relationship between entries belonging to a Hierarchical Group (i.e., need not be aware if the Subscriber Entry is a part of a Hierarchical Group). + +A subscriber entry may only belong to one hierarchical group. + +A search for subscribers shall return an indication against those entries that are part of a Hierarchical Group and if Subscriber Entries that are part of the same Hierarchical Group exist below the entry returned. + +When searching for subscribers the information returned in the response, for each entry that satisfies the search criteria, may be specified by the following parameter (HierarchySelection): + +| HierarchySelection | Information returned in the response | +|--------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Top | The Subscriber Entry with a HierarchyLevel zero for the Hierarchical Group that contains the entry that satisfies the search. | +| Self | The Subscriber Entries that satisfy the search. | +| Children | The Subscriber Entries with a HierarchyLevel one greater (i.e., lower in the hierarchy) than the HierarchyLevel of the entries that satisfy the search for the part of the subtree defined by the entry that satisfies the search. | +| Parent | The Subscriber Entries with a HierarchyLevel one less (i.e., higher in hierarchy) than the hierarchy level of the entries that satisfy the search. | +| Hierarchy | The Subscriber Entries that are lying directly above the entries satisfying the search up to and including the top entry. | +| Subtree | The Subscriber Entries with a Hierarchy Level greater or equal to the HierarchyLevel of the entries that satisfy the search for the part of the subtree defined by the entry that satisfied the search. | +| All | All the Subscriber Entries of the hierarchical group. | + +It shall be possible to reconstruct the correct hierarchical relationships between entries that are returned in a response. + +NOTE – It is assumed that a mechanism exists that enables the Service Provider to identify the entries specified by the HierarchySelection parameter (Top, All, etc.) but the specification of such a mechanism is not part of this service definition. + +For example, where the directory contains the following entries: + +| EntryIdentifier | HierarchyLevel | Subscriber name for entry | +|-----------------|----------------|--------------------------------------------------------------| +| A | 0 | Hilton Hotel | +| B | 1 | Hilton Hotel    Reservations | +| C | 2 | Hilton Hotel    Reservations    Rooms | +| D | 2 | Hilton Hotel    Reservations    Conference Facilities | +| E | 1 | Hilton Hotel    Reception | +| F | 1 | Hilton Hotel    Bar | + +*The following searches will return the entries indicated:* + +| Search For | HierarchySelection | Entries satisfying search criteria | Entries returned | +|----------------------------------------|---------------------------|-------------------------------------------|-------------------------| +| Hilton Hotel Reservations Rooms | Top | C | A | +| Hilton | All | A, B, C, D, E, F | A, B, C, D, E, F | +| Hilton Re | Self | B, C, D, E | B, C, D, E | +| Hilton Reservations | Self | B, C, D | B, C, D | +| Hilton Reservations | Children | B, C, D | C, D | + +*A HierarchyLevel identifying that the returned entry is part of a hierarchical group and specifying the precise level for the entry would be returned for A, B, C, D, E and F.* + +*A HierarchyBelow identifying that entries exist below the returned entry would be given for A and B.* + +## 9.6 Number of entries + +The limit on the number of entries to be returned in the response may be specified in a request. + +The limit, imposed by the Service Provider, on the maximum number of entries that can be returned in the response shall be at least 30. + +When the limit is not provided in the request then the limit is the one imposed by the Service Provider. + +## 9.7 Entry count + +The Service Provider shall return the number of entries matched by the request, or, if the total number of entries matched cannot be provided, for example, because searching is suspended, a number together with an indication that the number of entries matched exceeds this number should be returned. + +## 9.8 Alternate attribute values + +A Service Provider may permit alternative attribute values than those provided in a request, to match the attribute value of an object. In these cases the Service Provider may return a different attribute value to the one provided in the request. + +For example, the SubscriberName Heathrow, in the request, may be used to identify objects for which the SubscriberName is London Airport. The SubscriberName London Airport may be returned in the response. + +## 9.9 Weighted attribute values + +This would typically be used to relax the search conditions when no entries are matched using all search criteria. An attribute, in the request, may have an AttributeWeight assigned to indicate its significance for matching entries. Attributes, in the request, that are mandatory or have an AttributeWeight assigned equal to one (default) shall be used, by the Service Provider to identify matching entries. + +The Service Provider shall return any entries matched using all search criteria, i.e., applying an AttributeWeight of one to all attributes in the request. The Service Provider may return any entries (in the same response) matched by ignoring any of the attributes in the request that are optional and have an AttributeWeight of zero. + +For example, a search for SubscriberName *Lloyd* at HouseIdentifier 57 in StreetAddress *Acacia Avenue* where the HouseIdentifier has an AttributeWeight zero will return *Lloyd 57 Acacia Avenue* (if the entry exists) and may return *Lloyd 143 Acacia Avenue*. + +AttributeWeight is only relevant for parameters that are optional according to clause 10. If AttributeWeight is specified for a mandatory parameter, it shall be ignored by the service provider. + +## 9.10 Geographical extension + +Local knowledge may be applied by the Service Provider to define the geographical area on which the search will be extended. Geographical Extension is only valid where the search request identifies an unambiguous StateOrProvince, Locality or Small Locality. The amount of extension may or may not be related to geographical distance, as determined by the Service Provider. + +The Geographical Extension is specified in a request by two parameters: + +ExtendedArea            Integer value 1 or greater, which specifies the degree of extension, as determined by the Service Provider. + +IncludeAllAreas        Specifies if the geographical area to be searched includes all areas (degrees of extension), corresponding to 0 (no extension) to ExtendedArea, or if only the extension specified by ExtendedArea is to be searched. + +For example, a search of the town *St. Helier* (on the island of *Jersey*) may be extended to include the whole of the island of *Jersey* when a value of 1 is specified for ExtendedArea and may be extended to include the neighbouring island of *Guernsey* when a value of 2 is specified for ExtendedArea. + +## 9.11 Ignore if absent + +When a requester as part of the search criteria specified an attribute type that is not relevant for a particular entry, this attribute should be ignored by the Service Provider. As an example, if the GivenName is a search criterion, then this criterion should be ignored when matching against an organizational entry. + +## 9.12 Paging + +A requestor may specify that information should be returned in pages with an indication about the maximum number of entries within each page. + +# 10 White Pages DQ + +The White Pages directory is based primarily on the identification of directory entries by name and geography. + +This clause defines the White Pages Directory Service on the basis of the generic services and the common elements in clause 8. Other Named Services, such as Yellow Pages, may also be defined on the same basis. + +The geographic and subscriber attributes that are required to support the minimum service are identified as essential attributes. The attributes that may be provided, in order to support higher levels of service, are identified as additional attributes. + +## 10.1 Service Data Model + +The Service Data Model described in this subclause is formally defined in Annex C using the Object Modelling notation described in Annex D. + +### **10.1.1 Geographic Data Model** + +Locality, Small Locality, Locality Street Address, State Or Province and Country are types of Geographic Locality. Attributes of the same name are used to describe them. Each may have one or more alternate names. + +A Country is made up of a set of States or Provinces. + +A State or Province is made up of a set of Localities. + +One or more Small Localities may be associated with each Locality. + +A Subscriber Locality is either a Locality or a Small Locality. + +One or more Locality Street Addresses may be associated with each Subscriber Locality. + +### **10.1.2 Subscriber Data Model** + +One or more communications addresses may be associated with each subscriber entry. + +A subscriber entry is either a governmental, organizational or residential subscriber. + +A (hierarchical) group is made up of a set of one or more Subscriber Entries. + +### **10.1.3 Association between Subscriber Data and Geographic Data** + +One or more Subscriber Entries may be associated with each Subscriber Locality. + +### **10.1.4 Geographic Data** + +The geographic structure used to support the White Pages service is based on a formal subdivision of the country using geographic locality names commonly known to enquirers; this will generally be based on postal or political boundaries. + +**Geographic Locality**      A named geographic area. + +Each level of the geographic hierarchy is a particular type of Geographic Locality. Each instance is normally identified by its LocalityName, which needs to be unique within the larger geographic locality to which it belongs. + +Attributes: LocalityName +LocalityNameLanguage +LocalityCode +LocalityNameType +NoSubscriberInformation +LocalityDescription + +A Geographic Locality may be one of the following five types. + +| | | +|------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Country | As defined by ISO 3166-1. A Country is identified by its LocalityCode.
Essential Attribute: LocalityCode | +| StateOrProvince | The first level of division of a Country; this will be the States, Counties, or Provinces or equivalent units into which a country is subdivided. It may also be used to identify major metropolitan areas.
Essential Attribute: LocalityName
Additional Attributes: LocalityNameLanguage
LocalityCode
NoSubscriberInformation
LocalityDescription | +| Locality | The Geographic Locality level to which subscribers are assigned; this will correspond to the cities, major towns, or other areas into which each StateOrProvince (or Country, if StateOrProvince is absent) is subdivided.
Essential Attributes: LocalityName
NoSubscriberInformation
Additional Attributes: LocalityNameLanguage
LocalityCode
LocalityNDC
LocalityDescription

NOTE – The actual relationship between National Destination Codes and Localities is somewhat complex, due in part to National Destination Codes being related to exchanges, which cover areas which do not align precisely to the Localities. For the purpose of supporting the required service, then the simplification of National Destination Code as an attribute of Locality will suffice. | +| SmallLocality | A locally identified district or region of a larger town or city, or rural district. Within the data of some Service Providers, this simply acts as an alias in that it is used to identify the associated Locality to which subscribers are assigned. Where Service Providers are able to assign Subscriber entries to this level of detail then it will enable more refined searching.
Essential Attributes: LocalityName
NoSubscriberInformation
Additional Attributes: LocalityNameLanguage
LocalityCode
LocalityNDC
LocalityDescription | +| LocalityStreetAddress | A named street within either a Locality or a Small Locality.
Essential Attributes: LocalityName
Additional Attributes: LocalityNameLanguage
LocalityCode
LocalityDescription | + +### 10.1.5 Subscriber data + +The White Pages Service is supported by the following subscriber data. + +**SubscriberEntry** Directory entry for a subscriber. +Essential Attributes: SubscriberName +HierarchyBelow +HierarchyLevel +Additional Attributes: HouseIdentifier +StreetAddress +StreetAddressLanguage +Zip/PostCode +Profession +Language +SubscriberDescription +LinkedEntries + +The directory entry for a subscriber may be one of the following three types. + +**ResidentialSubscriber** Directory entry for a residential subscriber. +Essential Attributes: SecondFamilyName +GivenName + +**Organizational Subscriber** Directory entry for an Organization. +Additional Attributes: BusinessCategory + +**Governmental Subscriber** Directory entry for local or regional governmental organization/authority. +Additional Attributes: BusinessCategory + +**CommsAddress** Details of communications device and address. The Device Address may be an essential or additional attribute depending upon the type of address. +Essential Attributes: CommunicationsAddress +CommunicationsNetwork +Additional Attributes: CommunicationsService +CommunicationsAddress +CommunicationsNetwork +AddressValidity +AddressCoverage +AddressTariffClass +AddressRestriction +AddressDescription + +## 10.2 Search controls and response indicators + +### 10.2.1 Search controls + +The following search controls are defined here and used in 10.4, 10.5 and 10.6. + +**UserClass** Identifies the type of service requested from the Service Provider. The value of this parameter may be used in the context of access control and charging. The range of values for UserClass may be extended and is not limited to those given. + +| | | +|---------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| OperatorService | A type of service that is appropriate to an International Directory Enquiry operator. | +| PublicService | A type of service that is appropriate to a general user. | +| PerformExactly | Where there is an indication in the request that the matching rules in the request shall be performed exactly, the Service Provider shall return an error if any requested matching rule is not supported. Otherwise, the Service Provider is permitted to apply any matching rule, defined in 9.2, and return an indication that the matching rules in the request have not been performed as requested. | +| EntryLimit | The limit on the number of entries to be returned in the response (see 9.6). | +| HierarchySelection | The number of levels to be returned in the response may be specified as zero (only entries matched by the request), one (entries immediately below those entries matched by the request), all (entries matched by the request and all entries below those entries) or top (the top level entry corresponding to the matched entry). | +| ExtendedArea | Integer value 1 or greater, which specifies the degree of extension, as determined by the Service Provider. | +| IncludeAllAreas | Specifies if the geographical area to be searched includes all areas (degrees of extension), corresponding to 0 (no extension) to ExtendedArea, or if only the extension specified by ExtendedArea is to be searched. | +| TargetObject | The primary object that contains the information to which the search request applies. | + +### 10.2.2 Response indicators + +The following response indicators are defined here and used in 10.4, 10.5 and 10.6. + +| | | +|-------------------|----------------------------------------------------------------------------------------------------------------------------------------------| +| EntryCount | The number of entries matched by the request or a number together with an indication that the number of entries matched exceeds this number. | +|-------------------|----------------------------------------------------------------------------------------------------------------------------------------------| + +## 10.3 Predicates + +The following predicates are defined here and used in the tables in 10.5 and 10.6. + +| | | +|-----------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| P1 | The parameter shall be present in the response if the parameter was selected, in the request, and a value is present in the directory. In case no selection was performed by the requester, it is a Service Provider option what to return. The presence of parameter values in the response may also be subject to restrictions independent of the Service Provider, for example, to meet legal or regulatory requirements. | +| P2 | The parameter shall be present in the response if the parameter was selected, in the request, and a value is present in the directory. In case no selection was performed by the requester, it is a Service Provider option what to return. The presence of parameter values in the response may also be subject to restrictions imposed by the Service Provider, for example, restricting access according to UserClass. | + +- P3** One or both of SubscriberName and SecondFamilyName shall be present (either in the request or in the response). +- P4** Shall be present in in the response if the entry is part of a Hierarchical Group. +- P5** An indication is returned when PerformExactly was not present in the request and the matching rules in the request have not been performed as requested. +- P6** Toll Free is the only permitted request parameter value. + +## 10.4 Common query conditions + +The following conditions apply to the searches specified in 10.5 and 10.6. + +Search parameters are present in a search request and response. A search parameter in a search request is either: + +- a search control that governs the search operation; or +- an information item holding attribute information to be matched against the corresponding attribute in directory entries according to the applicable matching rule (see 9.2). + +An entry satisfies the search criteria if the attribute information items all match the corresponding attributes within that entry, i.e., different items are logically AND'ed. + +A search parameter in a response is either: + +- a response indicator expressing some characteristics about the response; or +- an attribute. + +The search parameters listed for the SearchTypes in 10.5 and 10.6 are those relevant for the particular SearchType. Search parameters are designated as either mandatory, optional or omitted (see 3.2). Search parameters not listed for a given SearchType shall not be present for that SearchType. + +In 10.5 the search parameters that are required to support the minimum service are identified, in the tables, as essential. + +The search parameters that may be provided, in order to support higher levels of service, are identified as additional. + +The response to a search request is a set of entries of type TargetObject, or one or more messages (defined in Annex A) or both. + +A message is identified by a code and may include one or more accompanying parameters and search controls. An accompanying parameter may be a SearchType (described in 10.5 and 10.6) identifying the search requested (or performed), an MatchingIdentifier for a matching rule (described in 9.2), a TargetObject type (described in 10.5 and 10.6), an AttributeType (described in 10.1) or a Response Indicator (described in 10.2.2). + +Where Subscriber attributes are specified as request parameters, then it is assumed that attributes in the search request and response are associated with the appropriate Subscriber type. + +A request parameter of each type should only appear, at most, once in a request. + +The matching rules specified in the tables in 10.5 and 10.6 shall be supported by the Service Provider. + +Any of the matching rules may be supported by the Service Provider for any Search Parameter. The listed matching rules are not appropriate for the search controls: PerformExactly, EntryLimit, EntryIdentifier and HierarchySelection parameters and the Service Provider need only support a test for equality for these search controls. + +A string matching rule shall be specified, either explicitly or implicitly by default, for each request parameter and, likewise, a word matching rule shall be specified for each word in each request parameter. + +The character matching rule may be requested for the following parameters: StateOrProvince LocalityName, Locality LocalityName, SmallLocality LocalityName, SubscriberName, SecondFamilyName, GivenName, StreetAddress and HouseIdentifier. The default character matching rule for these attributes is CharacterMap (CR1). For other parameters the CharacterExact (CR2) always applies. + +Where a Geographic Locality type is specified as a request parameter or a response parameter then the value required for the StateOrProvince, Locality, SmallLocality or LocalityStreetAddress is the LocalityName or LocalityCode and the value required for Country is the LocalityCode. + +Any of the attributes listed in 10.1, including the Geographic Locality types, may be selected in the request for return in the response (subject to predicate P1 or P2). + +The interpretation of the request parameter Locality or SmallLocality (as either a Locality or a SmallLocality) will depend on the Service Provider. Where a Service Provider has assigned Subscribers to a Locality and the request parameter values match a SmallLocality, then the Service Provider shall return Subscribers for the Locality that contains the SmallLocality. + +The following common request and response parameters apply to all searches (see Table 6): + +**Table 6/F.510 – Common parameters** + +| Search parameters | Request | Response | +|--------------------------------------|---------|----------| +| Essential search controls | | | +| SearchType | M | – | +| UserClass | M | – | +| Essential response indicators | | | +| EntryCount | X | M | +| Additional search controls | | | +| PerformExactly | O | P5 | +| EntryLimit | O | X | + +The TargetObject is implicitly given by the SearchType. + +## 10.5 Basic service offering + +The searches in this clause define the minimum service that must be supported. Table 7 specifies the SearchType that may be used to identify the purpose of a search. + +**Table 7/F.510 – Search Types for basic service** + +| SearchType | Defined in clause | +|--------------------------------------------|-------------------| +| Search for state or province | 10.5.1 | +| Search for locality | 10.5.2 | +| Search for subscribers within locality | 10.5.3 | +| Search for subscribers by entry identifier | 10.5.4 | + +### 10.5.1 Search for state or province + +The system shall provide users with a list of StateOrProvince names for a country according to the search specification in Table 8. The TargetObject for the search is StateOrProvince. + +**Table 8/F.510 – Search for state or province** + +| Search parameters | Request parameters | Response parameters | Word matching | String matching | +|------------------------------|--------------------|---------------------|---------------|-----------------| +| Essential attributes | | | | | +| Country | M | M | MR1 | MR4 | +| StateOrProvince | O | M | MR1, MR2 | MR9 | +| Additional attributes | | | | | +| LocalityDescription | X | P2 | – | – | + +### 10.5.2 Search for locality + +The system shall provide users with a list of localities for a specified country according to the search specification in Table 9. The TargetObject for the search is Locality or Small Locality. + +The values of StateOrProvince and Locality or SmallLocality provided in the response may be used (in a subsequent request), without change, to identify exactly an unambiguous locality (within the country). + +**Table 9/F.510 – Search for locality** + +| Search parameters | Request parameters | Response parameters | Word matching | String matching | +|------------------------------|--------------------|---------------------|---------------|-----------------| +| Essential attributes | | | | | +| Country | M | M | MR1 | MR4 | +| StateOrProvince | O | M | MR1, MR2 | MR9 | +| Locality or SmallLocality | M | M | MR1, MR2 | MR9 | +| NoSubscriberInformation | X | M | – | – | +| Additional attributes | | | | | +| LocalityNDC | X | P2 | – | – | +| LocalityDescription | X | P2 | – | – | + +### 10.5.3 Search for subscribers within locality + +The system shall provide users with a list of subscribers for a specified locality according to the search specification in Table 10. The TargetObject for the search is Subscriber Entry. + +If the combination of Locality or Small Locality name and StateOrProvince name (if provided) does not identify an unambiguous locality (within the country), the response is defined by *Search for locality* using only the request parameters for Country, StateOrProvince and Locality or Small Locality. + +**Table 10/F.510 – Search for subscribers within locality** + +| Search parameters | Request parameters | Response parameters | Word matching | String matching | +|-----------------------------------|---------------------------|----------------------------|----------------------|------------------------| +| Essential attributes | | | | | +| Country | M | M | MR1 | MR4 | +| StateOrProvince | O | M | MR1, MR2 | MR9 | +| Locality or SmallLocality | M | M | MR1, MR2 | MR9 | +| SubscriberName | P3 | P1 | MR1, MR2 | MR9 | +| SecondFamilyName | P3 | P1 | MR1, MR2 | MR9 | +| SubscriberType | O | P1 | MR1 | MR4 | +| GivenName | O | P1 | MR1, MR2 | MR9 | +| CommunicationsAddress | X | P1 | – | – | +| CommunicationsNetwork | O | P1 | MR1 | MR4 | +| HierarchyBelow | X | P4 | – | – | +| HierarchyLevel | X | P4 | MR1 | MR4 | +| EntryIdentifier | X | P4 | – | – | +| Essential search controls | | | | | +| HierarchySelection | M | – | – | – | +| Additional attributes | | | | | +| StreetAddress | O | P2 | MR1, MR2 | MR9 | +| HouseIdentifier | O | P2 | MR1, MR2 | MR9 | +| BusinessCategory | O | P2 | MR1, MR2 | MR9 | +| Profession | X | P2 | – | – | +| ZipOrPostCode | X | P2 | – | – | +| Language | X | P2 | – | – | +| SubscriberDescription | X | P2 | – | – | +| CommunicationsService | O | P2 | MR1 | MR4 | +| CommunicationsAddress | X | P2 | – | – | +| CommunicationsNetwork | O | P2 | MR1 | MR4 | +| AddressValidity | X | P2 | – | – | +| AddressCoverage | X | P2 | – | – | +| AddressTariffClass | P6 | P2 | MR1 | MR4 | +| AddressRestriction | X | P2 | – | – | +| AddressQualifier | X | P2 | – | – | +| LinkedEntries | X | P2 | – | – | +| Additional search controls | | | | | +| ExtendedArea | O | – | – | – | +| IncludeAllAreas | O | – | – | – | + +### 10.5.4 Search for subscribers by entry identifier + +The system shall provide users with a list of subscribers determined by the EntryIdentifier and HierarchySelection parameters according to the search specification in Table 11. The TargetObject for the search is Subscriber Entry. + +The EntryIdentifier may be taken from the EntryIdentifier or a LinkedEntries value for a Subscriber obtained from a previous search. + +**Table 11/F.510 – Search for subscriber by entry identifier** + +| Search parameters | Request parameters | Response parameters | Word matching | String matching | +|-----------------------------------|--------------------|---------------------|---------------|-----------------| +| Essential attributes | | | | | +| Country | X | M | – | – | +| StateOrProvince | X | M | – | – | +| Locality or SmallLocality | X | M | – | – | +| SubscriberName | X | P1 | – | – | +| SecondFamilyName | X | P1 | – | – | +| SubscriberType | X | P1 | – | – | +| GivenName | X | P1 | – | – | +| CommunicationsAddress | X | P1 | – | – | +| CommunicationsNetwork | X | P1 | – | – | +| HierarchyBelow | X | P4 | – | – | +| HierarchyLevel | X | P4 | – | – | +| EntryIdentifier | M | P4 | MR1 | MR4 | +| Essential search controls | | | | | +| HierarchySelection | M | – | – | – | +| Additional parameters | | | | | +| StreetAddress | X | P2 | – | – | +| HouseIdentifier | X | P2 | – | – | +| BusinessCategory | X | P2 | – | – | +| Profession | X | P2 | – | – | +| ZipOrPostCode | X | P2 | – | – | +| Language | X | P2 | – | – | +| SubscriberDescription | X | P2 | – | – | +| CommunicationsService | X | P2 | – | – | +| CommunicationsAddress | X | P2 | – | – | +| CommunicationsNetwork | X | P2 | – | – | +| AddressValidity | X | P2 | – | – | +| AddressCoverage | X | P2 | – | – | +| AddressTariffClass | X | P2 | – | – | +| AddressRestriction | X | P2 | – | – | +| AddressQualifier | X | P2 | – | – | +| LinkedEntries | X | P2 | – | – | +| Additional search controls | | | | | +| ExtendedArea | X | – | – | – | +| IncludeAllAreas | X | – | – | – | + +## 10.6 Enhanced service offering + +Over and above the facilities provided by the basic definition, this enhanced definition allows each Service Provider to optionally offer any number of the following features. + +Where a search service is supported, the table defines the minimum service that must be supported. + +Table 12 specifies the SearchType that may be used to identify a search. + +**Table 12/F.510 – SearchTypes for enhanced service** + +| SearchType | Defined in subclause | +|--------------------------------------------------|----------------------| +| Search for subscribers within state or province | 10.6.1 | +| Search for subscribers within country | 10.6.2 | +| Search for street address | 10.6.3 | +| Search for subscribers by street address | 10.6.4 | +| Search for subscribers by communications address | 10.6.5 | +| Search for subscribers by business category | 10.6.6 | + +### 10.6.1 Search for subscribers within state or province + +The system shall provide users with a list of subscribers for a specified state or province according to the search specification in Table 13. The TargetObject for the search is Subscriber Entry. + +If the StateOrProvince name does not identify an unambiguous locality (within the country), the response is defined by *Search for StateOrProvince* using only the request parameters for Country and StateOrProvince. + +**Table 13/F.510 – Search for subscribers within state or province** + +| Search parameters | Request parameters | Response parameters | Word matching | String matching | +|---------------------------|--------------------|---------------------|---------------|-----------------| +| Attributes | | | | | +| Country | M | M | MR1 | MR4 | +| StateOrProvince | M | M | MR1, MR2 | MR9 | +| Locality or SmallLocality | X | M | – | – | +| SubscriberName | P3 | P1 | MR1, MR2 | MR9 | +| SecondFamilyName | P3 | P1 | MR1, MR2 | MR9 | +| SubscriberType | O | P1 | MR1 | MR4 | +| GivenName | O | P1 | MR1, MR2 | MR9 | +| BusinessCategory | O | P2 | MR1, MR2 | MR9 | +| Profession | X | P2 | – | – | +| StreetAddress | O | P2 | MR1, MR2 | MR9 | +| HouseIdentifier | O | P2 | MR1, MR2 | MR9 | +| ZipOrPostCode | X | P2 | – | – | +| Language | X | P2 | – | – | +| SubscriberDescription | X | P2 | – | – | +| CommunicationsService | O | P2 | MR1 | MR4 | +| CommunicationsAddress | X | P2 | – | – | +| CommunicationsNetwork | O | P2 | MR1 | MR4 | + +**Table 13/F.510 – Search for subscribers within state or province** + +| Search parameters | Request parameters | Response parameters | Word matching | String matching | +|------------------------|--------------------|---------------------|---------------|-----------------| +| Attributes | | | | | +| AddressValidity | X | P2 | – | – | +| AddressCoverage | X | P2 | – | – | +| AddressTariffClass | P6 | P2 | MR1 | MR4 | +| AddressRestriction | X | P2 | – | – | +| AddressQualifier | X | P2 | – | – | +| LinkedEntries | X | P2 | – | – | +| HierarchyBelow | X | P4 | – | – | +| HierarchyLevel | X | P4 | MR1 | MR4 | +| EntryIdentifier | X | P4 | – | – | +| Search controls | | | | | +| ExtendedArea | O | – | – | – | +| IncludeAllAreas | O | – | – | – | +| HierarchySelection | M | – | – | – | + +### 10.6.2 Search for subscribers within country + +The system shall provide users with a list of subscribers without specifying the geographical area for a specified country according to the search specification in Table 14. The TargetObject for the search is Subscriber Entry. + +**Table 14/F.510 – Search for subscribers within country** + +| Search parameters | Request parameters | Response parameters | Word matching | String matching | +|---------------------------|--------------------|---------------------|---------------|-----------------| +| Attributes | | | | | +| Country | M | M | MR1 | MR4 | +| StateOrProvince | X | M | – | – | +| Locality or SmallLocality | X | M | – | – | +| SubscriberName | P3 | P1 | MR1, MR2 | MR9 | +| SecondFamilyName | P3 | P1 | MR1, MR2 | MR9 | +| SubscriberType | O | P1 | MR1 | MR4 | +| GivenName | O | P1 | MR1, MR2 | MR9 | +| BusinessCategory | O | P2 | MR1, MR2 | MR9 | +| Profession | X | P2 | – | – | +| StreetAddress | O | P2 | MR1, MR2 | MR9 | +| HouseIdentifier | O | P2 | MR1, MR2 | MR9 | +| ZipOrPostCode | X | P2 | – | – | +| Language | X | P2 | – | – | +| SubscriberDescription | X | P2 | – | – | +| CommunicationsService | O | P2 | MR1 | MR4 | + +**Table 14/F.510 – Search for subscribers within country** + +| Search parameters | Request parameters | Response parameters | Word matching | String matching | +|--------------------------|---------------------------|----------------------------|----------------------|------------------------| +| CommunicationsAddress | X | P2 | – | – | +| CommunicationsNetwork | O | P2 | MR1 | MR4 | +| AddressValidity | X | P2 | – | – | +| AddressCoverage | X | P2 | – | – | +| AddressTariffClass | P6 | P2 | MR1 | MR4 | +| AddressRestriction | X | P2 | – | – | +| AddressQualifier | X | P2 | – | – | +| LinkedEntries | X | P2 | – | – | +| HierarchyBelow | X | P4 | – | – | +| HierarchyLevel | X | P4 | MR1 | MR4 | +| EntryIdentifier | X | P4 | – | – | +| Search controls | | | | | +| HierarchySelection | M | – | – | – | + +### **10.6.3 Search for street address** + +The system shall provide users with a list of street addresses, specifying part of a street address according to the search specification in Table 15. The TargetObject for the search is Street Address. + +**Table 15/F.510 – Search for street address** + +| Search parameters | Request parameters | Response parameters | Word matching | String matching | +|---------------------------|---------------------------|----------------------------|----------------------|------------------------| +| Attributes | | | | | +| Country | M | M | MR1 | MR4 | +| StateOrProvince | O | M | MR1, MR2 | MR9 | +| Locality or SmallLocality | O | M | MR1, MR2 | MR9 | +| StreetAddress | M | M | MR1, MR2 | MR9 | +| ZipOrPostCode | X | P2 | – | – | +| LocalityDescription | X | P2 | – | – | + +### **10.6.4 Search for subscribers by street address** + +The system shall provide users with a list of subscribers for a specified street address according to the search specification in Table 16. The TargetObject for the search is Subscriber Entry. + +**Table 16/F.510 – Search for subscribers by street address** + +| Search parameters | Request parameters | Response parameters | Word matching | String matching | +|---------------------------|--------------------|---------------------|---------------|-----------------| +| Attributes | | | | | +| Country | M | M | MR1 | MR4 | +| StateOrProvince | O | M | MR1, MR2 | MR9 | +| Locality or SmallLocality | O | M | MR1, MR2 | MR9 | +| SubscriberName | O | P1 | MR1, MR2 | MR9 | +| SecondFamilyName | O | P1 | MR1, MR2 | MR9 | +| SubscriberType | O | P1 | MR1 | MR4 | +| GivenName | O | P1 | MR1, MR2 | MR9 | +| BusinessCategory | O | P2 | MR1, MR2 | MR9 | +| Profession | X | P2 | – | – | +| StreetAddress | M | P2 | MR1, MR2 | MR9 | +| HouseIdentifier | O | P2 | MR1, MR2 | MR9 | +| ZipOrPostCode | X | P2 | – | – | +| Language | X | P2 | – | – | +| SubscriberDescription | X | P2 | – | – | +| CommunicationsService | O | P2 | MR1 | MR4 | +| CommunicationsAddress | X | P2 | – | – | +| CommunicationsNetwork | O | P2 | MR1 | MR4 | +| AddressValidity | X | P2 | – | – | +| AddressCoverage | X | P2 | – | – | +| AddressTariffClass | P6 | P2 | MR1 | MR4 | +| AddressRestriction | X | P2 | – | – | +| AddressQualifier | X | P2 | – | – | +| LinkedEntries | X | P2 | – | – | +| HierarchyBelow | X | P4 | – | – | +| HierarchyLevel | X | P4 | MR1 | MR4 | +| EntryIdentifier | X | P4 | – | – | +| Search controls | | | | | +| ExtendedArea | O | – | – | – | +| IncludeAllAreas | O | – | – | – | +| HierarchySelection | M | – | – | – | + +### 10.6.5 Search for subscribers by communications address + +The system shall provide users with a list of subscribers for a communications address according to the search specification in Table 17. The TargetObject for the search is Subscriber Entry. + +**Table 17/F.510 – Search for subscribers by communications address** + +| Search parameters | Request parameters | Response parameters | Word matching | String matching | +|---------------------------|--------------------|---------------------|---------------|-----------------| +| Attributes | | | | | +| Country | M | M | MR1 | MR4 | +| StateOrProvince | X | M | – | – | +| Locality or SmallLocality | X | M | – | – | +| SubscriberName | X | P1 | – | – | +| SecondFamilyName | X | P1 | – | – | +| SubscriberType | X | P1 | – | – | +| GivenName | X | P1 | – | – | +| BusinessCategory | X | P2 | – | – | +| Profession | X | P2 | – | – | +| StreetAddress | X | P2 | – | – | +| HouseIdentifier | X | P2 | – | – | +| ZipOrPostCode | X | P2 | – | – | +| Language | X | P2 | – | – | +| SubscriberDescription | X | P2 | – | – | +| CommunicationsService | O | P2 | MR1 | MR4 | +| CommunicationsAddress | M | P2 | MR1 | MR4 | +| CommunicationsNetwork | O | P2 | MR1 | MR4 | +| AddressValidity | X | P2 | – | – | +| AddressCoverage | X | P2 | – | – | +| AddressTariffClass | X | P2 | – | – | +| AddressRestriction | X | P2 | – | – | +| AddressDescription | X | P2 | – | – | +| LinkedEntries | X | P2 | – | – | +| HierarchyBelow | X | P4 | – | – | +| HierarchyLevel | X | P4 | MR1 | MR4 | +| EntryIdentifier | X | P4 | – | – | +| Search controls | | | | | +| HierarchySelection | M | – | – | – | + +### 10.6.6 Search for subscribers by business category + +The system shall provide users with a list of subscribers for a business category according to the search specification in Table 18. The TargetObject for the search is Subscriber Entry. + +**Table 18/F.510 – Search for subscribers by business category** + +| Search parameters | Request parameters | Response parameters | Word matching | String matching | +|---------------------------|--------------------|---------------------|---------------|-----------------| +| Attributes | | | | | +| Country | M | M | MR1 | MR4 | +| StateOrProvince | O | M | – | – | +| Locality or SmallLocality | O | M | – | – | +| SubscriberName | O | P1 | – | – | +| SecondFamilyName | O | P1 | – | – | +| SubscriberType | O | P1 | – | – | +| GivenName | O | P1 | – | – | +| BusinessCategory | M | P2 | – | – | +| Profession | O | P2 | – | – | +| StreetAddress | O | P2 | – | – | +| HouseIdentifier | O | P2 | – | – | +| ZipOrPostCode | O | P2 | – | – | +| Language | O | P2 | – | – | +| SubscriberDescription | X | P2 | – | – | +| CommunicationsService | O | P2 | MR1 | MR4 | +| CommunicationsAddress | O | P2 | MR1 | MR4 | +| CommunicationsNetwork | O | P2 | MR1 | MR4 | +| AddressValidity | X | P2 | – | – | +| AddressCoverage | X | P2 | – | – | +| AddressTariffClass | X | P2 | – | – | +| AddressRestriction | X | P2 | – | – | +| AddressDescription | X | P2 | – | – | +| LinkedEntries | X | P2 | – | – | +| HierarchyBelow | X | P4 | – | – | +| HierarchyLevel | X | P4 | MR1 | MR4 | +| EntryIdentifier | X | P4 | – | – | +| Search Controls | | | | | +| HierarchySelection | M | – | – | – | + +# 11 Directory Service Management Service Profile + +## 11.1 Introduction + +The Directory Service Management Service Profile defines representation for directory functions for the purpose of management. The representation takes the form of a structured information set termed a "managed object". + +Additionally, the profile will define a number of management procedures which enable the information embodied within managed objects to be utilized by Directory Service Managers and Administrators. There are three types of management procedures. + +- Notifications: A notification procedure signifies that an event has occurred related to the provision of a query service, and the notification is a message broadcast by the Directory Service Provider Administrator of the service to peer service managers. +- Request: A request procedure allows a Directory Service Provider Administrator to request access to a service offered by another domain. The results of a request are conveyed within a separate notification. +- List: A list procedure is used by a Directory Service Provider Administrator to view the attributes of services offered by other domains. + +All management procedures are effected by the Service Manager function on behalf of the Directory Service Provider Administrator. + +## 11.2 Directory Service Managed Objects + +### 11.2.1 Directory Service Profile Managed Objects + +#### 11.2.1.1 Directory Query Service + +A Directory Query Service managed object represents a specific directory query service instance and is defined in Table 19. + +**Table 19/F.510 – Directory Query Service Managed Object** + +| Attribute | Description | +|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Service Descriptor | The Service Descriptor references a service definition. The service definition is that defined in clause 10. | +| Service Domain | The name of the management domain which is offering the service. | +| Service Usage Conditions | The identified service usage conditions are:
a) Service Availability;
b) Maximum number of concurrent sessions (Note 1). | +| Service Usage Tariff | The Service Usage Tariff component defines tariff information for the usage of the service. | +| Service Usage Statistics | The Service Usage Statistics component records information on accesses to the service (Note 2) and records statistics such as failed connections, exceptions to service level agreement. | +| Service Special Character Mappings | Defines special character mapping pairs (character in the request, character in the directory) that define equality for character match that is supported by the Service Provider, e.g., aa, Å and ü, ue. Special means matching that is not simply removal of a diacritical mark. | +| NOTE 1 – A definition of a user managed object is required if categorization of service usage for a Query Service Agent needs to be defined for different user groups or user types. | | +| NOTE 2 – Requirements for usage statistics information will be defined by Quality of Service (clause 14). | | + +#### 11.2.1.2 Directory Query Service Access Point + +A Directory Service Access Point managed object represents the access details required to use a specific Directory Service (both query and data Directory Services). The Directory Service Access Point managed object is defined in Table 20. + +NOTE – A service is not restricted to one access point. + +**Table 20/F.510 – Directory Query Service Access Point Managed Object** + +| Attribute | Description | +|--------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Service Identifier | The name of the service accessible through this access point, comprising a service domain name and service descriptor. | +| Service Address | The address at which the service is to be used. (The exact value of this address depends upon the type of network upon which the service is being offered. In the case of OSI/X.500 this service address will be a Presentation Address.) | + +### 11.2.2 Directory Service Operational Managed Objects + +#### 11.2.2.1 Query Service Administrator + +The Service Administrator managed object represents the domain administrator user, who is responsible for management of services offered by the domain. This managed object is defined in Table 21. + +**Table 21/F.510 – Query Service Administrator Managed Object** + +| Attribute | Description | +|---------------------------------------|-------------------------------------------------------------------------------------------| +| Service Administrator Common Name | Simply, the name of the administrator. | +| Service Administrator Contact Details | This is a list of communications addresses, i.e., telephone, fax and e-mail numbers, etc. | + +## 11.3 Directory Service Management Procedures + +The identified Directory Service Management procedures are described in the Table 22. + +**Table 22/F.510 – Directory Service Management Procedures** + +| Procedure type | Procedure name | Descriptions | +|-----------------------|--------------------------------|---------------------------------------------------------------------------------------------------------------------------------------| +| Request | Service Access | This procedure is a request by an administrator user for access to a specified service. | +| Notification | Creation of Service | Occurrence of creation and current availability of a service from a specific domain. Broadcast to all peer Service Managers. | +| | Modification of Service | Occurrence of a modification to a service. Broadcast to all Peer Service Managers. | +| | Deletion of Service | Occurrence of service deletion. Broadcast to all peer service managers. | +| | Modification of Service Access | Occurrence of a change to service access points for a particular service. Multicast only to those domains which use the access point. | +| | Grant of Service Access | Positive indication in response to a "request for service access". Message sent only to requester service manager. | +| | Grant of Service Access Denied | Denial indication in response to a "request for service access". Message sent only to requester service manager. | +| | Modification to Service Usage | Indication that service usage conditions have been modified. | +| | Modification to Service Tariff | Indication that service tariffs have been modified. | + +**Table 22/F.510 – Directory Service Management Procedures** + +| Procedure type | Procedure name | Descriptions | +|-----------------------|-----------------------|-------------------------------------------------| +| List | Offered Services | List services offered by a domain. | +| | Usage Statistics | List usage information for a specified service. | +| | Usage Conditions | List usage conditions for a specified service. | +| | Usage Tariffs | List usage tariffs for a specified service. | + +# **12 Operational issues** + +## **12.1 Fallback and User Support** + +The operational issues for Fallback and User Support are covered in ITU-T Rec. E.104. + +## **12.2 Security** + +The Service Provider shall be able to authenticate the Query Service Agent when initial connection is established between Query Service Agent and Service Provider. + +The actual authentication mechanism is based on ITU-T Rec. X.509 | ISO/IEC 9594-8. Simple Authentication is the minimum level that shall be used and the use of Strong Authentication is preferred. + +# **13 Charging and accounting** + +The accounting model to be applied for this Recommendation is the one defined in clause 4/D.37. + +The elements to be accounted for settlements are based on message codes (with possibly accompanying parameters) and optionally, the number of entries provided in each response and RequestedService. Message codes are defined in Annex A. + +The actual message codes (with accompanying parameters) that will be used, and the use of number of entries for accounting, between a Query Service Agent and Service Provider will be defined *bilaterally*. + +# **14 Quality of Service** + +For further study. + +# **Annex A** + +## **Messages** + +F.510 Messages include not only errors detected while performing searches but additional information which characterizes the response to any search. + +These messages will usually be used to provide an explanatory meaning to the user. The presentation of the messages to the user is a local matter. + +The messages include, when necessary, precise information, such as attribute types, attribute values or other identifiers, that relate to problems detected. + +F.510 Messages are categorized according to the following occurrences: + +- message codes specific for the specification realising the ITU-T Rec. F.510 service (# 0) +- the full service cannot be offered due to technical problems (# 1). +- the whole request cannot be performed as a service has been requested that is not supported (# 2). +- the request cannot be performed as a service parameter has been requested for a given attribute that is not supported (# 3, 4). +- the search within the database has been performed but a problem in identifying the geography has been detected and no entries are returned (# 5). +- the request has been performed but the entry set returned is empty (# 6). +- the request has been performed: the entry set returned is incomplete (# 7) or complete (# 8). + +NOTE 1 – F.510 messages will usually be generated by the Service Provider but some of them can also be directly generated by the Query Service Agent depending on the knowledge it has of the Service Provider offerings. + +NOTE 2 – A reference is provided to each of the Messages in this annex but the actual mapping code will be described in associated technical profiles. + +### **0 Realisation specific messages** + +- 0.x Messages codes that are specific for the particular realisation of this service definition. + +### **1 Unavailable Service Access** + +*By default: The F.510 service is out of order* + +- 1.1 System congestion +- 1.2 Destination database not accessible +- 1.3 Destination database not accessible + +### **2 Required service not supported by the Service Provider** + +*By default: An enhanced service has been requested that is not supported by the Service Provider* + +- 2.1 Search for service is not supported +- 2.2 Search for service supported but not bilaterally agreed + +### **3 Required functionality not supported by the Service Provider** + +*By default: A function has been requested or attribute parameter used that is not supported by the Service Provider* + +- 3.1 Matching rule not supported by Service Provider for the whole service +- 3.2 Matching rule not supported by Service Provider for this specific +- 3.3 Inappropriate matching rule for +- 3.4 Invalid combination of matching rules for +- 3.5 Geographical extension feature not supported +- 3.6 Level of geographical extension not supported +- 3.7 Search with the specified HierarchySelection value not supported +- 3.8 Filter on for service not supported + +- 3.9 Filter on for service not supported +- 3.10 required by Service Provider to perform the search +- 3.11 Value for is not bilaterally agreed + +### **4 Wrong input information** + +*By default: The search cannot be performed because insufficient or incorrect information has been provided* + +- 4.1 Mandatory are missing for requested service +- 4.2 Minimum size of truncated word is required for +- 4.3 Mandatory combination of are missing for requested service +- 4.4 Invalid value for because of incorrect syntax (e.g., no figures for a number) +- 4.5 Unrecognized value for or (e.g., HierarchySelection parameter, AddressTariffClass attribute, CommunicationsService attribute, CommunicationsNetwork attribute) + +### **5 Undetermined key attributes** + +*By default: The service is unable to determine the geographic locality or a business category* + +- 5.1 The parameter does not exist +- 5.2 The parameter is ambiguous +- 5.3 The combination of parameters , , etc. does not exist + +### **6 No entries returned** + +*By default: The request is performed but no information has been returned* + +- 6.1 No subscriber entries are available under geographical area +- 6.2 No subscriber entries have been found for the requested HierarchySelection value +- 6.3 No subscriber entries have been found but entries exist without constraint for +- 6.4 No entries have been found but entries exist without constraint +- 6.5 Too many entries selected: more selective information is necessary (needed?) +- 6.6 No entries have been found that match the search criteria +- 6.7 The requested page is not available + +### **7 List of non-subscribers attributes** + +*By default: Outcomes are partial when returning a list of subscribers or localities* + +- 7.1 Incomplete list of , more available +- 7.2 Incomplete list of no more available + +### 8 Entries returned + +*By default: Entries have been returned* + +- 8.1 Entries found +- 8.2 Entries found, more entries available + +# Annex B + +## List of symbols + +The symbols listed here are defined in clauses 3 and 9 and used in clause 10. + +- CR1 CharacterMap character matching rule +- CR2 CharacterCaseIgnore character matching rule +- CR3 CharacterExact character matching rule +- MR1 WordExact word matching rule +- MR2 WordTruncated word matching rule +- MR3 WordPhonetic word matching rule +- MR4 StringExact string matching rule +- MR5 WordDeletion string matching rule +- MR6 WordRestrictedDeletion string matching rule +- MR7 WordRotation string matching rule +- MR8 WordRotationAndDeletion string matching rule +- MR9 ProviderDefined string matching rule +- M Mandatory +- O Optional +- X Omitted +- Not Applicable +- Pn Predicate (number n) + +# Annex C + +## Service Data Model + +The Service Data Model is described in Figure C.1 using the notation described in Annex D. + +![UML Class Diagram for White Pages Service showing Geographic Data and Subscriber Data domains.](40a8c30f7ea5ecea4912e040c97c5b9c_img.jpg) + +The diagram illustrates the Service Data Model for the White Pages Service, organized into two main domains: **Geographic Data** and **Subscriber Data**. + +**Geographic Data Domain:** + +- Geographic Locality** is the central entity, which has a one-to-one relationship with **Geographic Locality Alias**. +- Country**, **State or Province**, **Locality**, and **Small Locality** form a hierarchical chain. **Country** has a one-to-many relationship with **State or Province**, which in turn has a one-to-many relationship with **Locality**, which finally has a one-to-many relationship with **Small Locality**. +- Subscriber Locality** is a subclass of **Locality** (indicated by a generalization relationship). +- Subscriber Locality** has a one-to-one relationship with **Locality Street Address**. + +**Subscriber Data Domain:** + +- Subscriber Locality** is the central entity, which has a one-to-one relationship with **Comms. Address**. +- Hierarchical Group** has a one-to-many relationship with **Subscriber Locality**. +- Governmental Subscriber**, **Organizational Subscriber**, and **Residential Subscriber** are subclasses of **Subscriber Locality** (indicated by a generalization relationship). + +A vertical line connects the **Subscriber Locality** entity in the Geographic Data domain to the **Subscriber Locality** entity in the Subscriber Data domain, indicating they represent the same concept across different data domains. + +UML Class Diagram for White Pages Service showing Geographic Data and Subscriber Data domains. + +F.510\_FC.1 + +Figure C.1/F.510 – CLASS DIAGRAM: White Pages Service + +# Annex D + +## Object Modelling Notation + +Figure D.1 illustrates the notation used to describe classes, object (class) instances, relationships between classes and multiplicity of associations between classes. + +![UML diagrams showing Class, Object Instances, Generalization, Aggregation, and Associations.](a0e8fe7862a6d7341faf5dac275277cc_img.jpg) + +The diagram illustrates various UML modeling notations: + +- Class:** A rectangular box with two compartments. The top compartment contains the text "Class Name". +- Object Instances:** A rounded rectangular box containing the text "(Class Name)". +- Generalization:** A "Super Class" box at the top is connected by a line with an open triangle arrowhead to two boxes below labeled "Subclass-1" and "Subclass-2". +- Aggregation:** An "Assembly Class" box at the top is connected by a line with an open diamond to two boxes below labeled "Part-1-Class" and "Part-2-Class". +- Associations:** Three examples of associations between "Class-1" (top) and "Class-2" (bottom): + - Exactly One:** A simple vertical line connects the two classes. + - Optional (zero or one):** A vertical line connects the two classes, with a small open circle at the "Class-2" end. + - Many (zero or more):** A vertical line connects the two classes, with a solid black circle at the "Class-2" end. + +F.510\_FD.1 + +UML diagrams showing Class, Object Instances, Generalization, Aggregation, and Associations. + +**Figure D.1/F.510 – Object Modelling Notation** + + + + + +# SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|--------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series B | Means of expression: definitions, symbols, classification | +| Series C | General telecommunication statistics | +| Series D | General tariff principles | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Construction, installation and protection of cables and other elements of outside plant | +| Series M | TMN and network maintenance: international transmission systems, telephone circuits, telegraphy, facsimile and leased circuits | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks and open system communications | +| Series Y | Global information infrastructure and Internet protocol aspects | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.511-201404-I_PDF-E/0b3d9fe35da3ee0c88f1420bb9ed7a03_img.jpg b/marked/F/T-REC-F.511-201404-I_PDF-E/0b3d9fe35da3ee0c88f1420bb9ed7a03_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f2f71320a9cbbbf520de1aa40ce5f192a802dc74 --- /dev/null +++ b/marked/F/T-REC-F.511-201404-I_PDF-E/0b3d9fe35da3ee0c88f1420bb9ed7a03_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ad358ee1cc0bd65d62531a6e8621740ff20eb449b1481199ba3497a0080bfb62 +size 42272 diff --git a/marked/F/T-REC-F.511-201404-I_PDF-E/19a5f0db57a21a0e82a7f326083e96fd_img.jpg b/marked/F/T-REC-F.511-201404-I_PDF-E/19a5f0db57a21a0e82a7f326083e96fd_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..177c550353ace8a2d178fddfefaa8fd40e6c6b79 --- /dev/null +++ b/marked/F/T-REC-F.511-201404-I_PDF-E/19a5f0db57a21a0e82a7f326083e96fd_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8c5ea7c94b2be2be6f1e92521cbdd6d515f2e890f1ad389dceb0f201e59510f1 +size 19380 diff --git a/marked/F/T-REC-F.511-201404-I_PDF-E/1b2ad37940c441d410002c05ff71c7c5_img.jpg b/marked/F/T-REC-F.511-201404-I_PDF-E/1b2ad37940c441d410002c05ff71c7c5_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7944a7cf9ff10550f5c20e4d22288052f997242a --- /dev/null +++ b/marked/F/T-REC-F.511-201404-I_PDF-E/1b2ad37940c441d410002c05ff71c7c5_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7401aae0ccdce229c762219157eee79cd6d37f32bf106e8570c2f9dceb399657 +size 37273 diff --git a/marked/F/T-REC-F.511-201404-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.511-201404-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9002ee68e2b9ebf63d27d39ba8c73bdafb0e6be6 --- /dev/null +++ b/marked/F/T-REC-F.511-201404-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8df701536c25ab8246d2868de3c20c8a4e25168bb52c5eb9078278ab1b5b4fd8 +size 6859 diff --git a/marked/F/T-REC-F.511-201404-I_PDF-E/34b047489058d6400b412cd0ae2334ba_img.jpg b/marked/F/T-REC-F.511-201404-I_PDF-E/34b047489058d6400b412cd0ae2334ba_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..28c74d1853c3b736d53c9bb9c50e38e7f2601084 --- /dev/null +++ b/marked/F/T-REC-F.511-201404-I_PDF-E/34b047489058d6400b412cd0ae2334ba_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1eb709c6d7f43596b0745cc7367aa3445b41f389729ffeca8b63d37ec1dd31bf +size 24686 diff --git a/marked/F/T-REC-F.511-201404-I_PDF-E/365b54f616aff249b4e6c0edafdcb9b3_img.jpg b/marked/F/T-REC-F.511-201404-I_PDF-E/365b54f616aff249b4e6c0edafdcb9b3_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..55060c3d35da5716fa24c97605044916da1a3796 --- /dev/null +++ b/marked/F/T-REC-F.511-201404-I_PDF-E/365b54f616aff249b4e6c0edafdcb9b3_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2928374f885e22f6fc118bb6d56f7dca9d34019a1c6b6f8e98c347e024265de5 +size 75026 diff --git a/marked/F/T-REC-F.511-201404-I_PDF-E/366a77fdefb0097b3289b4a011911390_img.jpg b/marked/F/T-REC-F.511-201404-I_PDF-E/366a77fdefb0097b3289b4a011911390_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c0fc8903599063b93292f467761785fd4c1e5688 --- /dev/null +++ b/marked/F/T-REC-F.511-201404-I_PDF-E/366a77fdefb0097b3289b4a011911390_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:de017804398e903fc740bcb8fd574acfdbc33a09749241d15005319987c1e8ef +size 103790 diff --git a/marked/F/T-REC-F.511-201404-I_PDF-E/5d782eeb9d1e5871d7f09e0ccdd4cdf1_img.jpg b/marked/F/T-REC-F.511-201404-I_PDF-E/5d782eeb9d1e5871d7f09e0ccdd4cdf1_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..168fbed6f0258cbe940c4bdb0e519896f8632ad7 --- /dev/null +++ b/marked/F/T-REC-F.511-201404-I_PDF-E/5d782eeb9d1e5871d7f09e0ccdd4cdf1_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3f89c561905f13bbbbdb9dbf73e3b80d2de956a1e6aa27fe3d2c9b39777d64a2 +size 47593 diff --git a/marked/F/T-REC-F.511-201404-I_PDF-E/67114ecf6da13c97a89aeff0d86885d5_img.jpg b/marked/F/T-REC-F.511-201404-I_PDF-E/67114ecf6da13c97a89aeff0d86885d5_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c6f7334e281de7f0d95962601710f29564989eb5 --- /dev/null +++ b/marked/F/T-REC-F.511-201404-I_PDF-E/67114ecf6da13c97a89aeff0d86885d5_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e3fa43b593718ba6b0723d9497dc47c43c8cf7ab310bda88d4b8d0ccc3eb142e +size 43588 diff --git a/marked/F/T-REC-F.511-201404-I_PDF-E/70a863dc6cd47bac82c12162a9479aac_img.jpg b/marked/F/T-REC-F.511-201404-I_PDF-E/70a863dc6cd47bac82c12162a9479aac_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..5071df821febec287338c0d7beb68e48cce154d4 --- /dev/null +++ b/marked/F/T-REC-F.511-201404-I_PDF-E/70a863dc6cd47bac82c12162a9479aac_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:da6a717fcae766ca3addc5448269ae101dbfbd2571312e1d8ecb59740afdff1b +size 44802 diff --git a/marked/F/T-REC-F.511-201404-I_PDF-E/7f25db95ce3916c0e09803b861a2f7bc_img.jpg b/marked/F/T-REC-F.511-201404-I_PDF-E/7f25db95ce3916c0e09803b861a2f7bc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..251212d437b9ebcc8434f8263a3503f18f843d43 --- /dev/null +++ b/marked/F/T-REC-F.511-201404-I_PDF-E/7f25db95ce3916c0e09803b861a2f7bc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6e8d5839e02518b66890dad1e747a2026c77ce3d405ea22c42c29e06669b73df +size 52900 diff --git a/marked/F/T-REC-F.511-201404-I_PDF-E/85e2327652d513a7fee8fdbf97ad06f1_img.jpg b/marked/F/T-REC-F.511-201404-I_PDF-E/85e2327652d513a7fee8fdbf97ad06f1_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d0c3f3b2ece0c5e122cb5e3aec113efecfbe42fd --- /dev/null +++ b/marked/F/T-REC-F.511-201404-I_PDF-E/85e2327652d513a7fee8fdbf97ad06f1_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:749b130d57ab853bcaac5148af04de65f4a6572dfbd66cd32546e3711f12459f +size 49010 diff --git a/marked/F/T-REC-F.511-201404-I_PDF-E/a1890b9a9b85f13e67ed59bbad623659_img.jpg b/marked/F/T-REC-F.511-201404-I_PDF-E/a1890b9a9b85f13e67ed59bbad623659_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c57087d2507b8dde4a8bcde3ac507e53d81462d3 --- /dev/null +++ b/marked/F/T-REC-F.511-201404-I_PDF-E/a1890b9a9b85f13e67ed59bbad623659_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c3c68ae70314ef0ec156efaa8ef08ccdab158986240aa3791bf69a157e9c22c8 +size 44398 diff --git a/marked/F/T-REC-F.511-201404-I_PDF-E/a5b9392ecb96e6b5e0b4ee0664210f72_img.jpg b/marked/F/T-REC-F.511-201404-I_PDF-E/a5b9392ecb96e6b5e0b4ee0664210f72_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..688b3c2b9a1404b284f7a054913cdad3eaf49cc4 --- /dev/null +++ b/marked/F/T-REC-F.511-201404-I_PDF-E/a5b9392ecb96e6b5e0b4ee0664210f72_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1d8e5bc7574a1260eaf3d75f3c11e510059da99aef7e6b289eb1507ada0403cf +size 26920 diff --git a/marked/F/T-REC-F.511-201404-I_PDF-E/b560268ea8f6526970f23f0da225b099_img.jpg b/marked/F/T-REC-F.511-201404-I_PDF-E/b560268ea8f6526970f23f0da225b099_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d5a0ebcdbad71bfd638975ccef17e01d313e9151 --- /dev/null +++ b/marked/F/T-REC-F.511-201404-I_PDF-E/b560268ea8f6526970f23f0da225b099_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:039f5d2be7b024d98bfb58e67b96d979aa736c35d732bde415d6d3ea6e9e6ded +size 33486 diff --git a/marked/F/T-REC-F.511-201404-I_PDF-E/c85b57b2414f341860dfc338e1cf2509_img.jpg b/marked/F/T-REC-F.511-201404-I_PDF-E/c85b57b2414f341860dfc338e1cf2509_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3a98c4ba597eadbaf295dc2611c82007e3bcfa11 --- /dev/null +++ b/marked/F/T-REC-F.511-201404-I_PDF-E/c85b57b2414f341860dfc338e1cf2509_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c1602bd30e87240cb11778d986673a83a6035e4884198bc62659ea8aeaae050c +size 22835 diff --git a/marked/F/T-REC-F.511-201404-I_PDF-E/dd5771673aececa53d42ece89218299d_img.jpg b/marked/F/T-REC-F.511-201404-I_PDF-E/dd5771673aececa53d42ece89218299d_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d5eb20f0969598ff60460072978dcf1a4e4548fa --- /dev/null +++ b/marked/F/T-REC-F.511-201404-I_PDF-E/dd5771673aececa53d42ece89218299d_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d6ca2168303169badd021ca3309e85fcbe247f80ba9037f96d8487760d2b3b9e +size 50957 diff --git a/marked/F/T-REC-F.511-201404-I_PDF-E/e354b57563dae469c00b412b2abdf765_img.jpg b/marked/F/T-REC-F.511-201404-I_PDF-E/e354b57563dae469c00b412b2abdf765_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d9ea7b349d5c11119991ace35d1e24fc0c0339a0 --- /dev/null +++ b/marked/F/T-REC-F.511-201404-I_PDF-E/e354b57563dae469c00b412b2abdf765_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ecf6f10b6cd758b6c189e765c6671c13252d4862f499837ce1b4ffd486491a3f +size 114132 diff --git a/marked/F/T-REC-F.511-201404-I_PDF-E/e69b9188aa2c14ec6b21c83f711fef65_img.jpg b/marked/F/T-REC-F.511-201404-I_PDF-E/e69b9188aa2c14ec6b21c83f711fef65_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..48c7977b725200315dc8d473a6bb78fd315872ec --- /dev/null +++ b/marked/F/T-REC-F.511-201404-I_PDF-E/e69b9188aa2c14ec6b21c83f711fef65_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:93488251ad8d63c5d93df43b9ca95fa712edeafc4e52a7c67a1e32c2a8c9fbed +size 53925 diff --git a/marked/F/T-REC-F.511-201404-I_PDF-E/e821c3d8a87ee2a9ff6b8644ffe6bdae_img.jpg b/marked/F/T-REC-F.511-201404-I_PDF-E/e821c3d8a87ee2a9ff6b8644ffe6bdae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..154af50a5383dc20058d2d6bb1a72e32560c3e7b --- /dev/null +++ b/marked/F/T-REC-F.511-201404-I_PDF-E/e821c3d8a87ee2a9ff6b8644ffe6bdae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:067401fab488747522e9926943ca33b5cb18d075c7f2e1719a55b786287676fb +size 42434 diff --git a/marked/F/T-REC-F.511-201404-I_PDF-E/e94cbc0895c6914a2bb53f150ab7495e_img.jpg b/marked/F/T-REC-F.511-201404-I_PDF-E/e94cbc0895c6914a2bb53f150ab7495e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7d2adb77652fec41c58c9daff9c699e0295550f9 --- /dev/null +++ b/marked/F/T-REC-F.511-201404-I_PDF-E/e94cbc0895c6914a2bb53f150ab7495e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:29519053ccde71512617232d58da20418854c43568a006ada75c449ee8f142ad +size 41443 diff --git a/marked/F/T-REC-F.511-201404-I_PDF-E/f1091147d93cee4dfa88498610e395a7_img.jpg b/marked/F/T-REC-F.511-201404-I_PDF-E/f1091147d93cee4dfa88498610e395a7_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f78c14b591a40e49cf6d1cbf7ebc9af21d9781b1 --- /dev/null +++ b/marked/F/T-REC-F.511-201404-I_PDF-E/f1091147d93cee4dfa88498610e395a7_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a056fdc86da2bbfc0c20c45bf29ab562140c7dbe623aa5770bbba1b8381b6ad4 +size 25680 diff --git a/marked/F/T-REC-F.511-201404-I_PDF-E/f2ea0f64a770b22b902820457d262265_img.jpg b/marked/F/T-REC-F.511-201404-I_PDF-E/f2ea0f64a770b22b902820457d262265_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c9a9d51207b4ae2010608cd058a072c21dd0594f --- /dev/null +++ b/marked/F/T-REC-F.511-201404-I_PDF-E/f2ea0f64a770b22b902820457d262265_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a2bd13fe512a07fcc17fbfd09563bdb8a9778824b332baa31e4e7525e7c3cc1a +size 38530 diff --git a/marked/F/T-REC-F.511-201404-I_PDF-E/fa859e4e468bfb2710a94527f2c504af_img.jpg b/marked/F/T-REC-F.511-201404-I_PDF-E/fa859e4e468bfb2710a94527f2c504af_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ec7565a7dba5408c59245e43271fa4128e6fc803 --- /dev/null +++ b/marked/F/T-REC-F.511-201404-I_PDF-E/fa859e4e468bfb2710a94527f2c504af_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a4a14c00abda68ac7526e0824e524c9a8b03bc1b503b02b7f41d559c4d3a263d +size 33402 diff --git a/marked/F/T-REC-F.511-201404-I_PDF-E/ff7ce44f3fdd51bae7b231f34df07c6a_img.jpg b/marked/F/T-REC-F.511-201404-I_PDF-E/ff7ce44f3fdd51bae7b231f34df07c6a_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..dcd7cf0616e009112e581b95e321b906aae792c3 --- /dev/null +++ b/marked/F/T-REC-F.511-201404-I_PDF-E/ff7ce44f3fdd51bae7b231f34df07c6a_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:050bb204f2175ed6b94cbaeb2d9e97f6382634117424729a66a3afe846e7b392 +size 48077 diff --git a/marked/F/T-REC-F.515-200304-I_PDF-E/raw.md b/marked/F/T-REC-F.515-200304-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..ddbaa69228436bf34a6773ed2ade43381f59df7d --- /dev/null +++ b/marked/F/T-REC-F.515-200304-I_PDF-E/raw.md @@ -0,0 +1,3504 @@ + + +![ITU logo](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +The logo of the International Telecommunication Union (ITU) features a globe with a lightning bolt superimposed on it, and the letters 'ITU' written across the center. + +ITU logo + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.515** + +(04/2003) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Directory services + +--- + +**Unified Directory Specification** + +ITU-T Recommendation F.515 + +--- + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| AUDIOVISUAL SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# **Unified Directory Specification** + +## **Summary** + +This Recommendation provides a directory specification to fulfil the service requirements of ITU-T Rec. F.510. The directory specification is named Unified Directory Specification (UDS) to reflect that it allows a great variety of database structures to be supported. The ITU-T Rec. F.510 service is realized by the Unified Directory Access Protocol (UDAP). This protocol is expressed using the eXtensible Markup Language (XML) schema notation. + +UDAP has been designed in a way that facilitates an easy mapping to ITU-T Rec. E.115. + +The structure of this Recommendation follows the structure of ITU-T Rec. F.510. Certain clauses in ITU-T Rec. F.510 do not require a corresponding specification and are left void. + +## **Source** + +ITU-T Recommendation F.515 was prepared by ITU-T Study Group 17 (2001-2004) and approved under the ITU-T Recommendation A.8 procedure on 22 April 2003. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications. The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure e.g. interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementors are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database. + +© ITU 2003 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## CONTENTS + +###### Page + +| | | | +|------|----------------------------------------------------|----| +| 1 | Scope ..... | 1 | +| 2 | Normative references..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | White pages service definitions..... | 1 | +| 3.2 | Unified directory definitions ..... | 2 | +| 4 | Abbreviations..... | 2 | +| 5 | Conventions ..... | 2 | +| 5.1 | Text conventions ..... | 2 | +| 5.2 | XML notation conventions..... | 3 | +| 6 | Directory service model..... | 3 | +| 7 | Directory information model ..... | 4 | +| 7.1 | Information structure ..... | 4 | +| 7.2 | Geography ..... | 9 | +| 7.3 | Business classification..... | 9 | +| 7.4 | Organizational structure ..... | 9 | +| 7.5 | Subscriber entry..... | 9 | +| 8 | Service specification..... | 14 | +| 9 | Common Protocol Specification element ..... | 14 | +| 9.1 | Character set, encoding and repertoire ..... | 14 | +| 9.2 | Matching rules ..... | 15 | +| 9.3 | Language differences..... | 15 | +| 9.4 | Ordering of entries within a response..... | 15 | +| 9.5 | Hierarchical groups ..... | 15 | +| 9.6 | Number of entries ..... | 17 | +| 9.7 | Entry count ..... | 17 | +| 9.8 | Alternative directory attribute values ..... | 17 | +| 9.9 | Weighted directory attribute values..... | 17 | +| 9.10 | Geographical extensions..... | 17 | +| 9.11 | Ignore if absent..... | 17 | +| 9.12 | Paged Results request..... | 17 | +| 10 | The Unified Directory Access Protocol (UDAP) ..... | 18 | +| 10.1 | Service model ..... | 18 | +| 10.2 | Search controls and indications ..... | 18 | +| 10.3 | Predicates..... | 18 | + +| | Page | +|-------------------------------------------------------|------| +| 10.4 Common query conditions ..... | 18 | +| 10.5 Basic service offering ..... | 19 | +| 10.6 Enhanced service offering ..... | 21 | +| 11 Operational issues ..... | 21 | +| 11.1 Security ..... | 21 | +| 12 Charging and accounting ..... | 21 | +| 13 Quality of Service ..... | 21 | +| Annex A – Message codes and notifications ..... | 21 | +| A.1 Notification attribute types ..... | 21 | +| A.2 Message codes ..... | 22 | +| Annex B – XML schema for UDAP ..... | 28 | +| B.1 Search request ..... | 28 | +| B.2 Search result ..... | 29 | +| B.3 Attribute Types ..... | 30 | +| Annex C – Use of Web Service facilities ..... | 42 | +| C.1 Use of Simple Object Access Protocol (SOAP) ..... | 42 | +| Annex D – ASN.1 version of UDAP ..... | 44 | + +# Introduction + +This Recommendation provides a directory specification to fulfil the service requirements of ITU-T Rec. F.510. The directory specification is named Unified Directory Specification (UDS) to reflect that it allows a great variety of database structures to be supported. The ITU-T Rec. F.510 service is realised by the Unified Directory Access Protocol (UDAP). This protocol is expressed using the eXtensible Markup Language (XML) schema notation. + +UDAP has been designed in a way that facilitates an easy mapping to ITU-T Rec. E.115. + +The structure of this Recommendation follows the structure of ITU-T Rec. F.510. Certain clauses in ITU-T Rec. F.510 do not require a corresponding specification and are left void. + +Annex A, which is an integral part of this Recommendation, list the messages codes as defined in Annex A/F.510, and it lists the conditions under which they are generated and what additional information is associated with each message code. + +Annex B, which is an integral part of this Recommendation, gives the formal specification of UDAP in XML schema notation. + +Annex C, which is an integral part of this Recommendation, specifies how UDS uses underlying services. + +Annex D, which is an integral part of this Recommendation, gives the formal specification of UDAP in ASN.1 notation. + +An implementation shall implement the encoding specified either in Annex B, or in Annex D, or in both. + + + +# Unified Directory Specification + +# 1 Scope + +This Recommendation only considers directory information retrieval using the search operation. Requirements on administrative workstations or administrative capabilities are outside the scope of this Recommendation. + +This Recommendation supports the Directory Service as specified by ITU-T Rec. F.510. However, it could be used in other environments. + +This Recommendation considers only client to directory server communication. It does not consider directory server to directory server communication. + +# 2 Normative references + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +- ITU-T Recommendation E.115 (1995), *Computerized directory assistance*. +- ITU-T Recommendation F.510 (2003), *Automated directory assistance – White pages service definition*. +- ITU-T Recommendation X.693 (2001) | ISO/IEC 8825-4:2002, *Information technology – ASN.1 encoding rules: XML Encoding Rules (XER)*. +- ISO/IEC 10646-1:2000, *Information technology – Universal Multiple-Octet Coded Character Set (UCS) – Part 1: Architecture and Basic Multilingual Plane*. +- ISO 3166 (all parts), *Codes for the representation of names of countries and their subdivisions*. + +# 3 Definitions + +## 3.1 White pages service definitions + +The following terms are defined in ITU-T Rec. F.510: + +- a) (directory) attribute; +- b) (directory) entry; +- c) directory service; +- d) hierarchical group; +- e) key attribute; +- f) query service entry. + +## 3.2 Unified directory definitions + +This Recommendation defines the following terms: + +**3.2.1 client:** An application that issues directory requests and receives directory information in responses. + +**3.2.2 directory server:** An application that can return directory information when receiving request over the Unified Directory Access Protocol. + +**3.2.3 entry:** A collection of directory attributes and families that comprises the information held about an object. + +**3.2.4 family:** A grouping of related directory attributes within an entry. + +**3.2.5 filter:** A construct carried in a search request to be matched against entry information. + +**3.2.6 filter item:** A component of a filter that carries information to be matched against directory attributes of a particular type. + +**3.2.7 object:** An entity, e.g., a person, that is represented by an entry of a directory server. + +# 4 Abbreviations + +This Recommendation uses the following abbreviations: + +| | | +|-------|--------------------------------------------| +| ASN.1 | Abstract Syntax Notation One | +| DMD | Directory Management Domain | +| GSM | Global System for Mobile communications | +| IANA | Internet Assigned Numbers Authority | +| ISDN | Integrated Services Digital Network | +| LDAP | Lightweight Directory Access Protocol | +| PSTN | Public Switched Telephone Network | +| SMS | Short Message Service | +| SOAP | Simple Object Access Protocol | +| UDAP | Unified Directory Access Protocol | +| UDS | Unified Directory Specification | +| UMTS | Universal Mobile Telecommunications System | +| XML | eXtensible Markup Language | + +# 5 Conventions + +## 5.1 Text conventions + +There is terminology collision between that of directories and that of eXtensible Markup Language (XML). The terms attributes and schema are used in both environments. To avoid confusions, those two terms are qualified by "directory" or "XML", as appropriate. + +This Recommendation presents XML notation in Courier New, 9-point typeface and ASN.1 in Courier New, 10-point typeface. When XML specifications are referenced in normal text, they are differentiated from other text by presenting them in the bold Courier New, 10-point typeface. + +## 5.2 XML notation conventions + +### 5.2.1 General on XML notation + +All complex and simple XML types defined by this Recommendation have a name that starts with an uppercase letter. + +All names of non-abstract XML elements defined by this Recommendation start with a lowercase letter. + +All names of XML attributes defined by this Recommendation start with a lowercase letter. + +All names of abstract elements start with uppercase letter. + +### 5.2.2 Syntaxes with upper bounds + +Most directory attribute types with string syntax has a maximum length, termed upper bound. Simple XML types, which specifies upper limits, are defined for that purpose. + +The name of such a simple XML type is the letters **ub** plus a number, for example **ub64**, where the number indicates the upper bound in number of characters for the type. Such types are derived from the XML built-in **string** type. + +### 5.2.3 Enumerated types + +Some elements and XML attributes may take one out of several predefined values. Special XML simple types are defined for such element types. + +The XML built-in type **NCName** is the base type for an enumerated type. + +An XML enumerated simple type is named by taking a *descriptive name*, e.g., **AttributeType**. The first character of the name of such a simple type is an uppercase letter. + +The value of an XML element or an XML attribute of the enumerated type shall be exactly one of the defined enumerated values. + +### 5.2.4 List types + +The XML list type facility is used where several values of an enumerated type may be required within a single XML element or XML attribute. An XML simple list type is given a *descriptive name* suffixed with an **s**, e.g., **AttributeTypes**. The first character of the name of such a simple type is an uppercase letter. + +A value of this type holds zero or more of the defined enumerated values. + +### 5.2.5 Abstract element schema definitions and substitution groups + +Abstract XML element schema definitions are used as "head" elements for XML schema substitution groups. The first character of a name of an abstract XML element is an uppercase letter. + +# 6 Directory service model + +Figure 1 illustrates the position of the UDAP with respect to different components defined within clause 6/F.510. The protocol between the access service and the query service agent could be any private or standardized protocol. In particular, it could be UDAP as defined by this Recommendation. The query service agent may, dependent on circumstances, provide some of the ITU-T Rec. F.510 service elements. However, the UDAP supports the full set of ITU-T Rec. F.510 service elements. + +![Figure 1/F.515 – Possible service scenarios. The diagram shows two service scenarios separated by a dashed vertical line labeled 'F.510 service interface'. The top scenario shows a 'Local agent' containing multiple 'User' circles, 'Operator' rectangles, and 'Operator workstation' rectangles, all connected to an 'Access service' rectangle. This 'Access service' connects to a 'Query Service Agent' rectangle, which in turn connects to a 'UDS server' rectangle via a 'UDAP' label. The bottom scenario shows a 'Local agent' containing a 'User' circle connected to a 'Workstation/access service' rectangle. This rectangle connects to a 'Query Service Agent' rectangle, which connects to a 'UDS server' rectangle via a 'UDAP' label. The label 'F.515_F01' is at the bottom right.](d4af765160d04ecef538e5066006dc77_img.jpg) + +Figure 1/F.515 – Possible service scenarios. The diagram shows two service scenarios separated by a dashed vertical line labeled 'F.510 service interface'. The top scenario shows a 'Local agent' containing multiple 'User' circles, 'Operator' rectangles, and 'Operator workstation' rectangles, all connected to an 'Access service' rectangle. This 'Access service' connects to a 'Query Service Agent' rectangle, which in turn connects to a 'UDS server' rectangle via a 'UDAP' label. The bottom scenario shows a 'Local agent' containing a 'User' circle connected to a 'Workstation/access service' rectangle. This rectangle connects to a 'Query Service Agent' rectangle, which connects to a 'UDS server' rectangle via a 'UDAP' label. The label 'F.515\_F01' is at the bottom right. + +**Figure 1/F.515 – Possible service scenarios** + +As the UDAP may be used in other places than indicated in Figure 1, the term *client* is used for the function that issues a UDAP request, and the term *directory server* is used for the function replying to a UDAP request with a response. + +# **7 Directory information model** + +## **7.1 Information structure** + +### **7.1.1 Relationship to ITU-T Rec. F.510** + +ITU-T Rec. F.510 is a service definition for a White Pages service. However, it does not provide details on how this service should be realized. This Recommendation gives a directory specification for the support of ITU-T Rec. F.510. It is also able to cope with some future extensions to that Recommendation. + +### **7.1.2 Entry structure** + +See Figure 2 + +![Figure 2/F.515 – Structure of a directory entry. The diagram shows an 'Entry:' box containing a sequence of 'Directory attribute' boxes. The first three are individual boxes. The next two are grouped under a 'Family' label in a shaded box. The final two are grouped under another 'Family' label in a shaded box. Ellipses indicate more attributes. The label 'F.515_F02' is at the bottom right.](fbfa653853daf5541118a9ddecb92284_img.jpg) + +Figure 2/F.515 – Structure of a directory entry. The diagram shows an 'Entry:' box containing a sequence of 'Directory attribute' boxes. The first three are individual boxes. The next two are grouped under a 'Family' label in a shaded box. The final two are grouped under another 'Family' label in a shaded box. Ellipses indicate more attributes. The label 'F.515\_F02' is at the bottom right. + +**Figure 2/F.515 – Structure of a directory entry** + +A directory holds information about objects. The following object classes are defined: + +- a) subscriber, which is either a residential person, an organization or a governmental institution; +- b) state-or-province; +- c) locality; +- d) street; and +- e) business category. + +The collection of directory information about some object is called an entry. The entry model is only used for defining how directory information is transferred on the protocol. It does not imply anything about the internal structure of a database. + +The information within an entry is modelled as a set of directory attributes and families. + +Each directory attribute holds a particular piece of information, e.g., a street address. A particular directory attribute is an instance of a directory attribute type. A directory attribute type is a definition of the type of information held by an (instance) directory attribute of that type. + +There may be multiple directory attributes of the same type within an entry. As an example, a locality name may be provided in different languages. + +The content of directory entries corresponding the above different object classes is outlined in 7.2, 7.3 and 7.5. + +A family is a grouping of related directory attributes that describes one particular entity, for example, a communications address (see 7.5.2). A particular family is an instance of a family type. + +### 7.1.3 Directory attributes + +#### 7.1.3.1 Directory attribute type definitions + +A directory attribute type is defined as an XML element type. The element name is a global unique identification of the directory attribute type. A directory attribute is an instance of this element type where the content of the element is the value of the directory attribute. + +The following is a typical directory attribute specification: + +``` +Nakkedraget +``` + +The syntax of the value is part of the element definition. Some directory attribute type definitions include the `xml:lang` XML attribute, which indicates the language of the directory attribute value. As an example, a street address in Brussels have both a French and a Flemish variant. + +All the defined directory attribute types belong to the XML `Attribute` substitution group. This means that directory attributes can be inserted into the protocol whenever the `Attribute` abstract element is specified within the XML schema (as an example, see 7.1.4). + +#### 7.1.3.2 Grouping of directory attributes (families) + +It is possible to define different family types. A family type is defined as an XML element type. The element name is a global unique identification of the family type. A family element has as child elements one or more elements of the XML `Attribute` substitution group. + +A family type definition is member of the `Family` substitution group. + +The only currently defined member of the `Family` substitution group is the `commsAddress` family type. + +NOTE – Other family types could be defined in the future, e.g., for postal addresses. + +### 7.1.4 Entry representation + +Entry information, as it is transmitted in the protocol, is represented by the **entry** XML element (see Figure 3). + +![Diagram of the entry XML element structure. The 'entry' element has a dashed box containing a vertical bar, with a '1..∞' multiplicity below it. To the right of the dashed box, there are three boxes: 'Attribute', 'Family', and 'commAddr'. Arrows from the dashed box point to each of these three boxes. The 'commAddr' box has an arrow pointing to an 'Attribute' box. Below this 'Attribute' box is a dashed box containing a vertical bar, with a '1..∞' multiplicity below it. To the right of this second 'Attribute' box is the text 'F.515_F03'.](4ee27dbf5ef12e7b58b0ef0937bc5a5e_img.jpg) + +Diagram of the entry XML element structure. The 'entry' element has a dashed box containing a vertical bar, with a '1..∞' multiplicity below it. To the right of the dashed box, there are three boxes: 'Attribute', 'Family', and 'commAddr'. Arrows from the dashed box point to each of these three boxes. The 'commAddr' box has an arrow pointing to an 'Attribute' box. Below this 'Attribute' box is a dashed box containing a vertical bar, with a '1..∞' multiplicity below it. To the right of this second 'Attribute' box is the text 'F.515\_F03'. + +Figure 3/F.515 – Transmitted entry information + +A child element of the **entry** element is a choice between an element of the **Attribute** substitution group and an element of the **Family** substitution group. This choice can be taken one or more times as indicated in above figure. This allows such elements to be present in any number and order. + +The **entry** element has the following XML attributes: + +- The **entryIdentifier** XML attribute allows an entry identifier to be assigned to the entry (see 7.1.5). +- The **hierarchyLevel** XML attribute shall be present for any entry that is a member of a hierarchical group. This XML attribute shall take the value zero for the hierarchical top. +- The **hierarchyBelow** XML attribute indicates whether the entry has any hierarchical children. A value of **true** indicates that hierarchical children exist. If the value is **false** or this XML attribute is absent, no hierarchical children exist. +- The **ReturnedObject** XML attribute indicates the type of entry returned. It can take the value **subscriber** (default), **stateOrProvince**, **locality**, **street** or **businessCategory**. + +### 7.1.5 Entry identifiers + +Clause 7.5/F.510 introduces the concept of *entry identifiers*. An entry identifier may be supplied in a UDAP search request (see 10.5.1.2). An entry identifier may also be supplied for some or all of the entries returned in a search result. + +ITU-T Rec. F.510 mandates the support of entry identifiers, but support is optional according to this Recommendation. However, if an implementation is able to return identification information in the results, it shall also support identification information in the requests. + +Entry identifier can be used for accessing a particular entry directly. It can also be used to establish relationship among members of a hierarchical group. + +Some databases have external visible identifiers that can be used as permanent entry identifiers; others have not. Although a database may not provide for such permanent entry identifiers, the directory server may assign temporary entry identifiers to entries returned. The client can then within a locally defined period use an entry identifier to access a particular entry. + +The entry identifier is given as an XML attribute with syntax **xsd:base64Binary** allowing any bit string to be represented. + +This XML attribute is mandatory for the **base** child element of the **searchRequest** and it is optional for the **entry** element of the **searchResult**. + +If the client in a request specified an expired or unknown entry identifier, the directory server shall return an empty result with message code 0.1. + +### 7.1.6 Search filter + +#### 7.1.6.1 Filter concept + +A **filter** element is part of a search request. It consists of *filter items*. A filter item holds directory attribute information about a particular directory attribute type. The directory server compares this information against corresponding directory attribute(s) within each entry of the directory server. If the filter item matches just one directory attribute of the type, whether that directory attribute is a direct part of the entry or is contained within a family, the matching returns a value of TRUE. Otherwise; it returns a value of FALSE. If the combination of filter element results in a TRUE condition, the entry is a candidate to be returned in the search result. Otherwise, it is not. + +NOTE 1 – Although an entry is a candidate to be returned, local policies may prevent that entry from being returned. + +NOTE 2 – Currently, only AND'ing of filter element is supported, i.e., that all filter items have to return TRUE for the filter to return TRUE. The filter specification is made in such a way that extension to more complex and powerful filter specifications is possible. + +A filter item can hold a complete directory attribute value. This shall be the case for all enumerated directory attribute types. In this case the filter item is represented by an XML element of the **Assertion** substitution group. A filter item can also hold one or more, possible truncated words to be matched against a string-type directory attribute. In this case the filter item is represented by an XML element of the **Substrings** substitution group. + +#### 7.1.6.2 Filter structure + +A filter is carried in a search request in the **filter** XML element. Its structure is indicated in Figure 4. A child element of the **filter** element is a filter item that is defined as a choice between an element of the **Assertion** substitution group and an element of the **Substrings** substitution group. This choice can be taken one or more times. This allows multiple of such elements to be specified in any order + +![Diagram of the filter structure. A 'filter' element box is connected to a dashed box containing a vertical bar and a minus sign, with a '1..∞' multiplicity indicator below it. This dashed box is connected to a choice between two boxes: 'Assertion' and 'Substrings'. The label 'F.515_F04' is at the bottom right.](a7c51c18111139f9aca2805114108565_img.jpg) + +``` +graph LR; filter[filter] --> choice(( )); choice --> Assertion[Assertion]; choice --> Substrings[Substrings]; style choice fill:none,stroke-dasharray: 5 5 +``` + +Diagram of the filter structure. A 'filter' element box is connected to a dashed box containing a vertical bar and a minus sign, with a '1..∞' multiplicity indicator below it. This dashed box is connected to a choice between two boxes: 'Assertion' and 'Substrings'. The label 'F.515\_F04' is at the bottom right. + +Figure 4/F.515 – Filter structure + +#### 7.1.6.3 Directory assertion definitions + +A separate version of the directory attribute definition is used for defining elements of the **Assertion** substitution group. The suffix **As** is added to the element name of the corresponding basic directory attribute definition to separate it from that definition. + +The following is a typical assertion specification: + +``` +residential +``` + +In this example, the **subscriberTypeAs** is an enumerated that can take the value **residential**, **organization** or **government**. + +An **Assertion** type element has the **weight** XML attribute, which may affect the result of the evaluation (see 9.9). + +#### 7.1.6.4 Directory substrings/word matching definitions + +A separate version of the directory attribute definition is used for defining elements of the **Substrings** substitution group. The suffix **Sub** is added to the element name of the corresponding basic directory attribute definition used to distinguish it from that definition. + +Figure 5 illustrates the XML schema for an XML element of the **Substrings** substitution group. It consists of a global XML element with a name following the rules indicated above and it has as child elements one or more **value** elements. + +![Figure 5/F.515 – Typical substrings definition. The diagram shows a sequence of three elements: 'localitySub', a placeholder box with four small squares, and 'value'. The 'value' element is connected to a bracket labeled '1..∞', indicating one or more occurrences. The label 'F.515_F05' is next to the bracket.](1a827b10290f33d4fec04d0e8ef7a897_img.jpg) + +``` +graph LR; localitySub[localitySub] --- Placeholder[■ ■ ■ ■]; Placeholder --- value[value]; value --- Bracket["1..∞"]; Bracket --- F515_F05[F.515_F05]; +``` + +Figure 5/F.515 – Typical substrings definition. The diagram shows a sequence of three elements: 'localitySub', a placeholder box with four small squares, and 'value'. The 'value' element is connected to a bracket labeled '1..∞', indicating one or more occurrences. The label 'F.515\_F05' is next to the bracket. + +**Figure 5/F.515 – Typical substrings definition** + +The content of each **value** child element is a word, possibly truncated. + +The global element has two XML attributes: + +- The **weight** XML attribute has the same significance as for an element of the **Assertion** substitution group. +- The **string** attribute indicates the string matching rule as requested by the client (see 9.2.4/F.510). + +A **value** element has two XML attributes: + +- The **wordMatch** XML attribute indicates the word matching rule as requested by the client (see 9.2.3/F.510). +- The **characterMatch** XML attribute indicates the character matching rule as requested by the client (see 9.2.2/F.510). + +Details on matching rule specifications can be found in 9.2. + +### 7.1.7 Defined directory attribute types + +This Recommendation defines directory attribute types in 7.2 and 7.5 corresponding to the attribute types defined in ITU-T Rec. F.510. + +The directory attributes are listed in tables where the columns have the following meanings: + +- The Directory attribute type column gives the name of the directory attribute type, which is also the name of the XML element type representing the directory attribute type. +- The Syntax column gives the syntax used for a directory attribute value of the type. +- The Match type column indicates what type of filter item is used for the directory attribute type. 'As' means that the element of the **Assertion** substitution group has been defined for the directory attribute type. 'Sub' means that the element of the **Substrings** substitution group has been defined for the directory attribute type. '-' means that matching is not relevant for the attribute type. +- The Language attribute column indicates whether the **xml:lang** XML attribute has been included in the definition of the base directory attribute type. 'Y' means it has been included, 'N' means it has not been included and '-' means that a language code is not relevant. +- The Use column gives a little explanation on the use of the directory attribute type. + +## 7.2 Geography + +See Table 1. + +**Table 1/F.515 – Geography Locality directory attributes** + +| Directory attribute type | Syntax (XML type) | Match type | Language attribute | Use | +|--------------------------|-------------------|------------|--------------------|--------------------------------------------------------------| +| Country | NMTOKEN | As | – | Specifies the ISO 3166-1 code for the country. | +| Locality | Ub128 | Sub | Y | Specifies the name of a locality. | +| stateOrProvince | Ub128 | Sub | Y | Specifies the name of a state or province. | +| description | Ub1024 | – | Y | Specifies free-form text that describes the Returned Object. | +| localityCode | Ub64 | As | – | Specifies a common code to identify a geographical locality. | +| localityNDC | Num-ub16 | As | – | Specifies the value of a field within the ITU-T Rec. E.164. | + +Table 1/F.510 lists the directory attributes corresponding to the attribute types as they are defined in 7.2/F.510. The ITU-T Rec. F.510 attribute type NoSubscriberInformation is indirectly signalled by the `returnedObject` XML attribute of the `entry` element having a value different from `subscriber`. + +## 7.3 Business classification + +No specification beyond ITU-T Rec. F.510 is needed. + +## 7.4 Organizational structure + +No specification beyond ITU-T Rec. F.510 is needed. + +## 7.5 Subscriber entry + +See Table 2. + +**Table 2/F.515 – Subscriber directory attributes** + +| Directory attribute type | Syntax (XML type) | Match type | Language attribute | Use | +|--------------------------|-------------------|------------|--------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| streetAddress | Ub128 | Sub | Y | Specifies a site for the local distribution and physical delivery in a postal address, i.e., the street name, place, avenue, etc. | +| houseId | Ub64 | Sub | N | Specifies a linguistic construct used to identify a particular building, for example a house number or house name relative to a street, avenue, town or city, etc. | +| givenName | Ub64 | Sub | N | Specifies the linguistic construct which is normally given to an individual by the individual's parent, or is chosen by the individual, or by which the individual is commonly known. | + +**Table 2/F.515 – Subscriber directory attributes** + +| Directory attribute type | Syntax (XML type) | Match type | Language attribute | Use | +|--------------------------|-------------------|------------|--------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| title | Ub64 | Sub | Y | Specifies the designated position or function of the object within an organization. | +| description | Ub1024 | – | Y | Specifies free-form text that describes the returned subscriber entry. | +| businessCategory | Ub128 | Sub | Y | Specifies information concerning the occupation of some common objects, e.g., people. For example, this directory attribute provides the facility to interrogate the directory about people sharing the same occupation. | +| postalCode | Ub40 | Sub | – | Specifies the postal code of the named object. | +| postOfficeBox | Ub40 | Sub | – | Specifies the Post Office Box by which the object will receive physical postal delivery. | +| dmdName | Ub64 | Sub | | Specified a service provider name that has to be unique within a certain scope, e.g., within a country or Europe. | +| localityCode | Ub64 | As | – | Specifies a common code to identify a geographical locality. | +| nationalDestination Code | Num-ub16 | As | – | Specifies the value of a field within ITU-T Rec. E.164. | +| subscriberName | Ub128 | Sub | Y | Specifies the name by which the Subscriber chooses to be known in the directory. The subscriberName is the surname for a residential subscriber and organizational name for an organizational or governmental subscriber. | +| subscriberType (Note) | SubscTypes | As | – | Specifies the type of subscriber. It is defined as a multi-valued directory attribute type allowing a subscriber to be simultaneously of multiple types. | +| secondFamilyName | Ub128 | Sub | N | Specifies a residential subscriber when subscriberName is not sufficiently unique | +| profession | Ub128 | Sub | Y | Specifies the profession or job of an individual, business or department. | +| language | language | As | – | Specifies the preferred language in which the subscriber entry shall be displayed. | + +NOTE – The XML syntax is a list of enumerated values. It is therefore possible to add several values to a single XML attribute element. + +### 7.5.1 Subscriber types + +The `subscriberType` directory attribute gives the type of the subscriber. The different subscriber types are described in 7.5.1/F.510. The directory attribute syntax is the `SubscrTypes` simple XML type. It is a list item type derived from the enumerated `SubscrType` XML simple type. The `SubscrType` defines the following enumerated values: + +- `residential;` +- `organization;` +- `government.` + +It is possible to specify multiple subscriber types, e.g., in the case the same communications address is for both organizational and private use. + +### 7.5.2 Communications address + +See Table 3. + +**Table 3/F.515 – Communication address directory attributes** + +| Directory attribute type | Syntax (XML type) | Match type | Language attribute | Use | +|-----------------------------------|-------------------|------------|--------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| telephone | Ub32 | Sub | – | Specifies a telephone number associated with an object.
A directory attribute value for Telephone Number is a string that complies with the internationally agreed format for showing international telephone numbers, ITU-T Rec. E.123 (e.g., "+ 44 582 10101"). | +| orAddress | ub1024 | Sub | – | Specifies a text encoding of an X.400 O/R address, as specified in RFC 987. | +| mail | ub256 | Sub | – | Specifies an electronic mailbox directory attribute following the syntax specified in RFC 2822. | +| url | ub1024 | Sub | | Specifies a URL following the syntax specified in RFC 1738. | +| commService
(See item 1 below) | ComServices | As | – | It specifies the type of service(s) associated with a communications address. It describes the class of service that the Communications Address provides access to, for example, telephone (voice), facsimile, electronic mail, SMS (short messaging service), EDI, file transfer, etc. | +| commNetwork | ComNetwork | As | – | It specifies the type of network for which a communications address is used. It describes the type of network where the Communications Address is allocated. For example, a Public Switch Telephone Network (PSTN), an ISDN network, or a GSM mobile phone network. It could also be an application oriented network, e.g., a banking network. | + +**Table 3/F.515 – Communication address directory attributes** + +| Directory attribute type | Syntax (XML type) | Match type | Language attribute | Use | +|--------------------------|----------------------|------------|--------------------|--------------------------------------------------------------------------------------------------------------------------------------------------| +| addrValidFrom | dateTime | – | – | See item 2 below. | +| addrValidUntil | dateTime | – | – | See item 2 below. | +| addrValidity | AddrValidity Type | As | – | See item 2 below. | +| addrCoverage | AddrCoverage Type | As | – | The address coverage directory attribute type specifies area of validity of a communications address. | +| addrTariff | AddrTariff Type | As | – | The Address Tariff directory attribute type specifies what tariff is associated with a communications address. | +| addrRestriction | addrRestriction Type | As | – | The Address Restriction directory attribute type provides information about the restriction of access to the Communications Address information. | + +- 1) The XML syntax is a list of enumerated. It is therefore possible to add several values to a single XML attribute element. +- 2) If neither of the directory attribute types is present, the communications address is *valid* and *current*. + +If the `addrValidFrom` directory attribute is present, but the `addrValidUntil` directory attribute is absent, and if the current date is equal or greater than the date given in the `addrValidFrom` directory attribute, then the communications address is *current*; otherwise it is *future*. + +If the `addrValidUntil` attribute is present, but the `addrValidFrom` directory attribute is absent, and if the current date is less or equal than the date given in the `addrValidUntil` directory attribute, then the communications address is *current*; otherwise it is *old*. + +If both the `addrValidFrom` directory attribute and the `addrValidUntil` directory attribute are present, and if the current date is greater or equal to the date given the `addrValidFrom` directory attribute and is less or equal to the date given in the `addrValidUntil` directory attribute, then the communications address is *temporary*. If the current date is less than the date given in the `addrValidFrom` directory attribute, then the communications address is *new*. If the current date is greater than the date given in the `addrValidFrom` directory attribute, then the communications address is *old*. + +NOTE 1 – This assumes that the `addrValidFrom` value is less or equal to `addrValidUntil` value. If that is not the case, the specifications are invalid and useless. + +The `addrValidity` is used primarily in the case where a database does not support inclusion of validity dates of communications addresses. If `addrValidFrom` and `addrValidUntil` directory attribute types are supported, the `addrValidity` directory attribute, if present, should be derived from those two other directory attribute types. + +NOTE 2 – Even when validity dates are supported by the database, the `addrValidity` could be derived to support matching against this directory attribute type. + +#### 7.5.2.1 The Communications Service Type directory attribute + +The `ComServiceTypes` simple XML type is a list item type derived from the enumerated `ComServiceType` XML simple type. The `ComServiceType` defines the following enumerated values: + +- `voice;` +- `fax:` + - `textPhone;` + - `videoPhone;` + - `publicPhoneBox;` + - `switchBoard;` + - `pager;` +- `data;` +- `email;` +- `web.` + +#### 7.5.2.2 Network Types simple XML type + +The `NetworkType` type is an enumerated XML simple type. The following enumerated values are defined: + +- `pstn;` +- `isdn;` +- `gsm;` +- `umts;` +- `internet.` + +#### 7.5.2.3 Address Validity Types simple XML type + +The `AddrValidityType` is an enumerated XML simple type. The following enumerated values are defined: + +- `current` indicates that the communications address is currently valid; +- `old` indicates that the communication address has been taken out of service; +- `future` indicates that the communication address has not been put into service; and +- `temporary` indicates that the communications address is currently valid, but its lifetime is limited. + +#### 7.5.2.4 Coverage Type simple XML type + +The `addrCoverageType` is an enumerated XML simple type. The following enumerated values are defined: + +- `international` indicates that the communications address has unlimited coverage. +- `national` indicates that the communications address is valid within the country of the subscriber. +- `stateOrProvince` indicates that the communications address is valid at least within the area indicated by the `stateOrProvince` directory attribute of the subscriber entry. +- `locality` indicates that the communications address is valid at least within the area indicated by the `localityName` directory attribute of the subscriber entry. + +#### 7.5.2.5 Tariff simple XML type + +The `AddrTariffType` is an enumerated XML simple type. The following enumerated values are defined: + +- `normal` indicates that standard tariff applies for the calling the number. +- `premium` indicates that some special tariff is used when calling the number. +- `toll-free` indicates that in general the called part pays for a call. + +NOTE – New values may be introduced later. Implementations should be prepared for extensions. + +#### 7.5.2.6 Address Restriction simple XML type + +The `AddrRestrictionType` is an enumerated XML simple type. The following enumerated values are defined: + +- `public` indicates that the communications address can be revealed without restrictions; +- `secret` indicates that the communications address shall not be revealed; +- `call-screen`; +- `no-marketing`; +- `complete-only`. + +The `call-screen` value indicates that an operator can call the subscriber to ask if it wants to receive the call, and then allow call completion. The communications address is not revealed. + +The `no-marketing` value indicates that the communications address can be revealed, but it shall not be used for unsolicited marketing. + +The `complete-only` value indicates that the communications address shall not be revealed, but the operator may put a call through for the inquirer. + +NOTE – New values may be introduced later. Implementations should be prepared for extensions. + +# 8 Service specification + +No specification beyond ITU-T Rec. F.510 is needed. + +# 9 Common Protocol Specification element + +## 9.1 Character set, encoding and repertoire + +The ISO/IEC 10646-1 character set with the UTF-8 encoding shall be used for encoding the UDAP. + +The ISO/IEC 10646-1 character repertoire is constrained to the following subset of plane 00: + +- Row 00, positions 20-7E (BASIC LATIN); +- Row 00, positions A0-FF (LATIN-1 SUPPLEMENT); +- Row 01, positions 00-13, 16-2B, 2E-4D, 50-7E (part of LATIN EXTENDED-A). + +It is expected that the repertoire will be extended as directory services based on this Recommendation spread outside Europe. + +To the extent that this repertoire cannot be supported by an implementation, it has to be clearly stated. + +NOTE – Service providers may make mutual agreements that further limits the character repertoire. + +## 9.2 Matching rules + +Matching as defined in 9.2/F.510 shall be supported in the protocol as XML attributes. + +The character mapping rule requirements shall be carried in the `characterMatch` XML attribute with the following enumerated values: `exact`, `caseIgnore` (default) and `mapped`. This XML attribute is associated with the `value` child element of an element of the `Substrings` substitution group and applied to the word provided in the `value` element. + +The word matching rule requirements shall be carried in the `wordMatch` XML attribute with the following enumerated values: `exact` (default), `truncated` and `phonetic`. This XML attribute is associated with the `value` child element of an element of the `Substrings` substitution group and applied to the word provided in the `value` element. + +The string matching rule requirements shall be carried in the `string` XML attribute with the enumerated values: `exact`, `deletion`, `restrictedDeletion`, `permutation`, `permutationAndDeletion` and `providerDefined`. This XML attribute is associated with an element of the `Substrings` substring group and applies to all the words supplied in the `value` child elements. + +Only string type directory attributes types use the matching rules above. Other types of directory attribute types require, e.g., enumerated directory attribute types, use a simple equality match (see 7.1.3.2). + +## 9.3 Language differences + +A language specification can be associated with string directory attribute types. Tables 1 to 3 indicate what directory attribute types may include a language specification. The XML language attribute is used for specifying the language. This requires that the has to be imported and a reference to the namespace has to be included (see B.3). + +## 9.4 Ordering of entries within a response + +It is for further study. + +## 9.5 Hierarchical groups + +### 9.5.1 Hierarchical group concept + +Entries may have hierarchical relationship. *Hierarchical groups* form such relationships by forming a logical tree with a root called the *hierarchical top*. + +By referring to hierarchical relationships, it is possible in a search operation to retrieve information from entries matching a search filter, but also from other entries within the same hierarchical group. + +To describe navigation within a hierarchical group it is convenient to define terms for the relationships that a given entry has with other entries within the group. An entry that is below some other entry in a hierarchical group is a *hierarchical child* of that other entry. If it is just below that other entry, it is also an *immediate hierarchical child* of that other entry. An entry that is above some other entry is a *hierarchical parent* of that other entry. If it is just above this other entry, it is an *immediate hierarchical parent* of that other entry. An entry that has no hierarchical children is a *hierarchical leaf*. A hierarchical top has no hierarchical parents. + +### 9.5.2 Hierarchy selection + +The hierarchy selection is specified in the `hierSelect` XML attribute of the `searchRequest` element. It may take the values `self` (default), `children`, `parent`, `hierarchy`, `top`, `subtree` and `all` with the semantics as described in 9.5/F.510. + +The **hierSelect** XML attribute is a list XML attribute type allowing several hierarchy selections to be provided within the same XML attribute as illustrated by the following example: + +``` +... hierSelect="self top subtree" +``` + +### 9.5.3 Return of hierarchical information + +Two XML attributes of the **entry** element are related to hierarchical groups: + +- The **hierarchyLevel** XML attribute holds an integer that gives the level of the entry within the hierarchical group. The top entry has level 0, all immediate children to the hierarchical top have level 1, etc. This XML attribute shall be included for an entry in the response if the entry is member of a hierarchical group. Otherwise, it shall be absent. +- The **hierarchyBelow** XML attribute indicates whether the entry has one or more hierarchical children. If the entry has hierarchical children, this directory attribute shall be present with the value **true**. If the entry is a hierarchical leaf, it shall either be present with the value **false** or be absent. If the entry is not member of a hierarchical group, this XML attribute shall be absent. + +When entries of a hierarchical group are returned, the start and end of the hierarchical group is signalled by the **hierarchy** element as indicated by below example. + +``` + + + + + + + + + +``` + +The **hierarchy** element shall contain one or more **entry** child elements for entries members of the same hierarchical group. It has two XML attributes: + +- a) **entries**, which signals how many entries that are to be returned for that hierarchical group. It shall be present if paged results are returned; otherwise, it is optional. +- b) **entryNumber**, which signals the relative number of the first entry returned within the **hierarchy** element with respect to the first entry to be returned. The first entry is number one, which is also the default. It shall be present if page results are returned and the page starts in the middle of the returned hierarchical group. + +### 9.5.4 Sequential ordering of a hierarchical group + +When transmitting a hierarchical group, a sequential ordering rule is required. The sequential order of a hierarchical group comes from following all the strands of the hierarchical group as follows: + +- a) The top entry is the first entry in the sequence followed by the remaining entries within a complete strand going down from top to a hierarchical leaf. It is a local choice which strand to select as the first one. +- b) The next strand to be selected is one that has not previously been selected and which has the maximum number of entries common with the previous selected strand. If several strands are identical in that respect, selection is a local matter. Only those entries not previously included are included in the sequence. +- c) The procedure in b) is repeated until all strands have been included. + +The rule above also applies when only part of a hierarchical is returned, except it may leave holes in the sequence. If the hierarchical top is returned, it will always be the first one returned. Otherwise, the first one returned is dependent on which strand is selected as the initial one according to the algorithm above. + +## 9.6 Number of entries + +The client can specify a limit for the maximum number of entries to be returned. It is specified as a positive integer in the `entryLimit` XML attribute of the `searchRequest`. + +## 9.7 Entry count + +The entry count of the response is given in the `entryCount` element of the `searchResult`. The `entryCount` element is an empty element with two XML attributes: + +- `count`, which provides the returned entry count; and +- `qualifier`, with the choice of the two values, where the `exact` choice shall be taken if the directory server can supply an exact count; otherwise, the `bestEstimate` choice shall be taken. + +## 9.8 Alternative directory attribute values + +This concept is not reflected in the protocol. For further explanation, see 9.8/F.510. + +## 9.9 Weighted directory attribute values + +Each `Assertion` type element and each `Substrings` type element has a `weight` XML attribute that can take the values `high` (default) and `low` corresponding to `AttributeWeight` one and zero as defined in 9.9/F.510. + +## 9.10 Geographical extensions + +The search controls for geographical extensions are defined in 9.10/F.510. The Extended Area control is provided as the `extendedArea` XML attribute of the `searchRequest` element. The Include All Areas control is provided as the `includeAllAreas` value of the `searchOptions` XML attribute of the `searchRequest` element. + +## 9.11 Ignore if absent + +This concept is not reflected in the protocol. For further explanation, see 9.11/F.510. + +## 9.12 Paged Results request + +``` + + + + + +``` + +The `PagedResults` complex XML type allows the client to ask for piecewise return of the result. In the initial request, the client specifies page size and a possible page number, if the client requires a page different from the first page. To retrieve more pages, the client has to issue a new request for each page to be returned. + +A directory server may operate in two different modes: + +- a) The server may keep status information about the operation progress. To operate in this mode, the server shall return a query reference to be used in subsequent request. In subsequent requests, the requester only needs to send the query reference to retrieve the next page. +- b) The server keeps no status information and returns no query reference. The requestor then has to explicitly specify a page number to get anything but the first page returned. + +A directory server may also operate in a hybrid mode, supporting both of the above modes. + +NOTE – This may occur when the directory server actually consists of several identical servers for load sharing. If a subsequent request gets to a server that has previously served this operation, then the query reference would be usable; otherwise, the page number will be used. For this to work properly, the client has to include both **pageNumber** and **queryReference**. + +The **pageSize** XML attribute indicates the number of entries that shall be returned by each paged result. This is a mandatory XML attribute when paged results are requested. + +The **pageNumber** XML attribute indicates the number of the page to be returned. If this XML attribute is absent, then: + +- the first page shall be returned if a **queryReference** XML attribute is not supplied or if the directory server does not support the **queryReference**; +- the next page shall be returned number if the **queryReference** is present and supported. + +The **queryReference** shall not be present in the initial request. However, the directory server may in the reply to a paged result request return a **queryReference** that may be used in subsequent requests. + +# 10 The Unified Directory Access Protocol (UDAP) + +## 10.1 Service model + +No specification beyond ITU-T Rec. F.510 is needed. + +## 10.2 Search controls and indications + +The **SearchType** search control from ITU-T Rec. F.510 is provided by the **searchType** XML element of the SOAP **Header** element (see C.1.1). + +The **UserClass** search control from ITU-T Rec. F.510 is provided by the **userClass** XML element of the SOAP **Header** element (see C.1.1). + +The **EntryLimit** search control from ITU-T Rec. F.510 is provided by the **entryLimit** XML attribute of the **searchRequest** (see 9.6). + +The **ExtendedArea** search control from ITU-T Rec. F.510 is provided by the **extendedArea** XML attribute of the **searchRequest** (see 9.10). + +The **IncludeAllAreas** and **PerformExactly** search controls from ITU-T Rec. F.510 are provided by the **searchOption** XML list attribute allowing the values **includeAllAreas** (see 9.10) and **performExactly**. + +The **EntryCount** respond indicator from ITU-T Rec. F.510 is provided by the **entryCount** child element of the **searchResult**. + +## 10.3 Predicates + +No specification beyond ITU-T Rec. F.510 is needed. + +## 10.4 Common query conditions + +The common query search controls and result indicators as listed in Table 6/F.510 are specified in 10.2. + +## 10.5 Basic service offering + +### 10.5.1 Search request + +``` + + + + + + + + +``` + +#### 10.5.1.1 XML attributes of the `searchRequest` element + +For the `hierSelect` XML attribute, see 9.5.2. For the `searchOptions` XML attributes, see 10.2. + +#### 10.5.1.2 XML child elements of the `searchRequest` element + +The `base` is an empty element having a single `entryIdentifier` mandatory XML attribute. This element is mandatory for the `subscriberByEntryIdentifier` search-type. It is not relevant for other search-types. + +The `filter` element is described in 7.1.6. + +![Diagram of information selection structure](356eb99ab9489bbd647223390a913903_img.jpg) + +The diagram illustrates the structure of the `infoSelect` element. It starts with a box labeled `infoSelect` connected to a hexagonal symbol containing three dots. This is followed by a box with a minus sign (`-`). A line branches from this box to three dashed rectangular boxes: `AttrTypes`, `Family`, and `commsAddress`. Below the `Family` box is the cardinality `0..∞`. Below the `commsAddress` box is the cardinality `0..∞`. To the right of the `commsAddress` box is the label `F.515_F06`. + +Diagram of information selection structure + +Figure 6/F.515 – Information selection + +The `infoSelect` element allows a client to select what information (families and directory attribute types) is to be returned from each entry. + +Only directory attributes of the types specified in the `attrTypes` element may be returned independent on whether they are in the main part of the entry or part of a family. Directory attributes of types not listed should not be returned. Not all directory attributes requested by the client may be returned, as some of the requested directory attributes may not be considered part of the service. + +The client can also limit the families to be returned. This is done using one or more elements of the `Family` substitution group holding one or more directory attributes. Such an element is compared with all the families of the entry of that type. If the family holds the same attribute types as included in the element of the request and the value matches, then the family shall be returned; otherwise it should not. If an attribute in the element of the request is empty, then the attribute matches independently of the value of the stored directory attribute. + +If the directory server does not support the `infoSelect` element or part of it, it shall be ignored what is not supported and the request shall be processed accordingly. + +The `pagedResults` element is of type `PagedResults` as described in 9.2. + +### 10.5.2 Search result + +#### 10.5.2.1 Search result schema + +``` + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +``` + +#### 10.5.2.2 XML attributes of the searchResult element + +The value of the **msgCode** XML attribute shall be used for a message code value as defined by ITU-T Rec. F.510. It is mandatory in an ITU-T Rec. F.510 environment for situations where a message code is defined. Otherwise, it is optional (see Annex A). + +For the **queryReference** XML attribute, see 9.12. + +#### 10.5.2.3 Information element + +The **information** element shall contain one or more **entry** (see 7.1.2) and/or **hierarchy** (see 9.5.3) child elements. If no information is returned in the **searchResult**, the **information** element shall be absent. + +#### 10.5.2.4 Notifications element + +This element shall contain one or more notification attributes. The conditions for return of notifications attributes are specified in Annex A. + +## 10.6 Enhanced service offering + +The same XML specification applies for both basic and advanced service offering (see 10.5). + +# 11 Operational issues + +## 11.1 Security + +This Recommendation does not specify any features. It is assumed that security will be provided by an underlying service. + +# 12 Charging and accounting + +This Recommendation does not have any special provision for charging and accounting. + +# 13 Quality of Service + +This Recommendation does not have any special provision for Quality of Service. + +# Annex A + +## Message codes and notifications + +## A.1 Notification attribute types + +Notification attributes are carried in the `notifications` element of the `searchResult`. They provide information to the user that can help rectify problems in a `searchRequest`. + +| Notification attribute type | Syntax (XML type) | Use | +|-----------------------------|---------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| limitProblem | NCName |

The limitProblem notification attribute shall be included if the directory server encounters some limit problem. It is defined as an enumerated. It can take the following values:

  • – The adminLimit value is used if the directory server has reached some limit set by an administrative authority.
  • – The permanentRestriction is used if an operation has caused the directory server to exceed some limit that causes the process to stop and a repeated operation is judged to encounter the same problem.
  • – The temporaryRestriction is used if an operation has caused the directory server to exceed some limit that causes the process to stop, but the reason is judged to be a temporary problem, e.g., resources depletion.
| +| serviceProblem | NCName | It describes problems relating to the service provider policy for the service in question. | + +| Notification attribute type | Syntax (XML type) | Use | +|-----------------------------|---------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| searchType | SearchType | It gives the search-type of the failing search. | +| attributeTypeList | AttributeTypes | It gives a list of directory attribute types related to a specific return code. | +| filterNot | Filter | It gives a filter specification (e.g., see message code 6.3 in A.1). | +| filterItem | complex | It gives a list of invalid filter items in a search filter. | +| providerName | ProviderId | | +| hierarchySelectList | HierarchySelections | It gives a list identifying one or more hierarchy selection options. When a value is set, it indicates that the corresponding hierarchy selection is invalid. Either a forbidden or unsupported selection has been requested, or the selection has not been requested when it is required. | +| searchControlOptionsList | SearchOptions | It gives a list identifying one or more search control options. When a value is set, it indicates that the corresponding search control option selection is invalid. Either a forbidden or unsupported option has been requested, or the option has not been requested when it is required. | +| attributeCombination | Filter | It gives a list of attribute combinations that were missing in a search filter. | +| wordRestriction | complex | It specifies word matching features not supported by the service provider and/or it specified minimum string length requirements. | +| notSupported | Options | It gives a list of optional functions not supported by the service provider. The directory server has made a best effort. This notification attribute shall be returned whenever one or more known facilities are not supported. | + +## A.2 Message codes + +| F.510 Service Definition Message code | Description | Associated notifications | +|-------------------------------------------|------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------| +| 0 Unified directory specific codes | | | +| 0.1 Invalid entry identifier | The search cannot be performed as the entry identifier is not recognized because it is either invalid or expired. | None. | +| 0.2 Invalid query reference | The search cannot be performed as the entry query reference is not recognized because it is either invalid or expired. | None. | +| 1 Unavailable service access | | | +| 1.1 System congestion | The search cannot be performed, as the directory server is congested. | Optionally, a providerName notification attribute may be returned. | + +| F.510 Service Definition Message code | Description | Associated notifications | +|-----------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| 1.2 Destination database not accessible | The search cannot be performed, as system that should perform the search cannot be reached. | Optionally, a providerName notification attribute may be returned. | +| 1.3 Destination database not accessible | The search cannot be performed as the data on which the search should be performed cannot be reached. | Optionally, a providerName notification attribute may be returned. | +| 2 Required service not supported by the service provider | | | +| 2.1 Search for service is not supported | The service provider does not support the requested search-type. | A searchType notification attribute having the search-type as the value shall be returned. | +| 2.2 Search for service supported but not bilaterally agreed | The service provider supports the requested search-type, but the search-type has not been bilaterally agreed. | A searchType notification attribute having the search-type as the value shall be returned. | +| 3 Required functionality not supported by the service provider | | | +| 3.1 Matching rule not supported by service provider for the whole service | A requested matching rule is or certain aspects of a matching rule are unknown to the service provider, as it is not supported by the directory server. Either performedExactly is specified in the request or no alternative matching has been defined by the directory server. | When an implementation does not support certain aspects of a matching, the following is returned:
– a wordRestriction notification attribute having as value a specification of the matching rules specified in the request that are not supported. An directory attribute type shall not be specified. | +| 3.2 Matching rule not supported by Service Provider for this specific | The requested matching rule and assertion with the specified directory attribute type are recognized and supported by the service provider, but as a policy, the service provider do not allow the requested matching in the particular situation. In addition, the performExactly is set or the service provider has not provided for alternative matching. | In case of an unsupported matching rule option for a particular directory attribute type, the following is returned:
– one or more wordRestriction notification attributes each indicating an directory attribute type and a specification of the matching rule(s) specified in the request that are not supported for the directory attribute type. | +| 3.3 Inappropriate matching rule for | The requested matching rule is recognized and supported by the service provider but is not appropriate for the particular directory attribute type. However, this situation will result in an invalid XML document (see 0.2). | None. | + +| F.510 Service Definition
Message code | Description | Associated notifications | +|------------------------------------------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| 3.4 Invalid combination of matching rules for | The specified combination of matching rules (String matching rule, word matching rule and character matching rule) is not supported by the service provider as expressed by applicable search-rule. In addition, the performExactly is set or the service provider has not provided for alternative matching. | In case of an unsupported combination of matching rule options for a particular directory attribute type, the following is returned:
– one or more wordRestriction notification attributes each indicating a directory attribute type and the invalid combination. | +| 3.5 Geographical extension not supported | The geographical extension functionality is not supported by the directory server. | None. | +| 3.6 Level of geographical extension not supported | The requested level of geographical extension functionality is not supported by the directory server. | None. | +| 3.7 Service relative to hierarchical group with the specified HierarchySelection value not supported | The requested hierarchy selection value is not supported by the directory server. | UDS recognizes three situations:
1) Hierarchy selection specification (except for self ) is not allowed for the type of service. If then included, the request is rejected with a serviceProblem notification attribute with the value hierSelectForbidden .
2) If one or more hierarchy selections are forbidden by the service provider, then the request is rejected with a serviceProblem notification attribute with the value hierSelectNotAvailableForService together a hierarchySelectList notification attribute indicating the invalid settings.
3) If one or more hierarchy selection options are not supported by the implementation and those options are not handled by the cases above, the request is rejected with a serviceProblem notification attribute with the value hierSelectNotSupported together a hierarchySelectList notification attribute indicating the settings in the requests that are unsupported. | + +| F.510 Service Definition
Message code | Description | Associated notifications | +|-----------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| 3.8 Filter on for service not supported | The search cannot be performed by the directory server, as one or more of the directory attribute types in the filter cannot be used for the given search (e.g., filtering by street address for a country wide search not supported by service provider). | The directory server will return:
  • – an attributeTypeList notification attribute having as values the illegal directory attribute types;
  • – a searchType notification attribute with the search-type as value.
| +| 3.9 Filter on for service not supported | The search cannot be performed by the directory server, as a value of one or more directory attribute types in the filter cannot be used for the given search. | The directory server shall return:
  • – a filterItem notification attribute with the failing filter items as value; and
  • – a searchType notification attribute with the search-type as value.
| +| 3.10 required by service provider to perform the Search | Equivalent to 4.1. | Equivalent to 4.1. | +| 3.11 The value of the parameter is not bilaterally agreed | The search cannot be performed as the directory server does not support the specified userClass . | None. | +| 4 Wrong input information | | | +| 4.1 Mandatory are missing for requested service | The search cannot be performed as one or more mandatory directory attributes are missing. | The directory server shall return:
  • – an attributeTypeList notification attribute having as values the missing directory attribute types;
  • – a searchType notification attribute with the search-type as value.
| +| 4.2 Minimum size of truncated word is required for | The search cannot be performed as the minimum number of characters for word truncation to be applied on the specified directory attribute is not respected. | The directory server shall return:
  • – one or more a wordRestriction notification attributes indicating the directory attribute type and what the minimum string length are for word matching for that directory attribute type. Two values are given, one for the first word in the sequence and one for the following words;
  • – a searchType notification attribute with the search-type as value.
| + +| F.510 Service Definition
Message code | Description | Associated notifications | +|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| 4.3 Mandatory combination of are missing for requested service | The search cannot be performed, as a required combination of directory attribute types is missing. As an example, a service provider may require that either givenName or businessCategory is specified in a search request, but the user has supplied neither. | The directory server shall return:
– an attributeCombinations notification attribute identifying the missing combination(s);
– a searchType notification attribute with the search-type as value. | +| 4.4 Invalid value for because of incorrect syntax (e.g., no figures for a number) | The search cannot be performed as the syntax used for one of the directory attribute is not valid. The directory machinery is not able to process the request. | Covered by 0.2. | +| 4.5 Unrecognized value for or
Unrecognized value for <Attribute_Type> or <Common_Query_Parameter> at the application level (e.g., GroupSelection parameter, DeviceTariff attribute, DeviceService attribute, DeviceAdressType attribute) | The search cannot be performed as the value provided for a specified parameter does not fit with one value from the set of predefined F.510 values. | Covered by 0.2. | +| 5 Undetermined geographical area | | | +| 5.1 The parameter does not exist | The search cannot be performed as the value provided for a key directory attributes does not exist. | The directory server will return a filterItem notification attribute having the failing filter item as value. | +| 5.2 The parameter is not unique | The search cannot be performed as the value provided for a key directory attribute is ambiguous. | None. | +| 5.3 The combination of parameters , , , etc., does not exist | The search cannot be performed as values provided for each geographical locality are in conflict (e.g., the locality does not exist under the specified Province). | The directory server will return a filterItem notification attribute having the failing filter items as values. | + +| F.510 Service Definition
Message code | Description | Associated notifications | +|-----------------------------------------------------------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| 6 No entries returned | | If relevant, a notSupported notification attribute may be returned specifying unsupported facilities. | +| 6.1 No subscriber entries are available under geographical area | No entries are provided due to the fact that for the geographical area in which the search should be performed, the requested service provider does not have any entries in its databases (e.g., it corresponds for instance to an area in the US where independent companies are operating and no electronic directory is available). | An empty result is returned. | +| 6.2 No subscriber entries have been found for the requested HierarchySelection value | No entries are provided. Entries would be returned if the HierarchySelection parameter value is relaxed. | An empty result is returned. | +| 6.3 No subscriber entries have been found but entries exist without constraint for | No entries are provided. Entries would be returned if the constraint on for the given directory attribute type is relaxed. | A result is returned without the information element , optionally including a filterNot notification attribute with a filter specification consisting of the filter supplied in the request with those filter items that that succeeded in matching some entries omitted. | +| 6.4 No entries have been found but entries exist without constraint | No entries are provided. Entries would be returned if the constraint on is relaxed. | As for 6.3. | +| 6.5 Too many entries selected: more selective information is necessary | The search cannot be performed as the scope of it is considered to be too large by the service provider (according to local knowledge). More information is requested from the user to allow the search to be applied. | Same as for 4.1. | +| 6.6 No entry matched the search filter | The search was performed, but no entry matched the filter. | None. | +| 6.7 The requested page is not available | The search was performed. However, the client asked for a page number higher than the number of pages returned by the search. | None. | + +| F.510 Service Definition
Message code | Description | Associated notifications | +|-------------------------------------------------------------|-----------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| 7 List of non-subscriber attributes | | If relevant, a notSupported notification attribute may be returned specifying unsupported facilities. | +| 7.1 Incomplete list of , more available | An administrative limit has been reached. | The directory server returns a limitProblem notification attribute with the value adminLimit if an administrator-imposed limit has been reached, and with permanentRestriction if a permanent implementation restriction has been reached. | +| 7.2 Incomplete list of , no more available | An administrative limit has been reached. | As for 7.1. | +| 8 Entries returned | | If relevant, a notSupported notification attribute may be returned specifying unsupported facilities. | +| 8.1 Entries found | Either all found entries are returned or the last entries of paged results are returned | None. | +| 8.2 Entries found, more entries available | A page-full of entries are returned. There are still more entries available. | None. | + +# Annex B + +## XML schema for UDAP + +## B.1 Search request + +``` + + + + + + + + + + + + + Unified Directory Search Request + + + + + + + + + + + +``` + +``` + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +``` + +## B.2 Search result + +``` + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +``` + +``` + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +``` + +## B.3 Attribute Types + +``` + + + + + + +``` + +``` + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +``` + +``` + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +``` + +``` + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +``` + +``` + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +``` + +``` + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +``` + +``` + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +``` + +``` + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +``` + +``` + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +``` + +``` + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +``` + +``` + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +``` + +``` + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +``` + +``` + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +``` + +# Annex C + +## Use of Web Service facilities + +## C.1 Use of Simple Object Access Protocol (SOAP) + +### C.1.1 Use of SOAP Header element + +The SOAP **Header** element for a search request shall hold the following header elements: + +- **requestRef** – This element allows a unique identifier to be assigned to a search request. The same value shall be returned in the returned result to be associated it with the corresponding request. An implementation shall ensure that it does not issue a request with a **requestRef** with the same value as an outstanding request. It is recommended not to + +re-use **requestRef** values for a period that is long compared to the maximum expected response time. + +- **searchType** – The SearchType search control from ITU-T Rec. F.510 is provided by this element. It has the XML **SearchType** simple type as syntax. This is an enumerated syntax with the following values listed in the same sequence as the corresponding SearchTypes as defined in Tables 8/F.510 to 11/F.510 and in Tables 13/F.510 to 18/F.510: + - **searchForStateOrProvince;** + - **searchForLocality;** + - **subscriberWithinLocality;** + - **subscriberByEntryIdentifier;** + - **subscriberWithinStateOrProvince;** + - **subscriberWithinCountry;** + - **searchForStreetAddress;** + - **subscriberByStreetAddress;** + - **subscriberByCommunicationsAddress;** and + - **subscriberByBusinessCategory.** +- **userClass** – The UserClass search control from ITU-T Rec. F.510 is provided by this element. It can take the values: + - **operator** (default); and + - **publicUser.** + +The SOAP **Header** element for a search result shall hold the following header element: + +- **requestRef** – This element shall take same value of as the **requestRef** element of the corresponding search request. + +### C.1.2 Use of SOAP Fault element + +The **faultcode** can only take the value **client** when reporting UDS problem. Use of **VersionMismatch** is outside the scope of this specification. + +The following situations shall be reported specifying the **client** error code in the **faultcode** SOAP element: + +- The document in the request was not well formed, in which case: + - the **faultstring** SOAP element shall have the content **notWellFormed**; and + - the **detail** SOAP element shall hold the detailed entry **explanation**, which can contain additional information in free text format, e.g., some diagnosis returned from an XML parser. +- The document in the request was well formed but did not comply with the XML schema specified in the request, in which case: + - the **faultstring** SOAP element shall have the content **invalidDocument**; and + - the **detail** SOAP element shall hold the detailed entry **explanation**, which can contain additional information in free text format, e.g., some diagnosis returned from an XML parser. +- The document in the request was well formed, but the XML schema specified in the request was unknown to the server, in which case: + - the **faultstring** SOAP element shall have as content the string **unknownSchema**; and + - the **detail** SOAP element shall be absent. + +- The `requestRef` of the header of the request had the same value as an outstanding request from the same client, in which case: + - the `faultstring` SOAP element shall have as content the string `duplicateRequestRef`; and + - the `detail` SOAP element shall be absent. + +# Annex D + +## ASN.1 version of UDAP + +This annex provides a specification of the UDS protocol using ASN.1. The XML documents produced by use of this specification with the ASN.1 XML encoding rules (defined in ITU-T Rec. X.693) are identical to those defined by the XSD specification in the body of this Recommendation. For compact binary encodings of this protocol this annex (together with the ASN.1 Packed Encoding Rules) provides the definitive specification of the protocol. + +Use of this specification enables ASN.1 tools to be used for UDS implementation and also enables relaying between XML encodings of UDS messages, and compact binary encodings of those messages with no loss of information. + +``` + +Uds DEFINITIONS AUTOMATIC TAGS ::= +BEGIN + +IMPORTS + Decimal, DateTime, Language, NMTOKEN, NCName + FROM XSD /* The XSD module is defined in the ASN.1 standards. */ + + Lang + FROM Xml; + +PagedResults ::= SEQUENCE { + pageSize INTEGER(1..MAX) OPTIONAL, + pageNumber INTEGER(1..MAX) OPTIONAL, + queryReference OCTET STRING OPTIONAL} + +-- Unified Directory Search Request +SearchRequest ::= SEQUENCE { + entryLimit INTEGER(1..MAX) OPTIONAL, + extendedArea INTEGER(0..MAX) DEFAULT 0, + searchOptions SearchOptions OPTIONAL, + hierSelect HierarchySelections DEFAULT {self}, + base SEQUENCE {entryIdentifier OCTET STRING OPTIONAL} OPTIONAL, + filter Filter OPTIONAL, + infoSelect + SEQUENCE {attrTypes AttributeTypes OPTIONAL, + family-list SEQUENCE OF Family} OPTIONAL, + pagedResults PagedResults OPTIONAL} + +ReqHead ::= SEQUENCE { + requestRef INTEGER, + searchType SearchType, + userClass ENUMERATED {operator, publicUser} DEFAULT operator} + +ProviderId ::= SEQUENCE {country Country, + dmdName DmdName OPTIONAL} + +NumericString-1 ::= IA5String(FROM ("0".."9")) (PATTERN "[0-9]") + +``` + +``` + +Ub16NumericString ::= IA5String(FROM ("0".."9")) (SIZE (0..16)) (PATTERN "[0-9]") + +Ub32 ::= UTF8String(SIZE (0..32)) + +Ub40 ::= UTF8String(SIZE (0..40)) + +Ub64 ::= UTF8String(SIZE (0..64)) + +Ub128 ::= UTF8String(SIZE (0..128)) + +Ub256 ::= UTF8String(SIZE (0..256)) + +Ub512 ::= UTF8String(SIZE (0..512)) + +Ub1024 ::= UTF8String(SIZE (0..1024)) + +AttributeType ::= ENUMERATED { + country, locality, stateOrProvince, streetAddress, houseId, givenName, + title, description, businessCategory, postalCode, postOfficeBox, dmdName, + localityCode, localityNDC, subscriberName, subscriberType, secondFamilyName, + profession, language, telephone, orAddress, mail, url, commService, + commNetwork, addrValidFrom, addrValidUntil, addrValidity, addrCoverage, + addrTariff, addrRestriction} + +AttributeTypes ::= SEQUENCE OF AttributeType + +SearchType ::= ENUMERATED { + searchForStateOrProvince, searchForLocality, subscriberWithinLocality, + subscriberByEntryIdentifier, subscriberWithinStateOrProvince, + subscriberWithinCountry, searchForStreetAddress, subscriberByStreetAddress, + subscriberByCommunicationsAddress, subscriberByBusinessCategory} + +HierarchySelection ::= ENUMERATED { + self, children, parent, hierarchy, top, subtree, all} + +HierarchySelections ::= SEQUENCE OF HierarchySelection + +SearchOption ::= ENUMERATED {performExactly, includeAllAreas} + +SearchOptions ::= SEQUENCE OF SearchOption + +Filter ::= + SEQUENCE (SIZE (1..MAX)) OF + CHOICE {assertion Assertion, + substrings Substrings} + +Attribute ::= CHOICE { + country Country, + locality Locality, + stateOrProvince StateOrProvince, + streetAddress StreetAddress, + houseId HouseId, + givenName GivenName, + title Title, + description Description, + businessCategory BusinessCategory, + postalCode PostalCode, + postOfficeBox PostOfficeBox, + dmdName DmdName, + localityCode LocalityCode, + localityNDC LocalityNDC, + subscriberName SubscriberName, + subscriberType SubscriberType, + secondFamilyName SecondFamilyName, +} + +``` + +``` + +profession Profession, +language Language, +telephone Telephone, +orAddress OrAddress, +mail Mail, +url Url, +commService CommService, +commNetwork CommNetwork, +addrValidFrom AddrValidFrom, +addrValidUntil AddrValidUntil, +addrValidity AddrValidity, +addrCoverage AddrCoverage, +addrTariff AddrTariff, +addrRestriction AddrRestriction} + +Family ::= CHOICE {commsAddress CommsAddress} + +Assertion ::= CHOICE { + countryAs CountryAs, + localityCodeAs LocalityCodeAs, + localityNDCAs LocalityNDCAs, + subscriberTypeAs SubscriberTypeAs, + languageAs LanguageAs, + commServiceAs CommServiceAs, + addrValidityAs AddrValidityAs, + addrCoverageAs AddrCoverageAs, + addrTariffAs AddrTariffAs, + addrRestrictionAs AddrRestrictionAs} + +AssertionAttr ::= SEQUENCE {weight ENUMERATED {low, high} DEFAULT high} + +Substrings ::= CHOICE { + localitySub LocalitySub, + stateOrProvinceSub StateOrProvinceSub, + streetAddressSub StreetAddressSub, + houseIdSub HouseIdSub, + givenNameSub GivenNameSub, + titleSub TitleSub, + businessCategorySub BusinessCategorySub, + postalCodeSub PostalCodeSub, + postOfficeBoxSub PostOfficeBoxSub, + cmdNameSub DmdNameSub, + subscriberNameSub SubscriberNameSub, + secondFamilyNameSub SecondFamilyNameSub, + professionSub ProfessionSub, + telephoneSub TelephoneSub, + orAddressSub OrAddressSub, + mailSub MailSub, + urlSub UrlSub} + +SubstringAttr ::= SEQUENCE { + string + ENUMERATED {exact, deletion, restrDeletion, permutation, + permutationAndDeletion, providerDefined} DEFAULT exact, + weight ENUMERATED {low, high} DEFAULT high} + +SubstringValueAttr ::= SEQUENCE { + wordMatch ENUMERATED {exact, truncated, phonetic} DEFAULT exact, + characterMatch ENUMERATED {exact, caseIgnore, mapped} DEFAULT caseIgnore} + +CommsAddress ::= SEQUENCE { + attribute-list SEQUENCE (SIZE (1..MAX)) OF attribute Attribute} + +Country ::= NMTOKEN + +``` + +``` +CountryAs ::= SEQUENCE {COMPONENTS OF AssertionAttr, + base NMTOKEN} + +Locality ::= Ub128 + +LocalitySub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value-list SEQUENCE (SIZE (1..MAX)) OF value Ub128} + +StateOrProvince ::= Ub128 + +StateOrProvinceSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value-list + SEQUENCE (SIZE (1..MAX)) OF value + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub128}} + +StreetAddress ::= Ub128 + +StreetAddressSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value-list + SEQUENCE (SIZE (1..MAX)) OF value + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub128}} + +HouseId ::= Ub64 + +HouseIdSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value-list + SEQUENCE (SIZE (1..MAX)) OF value + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub64}} + +GivenName ::= Ub64 + +GivenNameSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value-list + SEQUENCE (SIZE (1..MAX)) OF value + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub64}} + +Title ::= SEQUENCE {lang Lang OPTIONAL, + base Ub64} + +TitleSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value-list + SEQUENCE (SIZE (1..MAX)) OF value + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub64}} + +Description ::= SEQUENCE {lang Lang OPTIONAL, + base Ub1024} + +BusinessCategory ::= SEQUENCE {lang Lang OPTIONAL, + base Ub128} + +BusinessCategorySub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, +``` + +``` + +value-list + SEQUENCE (SIZE (1..MAX)) OF value + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub128}} + +PostalCode ::= Ub40 + +PostalCodeSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value-list + SEQUENCE (SIZE (1..MAX)) OF value + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub40}} + +PostOfficeBox ::= Ub40 + +PostOfficeBoxSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value-list + SEQUENCE (SIZE (1..MAX)) OF value + SEQUENCE {COMPONENTS OF SubstringValueAttr}} + +DmdName ::= Ub64 + +DmdNameSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value-list + SEQUENCE (SIZE (1..MAX)) OF value + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub64}} + +LocalityCode ::= Ub64 + +LocalityCodeAs ::= SEQUENCE {COMPONENTS OF AssertionAttr, + base Ub64} + +LocalityNDC ::= Ub16NumericString + +LocalityNDCAs ::= SEQUENCE { + COMPONENTS OF AssertionAttr, + base Ub16NumericString} + +SubscriberName ::= SEQUENCE {lang Lang OPTIONAL, + base Ub64} + +SubscriberNameSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value-list + SEQUENCE (SIZE (1..MAX)) OF value + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub64}} + +SubscrType ::= ENUMERATED {residential, organization, government} + +SubscrTypes ::= SEQUENCE OF SubscrType + +SubscriberType ::= SubscrTypes + +SubscriberTypeAs ::= SEQUENCE {COMPONENTS OF AssertionAttr, + base SubscrType} + +SecondFamilyName ::= Ub128 + +SecondFamilyNameSub ::= SEQUENCE { + +``` + +``` + +COMPONENTS OF SubstringAttr, +value-list + SEQUENCE (SIZE (1..MAX)) OF value + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub128}} + +Profession ::= SEQUENCE {lang Lang OPTIONAL, + base Ub128} + +ProfessionSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value-list + SEQUENCE (SIZE (1..MAX)) OF value + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub128}} + +Language ::= XSD.Language + +LanguageAs ::= SEQUENCE {COMPONENTS OF AssertionAttr, + base Language} + +Telephone ::= Ub32 + +TelephoneSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value-list + SEQUENCE (SIZE (1..MAX)) OF value + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub32}} + +OrAddress ::= Ub1024 + +OrAddressSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value-list + SEQUENCE (SIZE (1..MAX)) OF value + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub1024}} + +Mail ::= Ub256 + +MailSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value-list + SEQUENCE (SIZE (1..MAX)) OF value + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub256}} + +Url ::= Ub1024 + +UrlSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value-list + SEQUENCE (SIZE (1..MAX)) OF value + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub1024}} + +ComServiceType ::= ENUMERATED { + voice, fax, textPhone, videoPhone, publicPhoneBox, switchBoard, pager, + data, e-mail, web} + +ComServiceTypes ::= SEQUENCE OF ComServiceType + +CommService ::= ComServiceTypes + +``` + +``` +CommServiceAs ::= SEQUENCE {COMPONENTS OF AssertionAttr, + base ComServiceType} + +NetworkType ::= ENUMERATED {pstn, isdn, gsm, umts, internet} + +CommNetwork ::= NetworkType + +CommNetworkAs ::= SEQUENCE {COMPONENTS OF AssertionAttr, + base NetworkType} + +AddrValidFrom ::= DateTime + +AddrValidUntil ::= DateTime + +AddrValidityType ::= ENUMERATED {current, old, future, temporary} + +AddrValidity ::= AddrValidityType + +AddrValidityAs ::= SEQUENCE { + COMPONENTS OF AssertionAttr, + base AddrValidityType} + +AddrCoverageType ::= ENUMERATED { + international, national, stateOrProvince, locality} + +AddrCoverage ::= AddrCoverageType + +AddrCoverageAs ::= SEQUENCE { + COMPONENTS OF AssertionAttr, + base AddrCoverageType} + +AddrTariffType ::= ENUMERATED {normal, premium, toll-free} + +AddrTariff ::= AddrTariffType + +AddrTariffAs ::= SEQUENCE {COMPONENTS OF AssertionAttr, + base AddrTariffType} + +AddrRestrictionType ::= ENUMERATED { + public, secret, call-screen, no-marketing, complete-only} + +AddrRestriction ::= AddrRestrictionType + +AddrRestrictionAs ::= SEQUENCE { + COMPONENTS OF AssertionAttr, + base AddrRestrictionType} + +Notification ::= CHOICE { + limitProblem LimitProblem, + serviceProblem ServiceProblem, + searchType SearchType, + attributeTypeList AttributeTypeList, + filterNot FilterNot, + filterItem FilterItem, + providerName ProviderName, + hierarchySelectList HierarchySelectList, + searchControlOptionsList SearchControlOptionsList, + attributeCombinations AttributeCombinations, + wordRestriction WordRestriction, + notSupported NotSupported} + +LimitProblem ::= ENUMERATED {adminLimit, permanentRestriction} +``` + +``` + +ServiceProblem ::= ENUMERATED { + hierSelectForbidden, hierSelectNotAvailableForService, hierSelectNotSupported +} + +AttributeTypeList ::= AttributeTypes + +FilterNot ::= Filter + +FilterItem ::= + SEQUENCE (SIZE (1..MAX)) OF + CHOICE {assertion Assertion, + substrings Substrings} + +ProviderName ::= ProviderId + +HierarchySelectList ::= HierarchySelections + +SearchControlOptionsList ::= SearchOptions + +AttributeCombinations ::= + SEQUENCE (SIZE (1..MAX)) OF + CHOICE {attributeType AttributeType, + or Cor, + not AttributeCombinations} + +Cor ::= + SEQUENCE (SIZE (1..MAX)) OF + CHOICE {attributeType AttributeType, + and Cand, + not AttributeCombinations} + +Cand ::= + SEQUENCE (SIZE (1..MAX)) OF + CHOICE {attributeType AttributeType, + or Cor, + not AttributeCombinations} + +StringMatchType ::= ENUMERATED { + exact, deletion, restrictedDeletion, permutation, permutationAndDeletion, + providerDefined} + +StringMatchTypes ::= SEQUENCE OF StringMatchType + +WordMatchType ::= ENUMERATED {exact, truncated, phonetic} + +WordMatchTypes ::= SEQUENCE OF WordMatchType + +CharacterMatchType ::= ENUMERATED {exact, caseIgnore, mapped} + +CharacterMatchTypes ::= SEQUENCE OF CharacterMatchType + +WordRestriction ::= SEQUENCE { + attributeType AttributeType OPTIONAL, + initialMinimum INTEGER(1..MAX) OPTIONAL, + otherMinimum INTEGER(1..MAX) OPTIONAL, + stringMatchTypes StringMatchTypes OPTIONAL, + wordMatchTypes WordMatchTypes OPTIONAL, + characterMatchTypes CharacterMatchTypes OPTIONAL} + +Option ::= ENUMERATED {paging, weighting} + +Options ::= SEQUENCE OF Option + +NotSupported ::= Options + +``` + +``` + +/* +ENCODING-CONTROL XER +GLOBAL-DEFAULTS MODIFIED-ENCODINGS +NAMESPACE "http://www.itu.int/itu-t/Rec/f515/xsd" + +ATTRIBUTE ALL IN PagedResults, entryLimit, hierSelect, searchOptions, + extendedArea IN SearchRequest, SearchRequest.base.entryIdentifier, + ALL IN AssertionAttr, ALL IN SubstringAttr, ALL IN + SubstringValueAttr, WordRestriction.otherMinimum, + WordRestriction.initialMinimum, WordRestriction.attributeType +BASE64 SearchRequest.base.entryIdentifier +LIST AttributeTypes, HierarchySelections, SearchOptions, SubscrTypes, + ComServiceTypes, StringMatchTypes, WordMatchTypes, CharacterMatchTypes, + Options +NAME SearchRequest, ReqHead, CommsAddress, Country, CountryAs, + Locality, LocalitySub, StateOrProvince, StateOrProvinceSub, + StreetAddress, StreetAddressSub, HouseId, HouseIdSub, GivenName, + GivenNameSub, Title, TitleSub, Description, BusinessCategory, + BusinessCategorySub, PostalCode, PostalCodeSub, PostOfficeBox, + PostOfficeBoxSub, DmdName, DmdNameSub, LocalityCode, + LocalityCodeAs, LocalityNDC, LocalityNDCAs, SubscriberName, + SubscriberNameSub, SubscriberType, SubscriberTypeAs, + SecondFamilyName, SecondFamilyNameSub, Profession, ProfessionSub, + Language, LanguageAs, Telephone, TelephoneSub, OrAddress, + OrAddressSub, Mail, MailSub, Url, UrlSub, CommService, + CommServiceAs, CommNetwork, CommNetworkAs, AddrValidFrom, + AddrValidUntil, AddrValidity, AddrValidityAs, AddrCoverage, + AddrCoverageAs, AddrTariff, AddrTariffAs, AddrRestriction, + AddrRestrictionAs, LimitProblem, ServiceProblem, + AttributeTypeList, FilterNot, FilterItem, ProviderName, + HierarchySelectList, SearchControlOptionsList, WordRestriction, + NotSupported AS UNCAPITALIZED +NAME NumericString-1 AS "NumericString" +NAME Attribute.ALL, Family.ALL, Assertion.ALL, Substrings.ALL, + notification.ALL, country, dmdName IN ProviderId + AS REFERENCE UNCAPITALIZED +UNTAGGED SearchRequest.infoSelect.family-list, CountryAs.base, + value-list IN ALL, base IN ALL, StateOrProvinceSub.value- + list.value.base, + StreetAddressSub.value-list.value.base, HouseIdSub.value- + list.value.base, + GivenNameSub.value-list.value.base, TitleSub.value-list.value.base, + BusinessCategorySub.value-list.value.base, + PostalCodeSub.value-list.value.base, + PostOfficeBoxSub.value-list.value.base, DmdNameSub.value- + list.value.base, + SubscriberNameSub.value-list.value.base, + SecondFamilyNameSub.value-list.value.base, + ProfessionSub.value-list.value.base, TelephoneSub.value- + list.value.base, + OrAddressSub.value-list.value.base, MailSub.value-list.value.base, + UrlSub.value-list.value.base +TEXT AttributeType:localityCode AS CAPITALIZED + +*/ +END + +``` + +--- + +``` +Uds2 DEFINITIONS AUTOMATIC TAGS ::= +BEGIN + +IMPORTS + Decimal, DateTime, Language, NMTOKEN, NCName + FROM XSD -- The XSD module is defined in the ASN.1 standards. + + Lang + FROM Xml; + +Entry ::= SEQUENCE { + entryIdentifier OCTET STRING OPTIONAL, + returnedObject + ENUMERATED {subscriber, stateOrProvince, locality, street, businessCategory} + DEFAULT subscriber, + hierarchyLevel INTEGER(0..MAX) OPTIONAL, + hierarchyBelow BOOLEAN DEFAULT FALSE, + choice-list SEQUENCE OF CHOICE {attribute Attribute, + family Family}} + +SearchResult ::= SEQUENCE { + msgCode Decimal OPTIONAL, + queryReference OCTET STRING OPTIONAL, + performer ProviderId OPTIONAL, + information + SEQUENCE (SIZE (1..MAX)) OF + CHOICE {hierarchy + SEQUENCE {entries INTEGER(1..MAX) OPTIONAL, + firstEntry INTEGER(1..MAX) DEFAULT 1, + entry-list SEQUENCE (SIZE (1..MAX)) OF entry Entry}, + entry SEQUENCE {hierSeq INTEGER OPTIONAL, + base Entry}} OPTIONAL, + entryCount + SEQUENCE {count INTEGER(1..MAX) OPTIONAL, + qualifier ENUMERATED {exact, bestEstimate} OPTIONAL} OPTIONAL, + notifications + SEQUENCE {notification-list + SEQUENCE (SIZE (1..MAX)) OF notification Notification} OPTIONAL} + +ResHead ::= SEQUENCE {requestRef INTEGER} + +Explanation ::= UTF8String + +ProviderId ::= SEQUENCE {country Country, + dmdName DmdName OPTIONAL} + +NumericString-1 ::= IA5String(FROM ("0".."9")) (PATTERN "[0-9]") + +Ub16NumericString ::= IA5String(FROM ("0".."9")) (SIZE (0..16)) (PATTERN "[0-9]") + +Ub32 ::= UTF8String(SIZE (0..32)) + +Ub40 ::= UTF8String(SIZE (0..40)) + +Ub64 ::= UTF8String(SIZE (0..64)) + +Ub128 ::= UTF8String(SIZE (0..128)) + +Ub256 ::= UTF8String(SIZE (0..256)) + +Ub512 ::= UTF8String(SIZE (0..512)) +``` + +``` + +Ub1024 ::= UTF8String(SIZE (0..1024)) + +AttributeType ::= ENUMERATED { + country, locality, stateOrProvince, streetAddress, houseId, givenName, + title, description, businessCategory, postalCode, postOfficeBox, dmdName, + localityCode, localityNDC, subscriberName, subscriberType, secondFamilyName, + profession, language, telephone, orAddress, mail, url, commService, + commNetwork, addrValidFrom, addrValidUntil, addrValidity, addrCoverage, + addrTariff, addrRestriction} + +AttributeTypes ::= SEQUENCE OF AttributeType + +SearchType ::= ENUMERATED { + searchForStateOrProvince, searchForLocality, subscriberWithinLocality, + subscriberByEntryIdentifier, subscriberWithinStateOrProvince, + subscriberWithinCountry, searchForStreetAddress, subscriberByStreetAddress, + subscriberByCommunicationsAddress, subscriberByBusinessCategory} + +HierarchySelection ::= ENUMERATED { + self, children, parent, hierarchy, top, subtree, all} + +HierarchySelections ::= SEQUENCE OF HierarchySelection + +SearchOption ::= ENUMERATED {performExactly, includeAllAreas} + +SearchOptions ::= SEQUENCE OF SearchOption + +Filter ::= + SEQUENCE (SIZE (1..MAX)) OF + CHOICE {assertion Assertion, + substrings Substrings} + +Attribute ::= CHOICE { + country Country, + locality Locality, + stateOrProvince StateOrProvince, + streetAddress StreetAddress, + houseId HouseId, + givenName GivenName, + title Title, + description Description, + businessCategory BusinessCategory, + postalCode PostalCode, + postOfficeBox PostOfficeBox, + dmdName DmdName, + localityCode LocalityCode, + localityNDC LocalityNDC, + subscriberName SubscriberName, + subscriberType SubscriberType, + secondFamilyName SecondFamilyName, + profession Profession, + language Language, + telephone Telephone, + orAddress OrAddress, + mail Mail, + url Url, + commService CommService, + commNetwork CommNetwork, + addrValidFrom AddrValidFrom, + addrValidUntil AddrValidUntil, + addrValidity AddrValidity, + addrCoverage AddrCoverage, + addrTariff AddrTariff, + addrRestriction AddrRestriction} + +``` + +``` + +Family ::= CHOICE {commsAddress CommsAddress} + +Assertion ::= CHOICE { + countryAs CountryAs, + localityCodeAs LocalityCodeAs, + localityNDCAs LocalityNDCAs, + subscriberTypeAs SubscriberTypeAs, + languageAs LanguageAs, + commServiceAs CommServiceAs, + addrValidityAs AddrValidityAs, + addrCoverageAs AddrCoverageAs, + addrTariffAs AddrTariffAs, + addrRestrictionAs AddrRestrictionAs} + +AssertionAttr ::= SEQUENCE {weight ENUMERATED {low, high} DEFAULT high} + +Substrings ::= CHOICE { + localitySub LocalitySub, + stateOrProvinceSub StateOrProvinceSub, + streetAddressSub StreetAddressSub, + houseIdSub HouseIdSub, + givenNameSub GivenNameSub, + titleSub TitleSub, + businessCategorySub BusinessCategorySub, + postalCodeSub PostalCodeSub, + postOfficeBoxSub PostOfficeBoxSub, + dmdNameSub DmdNameSub, + subscriberNameSub SubscriberNameSub, + secondFamilyNameSub SecondFamilyNameSub, + professionSub ProfessionSub, + telephoneSub TelephoneSub, + orAddressSub OrAddressSub, + mailSub MailSub, + urlSub UrlSub} + +SubstringAttr ::= SEQUENCE { + string + ENUMERATED {exact, deletion, restrDeletion, permutation, + permutationAndDeletion, providerDefined} DEFAULT exact, + weight ENUMERATED {low, high} DEFAULT high} + +SubstringValueAttr ::= SEQUENCE { + wordMatch ENUMERATED {exact, truncated, phonetic} DEFAULT exact, + characterMatch ENUMERATED {exact, caseIgnore, mapped} DEFAULT caseIgnore} + +CommsAddress ::= SEQUENCE {attribute SEQUENCE (SIZE (1..MAX)) OF Attribute} + +Country ::= NMTOKEN + +CountryAs ::= SEQUENCE {COMPONENTS OF AssertionAttr, + base NMTOKEN} + +Locality ::= Ub128 + +LocalitySub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value SEQUENCE (SIZE (1..MAX)) OF Ub128} + +StateOrProvince ::= Ub128 + +StateOrProvinceSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value} + +``` + +``` + + SEQUENCE (SIZE (1..MAX)) OF + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub128}} + +StreetAddress ::= Ub128 + +StreetAddressSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value + SEQUENCE (SIZE (1..MAX)) OF + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub128}} + +HouseId ::= Ub64 + +HouseIdSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value + SEQUENCE (SIZE (1..MAX)) OF + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub64}} + +GivenName ::= Ub64 + +GivenNameSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value + SEQUENCE (SIZE (1..MAX)) OF + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub64}} + +Title ::= SEQUENCE {lang Lang OPTIONAL, + base Ub64} + +TitleSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value + SEQUENCE (SIZE (1..MAX)) OF + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub64}} + +Description ::= SEQUENCE {lang Lang OPTIONAL, + base Ub1024} + +BusinessCategory ::= SEQUENCE {lang Lang OPTIONAL, + base Ub128} + +BusinessCategorySub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value + SEQUENCE (SIZE (1..MAX)) OF + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub128}} + +PostalCode ::= Ub40 + +PostalCodeSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value + SEQUENCE (SIZE (1..MAX)) OF + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub40}} + +``` + +``` + +PostOfficeBox ::= Ub40 + +PostOfficeBoxSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value + SEQUENCE (SIZE (1..MAX)) OF SEQUENCE {COMPONENTS OF SubstringValueAttr}} + +DmdName ::= Ub64 + +DmdNameSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value + SEQUENCE (SIZE (1..MAX)) OF + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub64}} + +LocalityCode ::= Ub64 + +LocalityCodeAs ::= SEQUENCE {COMPONENTS OF AssertionAttr, + base Ub64} + +LocalityNDC ::= Ub16NumericString + +LocalityNDCAs ::= SEQUENCE { + COMPONENTS OF AssertionAttr, + base Ub16NumericString} + +SubscriberName ::= SEQUENCE {lang Lang OPTIONAL, + base Ub64} + +SubscriberNameSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value + SEQUENCE (SIZE (1..MAX)) OF + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub64}} + +SubscrType ::= ENUMERATED {residential, organization, government} + +SubscrTypes ::= SEQUENCE OF SubscrType + +SubscriberType ::= SubscrTypes + +SubscriberTypeAs ::= SEQUENCE {COMPONENTS OF AssertionAttr, + base SubscrType} + +SecondFamilyName ::= Ub128 + +SecondFamilyNameSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value + SEQUENCE (SIZE (1..MAX)) OF + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub128}} + +Profession ::= SEQUENCE {lang Lang OPTIONAL, + base Ub128} + +ProfessionSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value + SEQUENCE (SIZE (1..MAX)) OF + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub128}} + +``` + +``` + +Language ::= XSD.Language + +LanguageAs ::= SEQUENCE {COMPONENTS OF AssertionAttr, + base Language} + +Telephone ::= Ub32 + +TelephoneSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value + SEQUENCE (SIZE (1..MAX)) OF + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub32}} + +OrAddress ::= Ub1024 + +OrAddressSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value + SEQUENCE (SIZE (1..MAX)) OF + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub1024}} + +Mail ::= Ub256 + +MailSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value + SEQUENCE (SIZE (1..MAX)) OF + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub256}} + +Url ::= Ub1024 + +UrlSub ::= SEQUENCE { + COMPONENTS OF SubstringAttr, + value + SEQUENCE (SIZE (1..MAX)) OF + SEQUENCE {COMPONENTS OF SubstringValueAttr, + base Ub1024}} + +ComServiceType ::= ENUMERATED { + voice, fax, textPhone, videoPhone, publicPhoneBox, switchBoard, pager, + data, e-mail, web} + +ComServiceTypes ::= SEQUENCE OF ComServiceType + +CommService ::= ComServiceTypes + +CommServiceAs ::= SEQUENCE {COMPONENTS OF AssertionAttr, + base ComServiceType} + +NetworkType ::= ENUMERATED {pstn, isdn, gsm, umts, internet} + +CommNetwork ::= NetworkType + +CommNetworkAs ::= SEQUENCE {COMPONENTS OF AssertionAttr, + base NetworkType} + +AddrValidFrom ::= DateTime + +AddrValidUntil ::= DateTime + +``` + +``` + +AddrValidityType ::= ENUMERATED {current, old, future, temporary} + +AddrValidity ::= AddrValidityType + +AddrValidityAs ::= SEQUENCE { + COMPONENTS OF AssertionAttr, + base AddrValidityType} + +AddrCoverageType ::= ENUMERATED { + international, national, stateOrProvince, locality} + +AddrCoverage ::= AddrCoverageType + +AddrCoverageAs ::= SEQUENCE { + COMPONENTS OF AssertionAttr, + base AddrCoverageType} + +AddrTariffType ::= ENUMERATED {normal, premium, toll-free} + +AddrTariff ::= AddrTariffType + +AddrTariffAs ::= SEQUENCE {COMPONENTS OF AssertionAttr, + base AddrTariffType} + +AddrRestrictionType ::= ENUMERATED { + public, secret, call-screen, no-marketing, complete-only} + +AddrRestriction ::= AddrRestrictionType + +AddrRestrictionAs ::= SEQUENCE { + COMPONENTS OF AssertionAttr, + base AddrRestrictionType} + +Notification ::= CHOICE { + limitProblem LimitProblem, + serviceProblem ServiceProblem, + searchType SearchType, + attributeTypeList AttributeTypeList, + filterNot FilterNot, + filterItem FilterItem, + providerName ProviderName, + hierarchySelectList HierarchySelectList, + searchControlOptionsList SearchControlOptionsList, + attributeCombinations AttributeCombinations, + wordRestriction WordRestriction, + notSupported NotSupported} + +LimitProblem ::= ENUMERATED {adminLimit, permanentRestriction} + +ServiceProblem ::= ENUMERATED { + hierSelectForbidden, hierSelectNotAvailableForService, hierSelectNotSupported +} + +AttributeTypeList ::= AttributeTypes + +FilterNot ::= Filter + +FilterItem ::= + SEQUENCE (SIZE (1..MAX)) OF + CHOICE {assertion Assertion, + substrings Substrings} + +ProviderName ::= ProviderId + +``` + +``` + +HierarchySelectList ::= HierarchySelections + +SearchControlOptionsList ::= SearchOptions + +AttributeCombinations ::= + SEQUENCE (SIZE (1..MAX)) OF + CHOICE {attributeType AttributeType, + or Cor, + not AttributeCombinations} + +Cor ::= + SEQUENCE (SIZE (1..MAX)) OF + CHOICE {attributeType AttributeType, + and Cand, + not AttributeCombinations} + +Cand ::= + SEQUENCE (SIZE (1..MAX)) OF + CHOICE {attributeType AttributeType, + or Cor, + not AttributeCombinations} + +StringMatchType ::= ENUMERATED { + exact, deletion, restrictedDeletion, permutation, permutationAndDeletion, + providerDefined} + +StringMatchTypes ::= SEQUENCE OF StringMatchType + +WordMatchType ::= ENUMERATED {exact, truncated, phonetic} + +WordMatchTypes ::= SEQUENCE OF WordMatchType + +CharacterMatchType ::= ENUMERATED {exact, caseIgnore, mapped} + +CharacterMatchTypes ::= SEQUENCE OF CharacterMatchType + +WordRestriction ::= SEQUENCE { + attributeType AttributeType OPTIONAL, + initialMinimum INTEGER(1..MAX) OPTIONAL, + otherMinimum INTEGER(1..MAX) OPTIONAL, + stringMatchTypes StringMatchTypes OPTIONAL, + wordMatchTypes WordMatchTypes OPTIONAL, + characterMatchTypes CharacterMatchTypes OPTIONAL} + +Option ::= ENUMERATED {paging, weighting} + +Options ::= SEQUENCE OF Option + +NotSupported ::= Options + +/* +ENCODING-CONTROL XER + GLOBAL-DEFAULTS MODIFIED-ENCODINGS + NAMESPACE "http://www.itu.int/itu-t/Rec/f515/xsd" + + ATTRIBUTE entryIdentifier, hierarchyBelow, hierarchyLevel, returnedObject + IN Entry, entities, firstEntry IN SearchResult.information.hierarchy, + SearchResult.information.entry.hierSeq, ALL IN SearchResult.entryCount, + msgCode, queryReference IN SearchResult, ALL IN AssertionAttr, + ALL IN SubstringAttr, ALL IN SubstringValueAttr, + otherMinimum, initialMinimum, attributeType IN WordRestriction + BASE64 Entry.entryIdentifier + LIST AttributeTypes, HierarchySelections, SearchOptions, SubscrTypes, + ComServiceTypes, StringMatchTypes, WordMatchTypes, CharacterMatchTypes, + +``` + +``` + + Options +NAME SearchResult, ResHead, Explanation, CommsAddress, Country, CountryAs, + Locality, LocalitySub, StateOrProvince, StateOrProvinceSub, + StreetAddress, StreetAddressSub, HouseId, HouseIdSub, GivenName, + GivenNameSub, Title, TitleSub, Description, BusinessCategory, + BusinessCategorySub, PostalCode, PostalCodeSub, PostOfficeBox, + PostOfficeBoxSub, DmdName, DmdNameSub, LocalityCode, + LocalityCodeAs, LocalityNDC, LocalityNDCAs, SubscriberName, + SubscriberNameSub, SubscriberType, SubscriberTypeAs, + SecondFamilyName, SecondFamilyNameSub, Profession, ProfessionSub, + Language, LanguageAs, Telephone, TelephoneSub, OrAddress, + OrAddressSub, Mail, MailSub, Url, UrlSub, CommService, + CommServiceAs, CommNetwork, CommNetworkAs, AddrValidFrom, + AddrValidUntil, AddrValidity, AddrValidityAs, AddrCoverage, + AddrCoverageAs, AddrTariff, AddrTariffAs, AddrRestriction, + AddrRestrictionAs, LimitProblem, ServiceProblem, + AttributeTypeList, FilterNot, FilterItem, ProviderName, + HierarchySelectList, SearchControlOptionsList, WordRestriction, + NotSupported AS UNCAPITALIZED +NAME NumericString-1 AS "NumericString" +NAME Attribute.ALL, Family.ALL, Assertion.ALL, Substrings.ALL, + notification.ALL, country, dmdName IN ProviderId + AS REFERENCE UNCAPITALIZED +UNTAGGED Entry.choice-list, SearchResult.information.*.hierarchy.entry-list, + SearchResult.information.*.entry.base, + SearchResult.notifications.notification-list, CountryAs.base, + value-list IN ALL, base IN ALL, StateOrProvinceSub.value- +list.value.base, + StreetAddressSub.value-list.value.base, HouseIdSub.value- +list.value.base, + GivenNameSub.value-list.value.base, TitleSub.value-list.value.base, + BusinessCategorySub.value-list.value.base, + PostalCodeSub.value-list.value.base, + PostOfficeBoxSub.value-list.value.base, DmdNameSub.value- +list.value.base, + SubscriberNameSub.value-list.value.base, + SecondFamilyNameSub.value-list.value.base, + ProfessionSub.value-list.value.base, TelephoneSub.value- +list.value.base, + OrAddressSub.value-list.value.base, MailSub.value-list.value.base, + UrlSub.value-list.value.base + TEXT AttributeType:localityCode AS CAPITALIZED +*/ +END + +``` + +--- + +``` + +Xml DEFINITIONS AUTOMATIC TAGS ::= +BEGIN + +``` + +``` + +IMPORTS + +``` + +``` + + AnyURI, NCName, Language + FROM XSD; -- The XSD module is defined in the ASN.1 standards. + +``` + +``` + +Lang ::= Language + +``` + +``` + +Space ::= ENUMERATED {default, preserve} + +``` + +``` + +Base ::= AnyURI + +``` + +``` +SpecialAttrs ::= SEQUENCE { + base Base OPTIONAL, + lang Lang OPTIONAL, + space Space OPTIONAL} + +/* + +ENCODING-CONTROL XER + ATTRIBUTE ALL IN SpecialAttrs, Space, Lang, Base + TEXT Space + +*/ +END +``` + + + +# SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|--------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series B | Means of expression: definitions, symbols, classification | +| Series C | General telecommunication statistics | +| Series D | General tariff principles | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Construction, installation and protection of cables and other elements of outside plant | +| Series M | TMN and network maintenance: international transmission systems, telephone circuits, telegraphy, facsimile and leased circuits | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks and open system communications | +| Series Y | Global information infrastructure and Internet protocol aspects | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.581-199303-I_PDF-E/042733dc5e8e7f5f30b60adba3266cde_img.jpg b/marked/F/T-REC-F.581-199303-I_PDF-E/042733dc5e8e7f5f30b60adba3266cde_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..eb7ee2dc9149b2065b18fb2c902e6773e50e50c1 --- /dev/null +++ b/marked/F/T-REC-F.581-199303-I_PDF-E/042733dc5e8e7f5f30b60adba3266cde_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0bdee6baef7a794436d32df88e44afc8aefacdac78549f6ccf986297dbfbd8d6 +size 44196 diff --git a/marked/F/T-REC-F.581-199303-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.581-199303-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..0c0c73a6b1c1452b28bdb9ace3364ef8712b040d --- /dev/null +++ b/marked/F/T-REC-F.581-199303-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:918856d707d44d2b6816333bd3ab17245b7ca6e7d7dcbf81a5f92a45078c43f8 +size 8203 diff --git a/marked/F/T-REC-F.581-199303-I_PDF-E/5e92d9e8e9ce204e405bff2367f88176_img.jpg b/marked/F/T-REC-F.581-199303-I_PDF-E/5e92d9e8e9ce204e405bff2367f88176_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..4292fefad3d50355134ad8b3d4da0e579f939f86 --- /dev/null +++ b/marked/F/T-REC-F.581-199303-I_PDF-E/5e92d9e8e9ce204e405bff2367f88176_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:204031bf9fd9910271eee82315dade07b8331aec051e64252026002765feed6a +size 65197 diff --git a/marked/F/T-REC-F.581-199303-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.581-199303-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..433f7f3e41f17f8e33c8f8bbd1d084a8368942d7 --- /dev/null +++ b/marked/F/T-REC-F.581-199303-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e55e6bf8438b9822bae3862abec3666ebcb2809564e6794eba27598ec76ffa14 +size 69931 diff --git a/marked/F/T-REC-F.581-199303-I_PDF-E/raw.md b/marked/F/T-REC-F.581-199303-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..9d1db3c85ddab1cb1554d7063ee6a2dd2341e799 --- /dev/null +++ b/marked/F/T-REC-F.581-199303-I_PDF-E/raw.md @@ -0,0 +1,401 @@ + + +![ITU logo: A globe with a lightning bolt and the letters ITU.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with a lightning bolt and the letters ITU. + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.581** + +(03/93) + +**OPERATIONS AND QUALITY OF SERVICE +DIRECTORY SERVICES** + +--- + +**GUIDELINES FOR PROGRAMMING +COMMUNICATION INTERFACES (PCIs) +DEFINITION: SERVICE RECOMMENDATION** + +**ITU-T Recommendation F.581** + +(Previously "CCITT Recommendation") + +--- + +# FOREWORD + +The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of the International Telecommunication Union. The ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Conference (WTSC), which meets every four years, established the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics. + +ITU-T Recommendation F.581 was prepared by the ITU-T Study Group I (1988-1993) and was approved by the WTSC (Helsinki, March 1-12, 1993). + +--- + +# NOTES + +1 As a consequence of a reform process within the International Telecommunication Union (ITU), the CCITT ceased to exist as of 28 February 1993. In its place, the ITU Telecommunication Standardization Sector (ITU-T) was created as of 1 March 1993. Similarly, in this reform process, the CCIR and the IFRB have been replaced by the Radiocommunication Sector. + +In order not to delay publication of this Recommendation, no change has been made in the text to references containing the acronyms "CCITT, CCIR or IFRB" or their associated entities such as Plenary Assembly, Secretariat, etc. Future editions of this Recommendation will contain the proper terminology related to the new ITU structure. + +2 In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +# CONTENTS + +*Page* + +| | | | +|-----|--------------------------------------------|---| +| 1 | Introduction ..... | 1 | +| 2 | Scope ..... | 2 | +| 2.1 | General ..... | 2 | +| 2.2 | Applicability ..... | 2 | +| 3 | Possible benefits of the PCIs ..... | 2 | +| 3.1 | For implementors ..... | 2 | +| 3.2 | For users ..... | 3 | +| 3.3 | For carriers ..... | 3 | +| 4 | PCIs requirements ..... | 3 | +| 4.1 | Independence ..... | 3 | +| 4.2 | Dependency ..... | 3 | +| 4.3 | Flexibility ..... | 3 | +| 4.4 | Extensibility ..... | 3 | +| 4.5 | Multiple applications ..... | 3 | +| 4.6 | Easy implementation and testing ..... | 4 | +| 5 | General guidelines for defining PCIs ..... | 4 | +| 5.1 | Location of the interface ..... | 4 | +| 5.2 | Functional description ..... | 4 | +| 5.3 | Formal description ..... | 4 | +| 5.4 | Presentation of the information ..... | 4 | +| 6 | General functional model ..... | 4 | +| 7 | Application related interfaces ..... | 5 | +| 7.1 | Definitions ..... | 5 | +| 7.2 | General features ..... | 5 | +| 8 | Network/Protocol related interfaces ..... | 6 | +| 8.1 | Definition ..... | 6 | +| 8.2 | General features ..... | 6 | +| | Annex A – Definitions ..... | 6 | +| | Appendix I – Examples ..... | 7 | +| I.1 | Telefax 3 on PSTN ..... | 7 | +| I.2 | ISDN Lower Layer Module ..... | 8 | + + + +# GUIDELINES FOR PROGRAMMING COMMUNICATION INTERFACES (PCIs) DEFINITION: SERVICE RECOMMENDATION + +(Helsinki, 1993) + +# 1 Introduction + +## 1.1 Today, telecommunication needs in the microcomputer world are quickly increasing. + +ITU-T services are defined to meet these needs, but due to the technical and commercial context of the microcomputer world some concrete problems may arise. + +### a) *Technical aspects* + +In some implementations communication equipment may be composed of hardware and software modules which come from various manufacturers. These modules must exchange information to provide the user with access to ITU-T services, but manufacturers may use different working methods and vocabularies. + +Therefore, adding new software and hardware to existing configurations involves very detailed knowledge of different areas: telecommunication aspects, computer aspects, hardware and software aspects. + +Setting up such equipment requires the help of experts, whereas using ITU-T services only requires a very basic knowledge of the technical aspects. + +The consequence of such a situation is often a lack of integration of the communication dialogue with the local software dialogue. + +### b) *Commercial aspects* + +Using hardware and software modules without standardized interfaces may mean buying software and hardware pieces designed to fit together well. + +The definition of a public universal interface for a specific communication application (such as facsimile, Teletex, telex, etc.) gives to the user more freedom in the choice of the products. + +Most of these difficulties regarding softwares may be solved by the standardization of public inter-application interfaces. + +## 1.2 Interfaces allowing access to software providing a ITU-T communication service are called PCIs for "Programming Communication Interfaces\*." + +Two levels of PCIs are defined: + +- Application related interfaces which allow any local Application\* (LA) to access any Communication Application\* (CA). One example of an application related interface is the APPLI/COM, which is described in Recommendation T.611. +- Network/Protocol related interfaces, which allow any LA or Communication Management Entity\* (CME) to access any Lower Layer Module\* (LLM). Network/Protocol related interfaces are under study. + +NOTE – The definition of new terms indicated with (\*) may be found in Annex A. + +Examples are shown in Appendix I. + +## 1.3 The information to be exchanged through the interfaces and the way to exchange them is subject to the PCIs technical Recommendations. + +The exchange method describes the procedures that are performed to convey the information between software modules on each side of the interfaces. + +# 2 Scope + +## 2.1 General + +### 2.1.1 This Recommendation gives guidelines from a service point of view for the definition of PCIs. + +The applicability of PCIs requires the definition of: + +- a) the information to be exchanged through the interface. These messages are defined Operating System and programming language independent, but ITU-T service definition dependent. Therefore, they are defined in technical ITU-T Recommendations; +- b) the way to convey (Exchange method) this information through the interface. + +### 2.1.2 The main objective of this Recommendation is to define: + +- the possible benefits of these PCIs; +- a common vocabulary for all PCIs definitions; +- a general functional model allowing all PCIs to work together if required; +- general rules to be taken into account in the definition of PCIs. + +### 2.1.3 Technical definitions of PCIs are defined in related Recommendation(s): + +T.611: Programming Communication Interface (PCI) APPLI/COM for Facsimile Group 3, Facsimile Group 4, Teletex and telex services. + +NOTE – Other Recommendations are under study. + +## 2.2 Applicability + +The provision of Programming Communication Interfaces is not mandatory for participating in a ITU-T service. + +The intention is to help developers by giving them general guidelines for the integration of ITU-T services in their products. + +It must be kept in mind that PCIs leave a certain freedom to manufacturers, who may implement one or more of the possible PCIs and may extend them for private needs. + +# 3 Possible benefits of the PCIs + +Possible benefits which could be expected from the definition of the PCIs concern implementors, users and carriers. They are: + +## 3.1 For implementors + +- a) PCIs help developers by providing them with common standardized interfaces. Implementors of Communication Applications\* (CA) will find in PCIs Recommendations practical details concerning the implementation of standardized ITU-T services. +- b) On the other hand, implementors of Local Applications (LA) will no longer need to become experts of telecommunication protocols. +- c) For both categories, time taken to develop the implementation may be shortened and developments may become cheaper because developers will no longer need to adapt their products to many other products, whose features change frequently. +- d) As another expected result, the PCIs might shorten the development of new applications as these applications could benefit from existing communication applications that comply with ITU-T service rules (provided these existing applications also conform to PCI). + +NOTE – The definition of new terms indicated with (\*) may be found in Annex A. + +## **3.2 For users** + +- a) The main expected result of using the PCIs is gaining uniformity in the use of ITU-T services. Thanks to the Programming Communication Interfaces, users will be able, for instance, to send a document using the facsimile service, telex service or the Teletex service without noticeable differences. However, service constraints described in related Recommendations are still applying. +- b) PCIs are designed to provide the end user with the ability to set up, extend and use any type of machine configuration (i.e. stand-alone machines, LANs, etc.) in order to send and receive files using various ITU-T services in a very simple and uniform way. + +NOTE – PCIs are not in competition with LANs standards. + +PCIs provide the means to hide system, software and hardware variations from end-users, thus facilitating the use of ITU-T services. + +## **3.3 For carriers** + +Traffic should increase as more equipment would offer communication features resulting from the universal interfaces, and users should be encouraged to communicate due to better dialogue ergonomics. + +# **4 PCIs requirements** + +## **4.1 Independence** + +The functional definition of a PCI is independent of: + +- the computer type, the operating systems and programming languages; +- the system configuration. It may be implemented in every case of machine configuration, e.g. stand-alone machines, LAN, etc. + +Furthermore, the choice of a presentation of information, independent of operating systems and programming languages should allow PCIs to be easily portable from one operating system to another. The possible choices of programming languages are defined in the technical Recommendations. + +## **4.2 Dependency** + +The functional definition of a PCI depends on the ITU-T services provided by the communication module to be accessed. + +## **4.3 Flexibility** + +PCIs allow flexibility in implementations, for the sake of product differentiation. PCIs definition may include a submission mechanism which allows an application to request another for processing a specific function. In this case, PCIs technical Recommendations define the functions which should normally be performed by each application. + +## **4.4 Extensibility** + +Private extensions are possible without jeopardizing the overall performance of the PCIs. Implementation documentation will specify how these extensions may be used by other application softwares. Technical Recommendations should provide for such extensions. + +## **4.5 Multiple applications** + +Multiple software modules located above an interface may access one or multiple software modules located below the same interface. Since this requirement is operating systems dependant, it is completely managed by the PCI exchange method. + +Procedures allowing a correct transfer of requests and responses between two applications should be independent of the physical location of these applications. + +Multiple local applications may be interfaced to one or several communication applications. + +## **4.6 Easy implementation and testing** + +The implementation of a PCI should be easy in order to gain a widespread acceptance. + +It should give additional information for a better implementation of the concerned applications, in order to avoid errors. The testing of a PCI should be simple. + +NOTE – Technical Recommendations should point out how a “real” implementation has to behave, and should give guidelines for implementors (in Appendix I). + +# **5 General guidelines for defining PCIs** + +## **5.1 Location of the interface** + +As already stated in the introduction two general scopes of PCIs are defined, the Application related interfaces (located at the higher layers of the OSI model) and the Network/Protocol related interfaces (located at the network and protocol dependent layers). Non-OSI applications should be incorporated depending on the individual PCIs. + +In order to avoid any misunderstanding, the ITU-T services, protocols, networks which are covered by the PCI must be clearly identified. + +## **5.2 Functional description** + +The functions which can fulfill the interface should be described. The description of functionality shall be completely independent of the equipment. + +## **5.3 Formal description** + +Messages interchanged through the interface should be described. + +At this step, the interface definition is required to be equipment independent. + +## **5.4 Presentation of the information** + +For application related interfaces the presentation of information (coding scheme) independent of operating systems and languages should be chosen. The description of this may form a part of technical Recommendations. + +The description of the presentation of the information concerning network/protocol related interfaces is not necessarily independent of operating systems and languages. The description may be a part of the technical Recommendation, if it is operating and languages systems dependent. Otherwise, it may form an appendix. Technical Recommendations give more details on this aspect. + +# **6 General functional model** + +The concept of Programming Communication interfaces covers two different levels of interfaces: + +- application related interfaces, one example is the APPLI/COM interface T.611; +- network/protocol related interfaces, which are under study for the time being. + +The interfaces are independent of each other and one may be specified/implemented on its own. Furthermore, one level may contain more than one interface, which may coexist independently of each other. + +The location of each interface level relative to the other is given in Figure 1 “PCIs general model”. + +![Diagram of the PCIs general model showing the hierarchy from Local applications (LA) down to the Network layer.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +The diagram illustrates the PCIs general model as a hierarchical stack of layers and interfaces: + +- Local applications (LA):** The top layer, containing a box labeled "Local application". +- Application related interfaces:** The interface between the Local application and the Communication management entity (CME). +- Communication management entity (CME):** A box below the application related interfaces. +- Network/Protocol related interfaces:** The interface between the CME and the Lower layer module (LLM). +- Lower layer module (LLM):** A large box containing four sub-components: "Management plane", "User plane", and "Control plane" (each in its own box), and "Lower layer module (LLM)" (in a box to the left). +- Network:** The bottom layer, represented by a box labeled "Network". + +On the right side, two curly braces group the layers: + +- The top brace groups the "Local application" and is labeled "Local applications (LA)". +- The bottom brace groups the "Application related interfaces", "CME", and "Network/Protocol related interfaces" and is labeled "Communication application (CA)". + +T0102260-92/d01 + +Diagram of the PCIs general model showing the hierarchy from Local applications (LA) down to the Network layer. + +FIGURE 1/F.581 +PCIs general model + +# 7 Application related interfaces + +## 7.1 Definitions + +**application related PCIs:** Provide interfaces to access one or more ITU-T services. The interfaces are provided by the Communication Management Entity\* (CME). The CME may use Network/Protocol related PCIs for ITU-T network and protocol access. + +The functionality and behaviour of the interface shall effectively uncouple an application from ITU-T service properties. + +NOTE – The definition of new terms indicated with (\*) may be found in Annex A. + +## 7.2 General features + +Functionality and presentation of messages and the exchange method shall be designed independent of hardware and systems software whenever possible. + +Application related interfaces may provide error handling capabilities and shall be able to be tested. + +The functionality of application related PCIs may be task submission oriented. + +# 8 Network/Protocol related interfaces + +## 8.1 Definition + +**network/Protocol related PCIs:** Provide interfaces to access one or more ITU-T Protocol and network. The software modules providing the interfaces are called Lower layer Modules\* (LLM). + +Network/Protocol related PCIs shall be designed most efficiently in terms of speed and size. + +Protocol availability may be identified by Network/Protocol related PCIs. + +NOTE – The definition of new terms indicated with (\*) may be found in Annex A. + +## 8.2 General features + +The messages shall be designed to be independent of hardware and system software as far as the functionality is concerned. However, the presentation of the messages and the exchange method may depend to a large degree on the hardware and systems software. + +The functionality of Network/Protocol related PCIs may be real-time oriented. It may be grouped into three planes: + +- a) management plane, presenting functionality regarding management of message exchange; +- b) user plane, presenting functionality for accessing protocol stacks and management of the communications data stream; +- c) control plane, presenting functionality for network control and signalling. In the case of in-band-signalling this plane is empty. + +# Annex A + +## Definitions + +(This annex forms an integral part of this Recommendation) + +For the purpose of this Recommendation the following definitions apply. + +**A.1 programming Communication Interfaces:** Interfaces allowing access to a software providing a ITU-T communication service are called PCIs for “Programming Communication Interfaces”. + +**A.2 application related interface:** Application related PCIs provide access to one or more ITU-T services. + +**A.3 network/Protocol related interfaces:** Network/Protocol related PCIs provide access to one or more ITU-T protocols and networks. + +**A.4 local Application (LA):** A Local Application is a logical entity capable of generating documents or files and providing the user with a communication dialogue. Concretely, it may be composed either of one software (editor and integrated communication dialogue) or of two softwares (editor and separated software for communication dialogue). + +**A.5 communication Application (CA):** A CA is a virtual entity located below the local application. It is a fictive item, covering all modules (either hard or software) providing telecommunication means to the LA. + +**A.6 communication Management Entity (CME):** A Communication Management Entity is a logical entity capable of storing and performing the communication request coming from the LA. It is also capable of providing the user with communication management functions (logs, etc.). This entity is located within the CA. A CME may handle one or more ITU-T services. + +**A.7 lower Layer Module (LLM):** The Lower Layer Module is an entity providing the Network/Protocol related PCIs. + +# Appendix I + +## Examples + +(This appendix does not form an integral part of this Recommendation) + +NOTE – The figures of this appendix are only provisional and given for information. Technical Recommendations are under study. They may be changed depending on the results of the relevant technical Recommendations. + +## I.1 Telefax 3 on PSTN + +Figure I.1 presents the case where the two PCI levels are used in equipment providing the access to the Telefax 3 service on the Public Switched Telephone Network. + +![Diagram of Telefax 3 on PSTN architecture showing layers from Local application (LA) down to PSTN.](5e92d9e8e9ce204e405bff2367f88176_img.jpg) + +The diagram illustrates the architecture for Telefax 3 on PSTN, showing the interaction between different protocol layers. At the top is a box labeled 'Local application (LA)'. Below it, a horizontal line separates it from the next layer, labeled 'Application related interfaces'. Below this is a box labeled 'Communication management entity (CME)'. Another horizontal line separates this from the next layer, labeled 'Network/Protocol related interfaces'. Below this is a large box containing four sub-components: 'Lower layer module (LLM)', 'Management plane', 'User plane T.30', and 'Control plane PSTN signalling'. A horizontal line separates this from the bottom box, labeled 'PSTN'. On the right side, a large curly bracket groups the 'Communication management entity (CME)' and the 'Network/Protocol related interfaces' layer, labeling them as 'Communication application (CA)'. + +Diagram of Telefax 3 on PSTN architecture showing layers from Local application (LA) down to PSTN. + +T0102270-92/d02 + +NOTE – One interface may be implemented without the other interface. + +FIGURE I.1/F.581 + +Telefax 3 on PSTN + +## I.2 ISDN Lower Layer Module + +Figure I.2 presents the case where the Lower layer Module is used in an equipment providing the access to ISDN. + +![Diagram of the ISDN lower layer module architecture showing the flow from Local application (LA) or communication management entity (CME) down to the ISDN interface through the Network/Protocol related interfaces and the Lower layer module (LLM) which contains Management, User, and Control planes.](042733dc5e8e7f5f30b60adba3266cde_img.jpg) + +The diagram illustrates the architecture of the ISDN lower layer module. It is a vertical stack of components: + +- Top block:** Local application (LA) or communication management entity (CME) +- Interface:** Network/Protocol related interfaces +- Lower layer module (LLM):** A large box containing three sub-components: + - Management plane + - User plane OSI layer 3 + - Control plane ISDN D channel signalling +- Bottom block:** ISDN + +Vertical lines connect the top block to the interface, the interface to the LLM, and the LLM to the bottom ISDN block. + +Diagram of the ISDN lower layer module architecture showing the flow from Local application (LA) or communication management entity (CME) down to the ISDN interface through the Network/Protocol related interfaces and the Lower layer module (LLM) which contains Management, User, and Control planes. + +T0102280-92/d03 + +FIGURE I.2/F.581 +ISDN lower layer module \ No newline at end of file diff --git a/marked/F/T-REC-F.59-199610-I_PDF-E/raw.md b/marked/F/T-REC-F.59-199610-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..1f4eeb9d828892b5c73adebef795abd95e23a3c3 --- /dev/null +++ b/marked/F/T-REC-F.59-199610-I_PDF-E/raw.md @@ -0,0 +1,489 @@ + + +![ITU logo: A globe with a lightning bolt and the letters ITU.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with a lightning bolt and the letters ITU. + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.59** + +(10/96) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Telegraph service – The international telex service + +--- + +**General characteristics of the international telex +service** + +ITU-T Recommendation F.59 + +(Previously CCITT Recommendation) + +--- + +# ITU-T F-SERIES RECOMMENDATIONS NON-TELEPHONE TELECOMMUNICATION SERVICES + +| | | +|-----------------------------------------------------------------|------------------| +| TELEGRAPH SERVICE | F.1–F.109 | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | F.110–F.159 | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | F.160–F.399 | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | F.550–F.599 | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| AUDIOVISUAL SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to ITU-T List of Recommendations.* + +## FOREWORD + +The ITU-T (Telecommunication Standardization Sector) is a permanent organ of the International Telecommunication Union (ITU). The ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Conference (WTSC), which meets every four years, establishes the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics. + +The approval of Recommendations by the Members of the ITU-T is covered by the procedure laid down in WTSC Resolution No. 1 (Helsinki, March 1-12, 1993). + +ITU-T Recommendation F.59 was revised by ITU-T Study Group 1 (1993-1996) and was approved by the WTSC (Geneva, 9-18 October 1996). + +--- + +### NOTE + +In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +© ITU 1997 + +All rights reserved. No part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and microfilm, without permission in writing from the ITU + +## CONTENTS + +*Page* + +| | | | +|------|------------------------------------------------------------------------------------------------------------|---| +| 1 | Introduction ..... | 1 | +| 1.1 | Scope ..... | 1 | +| 2 | General characteristics of the international telex service ..... | 1 | +| 3 | Operational aspects and basic characteristics of telex terminals derived from the service principles ..... | 2 | +| 3.1 | Permanently ready to receive..... | 2 | +| 3.2 | Automatic print-out on continuous paper (e.g. roll) ..... | 2 | +| 3.3 | Effect on telex connections due to failures of resources..... | 2 | +| 3.4 | Answerback unit ..... | 2 | +| 3.5 | Character set and character presentation..... | 3 | +| 3.6 | Special character sequences..... | 3 | +| 3.7 | Presentation of messages at sending and receiving terminals..... | 3 | +| 3.8 | Conversational mode ..... | 3 | +| 3.9 | Stop of transmission by backward signals ..... | 3 | +| 3.10 | Verification of text..... | 3 | +| 4 | References ..... | 3 | +| 4.0 | Definitions ..... | 5 | +| 4.1 | Basic characteristics of the international telex service ..... | 5 | +| 4.2 | Additional facilities of the international telex service ..... | 5 | +| 4.3 | Charging and accounting, Quality of Service, statistics ..... | 6 | +| 4.4 | Service oriented network aspects..... | 6 | +| 4.5 | Mobile services (maritime, satellite)/Radiotelex ..... | 7 | +| 4.6 | Interworking with other services ..... | 7 | + +# **GENERAL CHARACTERISTICS OF THE INTERNATIONAL TELEX SERVICE** + +*(revised in 1996)* + +## **1 Introduction** + +### **1.1 Scope** + +This Recommendation summarizes the service characteristics of the international telex service and the service-related basic characteristics of telex terminal equipment. + +**1.2** This summary of all service-related principles for the international telex service is intended to facilitate a common understanding of this service, bearing in mind: + +- that the international telex service is the only text communication subscriber service available throughout the world; +- that the perceived special merits of this service are its technical reliability and operational simplicity; +- that the operational provisions for the international telex service can be found in Recommendation F.60, but other service principles and service-oriented network and terminal characteristics have to be selected from a number of F-, R-, S- and U-Series Recommendations in addition. + +**1.3** This summary should also be the basis for all considerations of interworking of the international telex service with other services. + +## **2 General characteristics of the international telex service** + +The **international telex service** is a service provided by telecommunication Administrations of almost all countries of the world for the purpose of text communication. In its basic form, it is used for directly transmitting telex messages between subscriber terminal equipment connected to the telex network or to/from public telex stations. + +In the international telex service: + +**2.1** The text of the transmitted and received message is identical with respect to the character sequence in a line and to the line sequence. There is no defined limit on the length of a telex message. There is no “page” defined in the service. + +**2.2** Basic characteristics of telex messages are the character repertoire of the International Telegraph Alphabet No. 2 (ITA2) and a maximum line length of 69 spacing characters per line. + +**2.3** Communication is by using half-duplex, asynchronous (start-stop) transmission, operating with a character structure of 7 1/2 elements per character at a modulation rate of 50 bauds (i.e. 6 2/3 characters/sec.). + +**2.4** Switched connections are used in the telex networks. + +**2.5** Continuous availability of the service, 24 hours a day, in principle, is given. + +**2.6** Telex terminals must be ready to receive messages, even if they are unattended or used in local mode. + +**2.7** An answerback unit allows the identification of the telex terminals during the whole process of communication. At the beginning and at the end of each telex connection the exchange of answerback codes between transmitting and receiving telex terminals is recommended. + +**2.8** The automatic print-out of messages on continuous paper at the transmitting and receiving terminal in the same form and sequence as the messages are transmitted over the line is traditionally recognized as a basic feature. + +**2.9** Another important feature is the possibility of a direct character-by-character conversation between the operators of the telex terminals. + +NOTE – There are telex terminals for which this is not possible, that is where “Conversation Impossible” (CI) is applied. + +## **3 Operational aspects and basic characteristics of telex terminals derived from the service principles** + +### **3.1 Permanently ready to receive** + +Telex terminals must be available to receive messages both whilst in unattended mode or being operated in local mode. This includes the automatic return of the answerback code in response to a WRU signal. + +There is automatic recovery to this state after a power or line failure. + +### **3.2 Automatic print-out on continuous paper (e.g. roll)** + +Telex messages, as traditionally recognized, are printed at the sending and receiving terminals in the character and line sequence and in the chronological order in which they are transmitted over the line. + +If character-by-character printing is not used, then print-out should be no later than immediately after the connection is released. + +If send/receive buffers are used, these, before print-out, must be non-volatile and not accessible, changeable or removable by the operator. + +### **3.3 Effect on telex connections due to failures of resources** + +Prevent establishment or cause clear-down of established connections in case of: end of recording media, power failure. + +### **3.4 Answerback unit** + +The answerback unit is permanently ready to send the answerback code on local or remote request. It is protected from access by the operator. + +Answerback codes exchanged by transmitting/receiving terminals during a connection must be printed in chronological order together with the associated telex message. + +A message prepared for automatic transmission must not contain a WRU combination, unless there is sufficient pause in transmission to receive an answerback after sending a WRU. + +### **3.5 Character set and character presentation** + +The ITA2 character set is used. + +Character presentation is either entirely upper-case or entirely lower-case, but not mixed. + +Reception of the “bell” signal causes an instant audible signal (operator alarm). + +### **3.6 Special character sequences** + +Special character sequences which invoke network responses should not be used within telex messages. Other special character sequences should only be used for their defined purposes. + +### **3.7 Presentation of messages at sending and receiving terminals** + +The complete printable ITA2 character set must be represented. + +There is a maximum of 69 spacing character positions per line. No page format is defined. + +### **3.8 Conversational mode** + +Character-by-character conversation must be possible, with the exception of terminals responding to a “bell” signal with “CI”. Printed documentation of conversational messages should normally be available. + +Transmitted and received characters should be represented differently for distinction (i.e. by shape, colour, etc.). + +ITA2 control characters must be available to the operator. + +There is a manual clear-down facility. The operator at the receiving terminal must also be able to clear down a connection. + +A called telex terminal must not start automatic transmission without being requested to do so by the operator of the calling terminal. + +### **3.9 Stop of transmission by backward signals** + +The operator at the receiving terminal should be able to stop the reception of an automatic transmission by transmitting the character sequence TTT. + +### **3.10 Verification of text** + +Means of verifying the text to be transmitted (in the format to be sent over the line) should be provided: + +- prior to transmission when sending from memory; or +- during transmission when sending from the keyboard. + +## **4 References** + +For detailed information on the service aspects of the international telex service, the following references are given (see also Figure 1): + +![A hierarchical diagram showing the family of recommendations for International telex service. The root node is 'International telex service', which branches into 'Definitions', 'Basic characteristics', 'Charging accounting statistics', 'Additional facilities', 'Service-oriented network aspects', 'Mobile/radiotelex', and 'Interworking with other services'. Each branch further subdivides into specific recommendation codes (U, S, F, R) and annexes. A legend at the bottom right indicates that grey boxes represent 'Draft Recommendation'.](3121ebddccf183ca63bb9781be440a7e_img.jpg) + +The diagram illustrates the family of recommendations for International telex service, organized into seven main categories: + +- Definitions:** + - U.140 + - S.140 + - R.140 + - Annexe B/F.60 +- Basic characteristics:** + - U.45 + - U.1 + - S.20 + - S.9 + - S.8 + - S.6 + - S.5 + - S.4 + - S.1 + - F.68 + - F.65 + - F.69 + - F.60 + - F.59 +- Charging accounting statistics:** + - F.70 + - F.41 + - F.61 +- Additional facilities:** + - U.81 + - U.80 + - U.46 + - U.44 + - U.43 + - U.41 + - U.40 + - S.23 + - S.22 + - S.21 + - S.19 + - S.17 + - S.16 + - S.15 + - S.11 + - F.72 + - F.30 + - F.63 + - F.80 +- Service-oriented network aspects:** + - U.15 + - U.12 + - U.10 + - U.8 + - U.7 + - U.6 + - R.58 bis + - R.58 + - F.64 +- Mobile/radiotelex:** + - U.63 + - U.62 + - U.61 + - U.60 + - F.131 + - F.130 + - F.125 + - F.110 + - U.220 + - U.210 + - U.208 + - U.207 + - U.206 + - U.205 + - U.204 + - U.203 + - U.202 + - U.200 + - U.75 + - U.74 + - U.70 + - S.34 + - S.18 + - F.127 + - F.89 + - F.87 + - F.86 + - F.75/F.421 + - F.83 + - F.82 + - F.74 + - F.71 +- Interworking with other services:** + - U.220 + - U.210 + - U.208 + - U.207 + - U.206 + - U.205 + - U.204 + - U.203 + - U.202 + - U.200 + - U.75 + - U.74 + - U.70 + - S.34 + - S.18 + - F.127 + - F.89 + - F.87 + - F.86 + - F.75/F.421 + - F.83 + - F.82 + - F.74 + - F.71 + +T0206850-97/d01 + +Legend:   Draft Recommendation + +A hierarchical diagram showing the family of recommendations for International telex service. The root node is 'International telex service', which branches into 'Definitions', 'Basic characteristics', 'Charging accounting statistics', 'Additional facilities', 'Service-oriented network aspects', 'Mobile/radiotelex', and 'Interworking with other services'. Each branch further subdivides into specific recommendation codes (U, S, F, R) and annexes. A legend at the bottom right indicates that grey boxes represent 'Draft Recommendation'. + +Figure 1/F.59 – International telex service – Family of Recommendations + +## 4.0 Definitions + +- CCITT Recommendation R.140 (1988), *Definitions of essential technical terms in the field of telegraph transmission*. +- CCITT Recommendation S.140 (1988), *Definitions of essential technical terms relating to apparatus for alphabetic telegraphy*. +- CCITT Recommendation U.140 (1988), *Definitions of essential technical terms relating to telegraph switching and signalling*. +- CCITT Recommendation F.60 (1992), *Operational provisions for the international telex service*. (Annex B). + +### 4.1 Basic characteristics of the international telex service + +- ITU-T Recommendation F.59 (1996), *General characteristics of the international telex service*. +- CCITT Recommendation F.60 (1992), *Operational provisions for the international telex service*. +- CCITT Recommendation F.65 (1988), *Time-to-answer by operators at international telex positions*. +- CCITT Recommendation F.68 (1988), *Establishment of the automatic intercontinental telex network*. +- ITU-T Recommendation F.69 (1994), *The international telex service – Service and operational provisions of telex destination codes and telex network identification codes*. +- ITU-T Recommendation S.1 (1993), *International Telegraph Alphabet No. 2*. +- ITU-T Recommendation S.4 (1993), *Special use of certain characters of the International Telegraph Alphabet No. 2*. +- CCITT Recommendation S.5 (1980), *Standardization of page-printing start-stop equipment and cooperation between page-printing and tape-printing start-stop equipment (ITA2)*. +- CCITT Recommendation S.6 (1984), *Characteristics of answerback units (ITA2)*. +- ITU-T Recommendation S.8 (1993), *Intercontinental standardization of the modulation rate of start-stop apparatus and of the use of combination No. 4 in figure-shift*. +- CCITT Recommendation S.9 (1988), *Switching equipment of start-stop apparatus*. +- ITU-T Recommendation S.20 (1993), *Automatic clearing procedure for a telex terminal*. +- ITU-T Recommendation U.1 (1993), *Signalling conditions to be applied in the international telex service*. +- ITU-T Recommendation U.45 (1993), *Response to the not-ready condition of the telex terminal*. + +## 4.2 Additional facilities of the international telex service + +- ITU-T Recommendation F.30 (1993), *Use of various sequences of combinations for special purposes*. +- ITU-T Recommendation F.63 (1993), *Additional facilities in the international telex service*. +- ITU-T Recommendation F.72 (1996), *The international telex service – General principles and operational aspects of a store and forward facility*. +- CCITT Recommendation S.11 (1980), *Use of start-stop reperforating equipment for perforated tape retransmission*. +- CCITT Recommendation S.15 (1968), *Use of the telex network for data transmission at 50 bauds*. +- ITU-T Recommendation S.16 (1993), *Connection to the telex network of an automatic terminal using a V.24 DCE/DTE interface*. +- CCITT Recommendation S.17 (1968), *Answerback unit simulators*. + +- CCITT Recommendation S.19 (1980), *Calling and answering in the telex network with automatic terminal equipment.* +- ITU-T Recommendation S.21 (1993), *Use of display screens in telex machines.* +- ITU-T Recommendation S.22 (1993), *“Conversation impossible” and or pre-recorded message in response to J/BELL signals from a telex terminal.* +- ITU-T Recommendation S.23 (1993), *Automatic request of the answerback of the terminal of the calling party, by the telex terminal of the called party or by the international network.* +- ITU-T Recommendation U.40 (1993), *Reactions by automatic terminals connected to the telex network in the event of ineffective call attempts or signalling incidents.* +- CCITT Recommendation U.41 (1988), *Changed address interception and call redirection in the telex service.* +- CCITT Recommendation U.43 (1988), *Follow-on calls.* +- CCITT Recommendation U.44 (1988), *Multi-address calls in real time for broadcast purposes in the international telex service.* +- ITU-T Recommendation U.46 (1993), *Interruption of automatic transmission and flow control in the international telex service.* +- ITU-T Recommendation U.80 (1993), *International telex store and forward access from a telex subscriber.* +- ITU-T Recommendation U.81 (1996), *International telex store and forward – Delivery to telex.* + +### 4.3 Charging and accounting, Quality of Service, statistics + +- CCITT Recommendation F.41 (1991), *Interworking between the telemessage service and the international public telegram service.* +- CCITT Recommendation F.61 (1988), *Operational provisions relating to the chargeable duration of a telex call.* +- CCITT Recommendation F.70 (1988), *Evaluating the quality of the international telex service.* + +### 4.4 Service oriented network aspects + +- CCITT Recommendation F.64 (1988), *Determination of the number of international telex circuits required to carry a given volume of traffic.* +- CCITT Recommendation R.58 (1964), *Standard limits of transmission quality for the gentex and telex networks.* +- CCITT Recommendation R.58 bis (1984), *Limits on signal transfer delay for telegraph, telex and gentex networks.* +- CCITT Recommendation U.6 (1964), *Prevention of fraudulent transit traffic in the fully automatic international telex service.* +- ITU-T Recommendation U.7 (1993), *Numbering schemes for automatic switching networks.* +- CCITT Recommendation U.8 (1984), *Hypothetical reference connections for telex and gentex networks.* +- ITU-T Recommendation U.10 (1993), *Equipment of an international telex position.* +- ITU-T Recommendation U.11 (1993), *Telex and gentex signalling on intercontinental circuits used for intercontinental automatic transit traffic (type C signalling).* +- ITU-T Recommendation U.12 (1993), *Terminal and transit control signalling system for telex and similar services on international circuits (type D signalling).* +- ITU-T Recommendation U.15 (1993), *Interworking rules for international signalling systems according to Recommendations U.1, U.11 and U.12.* + +### **4.5 Mobile services (maritime, satellite)/Radiotelex** + +- ITU-T Recommendation F.110 (1996), *Operational provisions for the maritime mobile service.* +- ITU-T Recommendation F.125 (1993), *Numbering plan for access to the mobile-satellite services of INMARSAT from the international telex service.* +- CCITT Recommendation F.130 (1988), *Maritime answer-back codes.* +- CCITT Recommendation F.131 (1988), *Radiotelex service codes.* +- CCITT Recommendation U.60 (1984), *General requirements to be met in interfacing the international telex network with maritime satellite systems.* +- ITU-T Recommendation U.61 (1993), *Detailed requirements to be met in interfacing the international telex network with maritime satellite systems.* +- ITU-T Recommendation U.62 (1993), *General requirements to be met in interfacing the international telex network with the fully automated maritime VHF/UHF radio system.* +- CCITT Recommendation U.63 (1984), *General requirements to be met in interfacing the international telex network with the maritime “direct printing” system.* + +### **4.6 Interworking with other services** + +- CCITT Recommendation F.80 (1991), *Basic requirements for interworking relations between the international telex service and other services.* +- CCITT Recommendation F.71 (1988), *Interconnection of private teleprinter networks with the telex network.* +- CCITT Recommendation F.74 (1992), *Intermediate storage devices accessed from the international telex service using single stage selection – Answerback format.* +- CCITT Recommendation F.82 (1991), *Operational provisions to permit interworking between the international telex service and the intex service.* +- CCITT Recommendation F.83 (1990), *Operational principles for communication between terminals of the international telex service and data terminal equipment on packet switched public data networks.* +- CCITT Recommendation F.85/F.421 (1988), *Intercommunication between the IPM service and the telex service.* +- CCITT Recommendation F.86 (1991), *Interworking between the international telex service and the videotex service.* +- CCITT Recommendation F.87 (1991), *Operational principles for the transfer of messages from terminals on the telex network to Group 3 facsimile terminals connected to the public switched telephone network.* +- CCITT Recommendation F.89 (1992), *Status inquiry function in the international telex service.* +- ITU-T Recommendation F.127 (1996), *Operational procedures for interworking between the telex service and the service offered by INMARSAT standard-C system.* +- CCITT Recommendation S.18 (1980), *Conversion between International Telegraph Alphabet No. 2 and International Alphabet No. 5.* +- ITU-T Recommendation S.34 (1993), *Intex terminals – Requirements to effect interworking with the international telex service.* +- CCITT Recommendation U.70 (1984), *Telex service signals for telex to teletex interworking.* +- CCITT Recommendation U.74 (1988), *Extraction of telex selection information from a calling telex answerback.* + +- ITU-T Recommendation U.75 (1993), *Automatic called telex answerback check.* +- ITU-T Recommendation U.200 (1993), *The international telex service – General technical requirements for interworking.* +- ITU-T Recommendation U.202 (1993), *Technical requirements to be met in providing the international telex service within an integrated services digital network.* +- ITU-T Recommendation U.203 (1993), *Technical requirements to be met when providing real-time bothway communications between terminals of the international telex service and data terminal equipments on a PSPDN or via the PSTN.* +- ITU-T Recommendation U.204 (1993), *Interworking between the international telex service and the public interpersonal messaging service.* +- ITU-T Recommendation U.205 (1993), *Store-and-retrieve facility for the delivery of messages from a terminal of the international telex service to a data terminal equipment which connects to a packet-switched public data network over the public switched telephone network.* +- ITU-T Recommendation U.206 (1993), *Technical requirements for interworking between the international telex service and the videotex service.* +- ITU-T Recommendation U.207 (1993), *Technical requirements to be met for the transfer of messages between terminals of the international telex service and Group 3 facsimile terminals connected to the PSTN.* +- ITU-T Recommendation U.208 (1996), *The international telex service – Interworking with the INMARSAT standard-C system using one-stage selection.* +- ITU-T Recommendation U.210 (1993), *Intex service network requirements to effect interworking with the international telex service.* +- ITU-T Recommendation U.220 (1993), *The international telex service – Technical requirements for a status enquiry function in an interworking scenario.* + +# ITU-T RECOMMENDATIONS SERIES + +- Series A Organization of the work of the ITU-T +- Series B Means of expression: definitions, symbols, classification +- Series C General telecommunication statistics +- Series D General tariff principles +- Series E Overall network operation, telephone service, service operation and human factors +- Series F Non-telephone telecommunication services** +- Series G Transmission systems and media, digital systems and networks +- Series H Audiovisual and multimedia systems +- Series I Integrated services digital network +- Series J Transmission of television, sound programme and other multimedia signals +- Series K Protection against interference +- Series L Construction, installation and protection of cables and other elements of outside plant +- Series M Maintenance: international transmission systems, telephone circuits, telegraphy, facsimile and leased circuits +- Series N Maintenance: international sound programme and television transmission circuits +- Series O Specifications of measuring equipment +- Series P Telephone transmission quality, telephone installations, local line networks +- Series Q Switching and signalling +- Series R Telegraph transmission +- Series S Telegraph services terminal equipment +- Series T Terminals for telematic services +- Series U Telegraph switching +- Series V Data communication over the telephone network +- Series X Data networks and open system communication +- Series Z Programming languages \ No newline at end of file diff --git a/marked/F/T-REC-F.60-199208-I_PDF-E/raw.md b/marked/F/T-REC-F.60-199208-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..35e3ad885c39a493bd4e5451ebe7b266efc36fd6 --- /dev/null +++ b/marked/F/T-REC-F.60-199208-I_PDF-E/raw.md @@ -0,0 +1,1516 @@ + + +![ITU logo: A globe with a lightning bolt and the letters ITU.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with a lightning bolt and the letters ITU. + +INTERNATIONAL TELECOMMUNICATION UNION + +**CCITT** + +**F.60** + +THE INTERNATIONAL +TELEGRAPH AND TELEPHONE +CONSULTATIVE COMMITTEE + +(08/92) + +**TELEGRAPH AND MOBILE SERVICES +OPERATIONS AND QUALITY OF SERVICE** + +--- + +**OPERATIONAL PROVISIONS +FOR THE INTERNATIONAL +TELEX SERVICE** + +**Recommendation F.60** + +--- + +![CCITT logo: A globe with the letters CCITT.](84a1d09fb489061482111515543b60dc_img.jpg) + +CCITT logo: A globe with the letters CCITT. + +Geneva, 1992 + +# FOREWORD + +The CCITT (the International Telegraph and Telephone Consultative Committee) is a permanent organ of the International Telecommunication Union (ITU). CCITT is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The Plenary Assembly of CCITT which meets every four years, establishes the topics for study and approves Recommendations prepared by its Study Groups. The approval of Recommendations by the members of CCITT between Plenary Assemblies is covered by the procedure laid down in CCITT Resolution No. 2 (Melbourne, 1988). + +Recommendation F.60 was revised by Study Group I and was approved under the Resolution No. 2 procedure on the 4th of August 1992. + +# --- CCITT NOTES + +- 1) In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized private operating agency. +- 2) A list of abbreviations used in this Recommendation can be found in Annex C + +# **OPERATIONAL PROVISIONS FOR THE INTERNATIONAL TELEX SERVICE** + +*(revised 1992)* + +## **1 Introduction** + +### **1.1 Scope** + +1.1.1 These provisions fix the rules to be followed in the international telex service. + +1.1.2 The International telex service is a telegraph service for subscribers whereby they can communicate directly and temporarily between themselves using equipment operating at a modulation rate of 50 bauds and presenting a start-stop telegraph interface to the international telex network and employing International Telegraph Alphabet No. 2. + +1.1.3 The general characteristics of the international telex service are described in Recommendation F.59 which provides a convenient guide to the various Recommendations dealing with the international telex service, including the R-, S- and U-Series Recommendations which deal with matters of an essentially technical nature. + +### **1.2 International telex circuits – Routes** + +1.2.1 International telex circuits are made up by using telegraph-type circuits. + +1.2.2 The networks of the countries participating in the international telex service shall, as far as practicable, be directly connected and shall thereby constitute the international telex network. + +1.2.3 In case of breakdown, any defective international circuit (or section of an international circuit) must be repaired with all possible speed and, pending repair, every attempt must be made to provide a replacement circuit with the minimum delay. + +1.2.4 For each relation, the Administrations concerned shall, by mutual agreement, decide upon one or more primary telex routes and, to the extent necessary and possible, upon secondary telex routes. + +1.2.5 In this respect, the Administrations shall conform, as far as possible, with the principles recommended by the CCITT as regards the constitution and maintenance of circuits and installations. In particular, the provisions of Recommendation U.8 regarding end-to-end performance objectives and call set-up time should be taken into account when international telex routes are being established between two countries, or via a transit country. + +1.2.6 A Table of international telex relations and traffic [1] is published in accordance with Recommendation F.95. + +### **1.3 Duration of service – Legal time** + +1.3.1 Each Administration shall fix the working hours of its centres. + +1.3.2 Automatic international telex centres are in principle continuously open. + +1.3.3 Manual international telex centres should, as far as possible, afford continuous service. + +1.3.4 Switching centres that are not open continuously are required to extend their service beyond the normal closing hours when there are calls in progress. + +1.3.5 Each centre shall use the legal time of its country or of its zone. + +## 2 Classes of telex call + +### 2.1 General + +2.1.1 Accepted classes of telex call are: + +- a) ordinary private telex calls between subscribers; +- b) service telex calls, including requests for directory information between telex centres; +- c) privilege telex calls (see Recommendation D.193). + +2.1.2 In the manual and semi-automatic services only, the following additional classes of call are accepted: + +- a) safety of life telex calls; +- b) Government telex calls. + +### 2.2 Service telex calls + +2.2.1 Service telex calls may be exchanged free of charge between the Administrations participating in the international telex service (see Recommendations F.17 and D.192). + +2.2.2 Service telex calls may be requested only by persons authorized to do so by their respective Administrations. + +### 2.3 Government telex calls (manual and semi-automatic service only) + +2.3.1 The person booking a Government telex call must state his name and rank on request. + +2.3.2 A Government telex call shall have priority only if priority has been specifically requested by the calling subscriber. + +## 3 Operation of the international telex service + +### 3.1 Operating systems + +3.1.1 The international telex service may operate: + +- automatically; +- semi-automatically; or +- manually. + +3.1.2 Administrations shall reach mutual agreement upon the most appropriate method of operation to be applied in the international relations that concern them. + +### 3.2 Automatic operation + +3.2.1 It is strongly recommended that the telex network of each country be on an automatic switching basis and that it be possible for subscribers to reach one another by fully automatic selection. + +3.2.2 To establish a connection in the automatic service the subscriber should normally select: an access code to the international telex network, if required, which is fixed according to national rules. + +3.2.3 In accordance with Recommendation U.1, through-connection is normally indicated to the calling subscriber by the return of the called subscriber's answerback code. In order to facilitate the checking of this code within the delay provided by Recommendations F.61 and U.1, insertion by the network of any signals between the call-connected signal and the called subscriber's answerback should be avoided. + +*Note* – For future systems on international connections, it is also desirable to avoid the addition by the called network of date, time and other signals after the called subscriber's answerback. However, service codes or other information in accordance with U-series Recommendations may be sent by the called network, for example, **NCH** or **RDI** in accordance with Recommendation U.41. + +3.2.4 In the automatic service no priority shall be given to the various classes of calls listed in § 2.1. + +3.2.5 The duration of calls in the automatic service should not be limited. However, for operational reasons, Administrations may set a limit on the maximum duration, for example, 12 hours or 24 hours. + +3.2.6 In the automatic operation of the international telex service, if, during an established connection, there is a prolonged period of idle condition, the originating network may initiate clearing. This provision does not apply to either transit or the terminating network. Where such protection against inadvertent holding of a circuit is provided, it is recommended that Administrations make any necessary arrangements for the facility not to operate where requested by the subscriber or for certain types of calls. + +3.2.7 For any given traffic relation between two countries, the number of circuits provided should be arranged such that, during the busy hour, the probability of lost calls due to the lack of international circuits should not exceed 1 call in 50. For the calculation of the number of circuits, the requirements of Recommendation F.64 should be applied. + +### 3.3 *Semi-automatic and manual operation* + +#### 3.3.1 *General provisions* + +3.3.1.1 Wherever fully automatic selection has not yet been adopted, it is recommended that semi-automatic operation should be introduced, whereby the operator of the originating international telex position receives the booking, sets up and controls the call. + +3.3.1.2 Where semi-automatic service is not possible, calls shall be established manually by means of two or more international telex positions in tandem whereby the operator of the originating international telex position normally receives the booking. + +3.3.1.3 The operator of the originating international position must be acquainted with the necessary operating particulars of the networks in the destination country. The incoming Administration will give all the necessary technical information to the outgoing Administration. + +3.3.1.4 Any faults in installations noted by international telex positions must be reported without delay to the technical service responsible for their maintenance. + +3.3.1.5 The technical services responsible for the maintenance of telex circuits are recommended to use the abbreviations given in the *List of service abbreviations for maintenance of telegraph circuits*, published in Annex A of Recommendation R.90. + +3.3.1.6 The number of circuits between two networks and the switching equipment should in all cases be calculated as far as possible for a no-delay telex service. + +#### 3.3.2 *Booking of telex calls* + +3.3.2.1 In the booking of a call, the telex installation of the subscriber required must be designated by the name of the country, the subscriber's exchange if necessary, and his national number. + +3.3.2.2 Bookings of telex calls not completed shall cease to be valid: + +Where all the offices concerned are open continuously: + +- a) at midnight if the telex call has been booked before 10 p.m. on the same day; +- b) at 8 a.m. if the telex call has been booked after 10 p.m. the previous evening; +- c) in each case the times indicated shall be those of the originating telex centre. + +Where all the offices concerned are not open continuously: + +- at the closing time of the originating telex centre. + +3.3.2.3 In the case of all bookings of telex calls, and subject to the provisions relative to the validity of bookings, the caller may, so long as the required subscriber has not been obtained: + +- a) cancel his booking; +- b) specify the time after which the booking should be cancelled; +- c) change the number of the station required within the destination country. + +3.3.2.4 Modifications of bookings shall be permitted free of charge; the origin Administration may, however, make a special charge covering the additional work of recording. This charge shall not enter into the international accounts. + +#### 3.3.3 *Priority of telex calls* + +3.3.3.1 When the manual telex service normally provides a demand service, no priority shall be given to the various classes of call. + +3.3.3.2 Under fault or congestion conditions, and in general when a national telex service does not provide a demand service, either normally or temporarily, international telex calls shall be set up in the following order: + +- a) calls concerning safety of life; +- b) service calls concerning the re-establishment of international telecommunication links following a major breakdown or disruption in service (see Recommendation A.30); +- c) government calls for which priority has specifically been requested; +- d) government calls for which priority has not been requested, ordinary private calls, service calls other than those mentioned in b); +- e) privilege calls. + +3.3.3.3 In the international telex centre, calls shall take their priority according to their class and time of receipt at the exchange. + +#### 3.3.4 *Establishment and disconnection of calls by the international telex positions* + +3.3.4.1 Telex calls established manually or semi-automatically will normally be controlled by the international telex position in the origin country. However, where a call is established over two or more international links and access to the second link is obtained manually in the transit country concerned, control of the call will be exercised by the operator in the transit country in the following circumstances: + +- a) if the first link is provided by landline, satellite, microwave link or submarine cable and the second or subsequent link by ARQ radio; +- b) if the call is booked with the operator in the transit country and connection with the subscriber in the origin country is established semi-automatically. + +3.3.4.2 International telex centres connected with each other by several international telex circuits may, by mutual agreement, allocate certain of these circuits for setting up transit calls or for the establishment of traffic in one direction only. + +3.3.4.3 For the operation of international telex circuits, the French or English language shall be used between Administrations having different languages, in the absence of special agreements between them for the use of other languages. + +3.3.4.4 In the manual service, all bookings, modifications of bookings and cancellation advices shall be transmitted as quickly as possible to the international telex centre charged with establishing the calls booked. + +3.3.4.5 In the manual service, calling signals on international circuits must be answered immediately. + +3.3.4.6 On bothway circuits, calls of the same class are established alternately. The international telex centres may, by mutual arrangement, temporarily change to one-way working to improve the flow of traffic. + +3.3.4.7 Telex calls already prepared must not be delayed for the benefit of calls of higher priority, with the exception of calls concerning safety of life. + +3.3.4.8 Without prejudice to the provisions of § 3.3.6, the operator directing the calls at the international telex position shall verify that transmission between the correspondents is satisfactory. He shall note the time when the call is established as well as the time when the call ends and/or its duration. He shall record service incidents and other items necessary for the preparation of the international accounts. + +3.3.4.9 With the exception of the cases where the duration of calls is limited, and of cases where an infringement of the present provisions or national instructions has been noted, operators are forbidden to cut off or break into an established call that is proceeding normally. + +#### 3.3.5 *Limitation of the duration of international telex calls* + +3.3.5.1 In general, the duration of ordinary private calls, service calls, and privilege calls shall not be limited. However, under congestion conditions, the international telex centres concerned may agree to limit the duration of such calls to twelve, or even six minutes (see also § 3.2.5). + +3.3.5.2 The duration of safety of life and Government calls shall not be limited. These calls are only available in the manual and semi-automatic service. + +3.3.5.3 However, transit Administrations shall have the right, in the case of breakdown, to limit the duration of Government calls to twelve minutes when these calls are established through the intermediary of one of their exchanges. In such a case the operator in the transit country shall advise the controlling operator that restrictions on duration are in force. + +3.3.5.4 If the duration of an operator-assisted call is limited, the caller shall be informed, when the call is about to be connected, that it will be cut off after the due time. + +#### 3.3.6 *Operating procedure on international telex positions* + +##### 3.3.6.1 *Single operator case* + +3.3.6.1.1 If the called subscriber can be obtained directly by the controlling international telex operator, this operator should: + +- a) hold the calling subscriber and select a free circuit; +- b) select the called subscriber using the information in accordance with § 3.3.2.1; +- c) set up the call to the called subscriber and obtain the called subscriber's answerback, which must also be transferred to the calling subscriber; +- d) obtain the calling subscriber's answerback, which must also be transferred to the called subscriber; +- e) operate the timing equipment; +- f) clear the connection on reception of the clear signal from either party. + +3.3.6.1.2 If the called subscriber is engaged, the controlling international telex operator should signal **OCC** and then release the calling subscriber. When the calling subscriber has to be recalled, the signal **RAP** is sent after the **OCC** signal before releasing. + +##### 3.3.6.2 *Two-operator case* + +3.3.6.2.1 If the called subscriber is obtained via two international telex positions: + +- a) the controlling international operator should hold the calling subscriber and select a free circuit; +- b) the operator at the second international position should announce himself by the abbreviated name of his1) telex exchange2); +- c) the controlling international operator should send his own answerback code and signal the particulars of the required subscriber received in § 3.3.2.1; + +--- + +1) The use of gender-specific terms within this Recommendation have no significance. + +2) It is recommended that, as far as possible, the abbreviated name of the telex exchange shall be transmitted by means of the answerback unit and shall be so constituted as to permit the identification of the operator's position concerned in the connection of an international call. + +- d) the operator of the second international position should: + - i) hold the circuit from the controlling international position; + - ii) select the required subscriber, using the information received in accordance with § 3.3.6.2.1, c); + - iii) establish the connection to the called subscriber; + - iv) signal **DF** to the controlling international position, which is not passed to the calling subscriber; +- e) the controlling international operator should then: + - i) establish the connection if necessary, with the calling subscriber and obtain the called subscriber's answerback which must also be transferred to the calling subscriber; + - ii) obtain the calling subscriber's answerback, which must also be transferred to the called subscriber; + - iii) operate the timing equipment; + - iv) clear down the connection on receiving the clear signal from either party. + +3.3.6.2.2 If the called subscriber is engaged, the operator of the second international position should signal **OCC** and clear down the international circuit. The controlling operator should then proceed as indicated in § 3.3.6.1.2. + +##### 3.3.6.3 *Multiple operator case* + +3.3.6.3.1 If the called subscriber is obtained via more than two international telex positions: + +- a) the controlling international operator should hold the calling subscriber and select a free circuit; +- b) the operator at the second international position should announce himself by the abbreviated name of his telex exchange3); +- c) the controlling international operator should send his own answerback and signal the particulars of the required subscriber to the second international position; +- d) the operator of the second international position should extend the call to the third international position and signal **THRU** to the calling international position; +- e) the operator of the third international position should announce himself by the abbreviated name of his telex exchange3); +- f) the controlling international operator should send his own answerback and signal the particulars of the required subscriber to the third international position; +- g) the operator of the third international position should: + - i) hold the circuit from the controlling international position; + - ii) select the required subscriber, using the information received in accordance with § 3.3.6.3.1, f); + - iii) establish the connection between it and the called subscriber; + - iv) signal the letters **DF** to the controlling international position, which are not passed to the calling subscriber; +- h) the controlling international operator should: + - i) establish the connection with the calling subscriber and obtain the called subscriber's answerback, which must also be transferred to the calling subscriber; + - ii) obtain the calling subscriber's answerback, which must also be transferred to the called subscriber; + - iii) operate the timing equipment; + - iv) clear down the connection on receiving the clear signal from either party. + +3) It is recommended that, as far as possible, the abbreviated name of the telex exchange shall be transmitted by means of the answerback unit and shall be so constituted as to permit the identification of the operator's position concerned in the connection of an international call. + +3.3.6.3.2 If the operator of the second international telex position finds all the circuits to the third position engaged, he should signal **NC** and clear down the international circuit. + +3.3.6.3.3 If the called subscriber is engaged, the operator of the destination international telex position should signal **OCC** and clear down the international circuit. + +##### 3.3.6.4 *Subscriber recall* + +3.3.6.4.1 When a telex connection has to be established by recalling the calling subscriber (§ 3.3.6.1.2) the operator of the position controlling the call will first select from the two correspondents the one he can reach more easily. The procedure will be analogous to that described in §§ 3.3.6.1, 3.3.6.2 and 3.3.6.3, but before connecting the two subscribers the controlling operator will transmit **DF** to the calling subscriber to advise him that he is receiving a call that he has previously booked. + +3.3.6.4.2 The operator should not hold an international telex circuit while awaiting clearance of an occupied subscriber line. + +##### 3.3.6.5 *Operator recall* + +3.3.6.5.1 It is not possible to recall the operator of an international telex position to a connection already set up, except when applying Recommendation U.21 by agreement between Administrations. The operator-recall signal shall be acted upon by the controlling operator only. In the event of the assistance of any other operator being required, it will be obtained by the controlling operator. + +##### 3.3.6.6 *Instructions for foreign subscribers* + +3.3.6.6.1 All instructions necessary for the efficient handling of a subscriber's international telex traffic may be given to that subscriber only through the medium of the international terminal exchange to which he is connected. + +### 3.4 *Characteristics of subscribers' equipment* + +#### 3.4.1 *Network interface* + +3.4.1.1 The signals sent by the start-stop equipment used in the international telex service are those of International Telegraph Alphabet No. 2 as shown in Recommendation S.1. + +3.4.1.2 Where the subscriber's equipment is automatic, e.g. where a computer port simulates the functions of a teleprinter, the provisions of Recommendation U.40 should be observed, particularly concerning the number and timing of call attempts into the international telex network. + +3.4.1.3 In the particular case of an automatic terminal with several messages for the same telex destination, the provisions of Recommendation U.40 shall be applied to the first message. If this message is undeliverable, it is recommended that call attempts not be initiated in respect of the remaining messages. + +3.4.1.4 Where the subscribers equipment consists of a multiple-line terminal, the reaction to ineffective call attempts should be the same as that for a single line terminal and the provisions of Recommendation U.40 should be applied. + +3.4.1.5 The application of the above procedures is recommended in order to safeguard the quality of the international telex service. + +3.4.1.6 The general operational aspects and basic characteristics of telex terminals, derived from the service principles, are detailed in Recommendation F.59. + +#### 3.4.2 *Terminal availability* + +3.4.2.1 In the international telex service all terminals, like exchanges, shall provide continuous service. The terminal equipment of a free telex line shall accordingly be available at all times to answer a call and record a message from the calling subscriber whether or not an operator is present at the called terminal. + +3.4.2.2 The subscriber's equipment must be arranged in such a way that a call can be received, the answerback taken, the message transmitted and the connection cleared without the intervention of the called subscriber. + +Failure to abide by this condition should be indicated by failure to return the call connected signal in response to a valid call signal and results in the service signal **DER** being transmitted to the caller, unless the called terminal has requested temporary interruption of its service by reporting absent in which case that service signal may be replaced by **ABS**. The **DER** service signal may also take the expanded form as shown in Table 1/F.60. + +TABLE 1/F.60 +**Expanded form of DER code expression** + +| Code expression | Additional information characters a) | Meaning b) | +|-----------------|--------------------------------------|-----------------------------------------------------| +| DER | EXM | Out of order due to no text recording medium | +| | NAB | Out of order due to failure of answerback mechanism | +| | PFL | Out of order due to no power at the called terminal | + +a) These characters may appear anywhere on the line preceding the **DER** signal and are an integral part of the expanded service signal. + +b) Or technical failure presenting the same condition at the exchange. + +*Note 1* – The implementation of these expanded forms of the service signal is a national matter. + +*Note 2* – See also Recommendation U.45. + +3.4.2.3 In exceptional cases, Administrations may allow subscribers to dispense with the stipulation of § 3.4.2.2 for periods previously notified. In such cases, means must be provided for the transmission of one of the appropriate code expressions either automatically or, in the case of a manual exchange, by the incoming switchboard operator. + +3.4.2.4 While a call is established, the subscriber's equipment must be continuously ready to receive signals. Where applicable, the teleprinter motor must rotate continuously for the duration of an established connection. + +3.4.2.5 The subscriber's equipment should return its answer-back promptly in response to a **WRU** signal at any stage while the call is established. Nevertheless, following the initial exchange of answerbacks and in accordance with the S-Series Recommendations, a special sequence may be used to inhibit the answerback mechanisms where transfer to another alphabet is desired after call establishment. + +#### 3.4.3 Answerback composition + +3.4.3.1 The answerback code shall consist of 20 International Telegraph Alphabet No. 2 (ITA2) characters which should include: + +- a) the subscriber's number; +- b) if required, the machine identity letter or letters; +- c) optionally, an (abbreviated) name designating the subscriber; +- d) the telex network identification code, in accordance with the Recommendation F.69, preferably preceded by a space. + +3.4.3.2 Preferably, the various parts of the answerback code should be arranged in the order shown in § 3.4.3.1. Nevertheless, if Administrations alter on a network national basis the form of existing answerback codes or open new networks, they must ensure that the answerback code is composed in the form shown above. It should be noted, however, that there are some networks in existence which do not follow the prescribed order shown, and allowance should be made for these in any international relation. + +3.4.3.3 Where a telex subscriber has more than one telex line and automatic hunting facilities are provided by the terminating exchange, the answerback code of each machine of the group should, apart from the machine identification letter(s), be identical. + +3.4.3.4 If the order shown in § 3.4.3.1 is applied, the series of 20 characters in the answerback code, as shown in Recommendation S.6, should be as follows: + +- a) for machines without identification letters: + - figure-shift or (if permanently fitted or required by the network) letter-shift; + - carriage-return; + - line-feed; + - the national call number of the subscriber or (if letter-shift is fitted in the first position) figure-shift followed by the national call number of the subscriber; + - letter-shift; + - space; + - letters indicating as explicitly as possible the name of the telex subscriber; + - space; + - the one or two letters of the telex network identification code as listed in Recommendation F.69; + - letter-shift (if permanently fitted or required by the network); +- b) for machines with identification letters: + - figure-shift or (if permanently fitted or required by the network) letter-shift; + - carriage return; + - line-feed; + - the national number of the subscriber, or (if letter-shift is fitted in the first position) figure-shift followed by the national number of the subscriber; + - letter-shift; + - machine identification letter(s); + - space; + - letters indicating as explicitly as possible the name of the telex subscriber; + - space; + - the one or two letters of the telex network identification code as listed in Recommendation F.69; + - letter-shift (if permanently fitted or required by the network); +- c) for machines without identification letters and whose answerback code does not include letters indicating the (abbreviated) name of the subscriber: + - figure-shift or (if permanently fitted or required by the network) letter-shift; + - carriage return; + - line-feed; + - the national number of the subscriber, or (if letter-shift is fitted in the first position) figure-shift followed by the national number of the subscriber; + - letter-shift; + - space; + - the one or two letters of the telex network identification code as listed in Recommendation F.69; + - carriage-return; + - line-feed; + - letter-shift (if permanently fitted or required by the network); + +- d) for machines with identification letters, but whose answerback codes do not include letters indicating the (abbreviated) name of the subscriber: + - figure-shift or (if permanently fitted or required by the network) letter-shift; + - carriage-return; + - line-feed; + - the national number of the subscriber, or (if letter-shift is fitted in the first position) figure-shift followed by the national number of the subscriber; + - letter-shift; + - machine identification letter(s); + - space; + - the one or two letters of the telex network identification code as listed in Recommendation F.69; + - carriage-return; + - line-feed; + - letter-shift (if permanently fitted or required by the network). + +3.4.3.5 Unused places should be filled by the necessary number of letter-shifts, which should preferably be inserted before the telex network identification code. + +3.4.3.6 For the particular case of answerback codes generated by teleprinters (or equivalent terminal devices) on ships, see Recommendation F.130. + +### 3.5 *Restriction on the use of a telex station* + +3.5.1 Pursuant to Articles 19 and 20 of the Convention [2], Members may exercise their rights concerning stoppage of telecommunications and suspension of services in certain exceptional circumstances. + +### 3.6 *Subscribers' operating procedure for telex calls* + +3.6.1 Administrations may wish to advise their customers on how to make best use of the international telex service. To that end, an example for such provisions is given in Annex A. Such instructions may also include information regarding the code expressions used in the international telex service, which are listed in § 4.1 and which could usefully be inserted within national telex directories. + +### 3.7 *Directories* + +#### 3.7.1 *Compilation of directories* + +3.7.1.1 As far as possible each Administration shall publish a directory of its subscribers at least once a year. + +3.7.1.2 Directories should not be larger than 210 × 297 mm (A4). + +3.7.1.3 The directory shall be composed of two separate lists, a list of subscribers and a list of answerback codes. + +3.7.1.3.1 The list of subscribers shall be drawn up as follows: + +either    a) places where stations are located, classified in alphabetical order, and +               within that classification, subscribers' names arranged in alphabetical order; + +*Example:* + +| Place | Subscriber's name and address | Subscriber's exchange
(where necessary) | National
number | Answer-back
code | +|-------|-------------------------------|--------------------------------------------|--------------------|---------------------| +|-------|-------------------------------|--------------------------------------------|--------------------|---------------------| + +or        b) subscriber's names only, arranged in alphabetical order (subscribers of the same name being classified in the alphabetical order of the place in which they are located). + +*Example:* + +| Subscriber's name and address,
including the locality | Subscriber's exchange
(where necessary) | National number | Answer-back
code | +|----------------------------------------------------------|--------------------------------------------|-----------------|---------------------| +|----------------------------------------------------------|--------------------------------------------|-----------------|---------------------| + +3.7.1.3.2 The list of answerback codes shall be compiled in numerical order. + +*Example:* + +| Answer-back
code | Subscriber's name and place | Subscriber's exchange
(where necessary) | National number
(where necessary) | +|---------------------|-----------------------------|--------------------------------------------|--------------------------------------| +|---------------------|-----------------------------|--------------------------------------------|--------------------------------------| + +3.7.1.3.3 However, in cases where the answerback codes are not yet arranged in the order laid down in § 3.4.2.1, the list of answerback codes may be compiled in alphabetical order. + +3.7.1.4 The directories sent to Administrations shall be set up in Roman letters. The call-number published shall be that which the calling subscriber has to transmit in order to obtain the called subscriber after he has followed the procedure prescribed in his own country to gain access to an international circuit. + +3.7.1.5 When directories are written in a language other than the language used in that country, they shall be accompanied by an explanatory note to facilitate the use of such directories. This note shall be drawn up in whatever official language of the Union has been agreed upon by the Administrations concerned. + +3.7.1.6 Each directory should also contain: + +- the list of destination codes for the countries to which the national subscribers have access. These codes shall be supplemented by the access prefix for the international telex network; +- a list of the telex network identification codes of these countries; +- optionally, subscribers operating procedures in accordance with § 3.6.1. + +3.7.1.7 The telex network identification code of the country (or network) should be shown in large type on the cover and on the spine (i.e. the bound edge) of each directory. + +3.7.1.8 To facilitate reference in international telex centres to a number of directories issued by various Administrations, the orientation of the printing on the spine should be identical. When directories are stored vertically on a bookshelf, identification codes should all be horizontal and other particulars, which cannot conveniently be printed horizontally, should be read from bottom to top. + +#### 3.7.2 *Supply of directories* + +3.7.2.1 Each Administration shall supply, free of charge, to the Administrations with which a telex service exists, a sufficient number of copies of its subscribers' directories for official use. The number of such copies shall be fixed in advance by mutual agreement and shall be regarded as applying until a request to change it is received. Such request must be made not later than 1 February each year. + +3.7.2.2 Each Administration shall supply, against payment, to the Administrations and recognized private operating agencies with which a telex service exists, a number of its subscribers' directories to be put on sale. The number of copies intended for sale shall be fixed in advance by mutual agreement and shall be regarded as applying until a request to change it is received. Such requests must be made not later than 1 February each year. + +3.7.2.3 A subscriber wishing to obtain a copy of the telex directory of another country must apply to his own Administration. If an application for its directory is received direct by an Administration from a subscriber in a foreign country, the request shall be forwarded by that Administration to the Administration of the subscriber's country. + +3.7.2.4 An Administration that has supplied directories of its country intended for sale to another Administration shall indicate the equivalent in special drawing rights (SDR) or gold francs of the sale price of the directories applied in the country of origin plus any postal charges. + +## 4 Miscellaneous provisions + +### 4.1 *Code expressions used in the international telex service* + +4.1.1 The following code expressions may be used in operator-to-operator correspondence, generated by the network or used by the originating subscriber for special purposes. In certain cases, they may also be followed by the clear signal. Subscribers may also find it useful to use some of these codes during a conversational call. + +| | | +|-------------------------|--------------------------------------------------------------------------------| +| ABS | Absent subscriber/office closed | +| ADD | Please input your international telex number | +| ANUL | Delete | +| BCT | Broadcast call | +| BK | I cut off | +| BMC | No end of message or end of transmission received, therefore message cancelled | +| CFM | Please confirm/I confirm | +| CI4) | Conversation impossible | +| COL5) | Collation please/I collate | +| CRV | Do you receive well?/I receive well | +| DER | Out of order (see Table 1/F.60) | +| DF | You are in communication with the called subscriber | + +--- + +4) This code expression is intended to be generated only by automatic means and not normally used in service correspondence between operators. + +5) Repetition of, for example, isolated figures, mixed figures or other sensitive information. + +| | | +|------------------------------------------------------|----------------------------------------------------------------------------------------------------------------| +| EXM | Connection cleared due to exhaustion of text recording medium at either the called or calling terminal | +| FMT | Format error | +| GA | You may transmit/May I transmit? | +| IAB | Invalid answerback from destination | +| IMA | Input message acknowledgement | +| INF | Subscriber temporarily unobtainable, call the information service | +| ITD | Input transaction accepted for delivery | +| ITL | I transmit later | +| JFE | Office closed because of holiday | +| LDE | Maximum acceptable message length or duration has been exceeded | +| MNS | Minutes | +| MOM | Wait/waiting | +| MUT | Mutilated | +| NA | Correspondence with this subscriber is not admitted | +| NC | No circuits | +| NCH | Subscriber's number has been changed | +| NDN | Non-delivery notification | +| NI | No line identification available | +| NP | The called party is not, or is no longer, a subscriber | +| NR | Indicate your call number/My call number is . . . | +| OCC | Subscriber is engaged | +| OK | Agreed/Do you agree? | +| PPR | Paper | +| R | Received | +| RAP | I shall call you back | +| RDI | Redirected call | +| REF | Reference of the message delivered to the telex side from a conversion facility for telex/teletex interworking | +| REI | Address validation failure/non-compliant answerback received | +| RPT | Repeat/I repeat | +| RSBA | Retransmission still being attempted | +| SSSS | Change of alphabet | +| SVP | Please | +| T6)
(or figure 5) 6) | Stop your transmission | +| TAX | What is the charge?/The charge is . . . | +| TEST MSG | Please send a test message | +| THRU | You are in communication with a telex position | + +--- + +6) To be repeated until the transmission is brought to a stop. + +| | | +|--------------|----------------------------------------------------------------------------| +| TMA | Maximum number of addresses exceeded | +| TPR | Teleprinter | +| TTX | Designation of the conversion facility (CF) for telex-teletex interworking | +| VAL | Validation response | +| W | Words | +| WRU | Who is there? | +| XXXXX | Error | + +4.1.2 Code expressions received when interworking with users of the IPM service are given in Recommendation F.421(F.84). + +4.1.3 Code expressions received when interworking with data terminal equipments connected to a packet switched public data network (PSPDN) are given in Recommendations F.83 and U.203. + +### 4.2 *Printing of telex numbers* + +4.2.1 Standardized printing of telex numbers on letterheads is especially valuable for international purposes. It is recommended that this printing contain the word *Telex*, followed by the subscriber's answerback code, for example: + +Telex 31005 SHELL NL + +4.2.2 In those cases where there is no number and/or no telex network identification code in the answerback code the word *Telex* should be followed by the *Telex* number and the complete answerback within inverted commas, for example: + +Telex 24935 "LAPORTCHEM LDN" + +## ANNEX A + +(to Recommendation F.60) + +### **Operating procedure for telex calls** + +### A.1 *Setting up a telex call* + +A.1.1 In the automatic service the subscriber selects the telex number required. The establishment of a connection is recognized by the reception of the answerback from the wanted subscriber. The caller should check whether the answerback he has received is in fact that of the appropriate subscriber. If it is not, he should disconnect and reselect the number of the subscriber required. When the caller has received the correct answerback he should release his own answerback before starting the transmission of a message. + +A.1.2 In the manual or semi-automatic service a connection is established through the intermediary of an international telex position. The establishment of a connection is indicated by the reception of the answerback from the called subscriber, followed by that of the calling subscriber. The subscribers must not intervene during this procedure. The caller checks whether the answerback he has received is in fact that of the wanted subscriber. If it is not, he should disconnect and inform the international telex position accordingly. + +A.1.3 If the called telex station is continuously unattended, e.g. automatic answering equipment, store-and-forward equipment or other reception equipment, a dialogue with the called telex station is impossible. + +A.1.4 If the telex machine of the called subscriber is attended, a dialogue between subscribers is possible, in which case the end of each transmission should be indicated by the plus sign and question mark (+?) followed by a letter-shift, thus inviting the other party to transmit in his turn. + +### A.2 *Setting out the message* + +A.2.1 Where the recipient may be in doubt about the identity of the caller, it is recommended that the calling subscriber indicate: + +- a) name and place of the sender, preceded by the word **FROM**; +- b) name and place of the addressee, preceded by the word **TO**; +- c) if required, name and place of information addressee(s), preceded by the word **COPY**. + +Practice has shown that the observation of this recommendation will often save additional work for the recipient especially when messages have to be distributed for action. + +A.2.2 After the exchange of answerbacks as set out in § A.1 and following the recommendation in § A.2.1, the calling subscriber can transmit his message, for which the following uniform procedure is recommended: + +- a) start a new line and mention own reference, if any, and the date of dispatch; +- b) start a new line and indicate the priority of the message, if desirable, such as **URGENT**, **VERY URGENT**, etc.; +- c) start a new line and indicate the subject if appropriate and/or the name of the person or department for whose attention the message is intended; +- d) start a new line and mention any references, such as **REF YOUR TELEX 123 OF 15.7**, **REF YOUR LETTER 456 OF 25.7**, **REF OUR TELECON**, etc.; +- e) start a new line and transmit the text of the message; +- f) after having completed the message, start a new line and transmit a plus sign (+) indicating the end of the message or **NNNN**; +- g) obtain the answerback of the called subscriber, check it in order to be sure that the connection is still in good order and generate own answerback; +- h) if there are more messages, they should be separated from each other by at least 8 line-feeds, after the exchange of the answerbacks as mentioned in g); +- i) after transmission of the last message and the exchange of the answerbacks send at least 8 line-feeds and give the clearing signal. + +### A.3 *Additional instructions* + +A.3.1 When a group, or part of a group, is composed of a whole number and an ordinary fraction, the fraction should be separated from the number by means of a dash without space. + +Example: for *one and three quarters*: **1-3/4** + +A.3.2 In order to avoid misunderstanding, a whole number, a fractional number, or a fraction followed by a % or ‰ should be transmitted by joining them up to the % or ‰. sign by a dash, or transmitted in full as appropriate. + +Examples: for 2% transmit **2-0/0** or **2 PER CENT** + +for 4 1/2 ‰, transmit **4-1/2-0/00** or **4-1/2 PER MILLE** + +A.3.3 When important figures or words appear in the text it is desirable to repeat them immediately after the group followed by a space either in brackets or preceded by the word **REPEAT**. + +Examples: **1500 (1500)** + +**1500 REPEAT 1500** + +**NOT REPEAT NOT** + +A.3.4 To pass to the beginning of the next line, i.e. to start a new line, first press carriage-return and then line-feed. Some terminals may also provide a new line key which combines these functions. + +A.3.5 An error is corrected in the following manner: + +- a) In manual transmission, by the sequence **XXXXX** (letter **X** repeated five times and followed by a space)7) joined to the erroneous word. + +Example: **PLEASE DISPATCH FITXXXXX FIVE PARCELS** + +- b) In automatic transmission, when preparing perforated tape, by *backspacing* (if necessary by counting the number of characters to be erased, including spaces and shifts, and backspacing by that number) to the erroneous character and then operating the letter-shift key to erase all the characters up to and including the last punched character. Then start again with the character to be sent immediately after the last correctly punched character. +- c) If the procedure mentioned under b) for one reason or other cannot be followed, an error could be corrected as indicated under a). +- d) If an error is detected after the transmission of a message but before the exchange of the answerbacks it should be corrected by clearly indicating under the text of the message what change is required, e.g.: + +**CORRECT 4TH WORD 2ND LINE TO READ NOT REPEAT NOT** + +**DELETE 4TH WORD 2ND LINE** + +**INSERT THE WORD "WITH" BETWEEN THE 4TH AND 5TH WORD OF 2ND LINE** + +A.3.6 If, for some reason or other, a message has to be cancelled during transmission this should be clearly indicated on a new line by transmitting three times the word **ANUL**. + +A.3.7 In preparing a perforated tape for automatic transmission, care should be taken that: + +- a) the signal *Who are you?* (figure case D) does not appear on the tape, in order to prevent the text from being garbled by the returned answerback of the other party; +- b) the procedure mentioned in § A.3.4 above is followed; +- c) the tape is perforated to the end with a series of letter-shifts. + +A.3.8 Since figure case signs or letters coupled with the letters **F**, **G** and **H** are not universally standardized, they must not be used in international communications, but should be transmitted in full, e.g. **DOLLARS**, **POUNDS STERLING**, etc. + +A.4 *Ineffective call attempts when calling from a manual terminal* + +A.4.1 If an attempt to set up a call is unsuccessful (for example, if the wanted subscriber is engaged), the network will return a service code indicating the reason. It will also disconnect automatically except in the particular cases of changed number interception and call redirection as described in Recommendation U.41. + +A.4.2 The commonly used service codes, their meaning and the appropriate action for the subscriber to take are given in Table A-1/F.60. + +A.5 *Idle connections* + +A.5.1 To avoid incurring unnecessary charges, care should always be taken to clear the connection once all messages have been sent (see § A.2.2). + +A.5.2 In some national telex networks a prolonged idle condition may cause a warning message to appear followed, possibly after a short delay, by clearing. Normally, this action will not commence until the idle condition has persisted for a period to be determined by the Administration. + +--- + +7) It is to be noted that the sequence **E E E** (space and letter **E** repeated three times followed by a space and repetition of the last correct word) is in use but is not preferred. + +A.5.3 Should a subscriber wish to maintain an idle condition for a period in excess of that stipulated by the Administration, it is advisable to make that intention clear by sending at least one character periodically. + +TABLE A-1/F.60 + +### **Procedure after ineffective call attempts when calling from a manual terminal** + +| Service code | Meaning | What to do | +|----------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------| +| OCC
NC | The called subscriber is engaged
No circuits (or equipment) are available at the moment | Wait for at least one minute, then call again | +| DER | Called subscriber's terminal is out of order, or temporarily out of service whilst paper, ribbon or tape is replaced | Check the number and try again after about 5 minutes. If DER persists, refer the problem to the Telex Enquiries Service | +| ABS
NA
NP
NCH | Called subscriber is absent. Office is closed
Access to called service not admitted
The called number is not, or is no longer, a working line
Called subscriber's number has been changed ( NCH may be followed by the new number) | Check the number. If correct, try again. If the same service signal is returned refer the problem to the Telex Enquiries Service | + +*Note* – Recommendation U.40 describes the procedure after ineffective call attempts when calling from an automatic telex terminal. + +## ANNEX B + +(to Recommendation F.60) + +## **Terms and definitions relating to the international telex service** + +### **B.1 emergency routes** + +*F: voies de secours* + +*S: rutas de emergencia* + +The circuit(s) to be used in case of complete interruption or major breakdown of the primary and secondary routes. The emergency routes may pass through any country. + +### **B.2 Government telex calls** + +*F: communications télex d'Etat* + +*S: comunicaciones télex de Estado* + +Those telex calls originating with one of the authorities which enjoy the advantages of Government telecommunications, in accordance with the International Telecommunication Convention [2]. + +### **B.3 international telex position** + +*F: position télex internationale* + +*S: posición télex internacional* + +Manual position in an international telex centre for establishing telex calls between two countries. + +### **B.4 ordinary private telex calls** + +*F: communications télex privées ordinaires* + +*S: comunicaciones télex privadas ordinarias* + +All telex calls other than: + +- i) service telex calls, including requests for information; +- ii) safety of life telex calls; +- iii) Government telex calls; +- iv) privilege telex calls. + +### **B.5 primary routes** + +*F: voies primaires* + +*S: rutas primarias* + +The circuits normally used in a given relation. + +### **B.6 safety of life telex calls** + +*F: communications télex relatives à la sécurité de la vie humaine* + +*S: comunicaciones télex relativas a la seguridad de la vida humana* + +Those telex calls requested in accordance with Article 5.1 of the *International Telecommunication Regulations* (Melbourne, 1988). + +### **B.7 secondary routes** + +*F: voies secondaires* + +*S: rutas secundarias* + +The circuits to be used when the primary routes are congested. The secondary route(s) may pass through the same countries as the primary routes or through different countries. In manual and semi-automatic operation, secondary routes may also be used when the transmission on the primary route is not sufficiently good, or if traffic is to be handled outside the normal hours of service on the primary routes. + +### **B.8 service telex calls** + +*F: communications télex de service* + +*S: comunicaciones télex de servicio* + +Those telex calls that relate to the working of the international telecommunication services and are exchanged among authorized bodies or officials (see Recommendation F.17). + +### B.9 **(telex) relation** + +*F: relation (téléx)* + +*S: relación (téléx)* + +Exchange of traffic between two terminal countries, always referring to a specific service if there is between their Administrations: + +- a) a means for the exchange of traffic in that specific service: + - over direct circuits (direct relation); or + - via a point of transit in a third country (indirect relation), and +- b) normally, the settlement of accounts. + +### B.10 **privilege telex calls** + +*F: communications téléx privilégiées* + +*S: comunicaciones téléx privilegiadas* + +As defined in Article 2.5.1 of the *International Telecommunication Regulations* (Melbourne, 1988). + +### B.11 **international store-and-forward** + +*F: enregistrement et retransmission en international* + +*S: almacenamiento y retransmisión para internacional* + +Where a subscriber in country A accesses the store-and-forward unit in country B for the transmission of messages to that country. + +### B.12 **interconnected store-and-forward** + +*F: enregistrement et retransmission avec interconnexion* + +*S: almacenamiento y retransmisión con interconexión* + +Where the store-and-forward unit in country A is connected to the store-and-forward unit in country B for the transmission of messages between the two countries. + +### B.13 **interworking** + +*F: interfonctionnement* + +*S: interfuncionamiento* + +The facility of sending and receiving information between a telex equipment and an equipment/user of another service, such as interpersonal messaging, videotex, etc. + +### B.14 **conversion facility (CF)** + +*F: unité de conversion* + +*S: unidad de conversión* + +Fully automatic system performing the necessary conversion between the teletex service and the telex service (see Recommendation F.201). + +### B.15 **one-stage/two-stage selection procedure for telex to teletex direction of interworking** + +*F: procédures avec la sélection en une ou deux étapes pour l'interfonctionnement dans le sens téléx vers télétex* + +*S: procedimientos con marcación monoetapa o bietapa para el interfuncionamiento de téléx a teletex* + +Addressing of the teletex terminal by the telex terminal can be done, either by sending the total selection information in one phase to the CF or by calling first the CF (first stage of the selection), and by sending the teletex address after the connection to the CF has been established (second stage of the selection). + +### **B.16 store-and-forward conversion facility (CF using store-and-forward principles)** + +*F: unité de conversion avec enregistrement et retransmission (unité de conversion) utilisant les principes d'enregistrement et retransmission* + +*S: unidad de conversión con almacenamiento y retransmisión (unidad de conversión) que utiliza los principios de almacenamiento y retransmisión* + +CFs that “store” the received telex (or teletex) messages before “forwarding” them to the called teletex (or telex) terminal (see Recommendation F.201). + +### **B.17 real-time conversion facility (real-time interworking)** + +*F: unité de conversion en temps réel (interfonctionnement en temps réel)* + +*S: unidad de conversión en tiempo real (interfuncionamiento en tiempo real)* + +Such a CF shall transfer a message, in a unique communication, from a telex terminal to a teletex terminal, and from a teletex terminal to a telex terminal, without storage of the message (see Recommendation F.201). + +### **B.18 validation of the called teletex terminal [validation result (positive or negative)]** + +*F: validation du terminal télétex demandé [résultat de la validation (positif ou négatif)]* + +*S: validación del terminal teletex llamado [resultado de validación (positivo o negativo)]* + +This validation is performed by the CF to verify that the teletex terminal is an available one, i.e. either the teletex terminal has been called with this address (validation call) or this address has been controlled in a data base (see Recommendation F.201). + +### **B.19 message deposit/message delivery (text deposit/delivery)** + +*F: dépôt du message/remise du message (dépôt/remise du texte)* + +*S: depósito de mensaje/entrega de mensaje (depósito/entrega de texto)* + +The message “deposit” is the sending by the calling terminal of the whole message to the store and forward CF before its further “delivery” to the called terminal (see Recommendation F.201). + +### **B.20 on-line delivery acknowledgement** + +*F: avis de remise en ligne* + +*S: acuse de recibo de entrega en línea* + +The on-line delivery acknowledgement facility gives to the waiting telex (i.e. having maintained the connection with the CF after its message deposit) the opportunity to receive “on-line” a proof of the CF’s message delivery to the teletex terminal, provided the call establishment to the teletex terminal has been performed within 30 seconds counted after the end of the message input (see Recommendation F.201). + +### **B.21 non-delivery notification: (NDN)/positive delivery notification (PDN)** + +*F: avis de non remise/avis de remise positive* + +*S: notificación de no entrega/notificación de entrega positiva* + +If the CF has not been able to deliver the message to the called terminal despite the performance of a defined cycle of delivery attempts on the called terminal network (each network has a specific cycle) and within a maximum of a T2-defined duration, the CF should send an NDN to the calling user to indicate to him that his message has not been delivered to the called terminal and that no further delivery action will be taken by the CF (see Recommendation F.201). + +*Note 1* – The NDN facility is not provided in the first method of interworking for the telex to teletex direction (see Recommendation F.201). + +*Note 2* – The PDN facility, i.e. the ability of the CF to send back a proof of the delivery, is for further study. + +### B.22 **CF prefix** + +*F: préfixe d'unité de conversion* + +*S: prefijo de unidad de conversión* + +In the first method of interworking, the “CF prefix” is the special number (up to 7 digits) to be put before the called teletex number, to indicate that the total telex selection is for reaching a teletex terminal (see Recommendation F.201). + +### B.23 **CF national number** + +*F: numéro national d'unité de conversion* + +*S: número nacional de unidad de conversión* + +In the second method of interworking, the “CF national number” is the national telex number of the CF, given to the called telex users at the beginning of the telex delivery phase of the teletex to telex exchange for further use of interworking with the teletex of the CF’s country (see Recommendation F.201). + +### B.24 **input message acknowledgement (IMA)** + +*F: accusé de dépôt* + +*S: acuse de recibo de mensaje introducido* + +The IMA message sent by the CF to the telex user is used to indicate that the message has been well received by the CF and to give to the telex user a unique reference for this message. This reference should be used again when sending an NDN (see Recommendation F.201). + +The following terms bear an international electrotechnical vocabulary (IEV) number which follows the definition. + +### B.25 **telex conversation mode** + +*F: mode conversationnel télex* + +*S: modo conversacional télex* + +The use of a telex connection for a dialogue or exchange of information between two *terminals*. + +721.53.05 + +### B.26 **access to the public telegram service** + +*F: accès au service télégraphique public* + +*S: acceso al servicio público de telegramas* + +Provision for a *telex terminal* to send and receive *telegrams* to and from the *public telegram service*. + +721.53.07 + +### B.27 **user class-of-service** + +*F: catégorie d'usager* + +*S: clase de servicio de usuario* + +The category that defines the characteristics of a call available to a user of a public telecommunication service. + +*Note* – The characteristics for a user class of service could be, for example, *binary rate*, terminal operating mode, code structure, *access barred*. + +721.53.08 + +### B.28 **public telex booth** + +*F: cabine télex publique* + +*S: cabina télex pública* + +Telex *terminal* available to the public (i.e. non-subscribers). + +721.53.09 + +### **B.29 outgoing only terminal** + +*F: terminal spécialisé en sortie* + +*S: terminal de salida solamente* + +A *terminal* that can make outgoing calls to the network but which is prevented from receiving incoming calls. + +721.53.10 + +### **B.30 incoming only terminal** + +*F: terminal spécialisé en entrée* + +*S: terminal de llegada solamente* + +A *terminal* that can receive incoming calls from the network but which is prevented from making outgoing calls. + +721.53.11 + +### **B.31 access barred** + +*F: interdiction d'accès* + +*S: prohibición de acceso* + +A function of a telecommunication network that bars calls to or from certain subscribers, from or to certain services, routes or *terminals*. + +721.53.12 + +### **B.32 restricted service** + +*F: service restreint* + +*S: servicio restringido* + +A service whereby a subscriber may have *access barred* from his terminal installation to certain services, routes or *terminals* which would normally be accessible to all customers. + +721.53.13 + +### **B.33 priority** + +*F: priorité relative* + +*S: prioridad relativa* + +The possibility of setting up a call from a nominated *terminal* on a *private network* or *closed user group*, by assigning to it, at each stage of selection, priority over all other calls of lower priority that are in the process of being established. The possibility may apply either to every call or only to nominated calls from such a privileged terminal. + +721.53.14 + +### **B.34 absolute priority** + +*F: priorité absolue* + +*S: prioridad absoluta* + +The possibility of setting up a call from a nominated *terminal* on a *private network* or *closed user group*, by assigning to it at each stage or certain stages of selection, priority over all other calls of lower priority that are established. The possibility may apply either to every call or only to nominated calls from such a privileged terminal. + +721.53.15 + +### **B.35 priority for called subscriber** + +*F: abonné prioritaire en demandé* + +*S: prioridad del abonado llamado* + +A subscriber who has the facility of *priority* or *absolute priority* for all calls or for certain calls only to his *terminal*. This facility is activated by the sending of an appropriate signal by the calling terminal. + +*Note* – There may be several priority levels, each conferring relative or absolute priority with respect to lower levels. + +721.53.16 + +### **B.36 in-local override** + +*F: prise de pas sur le fonctionnement en local* + +*S: contraorden del funcionamiento en local* + +A facility of the network to override a *terminal* working *in local*, for the purpose of connecting an incoming call to that terminal. + +721.53.17 + +### **B.37 direct outgoing selection** + +*F: prise directe* + +*S: selección directa de salida* + +A facility that permits a *terminal* in a *private network* to set up a call to another network without human intervention in the private network. + +721.53.18 + +### **B.38 direct incoming selection** + +*F: sélection directe à l'arrivée* + +*S: selección directa de llegada* + +A facility that permits a *terminal* in a telex network to set up a call to a terminal designated by the caller in a *private network* without human intervention in the private network. + +721.53.19 + +### **B.39 direct incoming selection with integrated numbering** + +*F: sélection directe à l'arrivée avec numérotation intégrée* + +*S: selección directa de llegada con numeración integrada* + +*Direct incoming selection* using a single *selection sequence* made up from certain figures (digits) identifying the *private network* followed by certain figures identifying the called *terminal* in that network. The complete sequence of figures constitute a complete *address* integrated into the numbering plan of the telex network. + +721.53.20 + +### **B.40 direct incoming selection with two-stage selection** + +*F: sélection directe à l'arrivée avec numérotation en deux temps* + +*S: selección directa de llegada con marcación bietapa* + +*Direct incoming selection* using two *selection sequences* to select the required *terminal* in the *private network*. The first sequence identifies the private network, the second sequence identifies the terminal in this network. Only the first sequence is integrated into the numbering plan of the telex network. + +721.53.21 + +### **B.41 closed user group** + +*F: groupe fermé d'usagers* + +*S: grupo cerrado de usuarios* + +A user group on the public switched network whose *terminals* have the facility to communicate only with each other. + +*Note –* A terminal may belong to more than one closed user group. + +721.53.22 + +### **B.42 partially closed user group** + +*F: groupe partiellement fermé d'usagers* + +*S: grupo de usuarios parcialmente cerrado* + +A user group where certain terminals may make calls to or receive calls from any other terminals normally accessible in the public switched network, the other terminals having the facility to communicate only with the user of the group. + +*Note –* In some cases the external access for nominated terminals is limited to outgoing calls. + +721.53.23 + +### **B.43 user facility** + +*F: service complémentaire* + +*S: facilidad de usuario* + +A facility which may be provided on request to a user of the telecommunication network in addition to the normal service provided. + +*Note –* A user facility may be provided on a per call basis or for an agreed period of time. + +721.53.25 + +### **B.44 automatic calling** + +*F: appel automatique* + +*S: llamada automática* + +The sequence of operations required by the network procedure to set up a connection without manual intervention at the calling *terminal*. + +721.53.26 + +### **B.45 automatic answering** + +*F: réponse automatique* + +*S: respuesta automática* + +Answering in which the called terminal automatically responds to the *calling signal* and the call may be established whether or not the called terminal is attended. + +721.53.27 + +### **B.46 manual answering** + +*F: réponse manuelle* + +*S: respuesta manual* + +Answering in which a call is established only if the called user signals his readiness to receive it by means of a manual operation. + +721.53.28 + +### **B.47 automatic identification** + +*F: identification automatique* + +*S: identificación automática* + +The transmission without manual intervention of the identification of the calling *terminal* to the connected terminal or vice versa, or the identification of terminals to one another when a connection is established. + +*Note* – The identification may be provided by the network or by the terminal. + +721.53.29 + +### **B.48 line identification by the network** + +*F: identification de ligne par le réseau* + +*S: identificación de línea por la red* + +Transmission by the network, in response to a request from either of two connected parties, of an appropriate line or address identification. + +721.53.30 + +### **B.49 automatic date and time indication** + +*F: indication automatique de date et d'heure* + +*S: indicación automática de fecha y hora* + +Automatic indication by the network of date and time of the commencement of a call either to the calling *terminal* or to both the calling and the called terminals. + +721.53.31 + +### **B.50 indication of duration** + +*F: indication de durée* + +*S: indicación de duración* + +The indication by the network to the paying *terminal* of the chargeable time of a call prior to the release of the paying terminal or by recall at a convenient time. + +*Note* – This information may be provided automatically or on demand. + +721.53.32 + +### **B.51 indication of charge** + +*F: indication de taxe* + +*S: indicación del importe de la comunicación* + +The indication by the network to the paying *terminal* of the charge of a call prior to the release of the paying terminal or by recall at a convenient time. + +*Note* – This information may be provided automatically or on demand. + +721.53.33 + +### **B.52 statement of call account** + +*F: décompte de taxes de communications* + +*S: estado de cuentas de comunicaciones* + +The sending by the network upon request of a subscriber, an Administration, *closed user group* or *private network*, of a detailed account of his call charges either since his last request or over a nominated period. + +721.53.34 + +### **B.53 shared terminal** + +*F: terminal partagé* + +*S: terminal compartido* + +A facility offered to certain subscribers permitting the use of the same *terminal*, sharing the corresponding costs and charges. + +721.53.35 + +### **B.54 accounts for shared terminal** + +*F: décompte pour terminal partagé* + +*S: cuentas de un terminal compartido* + +Provision of separate accounts to users of a *shared terminal*. + +721.53.36 + +### **B.55 storage of call content** + +*F: archivage des messages* + +*S: almacenamiento del contenido de las comunicaciones* + +The storage for a specified length of time by the network at the subscriber's request of the contents of some or all of his calls sent or received. + +721.53.37 + +### **B.56 retrieval of stored call content** + +*F: consultation d'archives* + +*S: consulta del contenido almacenado de las comunicaciones* + +The transmission of the call contents to subscribers who had previously requested *storage of all content*. + +721.53.38 + +### **B.57 statistics on request** + +*F: statistiques sur demande* + +*S: estadísticas a petición* + +Provision for the network to send to the subscriber at his request, details of his calls under defined headings, e.g. international calls, national calls, calls to certain subscribers or total of all calls. + +721.53.39 + +### **B.58 recorded message** + +*F: réponse par message enregistré* + +*S: mensaje registrado* + +A facility provided by the called subscriber or terminating network, whereby incoming calls to that subscriber may be connected to a transmitter for recorded *messages*. + +721.53.40 + +### **B.59 camp-on: connect when free** + +*F: attente sur occupation* + +*S: conexión tras liberación* + +The holding by the network of a call attempt that was unsuccessful due to the called *terminal(s)* being busy or due to network congestion, with subsequent automatic connection as soon as possible. + +721.53.41 + +### **B.60 camp-on with recall** + +*F: attente sur occupation avec rappel* + +*S: conexión tras liberación con rellamada* + +A *camp-on* with the release of the calling *terminal* and recall as soon as possible. + +721.53.42 + +### **B.61 absent subscriber service (in telegraphy and data communication)** + +*F: service des abonnés absents (en télégraphie et transmission de données)* + +*S: servicio de abonado ausente (en telegrafía y comunicación de datos)* + +A facility that permits the calling *terminal* to be advised automatically by a *service signal* that, due to an action of the called subscriber, the latter's terminal is not available for calls. + +721.53.43 + +### **B.62 call re-direction** + +*F: renvoi d'appel* + +*S: redireccionamiento de la llamada* + +A facility that permits a call to be redirected to a previously nominated alternative destination upon the request of the called subscriber with advice by a *service signal* to the calling terminal. + +721.53.44 + +### **B.63 changed address interception** + +*F: interception suite à un changement de numéro* + +*S: interceptación de cambios de dirección* + +Automatic advice provided by the network to a calling *terminal* of a called terminal's new address followed either by *call redirection* or by release of the calling terminal. + +*Note* – It is also possible to simply send a *service signal* followed by release. + +721.53.45 + +### **B.64 store-and-forward** + +*F: enregistrement et retransmission* + +*S: almacenamiento y retransmisión* + +The process of storing *messages* or parts of messages and their subsequent transmission to the designated address or addresses. + +721.53.46 + +### **B.65 storage installation** + +*F: installation d'enregistrement* + +*S: instalación de almacenamiento* + +An installation that provides a *store-and-forward* function. + +*Note –* This installation may be provided at a *terminal* or at a centralized installation. + +721.53.47 + +### **B.66 redirection address** + +*F: adresse de réacheminement* + +*S: dirección de redireccionamiento* + +Information sent in the backward direction consisting of a number of *address* signals indicating the complete address to which the call is to be or has been redirected. + +721.53.48 + +### **B.67 delayed delivery** + +*F: remise différée* + +*S: entrega diferida* + +A *store-and-forward* process in which the re-transmission of stored *messages* is delayed until a predetermined period. + +721.53.49 + +### **B.68 message priority** + +*F: priorité des messages* + +*S: prioridad de los mensajes* + +A facility within a *store-and-forward*, or *message switching* system that enables a subscriber to attach to his *message* one of a number of levels of priority which has been provided. + +721.53.50 + +### **B.69 message spacing** + +*F: espacement des messages* + +*S: separación de los mensajes* + +A facility whereby a subscriber may request the network to transmit several *line feed* characters to his *terminal* at the end of each successful call, before clear down, for the purpose of providing a blank space between adjacent printed *messages*. + +721.53.51 + +### **B.70 header** + +*F: en-tête* + +*S: encabezamiento* + +The initial part of a *message* or *packet* which contains the service information. + +721.53.52 + +### **B.71 booked call** + +*F: appel à heure fixe, vacation* + +*S: llamada a hora convenida* + +A process whereby a subscriber may have his *terminal* called by the network at a given time, with or without an audible signal. + +721.53.53 + +### **B.72 network recall** + +*F: rappel du réseau* + +*S: rellamada a la red* + +The recall of the network by a subscriber during the *message* phase of the call to request facilities. + +721.53.54 + +### **B.73 multi-address call** + +*F: communication à destinations multiples* + +*S: comunicación multidireccional* + +A call set up by the network in which more than one called *terminal* is involved. + +721.53.55 + +### **B.74 conference call** + +*F: communication de conférence* + +*S: comunicación conferencia* + +A *multi-address call* in which the signals which may be transmitted by any one of the terminals are received simultaneously by all other *terminals*. + +*Note* – The order in which the terminals may transmit shall be mutually agreed. + +721.53.56 + +### **B.75 broadcast call** + +*F: communication de diffusion* + +*S: comunicación de difusión* + +A *multi-address call* in which signals are transmitted simultaneously by the calling *terminal* to all the called terminals. + +721.53.57 + +### **B.76 restricted conference call** + +*F: conférence restreinte* + +*S: comunicación conferencia restringida* + +A *conference call* in which certain nominated *terminals* may only transmit to one, or some, of the terminals involved, or may not transmit at all. + +721.53.58 + +### **B.77 broadcast conference call** + +*F: conférence-diffusion* + +*S: comunicación conferencia de difusión* + +A *restricted conference call* in which only one nominated *terminal* can transmit to and receive from the other terminals. + +721.53.59 + +### **B.78 prefix** + +*F: préfixe* + +*S: prefijo* + +An indicator, consisting of one or more digits, that allows the selection of different types of address formats (e.g. local, national or international address formats), transit network and/or service selection. Prefixes are not part of the national subscriber number and are not signalled over inter-network or international boundaries. + +### **B.79 escape code** + +*F: code d'échappement, séquence d'échappement* + +*S: código de escape* + +An indicator consisting of one or more digits. The indicator is defined in a given numbering plan and is used to indicate that the following digits are a number from a different numbering plan. Escape codes are currently used within Recommendation X.121 numbering plans. + +*Note* – An escape code may be carried forward through the originating network and can be carried across inter-network and international boundaries. Therefore, the values of escape codes should be standardized. + +## ANNEX C (to Recommendation F.60) + +### **Alphabetical list of abbreviations used in this Recommendation** + +| | | +|-------|-------------------------------------------| +| CF | Conversion facility | +| IEV | International electrotechnical vocabulary | +| IMA | Input message acknowledgement | +| ITA2 | International Telegraph Alphabet No. 2 | +| NDN | Non-delivery notification | +| ODA | On-line delivery acknowledgement | +| PDN | Positive delivery notification | +| PSPDN | Packet switched public data network | +| SDR | Special drawing rights | + +## **References** + +- [1] *Table of international telex relations and traffic*, ITU, Geneva, (yearly publication). +- [2] *International Telecommunication Convention*, Nairobi, 1982. \ No newline at end of file diff --git a/marked/F/T-REC-F.600-200404-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.600-200404-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..939eb6a57fac47a375526475ba6c31bf67e08b30 --- /dev/null +++ b/marked/F/T-REC-F.600-200404-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:52a289bbb49d106b16f646b5050f62e5c7cbe1bb201f8c767547c3039abbf359 +size 8245 diff --git a/marked/F/T-REC-F.600-200404-I_PDF-E/3121ebddccf183ca63bb9781be440a7e_img.jpg b/marked/F/T-REC-F.600-200404-I_PDF-E/3121ebddccf183ca63bb9781be440a7e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..14d1e1f077aeb25cd0f74000f54a008a1e24bb7b --- /dev/null +++ b/marked/F/T-REC-F.600-200404-I_PDF-E/3121ebddccf183ca63bb9781be440a7e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:44da4b3632b9ac8d1a70df4aa097b54c03eefd8d3e6a36308d4709f7d0b28fbc +size 39044 diff --git a/marked/F/T-REC-F.600-200404-I_PDF-E/97fe9069356cc2a84e8e70673e405958_img.jpg b/marked/F/T-REC-F.600-200404-I_PDF-E/97fe9069356cc2a84e8e70673e405958_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..8cca5ae94bfecc0e4c5cdd20dfeaa87133e190b3 --- /dev/null +++ b/marked/F/T-REC-F.600-200404-I_PDF-E/97fe9069356cc2a84e8e70673e405958_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1cca9f8e3ce36bdd2d7a9279b1ab28a752f091d91183d57bcd30573eed565580 +size 9375 diff --git a/marked/F/T-REC-F.600-200404-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg b/marked/F/T-REC-F.600-200404-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3134df864b5f5a704bfcbee421121a56a0714e43 --- /dev/null +++ b/marked/F/T-REC-F.600-200404-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cc98bd479070179de814c23e962d6ce9c06280cc9a0af25594bb29c1759be998 +size 49498 diff --git a/marked/F/T-REC-F.61-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.61-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..64672153e39c851d9e16fdbc702e526c7ba7e8a7 --- /dev/null +++ b/marked/F/T-REC-F.61-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:454a9958ffe168868cb7d38a0eb24418dafe31a7a4245c992089b2316ac37d3e +size 7392 diff --git a/marked/F/T-REC-F.61-198811-I_PDF-E/raw.md b/marked/F/T-REC-F.61-198811-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..eb93cb9c1e7e07b6636c39c4cc15e3bad4f6a99e --- /dev/null +++ b/marked/F/T-REC-F.61-198811-I_PDF-E/raw.md @@ -0,0 +1,73 @@ + + +![ITU logo](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +The logo of the International Telecommunication Union (ITU) features the letters 'ITU' in a bold, sans-serif font, superimposed on a stylized globe with latitude and longitude lines. A lightning bolt symbol is positioned to the right of the globe. + +ITU logo + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +**F.61** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +# **OPERATIONS AND QUALITY OF SERVICE TELEGRAPH SERVICES** --- + +## **OPERATIONAL PROVISIONS RELATING TO THE CHARGEABLE DURATION OF A TELEX CALL** + +**ITU-T Recommendation F.61** + +(Extract from the *Blue Book*) + +--- + +## NOTES + +1 ITU-T Recommendation F.61 was published in Fascicle II.4 of the *Blue Book*. This file is an extract from the *Blue Book*. While the presentation and layout of the text might be slightly different from the *Blue Book* version, the contents of the file are identical to the *Blue Book* version and copyright conditions remain unchanged (see below). + +2 In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +## **OPERATIONAL PROVISIONS RELATING TO THE CHARGEABLE DURATION OF A TELEX CALL1)** + +### **1 Fully automatic service** + +1.1 With fully automatic international telex operation, call durations are registered automatically. Administrations may not have tickets available for working out the distribution of charges on the basis of the chargeable duration of calls. + +1.2 In accordance with the relevant Recommendations for signalling in the telex service, the call-connected signal should start the equipment for determining the chargeable duration of the call in the case of automatic switching between subscribers. + +1.3 Some networks delay the start of charging in order to avoid charging ineffective calls when preceded by a *call-connected* signal. + +1.4 Consequently, the start of charging for calling subscribers may differ considerably from one network to another. It has been possible to note differences of 15 seconds between these starts for different networks. + +1.5 In accordance with Recommendation D.61, the chargeable duration of a call in the fully automatic international telex service should be calculated from a conventional start fixed at between 5 and 7 seconds after the start of the call-connected signal. The end of the chargeable duration should be fixed not later than 1 second after the start of the clearing signal. The conventional start of charging is valid for all calls whether charged: + +- a) minute by minute, any fraction of a minute being counted as one minute; or +- b) or by shorter periods, either derived by periodic pulse metering of the type used in national automatic services or derived from an automatic recording of all details, which would normally include the identification of the calling and called subscribers, the time at which the call began, the called duration and/or the time at which the call terminated. + +1.6 The degree of accuracy of the call-duration measuring equipment should be $\pm 2\%$ for a set of measurements covering an adequate number of calls which, in light traffic relations, may lead to acceptance of the fact that 2% accuracy should be obtained on the overall measurements for a year but not for each of the partial measurements made during that year (monthly measurements, for example, if the monthly interval is retained for the establishment of international accounts). + +### **2 Semi-automatic and manual service** + +2.1 The chargeable duration of a telex call begins at the moment the connection is established between the calling and the called subscribers. + +2.2 It ends at the moment when the clearing signal given by the calling or called subscriber is transmitted over the international circuit. To this end, the international telex position must be able to receive the clearing signal from both sides. + +2.3 With manual or semi-automatic operation, the operator of the controlling international telex position shall determine the chargeable duration, unless other arrangements have been made by agreement between the Administrations concerned, taking into account, where necessary, any difficulties in transmission or any irregularities that he may observe. + +--- + +1) See also Recommendation D.61. + +2.4 When a telex call is controlled by an operator at a telex position in a transit country, the chargeable duration of the call shall be notified to the international telex exchange of origin within 24 hours, giving the following details: + +- the locality and number of the calling subscriber; +- the locality and number of the called subscriber; +- the time at which the call began; +- its chargeable duration in minutes. + +Example: **STOCKHOLM 1846 TO ATHENS 21460 AT 1546Z 3 RPT 3 MNS** \ No newline at end of file diff --git a/marked/F/T-REC-F.64-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.64-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..64672153e39c851d9e16fdbc702e526c7ba7e8a7 --- /dev/null +++ b/marked/F/T-REC-F.64-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:454a9958ffe168868cb7d38a0eb24418dafe31a7a4245c992089b2316ac37d3e +size 7392 diff --git a/marked/F/T-REC-F.65-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.65-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..64672153e39c851d9e16fdbc702e526c7ba7e8a7 --- /dev/null +++ b/marked/F/T-REC-F.65-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:454a9958ffe168868cb7d38a0eb24418dafe31a7a4245c992089b2316ac37d3e +size 7392 diff --git a/marked/F/T-REC-F.68-198811-I_PDF-E/1be8e9cad5f38fa47bdb81e549a3bec9_img.jpg b/marked/F/T-REC-F.68-198811-I_PDF-E/1be8e9cad5f38fa47bdb81e549a3bec9_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..28f4fdd43a89c7f5259271650557e0caccaceed5 --- /dev/null +++ b/marked/F/T-REC-F.68-198811-I_PDF-E/1be8e9cad5f38fa47bdb81e549a3bec9_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:93b347a5430d6ac866c62b45b818317e4cce741535f8f4d643a5652fa733334e +size 23053 diff --git a/marked/F/T-REC-F.68-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.68-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..64672153e39c851d9e16fdbc702e526c7ba7e8a7 --- /dev/null +++ b/marked/F/T-REC-F.68-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:454a9958ffe168868cb7d38a0eb24418dafe31a7a4245c992089b2316ac37d3e +size 7392 diff --git a/marked/F/T-REC-F.68-198811-I_PDF-E/5d76c2e9976aaf97001ee189908830fb_img.jpg b/marked/F/T-REC-F.68-198811-I_PDF-E/5d76c2e9976aaf97001ee189908830fb_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..b138faf684b35eecd92ef0b27f61a75bc93770f5 --- /dev/null +++ b/marked/F/T-REC-F.68-198811-I_PDF-E/5d76c2e9976aaf97001ee189908830fb_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a961a1bd70c715d39ae1b5fea52a98b5f20a3cc9622a3a292bbe4cfe1bd2a0d3 +size 26308 diff --git a/marked/F/T-REC-F.68-198811-I_PDF-E/a738993919a50143787084ee7ce6e2f2_img.jpg b/marked/F/T-REC-F.68-198811-I_PDF-E/a738993919a50143787084ee7ce6e2f2_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..18b7f3335969c3b626dbf30502583576eaedbc12 --- /dev/null +++ b/marked/F/T-REC-F.68-198811-I_PDF-E/a738993919a50143787084ee7ce6e2f2_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cfab8bc20d34285c1adef1bd5466be5b6c54435236aa25c081f5ecc483312e68 +size 6766 diff --git a/marked/F/T-REC-F.68-198811-I_PDF-E/raw.md b/marked/F/T-REC-F.68-198811-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..6589226090d7929fee7cf7f8b65a15f20ea771a7 --- /dev/null +++ b/marked/F/T-REC-F.68-198811-I_PDF-E/raw.md @@ -0,0 +1,234 @@ + + +![ITU logo](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +The logo of the International Telecommunication Union (ITU) features the letters 'ITU' in a bold, sans-serif font, superimposed on a stylized globe with intersecting lines. + +ITU logo + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +**F.68** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**OPERATIONS AND QUALITY OF SERVICE +TELEGRAPH SERVICES** + +--- + +**ESTABLISHMENT OF THE AUTOMATIC +INTERCONTINENTAL TELEX NETWORK** + +**ITU-T Recommendation F.68** + +(Extract from the *Blue Book*) + +--- + +## NOTES + +1 ITU-T Recommendation F.68 was published in Fascicle II.4 of the *Blue Book*. This file is an extract from the *Blue Book*. While the presentation and layout of the text might be slightly different from the *Blue Book* version, the contents of the file are identical to the *Blue Book* version and copyright conditions remain unchanged (see below). + +2 In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +# ESTABLISHMENT OF THE AUTOMATIC INTERCONTINENTAL TELEX NETWORK + +## **1 Preamble – Definitions connected with the numbering of telex subscribers and the routing of telex and gentex calls in intercontinental service** + +These definitions are proposed to facilitate the reading of recommendations and surveys on the question of intercontinental telex and gentex traffic; they have been derived to large extent from the definitions submitted by telephone experts for studying the analogous problem in intercontinental telephone operation and adapted to the special features of the telex and gentex services. + +*Preliminary note* – The word *continent* is not necessarily used in its geographical sense: traffic characteristics may cause countries of geographically different continents to be included in one continent (within the meaning of these definitions). + +### **1.1 Circuits** + +1.1.1 A **national circuit** is one connecting two exchanges in the same country. + +1.1.2 An **international circuit** is one connecting two exchanges in different countries, whether or not they are in different continents. + +1.1.3 A **continental circuit** is one established between two exchanges in the same continent. + +1.1.4 An **intercontinental circuit** is one connecting two exchanges situated in different countries in different continents. + +1.1.5 An **intercontinental transit circuit** is an intercontinental circuit used primarily for routing intercontinental transit traffic. + +### **1.2 Exchanges** + +1.2.1 A **national exchange** is the termination centre for national circuits only. + +1.2.2 An **international exchange** is a centre where international circuits, and in general national circuits, terminate. + +1.2.3 A **continental exchange** is an international centre where the international circuits terminating there are solely continental circuits. + +1.2.4 **intercontinental transit exchange**: An exchange of this type would be directly connected to intercontinental transit circuits and would provide facilities to interconnect intercontinental transit circuits and trunks to terminal exchanges. It would also provide facilities for the interconnection of intercontinental transit circuits. + +1.2.5 **terminal international exchange**: An international exchange of this type would not be connected directly to intercontinental transit circuits, but would gain access to the intercontinental transit network through one (or more) intercontinental transit exchanges. + +### **1.3 Connections** + +1.3.1 **international connection**: Any connection between two stations situated in different countries, whether established between different continents or one continent. + +1.3.2 **continental connection**: Connection established between stations within the same continent. + +1.3.3 **intercontinental connection**: Connection established between two different continents. + +### 1.4 Numbering + +1.4.1 **subscriber's national telex number:** Set of figures to be selected by a caller in the same country to obtain this subscriber. + +1.4.2 **local number:** In national telex networks, when abridged call numbers are used for local or short-distance traffic, the abridged number is called the *local number*. + +1.4.3 **prefix giving access to the long-distance automatic telex network:** In national telex networks, when abridged call numbers are used for local or short-distance traffic, an access prefix should be selected to give access to the higher level network (long-distance level). + +1.4.4 **prefix giving access to the international automatic telex network:** This expression is taken to mean the digit or digits that a subscriber must select (if necessary after the prefix giving access to the automatic long-distance telex network) to obtain access to the automatic telegraph switching equipment for international telex traffic. + +1.4.5 **prefix giving access to the intercontinental automatic transit telex network:** This expression is taken to mean the digit or digits that a subscriber must select (if necessary after the prefix giving access to the international telex network) to obtain access to automatic telegraph switching equipment for intercontinental transit telex traffic. + +1.4.6 The origin country is free to use only a common *access prefix to the international network* instead of two different prefixes for access to the international network and the intercontinental network. + +1.4.7 **telex network identification code:** Letter or group of two letters serving to identify the subscribers or stations of a country (or a network in a country). + +1.4.8 **telex destination code:** A group of digits characterizing, for routing purposes, the subscribers or stations of a country, or of a network in a country. + +### 1.5 Routing + +1.5.1 **automatic alternative routing:** A facility whereby a call, which cannot find a free circuit on the primary route at an international outgoing exchange, is automatically diverted to a secondary route. + +1.5.2 **emergency routes:** The circuit(s) to be used in case of complete interruption or major breakdown of the primary and secondary routes. The emergency routes may pass through any country. + +1.5.3 **primary routes:** The circuits normally used in a given relation. + +1.5.4 **rerouting:** When congestion occurs at an intermediate transit exchange, rerouting permits a call to be remade via a secondary route from the outgoing international exchange. + +1.5.5 **secondary routes:** The circuits to be used when the primary routes are congested. The secondary route(s) may pass through the same countries as the primary routes or through different countries. In manual and semi-automatic operation, secondary routes may also be used when the transmission on the primary route is not sufficiently good, or if traffic is to be handled outside the normal hours of service on the primary routes. + +## 2 Recommendation for the establishment of the automatic intercontinental telex network + +The CCITT, + +considering + +(a) that intercontinental telex traffic is rapidly growing; in particular, the development of automatic subscriber selection in intercontinental relations has been made possible. The time differences between terminal countries in such relations and the consequent differences in the hours of peak traffic loading may take it economical to employ tandem transit routing to a much greater extent than has been necessary in the European network. The development of a comprehensive plan for the economical employment of tandem routing depends among other considerations, on agreement on numbering and routing plans; + +(b) that a worldwide service includes countries that are served by several telex networks. A telex subscriber's call number in a worldwide service must contain all the digits to be transmitted by the caller in order to establish the connection, irrespective of the routing channel; + +(c) that to facilitate automatic routing and charging for calls, the number of digits to be examined by the charging equipment must be limited; + +*unanimously declares the following:* + +### 2.1 *General characteristics of the network* + +2.1.1 It must be possible to establish the intercontinental network by means of: + +- a) submarine or underground cable telegraph circuits; +- b) telegraph circuits via telecommunication satellites; and +- c) telegraph circuits on radio channels. + +2.1.2 When circuits via various transmission facilities exist between two intercontinental transit exchanges, all such circuits must, for automatic selection purposes, be regarded as included in a single system. + +2.1.3 Administrations will agree on whether a given group of circuits should be operated on a one-way, bothway or partially divided basis. + +2.1.4 The traffic to be routed over these circuits may be either telex or gentex traffic; it may be either transit or terminal traffic. + +2.1.5 Countries (or networks) should be connected by direct circuits where this can be justified taking into account the relative economics of transit switching and bothway working where the time difference between the terminal centres makes this a significant factor. + +2.1.6 Where it is not practicable to provide direct circuits, the number of transit exchanges involved in a normally routed call should be reduced in so far as possible. + +2.1.7 Where the same group of circuits carries traffic originated by subscribers in the country providing facilities and transit traffic originated by another country, the Administration providing the transit exchange shall ensure that the transit calls receive a grade of service not inferior to that given to their own subscribers. + +### 2.2 *Identification of telex subscribers* + +2.2.1 For international purposes, a subscriber's national number should be accompanied by one or two letters, called the *telex network identification code*, characterizing either: + +- a) the subscriber's country, if in that country there is only one telex network; or +- b) the telex network to which the subscriber belongs in a country where there are several networks operated by different agencies. + +2.2.2 An identification code is especially valuable for countries possessing several telex networks operated by different companies and when national numbers do not clearly distinguish between such networks. In such circumstances, it is recommended that the identification code should be clearly published in national directories. Furthermore, Administrations shall ask subscribers to give every possible publicity to their telex identification letters (by including them in the letterheads of their correspondence for example). + +2.2.3 The answerback codes for subscriber equipment used in intercontinental telex services should include the telex network identification code allocated to the country or network concerned (see Recommendation F.60, § 3.4.2). + +2.2.4 For Administrations using two-character telex network identification codes these codes should be the same as the identification codes of their country (or network) for the telegram retransmission system (see Recommendations F.31 and F.96). + +2.2.5 The one-character telex network identification code **X** is used to identify mobile stations that may be connected to the international telex network, regardless of the transmission medium used (maritime mobile-satellite, VHF, HF). The two-character codes commencing with **X** will not be allocated to national telex networks. + +2.2.6 The list of telex network identification codes has been compiled by the CCITT and published in accordance with Recommendation F.96. + +*Note* – If in any country the telex and gentex networks are separate, two identification codes might be necessary, one for telex and the other for gentex. + +### 2.3 Routing + +2.3.1 On international circuits digits only will be transmitted for selection control. + +2.3.2 For each country, or for each network in countries possessing several telex networks, a group of two or three digits – *the telex destination code* – will uniformly characterize each country or network for the purposes of selection in intercontinental transit circuits. (See Recommendation F.69 for the list of telex destination codes.) + +2.3.3 The access prefix to be selected in an outgoing country by a subscriber wanting to put through a call to another country via the intercontinental transit network shall be decided on by the Administration responsible for the calling subscriber. This is a matter for internal regulation. + +2.3.4 There are two possibilities in relations between the international exchange of the outgoing country and an intercontinental exchange: + +2.3.4.1 There are direct trunk circuits between the international exchange in the outgoing country and the intercontinental exchange (see Figure 1/F.68). On these circuits, it should suffice to transmit the destination code of the country required, followed by the national number of the subscriber required; + +![Diagram of Figure 1/F.68 showing a direct trunk circuit between the international exchange in Country A and the intercontinental transit exchange in Country C.](1be8e9cad5f38fa47bdb81e549a3bec9_img.jpg) + +Diagram illustrating the routing path for a call from Country A (country of origin) to Country C (country of transit) via a direct trunk circuit. The path starts at the Subscriber (CCITT-52090) in Country A, connects to a National exchange, then to an International terminal exchange. A direct trunk circuit connects the International terminal exchange in Country A to the Intercontinental transit exchange in Country C, crossing the Frontier. The path ends at the Intercontinental transit exchange in Country C. + +Diagram of Figure 1/F.68 showing a direct trunk circuit between the international exchange in Country A and the intercontinental transit exchange in Country C. + +FIGURE 1/F.68 + +2.3.4.2 There are no such direct trunk circuits (see Figure 2/F.68). There are then direct circuits between the international exchange in the outgoing country and the international continental exchange in the transit country, adjacent to the intercontinental exchange. Hence this adjacent exchange will have to be traversed to reach the intercontinental network. + +![Diagram of Figure 2/F.68 showing a call path from Country A to Country C via an international exchange and a continental exchange adjacent to the intercontinental transit exchange.](5d76c2e9976aaf97001ee189908830fb_img.jpg) + +Diagram illustrating the routing path for a call from Country A (country of origin) to Country C (country of transit) via an international exchange and a continental exchange. The path starts at the Subscriber (CCITT-52100) in Country A, connects to a National exchange, then to an International exchange. A direct circuit connects the International exchange in Country A to the Continental exchange adjacent to the intercontinental transit exchange in Country C, crossing the Frontier. The path then continues from the Continental exchange to the Intercontinental transit exchange in Country C, ending at the Subscriber. + +Diagram of Figure 2/F.68 showing a call path from Country A to Country C via an international exchange and a continental exchange adjacent to the intercontinental transit exchange. + +FIGURE 2/F.68 + +2.3.5 Code 00 should be used as the standard access prefix for traversing a continental exchange. A country that might experience difficulty in accepting this 00 code may choose another code for traversing its continental exchange, subject to a bilateral agreement with the other Administration concerned. + +### 2.4 *Automatic alternative routing* + +2.4.1 Provision must be made for the possibility of using automatic alternative routing (see Figure 3/F.68). The putting into operation of automatic alternative routing is a question of the network situation, as it will often be preferable to create new telegraph circuits on a congested route rather than to bring automatic alternative routing into service. Be it noted that the automatic alternative routing method should be considered only if the peak hours on CD are different from those on CF and FD; otherwise, it is to be feared that transit switching equipment F will become saturated. + +![Diagram of automatic alternative routing showing nodes A, B, C, D, and F. A horizontal line connects A, C, D, and B. A triangle is formed by nodes C, F, and D, with F above the line CD. The label CCITT-52110 is below node D.](a738993919a50143787084ee7ce6e2f2_img.jpg) + +``` +graph LR; A --- C; C --- F; F --- D; D --- B; C --- D; +``` + +Diagram of automatic alternative routing showing nodes A, B, C, D, and F. A horizontal line connects A, C, D, and B. A triangle is formed by nodes C, F, and D, with F above the line CD. The label CCITT-52110 is below node D. + +FIGURE 3/F.68 + +### 2.5 *Automatic re-routing* + +2.5.1 The complications resulting from automatic re-routing would be out of all proportion to the benefits to be expected therefrom. + +### 2.6 *Call recording* + +2.6.1 In principle the originating exchange is responsible for timing calls, booking calls, repeated attempts, etc. The responsibility of an intercontinental transit exchange should be limited to providing a connection between the calling exchange or subscriber and the required exchange or subscriber. + +2.6.2 In accordance with Recommendation U.23, the elapsed time is normally taken as the basis for fixing the chargeable duration even where calls are routed over HF radio circuits fitted with ARQ equipment on the first or subsequent links in the connection. + +### 2.7 *Grade of service* + +2.7.1 Refer to Recommendation F.64. + +### 2.8 *Use of radiotelegraph circuits with ARQ equipment* + +2.8.1 Fully automatic operation on a radiotelegraph circuit incorporating ARQ equipment can be considered only if this circuit possesses adequate stability. + +2.8.2 Before incorporating a circuit with ARQ equipment in the fully automatic switched network, the Administrations must carry out extended trials. + +2.8.3 These trials should be made under normal traffic conditions, over a minimum period of three consecutive hours chosen from the busy period (or periods), when heavy traffic is foreseen to occur on the route under consideration (allowing for the traffic, whether terminal or transit, that prevails on the route according to the season). + +2.8.4 The condition that must be fulfilled before a circuit can be accepted for use in a fully automatic network is that its mean efficiency factor measured over periods of 20 consecutive seconds each, shall not fall below 80% for more than 10% of the total time involved in the measurements. The measurements must be repeated as often as will be necessary for the Administration to have an assessment of the suitability of the circuit. + +2.8.5 The attention of the Administrations is drawn to the fact that, before offering fully automatic transit working on a radio route incorporating ARQ equipment, the grade of service on the route under consideration must be only one call lost in 50. + +2.8.6 If these conditions are not complied with, it would be better to retain semi-automatic operation. \ No newline at end of file diff --git a/marked/F/T-REC-F.69-199406-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.69-199406-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7e86be364c2e70dc0688295c12b1f0dfde871be1 --- /dev/null +++ b/marked/F/T-REC-F.69-199406-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9f5a9bf4a6fdd6b2b1432e4a58154012788a8fe0b6a49989a97624d32d6d507c +size 8198 diff --git a/marked/F/T-REC-F.69-199406-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.69-199406-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..fa1508bfb5ca46ea2f99becbb13631b2ed361bf2 --- /dev/null +++ b/marked/F/T-REC-F.69-199406-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2ddc83688d52ddd6c7e7aadf31dd6b9afcb79d8d627a210fe52da04ab19699db +size 147302 diff --git a/marked/F/T-REC-F.70-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.70-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..64672153e39c851d9e16fdbc702e526c7ba7e8a7 --- /dev/null +++ b/marked/F/T-REC-F.70-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:454a9958ffe168868cb7d38a0eb24418dafe31a7a4245c992089b2316ac37d3e +size 7392 diff --git a/marked/F/T-REC-F.700-200011-I_PDF-E/10781f43062bf3e9601a1e086710556c_img.jpg b/marked/F/T-REC-F.700-200011-I_PDF-E/10781f43062bf3e9601a1e086710556c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d4d3b81317040c9aabe114a3dc6c46e75c60416c --- /dev/null +++ b/marked/F/T-REC-F.700-200011-I_PDF-E/10781f43062bf3e9601a1e086710556c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2775ab034c9d5edd6960f8d7ce399fa81beabcd129672a457ce0b4b105a9d483 +size 4536 diff --git a/marked/F/T-REC-F.700-200011-I_PDF-E/1439cb942d9e363bbb3161b5540dd8c6_img.jpg b/marked/F/T-REC-F.700-200011-I_PDF-E/1439cb942d9e363bbb3161b5540dd8c6_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..e85d8904f524604c0fc8277280481a6f702ea868 --- /dev/null +++ b/marked/F/T-REC-F.700-200011-I_PDF-E/1439cb942d9e363bbb3161b5540dd8c6_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c1f16ace4737426b060116ce9a197dc85fdc97e1277c13ba6367174987a88389 +size 21688 diff --git a/marked/F/T-REC-F.700-200011-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.700-200011-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ab2cbdd4af6bc1ea9ffa8cca87f9d455eb147008 --- /dev/null +++ b/marked/F/T-REC-F.700-200011-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:94ee226724314e8b36ac3c73e99f9ed2a4fbb4786e0c7f3f43c1debd5966aa00 +size 8239 diff --git a/marked/F/T-REC-F.700-200011-I_PDF-E/41a438d7e4adc17c3a4005e7c9500091_img.jpg b/marked/F/T-REC-F.700-200011-I_PDF-E/41a438d7e4adc17c3a4005e7c9500091_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..0ea2e5bdf8059da5f9f94ee511f5ae97f646a75b --- /dev/null +++ b/marked/F/T-REC-F.700-200011-I_PDF-E/41a438d7e4adc17c3a4005e7c9500091_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:31a11b23b75ecb997d234fe3ed4d8808686eaf8339d02190ca222344d32eda52 +size 11599 diff --git a/marked/F/T-REC-F.700-200011-I_PDF-E/4356776ca004ecba5d599667a155d7d4_img.jpg b/marked/F/T-REC-F.700-200011-I_PDF-E/4356776ca004ecba5d599667a155d7d4_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..eda6d3909dbecf5639c618e360f283fb3d66348c --- /dev/null +++ b/marked/F/T-REC-F.700-200011-I_PDF-E/4356776ca004ecba5d599667a155d7d4_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d14d22fc4b1bde8f684c59c202fad14f3938578aa5fd0e921eead5fbf6271aa0 +size 5133 diff --git a/marked/F/T-REC-F.700-200011-I_PDF-E/96b0240f56d14453b5da05ec30fd5c6e_img.jpg b/marked/F/T-REC-F.700-200011-I_PDF-E/96b0240f56d14453b5da05ec30fd5c6e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..fc11d88799ea6a00879e2a3da694afc94a8a56b6 --- /dev/null +++ b/marked/F/T-REC-F.700-200011-I_PDF-E/96b0240f56d14453b5da05ec30fd5c6e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:df249ff63145c3dd5f252c34de496a9ece151e4d74a651f97a68cc13598bb139 +size 6066 diff --git a/marked/F/T-REC-F.700-200011-I_PDF-E/9e8ebf03cae78f4f81b697548c2d7250_img.jpg b/marked/F/T-REC-F.700-200011-I_PDF-E/9e8ebf03cae78f4f81b697548c2d7250_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a2f1ec1424d80406d808d1eccd52a3ba34cc55d5 --- /dev/null +++ b/marked/F/T-REC-F.700-200011-I_PDF-E/9e8ebf03cae78f4f81b697548c2d7250_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bf27e96270b30949f9d1bb63c6123654c6aa64359022f0c7da316249d3ae01b2 +size 32020 diff --git a/marked/F/T-REC-F.700-200011-I_PDF-E/a74294f34a0c736b7ee0f5f0cdca7e28_img.jpg b/marked/F/T-REC-F.700-200011-I_PDF-E/a74294f34a0c736b7ee0f5f0cdca7e28_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..43c45691f4d617abbce761bf479108b95ade5408 --- /dev/null +++ b/marked/F/T-REC-F.700-200011-I_PDF-E/a74294f34a0c736b7ee0f5f0cdca7e28_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e2e157ef78d16c96ff8faee08175c1cb2575e48fff16111391d365092db26297 +size 214407 diff --git a/marked/F/T-REC-F.700-200011-I_PDF-E/b6671cfafda3820aafe9a24fa7a4d8c7_img.jpg b/marked/F/T-REC-F.700-200011-I_PDF-E/b6671cfafda3820aafe9a24fa7a4d8c7_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..69123f97af5d139b5b3b96ffad96ee85a8b59c79 --- /dev/null +++ b/marked/F/T-REC-F.700-200011-I_PDF-E/b6671cfafda3820aafe9a24fa7a4d8c7_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:32588c599cdd18da5616277a809050c1f470d7c5ea0c696544191c99870fddd1 +size 9552 diff --git a/marked/F/T-REC-F.700-200011-I_PDF-E/c5452f95f3b28f1bfe29e84fbc2e1267_img.jpg b/marked/F/T-REC-F.700-200011-I_PDF-E/c5452f95f3b28f1bfe29e84fbc2e1267_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..2462ad28bd5a1e77df32a3f0edaa1fbcc126cc59 --- /dev/null +++ b/marked/F/T-REC-F.700-200011-I_PDF-E/c5452f95f3b28f1bfe29e84fbc2e1267_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0442a2cc7e3e9e4437073007406a3ce426644a0bf109662ecf3dca86976dcfe2 +size 198422 diff --git a/marked/F/T-REC-F.700-200011-I_PDF-E/d4af765160d04ecef538e5066006dc77_img.jpg b/marked/F/T-REC-F.700-200011-I_PDF-E/d4af765160d04ecef538e5066006dc77_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..330e863638a607d9256d2fe1cdecff47ccff5cd2 --- /dev/null +++ b/marked/F/T-REC-F.700-200011-I_PDF-E/d4af765160d04ecef538e5066006dc77_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:20329053eced600091d9924cb14ebb44418d53cc1d10758e5063b2624fc9a223 +size 39813 diff --git a/marked/F/T-REC-F.700-200011-I_PDF-E/ddc7460821484f1ae2835c67955c554c_img.jpg b/marked/F/T-REC-F.700-200011-I_PDF-E/ddc7460821484f1ae2835c67955c554c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1ab52350cd07433522007cb2a932f2b1564b7c52 --- /dev/null +++ b/marked/F/T-REC-F.700-200011-I_PDF-E/ddc7460821484f1ae2835c67955c554c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:246cff72a91b07ba162024ba30feeca4d92fbf7c3ac3f51b94666a5cb96242fe +size 43747 diff --git a/marked/F/T-REC-F.700-200011-I_PDF-E/df82d77a0d2637cbf2da9ea920a554fa_img.jpg b/marked/F/T-REC-F.700-200011-I_PDF-E/df82d77a0d2637cbf2da9ea920a554fa_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..8885add2c837406e5762901d4c3d14b66731e3e6 --- /dev/null +++ b/marked/F/T-REC-F.700-200011-I_PDF-E/df82d77a0d2637cbf2da9ea920a554fa_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c4442eb6d45c5b9a80d81ebe90d915b36df1e177e34877e24003c7575626a34e +size 11263 diff --git a/marked/F/T-REC-F.700-200011-I_PDF-E/e928f4874ed492d3ad4c6fa2d29aedbc_img.jpg b/marked/F/T-REC-F.700-200011-I_PDF-E/e928f4874ed492d3ad4c6fa2d29aedbc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..059df984c839ce68302927db90b37e7ec317a813 --- /dev/null +++ b/marked/F/T-REC-F.700-200011-I_PDF-E/e928f4874ed492d3ad4c6fa2d29aedbc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:95d3e7445498b7b7c5899ac0fb77d87a342ddfebb6dbeadbdddcbb96a09fc161 +size 10377 diff --git a/marked/F/T-REC-F.701-200011-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.701-200011-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..b18e5b07ba6a266813c0ce76e92f4e93c0d9fd4b --- /dev/null +++ b/marked/F/T-REC-F.701-200011-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c7a97195c2fd68ad39d82d19b808d1738dce3aa52e94325870a031b5a2ea7c01 +size 8275 diff --git a/marked/F/T-REC-F.701-200011-I_PDF-E/367926125450c2bc3f4bdca9d59a62ba_img.jpg b/marked/F/T-REC-F.701-200011-I_PDF-E/367926125450c2bc3f4bdca9d59a62ba_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..26e72ccf86704bc121f0be195643d8f238f8c8ea --- /dev/null +++ b/marked/F/T-REC-F.701-200011-I_PDF-E/367926125450c2bc3f4bdca9d59a62ba_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:805e0a4c9df7dccb9b3679438f6ee6b24ce396bd918c43a41ef5abec08a96ddf +size 44415 diff --git a/marked/F/T-REC-F.701-200011-I_PDF-E/9791722d75115ddcc599b07d7bc35d73_img.jpg b/marked/F/T-REC-F.701-200011-I_PDF-E/9791722d75115ddcc599b07d7bc35d73_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..4539460578f11fdec86e4610d7826fed153e1f2b --- /dev/null +++ b/marked/F/T-REC-F.701-200011-I_PDF-E/9791722d75115ddcc599b07d7bc35d73_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c41c2c5c06750afe65d2e3f8119bd653b494795b422c993866f0d56950a07d2a +size 54282 diff --git a/marked/F/T-REC-F.702-199607-I_PDF-E/raw.md b/marked/F/T-REC-F.702-199607-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..6c5eaea3c672f73b433dc9e08990a3b598a05d7b --- /dev/null +++ b/marked/F/T-REC-F.702-199607-I_PDF-E/raw.md @@ -0,0 +1,1278 @@ + + +![ITU logo: A globe with a lightning bolt and the letters ITU.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with a lightning bolt and the letters ITU. + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.702** + +(07/96) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Audiovisual services + +--- + +**Multimedia conference services** + +ITU-T Recommendation F.702 + +(Previously "CCITT Recommendation") + +--- + +# ITU-T F-SERIES RECOMMENDATIONS NON-TELEPHONE TELECOMMUNICATION SERVICES + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| AUDIOVISUAL SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to ITU-T List of Recommendations.* + +## FOREWORD + +The ITU-T (Telecommunication Standardization Sector) is a permanent organ of the International Telecommunication Union (ITU). The ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Conference (WTSC), which meets every four years, establishes the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics. + +The approval of Recommendations by the Members of the ITU-T is covered by the procedure laid down in WTSC Resolution No. 1 (Helsinki, March 1-12, 1993). + +ITU-T Recommendation F.702 was prepared by ITU-T Study Group 1 (1993-1996) and was approved under the WTSC Resolution No. 1 procedure on the 19th of July 1996. + +--- + +## NOTE + +In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +© ITU 1997 + +All rights reserved. No part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and microfilm, without permission in writing from the ITU. + +# CONTENTS + +| | | Page | +|-------|---------------------------------------------------------|-------------| +| 1 | Introduction ..... | 1 | +| 2 | Definition ..... | 1 | +| 3 | References and terminology ..... | 1 | +| 3.1 | References ..... | 1 | +| 3.2 | Terminology ..... | 2 | +| 4 | Description ..... | 2 | +| 4.1 | General description ..... | 2 | +| 4.2 | Functional model ..... | 3 | +| 4.3 | Configuration ..... | 4 | +| 4.4 | Roles of the participants ..... | 5 | +| 4.5 | Terminal aspects ..... | 7 | +| 4.6 | Applications ..... | 8 | +| 4.7 | Supplementary services ..... | 8 | +| 5 | Static aspects ..... | 8 | +| 5.1 | Service level ..... | 8 | +| 5.1.1 | General aspects of the service ..... | 8 | +| 5.1.2 | Quality of service ..... | 8 | +| 5.1.3 | Security aspects ..... | 10 | +| 5.1.4 | General charging principles ..... | 10 | +| 5.2 | Communication task level ..... | 10 | +| 5.2.1 | Communication tasks used for the service ..... | 10 | +| 5.2.2 | Quality of service ..... | 10 | +| 5.3 | Media component level ..... | 11 | +| 5.3.1 | Media components used for the service ..... | 11 | +| 5.3.2 | Quality of service ..... | 11 | +| 6 | Dynamic aspects ..... | 12 | +| 6.1 | Activation phase ..... | 12 | +| 6.1.1 | Provision, withdrawal ..... | 12 | +| 6.1.2 | Call establishment ..... | 12 | +| 6.1.3 | Reservation ..... | 12 | +| 6.1.4 | On demand ..... | 13 | +| 6.2 | Communication phase ..... | 13 | +| 6.2.1 | Call set-up ..... | 13 | +| 6.2.2 | Addition of another party during the call ..... | 13 | +| 6.2.3 | Conference splitting ..... | 14 | +| 6.2.4 | Change of communication mode ..... | 14 | +| 6.2.5 | Call prolongation in the case of reservation ..... | 14 | +| 6.2.6 | Terminal disconnection ..... | 15 | +| 6.2.7 | Conference management procedures ..... | 15 | +| 6.3 | Termination phase ..... | 16 | +| 6.3.1 | Call termination ..... | 16 | +| 7 | Service profiles ..... | 17 | +| 7.1 | Different types of multimedia conference services ..... | 17 | +| 7.2 | List of service profiles ..... | 17 | + +| | | | +|----------------------------------------------------|------------------------------------------------------------------------------------------------------------|----| +| 8 | Interworking and intercommunication ..... | 19 | +| 8.1 | General..... | 19 | +| 8.2 | Interworking/intercommunication between two terminals ..... | 19 | +| 8.2.1 | Different types of audiographic conference terminals ..... | 19 | +| 8.2.2 | Audiographic conference and videoconference terminals ..... | 19 | +| 8.2.3 | Different types of videoconference terminals ..... | 19 | +| 8.2.4 | CDH and other types of terminals..... | 19 | +| 8.3 | Interworking/intercommunication between several terminals..... | 19 | +| 8.3.1 | Different types of multimedia conference terminals..... | 19 | +| 8.3.2 | Videophone and other multimedia conference terminals, multimedia conference and videophone terminals ..... | 20 | +| 8.3.3 | Addition of a telephone terminal ..... | 20 | +| 8.3.4 | Communication with a facsimile terminal ..... | 20 | +| Annex A – Multipoint Conference Units (MCUs) ..... | | 20 | +| A.1 | General..... | 20 | +| A.2 | Call control functions ..... | 22 | +| A.3 | Conference control functions..... | 22 | +| A.4 | Handling the various media components ..... | 23 | +| A.5 | Distributed MCU functions ..... | 23 | +| A.6 | Ports and interfaces..... | 25 | +| A.7 | Interworking/intercommunication ..... | 25 | +| Appendix I – Control of multiple windows ..... | | 25 | +| I.1 | General..... | 25 | +| I.2 | Starting, stopping and organizing multiple windowing ..... | 25 | +| I.3 | Selection of the windows' contents ..... | 25 | + + + +# **MULTIMEDIA CONFERENCE SERVICES** + +*(Geneva, 1996)* + +# **1 Introduction** + +The purpose of this Recommendation is to define the generic multimedia conference service and to describe their general features, regardless of the network environment in which these services are provided. These services allow conversational communication between several users in two or more locations. The generic multimedia conference service is one of the generic service types identified in Recommendation F.700, and its description follows the methodology described therein. The user requirements for the various applications supported by the service are translated into generic service specifications, independent of the constraints of specific implementations. Thus this Recommendation sets out the requirements that will satisfy the users' needs and allow proper intercommunication on an international basis of services offered by different providers and of equipment from different manufacturers. + +Various instances (profiles) of this generic service are considered here with a network independent view. Specific service descriptions for each network will be issued in other Recommendations. The detailed technical specifications of the terminal, network and protocol aspects for each of them are described in dedicated Recommendations of the H.200-, H.300-, H.400-Series, and of the T.120-Series for specific multipoint protocols. + +Along with the F.700 methodology, the description relies on the media components and the communication tasks described respectively in Annex A/F.700 and Annex B/F.700. These are service independent modular communication capabilities. + +# **2 Definition** + +The multimedia conference services provide real-time transmission of voice together with motion video and/or various types of multimedia information between groups of users in two or more locations. The documents exchanged may contain all information types. When moving pictures are present, their quality must be at least sufficient for the adequate representation of the fluid movements of a small group of participants. The media components used are described in Annex A/F.700 on audiovisual/multimedia services. Media component audio (A.1/F.700) is mandatory, and one or more of the media components video (A.2/F.700), text (A.3/F.700), graphics (A.4/F.700) and still pictures (A.5/F.700) should be present. + +# **3 References and terminology** + +The following Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision: all users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid-ITU-T Recommendations is regularly published. + +## **3.1 References** + +- ITU-T Recommendation F.700 (1996), *Framework Recommendation for audiovisual/multimedia services*. +- ITU-T Recommendation G.114 (1996), *One-way transmission time*. +- CCITT Recommendation G.711 (1988), *Pulse Code Modulation (PCM) of voice frequencies*. +- CCITT Recommendation G.722. (1988), *7 kHz audio-coding within 64 kbit/s*. +- ITU-T Recommendation T.120 (1996), *Data protocols for multimedia conferencing*. + +## 3.2 Terminology + +For the purposes of this Recommendation, the following definitions apply. + +**3.2.1 audiographic conference service:** An international service, offered by Administrations, enabling participants to conduct a real-time multimedia conference in which audio signals are exchanged together with non-voice graphics information, without motion video. + +**3.2.2 chairman:** The user who chairs the meeting and manages the conference. + +**3.2.3 collaborative document handling service (CDH):** A service that provides bidirectional transfer of data between two or more locations so that users are able to work on a common document, for drafting or amending it collectively. + +**3.2.4 communication modes:** A communication mode is defined by the various channels supporting the media used for the conference. Changes in communication mode may occur during the course of the conference, in order to set up or eliminate one of the media, or to change its quality level and thus the bit-rate allocation. It may be used for instance to temporarily add a channel for transmitting still pictures. + +**3.2.5 conference controller:** The user who manages the call during the conference. + +**3.2.6 conference convener:** The user who has organized the conference. + +**3.2.7 conference room:** In this Recommendation, the expression, conference room, refers to any location where a multimedia conference terminal is located. This may be a specially prepared group conference facility or an individual desktop terminal. + +**3.2.8 multipoint conference unit (MCU):** Equipment that provides multipoint connections among three or more conference rooms. + +**3.2.9 muting:** Preventing sound to be transmitted from a terminal equipment. + +**3.2.10 presenter:** The participant in a conference who controls the visual information being actively discussed. + +**3.2.11 videoconference service:** An audiovisual conversational conference service providing bidirectional real-time transfer of voice and motion video between groups of users in two or more separate locations. Although the audio and motion video information are the essential part of the service, other types of information, such as high resolution still pictures, text or graphics may also be exchanged. + +**3.2.12 videophone service:** An audiovisual conversational service providing bidirectional symmetric real-time transfer of voice and motion video between two locations. The minimum requirement is that under normal conditions the picture information transmitted is sufficient for the adequate representation of fluid movements of a person displayed in head and shoulders view. + +# 4 Description + +## 4.1 General description + +A multimedia conference service provides real-time communication between several users in different locations, combining a good audio facility with motion video of participants and/or transmission of multimedia information. The service is applicable to companies' private conference rooms as well as to public-access conference rooms for hire on an occasional basis. It is applicable to a variety of types of multimedia conference terminals such as: + +- dedicated studios equipped for multimedia conferences; +- multipurpose meeting rooms used only part-time for teleconferencing; +- portable or roll-about equipment which can be moved from one room to another to provide temporary service; +- terminal equipment for individual participants, e.g. microcomputer based terminals. + +The service is bidirectional via telecommunication networks, and provides for interconnection of two or more multimedia conference terminals on an equal basis. Other types of terminals may be added to the conference, such as videotelephones or even plain telephones; although they will usually have some limitations on the capability to send and receive all the different types of information used in a multimedia conference call, they will at least be able to exchange speech allowing their users to take part in the discussion; this is described in clause 8 on intercommunication. + +When the conference includes more than two terminals, a Multipoint Conference Unit (MCU) is usually required. All locations are connected individually to an MCU which elaborates a selection or the appropriate combination of these signals for each of the locations, and manages the signalling and the optional channels. + +Multimedia conference services are essentially built around the communication task conferencing described in B.2/F.700. Other communication tasks (receiving and sending) are optional. + +## 4.2 Functional model + +In a multimedia conference, two or more terminals exchange multimedia information through an interconnection system, under the control of a control unit (Figure 1). The interconnection system includes equipment for switching and/or combining the information from the different terminals and one or several networks. + +![Functional model diagram showing four terminal equipments connected to a central interconnection system, which is controlled by a control unit.](367926125450c2bc3f4bdca9d59a62ba_img.jpg) + +The diagram illustrates the functional model of a multimedia conference. At the center is an oval labeled "Interconnection system". Four rectangular boxes, labeled "Terminal equipment 1", "Terminal equipment 2", "Terminal equipment 3", and "Terminal equipment 4", are arranged around this central system. Solid lines connect each terminal equipment to the interconnection system, representing "User information". Dashed lines connect each terminal equipment to a rectangular box labeled "Control unit" positioned above the interconnection system, representing "Control signals (call control and conference control)". A dashed line also connects the control unit directly to the interconnection system. A legend at the bottom left shows a solid line for "User information" and a dashed line for "Control signals (call control and conference control)". A small code "T0104700-95/d01" is located at the bottom right of the diagram. + +Functional model diagram showing four terminal equipments connected to a central interconnection system, which is controlled by a control unit. + +NOTE – The physical boundaries of equipment are independent from the functional boundaries shown on the figure. For instance an MCU usually includes a control unit and other units, dedicated to the various media components, which perform switching and/or combining actions and functionally belong to the interconnection system. + +In the special case where there are only two terminals, they are connected point-to-point and a control unit is not needed. + +FIGURE 1/F.702 +**Functional model** + +## 4.3 Configuration + +The configuration may be point-to-point between two multimedia conference rooms, or multipoint-to-multipoint between several. In the latter case the terminals are usually connected through one or more Multipoint Conference Units (MCUs). The MCUs have to fulfill three functions: + +- managing the call, setting up and closing the connections; +- managing the conference, through control and indication signals exchanged with the terminals; +- handling the signals received and sent on each connection, switching, distributing, multiplexing and when necessary adapting and combining them as appropriate. + +### Types of configurations + +The multipoint configurations can be subdivided into: + +- multichannel multipoint; +- shared channel multipoint; +- switched multipoint. + +Including the point-to-point configuration, the following four types of configurations may thus be set up (see Figure 2). + +![Figure 2/F.702: Types of configurations. The diagram shows three network topologies: 1. Point-to-point configuration (case a): Two nodes, A and B, connected by a single line. 2. Multichannel multipoint configuration (case b): Four nodes, A, B, C, and D, arranged in a square with all possible connections between them (fully connected mesh). 3. Star configuration around an MCU (cases c) and d): A central square node labeled 'MCU' connected to four peripheral nodes, A, B, C, and D. Below the diagrams, a legend shows two nodes, A and B, with a dashed line between them, labeled 'terminal equipment'. A small code 'T0104710-95/d02' is present in the bottom right of the diagram area.](c2fc2621e8206d24427b56bcb2398fc0_img.jpg) + +Figure 2/F.702: Types of configurations. The diagram shows three network topologies: 1. Point-to-point configuration (case a): Two nodes, A and B, connected by a single line. 2. Multichannel multipoint configuration (case b): Four nodes, A, B, C, and D, arranged in a square with all possible connections between them (fully connected mesh). 3. Star configuration around an MCU (cases c) and d): A central square node labeled 'MCU' connected to four peripheral nodes, A, B, C, and D. Below the diagrams, a legend shows two nodes, A and B, with a dashed line between them, labeled 'terminal equipment'. A small code 'T0104710-95/d02' is present in the bottom right of the diagram area. + +FIGURE 2/F.702 +**Types of configurations** + +#### Case a) Point-to-point configuration + +Two conference rooms are directly connected (without any MCU). Conference management is by bilateral negotiation between the terminals. + +#### Case b) Multichannel multipoint configuration + +Three or more multimedia conference rooms are connected two-by-two so that each of them receives the signals from all the others, and may use them in various ways. For instance the sounds may be mixed or directed to separate loudspeakers. If video is present, each terminal permanently receives the images from all other locations and displays them simultaneously on separate screens or on different windows of a unique screen. Various schemes may be used for exchanging documents. Limitation to the number of participants comes from the number of channels available in each location and the number of equipment for presenting the information to the users (or the number of inputs to these equipment), e.g. the number of images that can be displayed simultaneously by the terminal equipment. An MCU is not mandatory, but it may be used for managing the conference. + +#### **Case c) Shared channel multipoint** + +This configuration requires a multipoint conference unit. This MCU receives signals from every terminal, and combines them to elaborate the signals sent out to each terminal. This may be done by multiplexing them in a higher bit-rate channel (e.g. four or five H0 channels into one H1 channel). It may also be achieved by mixing the sound signals, putting the various pictures into different windows (when applicable) and broadcasting the data channels if they are present. The MCU also processes the control and indication signals. + +If the number of available multiplexed channels or the number of available windows for the video is smaller than the number of terminals minus one, then a selection must be made inside the MCU. This applies mainly to the pictures in a videoconference. The pictures may be chosen by each user, or by the Chairman if there is one, or they may be those of the latest speakers, or any combination of these. More details on this case are for further study. + +#### **Case d) Switched multipoint configuration** + +This configuration requires at least one multipoint conference unit. This MCU receives signals from every terminal; it selects according to predetermined rules or to specific commands the signals sent to each terminal; it handles the signalling, commands and indications, forwards them when necessary and returns the appropriate answers; it manages the optional channels and broadcasts the signals received on these channels. + +Most often the switching only applies to the video signals, because data may be multiplexed into a common packet switched data channel and the sound signals are usually added in the MCU so that each terminal receives the sounds from all other terminals excluding its own; however, this addition process may be restricted to a few terminals in order to limit the noise level or the unwanted disturbances if the number of participants is large. Alternately, the sound may also be switched together with the image. + +Several MCUs may be required for some configurations either for technical or for economical reasons. In this case, each terminal is connected to one of the MCUs; the MCUs are interconnected and handle the signals to and from other MCUs similarly as to and from an ordinary terminal. + +From the network point of view, the connections that have to be established are as follows: + +- case a) point-to-point; +- case b) several point-to-point connections linking all locations two-by-two; +- case c) several point-to-point connections between each location and the MCU, which may be non-symmetrical if the network supports this type of connection; +- case d) several point-to-point connections between each location and the MCU, and possibly between MCUs if several are used; a single or multiple point-to-multipoint connection(s) may also be used in networks that support them. + +The four types of configurations are shown in Figure 2. Cases c) and d) are represented by the same diagram, but the functions of the MCU and the types of connections are different. + +When several MCUs are involved, this should be transparent to the users, except through the increased transmission delays and the possible limitations due to the capability of the inter-MCU connections. In the following subclauses the term MCU will be used indifferently for a unique MCU or for several interconnected MCUs, unless otherwise stated. Figure 3 shows several functionally equivalent configurations where the MCU function is either concentrated in one location or distributed between several. + +## **4.4 Roles of the participants** + +A participant may be allocated two types of particular privileges that allow his terminal to issue special commands to which the MCU will respond by the appropriate actions. These give him respectively the roles of the controller and the Chairman. + +The controller manages the call, and may perform the following: + +- accept new participants; +- disconnect a participant from the call; +- split the conference if this possibility is offered; + +- terminate the call; +- ask for a continuation of the call beyond the reserved time. + +![Three diagrams illustrating different MCU configurations: Star configuration around an MCU, Configuration with 2 MCUs, and Multi-MCU configuration.](007b053fe94a8348f75128a584503fd0_img.jpg) + +The diagram illustrates three functionally equivalent configurations for a conference call using Multipoint Control Units (MCUs): + +- Star configuration around an MCU:** A central MCU is connected to four terminals labeled A, B, C, and D. +- Configuration with 2 MCUs:** Terminal A is connected to MCU 1, which is connected to MCU 2. Terminal B is connected to MCU 1, and terminals C and D are connected to MCU 2. +- Multi-MCU configuration:** A chain of five MCUs (MCU 1 to MCU 5) is shown. Terminal A is connected to MCU 1, which is connected to MCU 3. Terminal B is connected to MCU 1, and terminal C is connected to MCU 1. Terminal D is connected to MCU 1, and terminal E is connected to MCU 1. Terminal F is connected to MCU 2, which is connected to MCU 3. Terminal G is connected to MCU 2. Terminal H is connected to MCU 3, which is connected to MCU 4. Terminal I is connected to MCU 3, and terminal J is connected to MCU 3. Terminal K is connected to MCU 4, which is connected to MCU 5. Terminal L is connected to MCU 4. Terminal M is connected to MCU 5, which is connected to terminal N. Terminal O is connected to MCU 5. + +T0104720-95/d03 + +Three diagrams illustrating different MCU configurations: Star configuration around an MCU, Configuration with 2 MCUs, and Multi-MCU configuration. + +FIGURE 3/F.702 +**Functionally equivalent configurations with single or multiple MCUs** + +The Chairman manages the conference, and may perform the following: + +- give the floor to a participant; +- mute another terminal; +- organize or allow a private talk; +- possibly allocate the right to transmit other information, although this facility control function may also be separately assigned. + +These two roles may be assigned to the same terminal or to different terminals. At the beginning of the call, the controller is usually the convener, but he may transfer this role to another terminal (see Note). + +Other participants may ask for the floor or for the authorization to transmit data, and the MCU forwards these requests to the chair-terminal. + +NOTE – Some MCUs may not be able to separate the two roles of controller and Chairman. In that case they should then be able to allocate the same joint controller-Chairman function to two different terminals (e.g. by issuing two identical tokens), and it will be left to both users to each perform only the functions pertaining to his role. Indeed this may be a preferred solution, because it allows the Chairman to leave to the controller some of the functions (for instance data management) that he may not be willing to perform; the Chairman may then have an easy terminal, to use with only very few controls. + +## 4.5 Terminal aspects + +In order to perform the basic functions necessary for multimedia conference services, the terminal equipment must include the following units necessary for audio communication: + +- one (or more) microphone(s); +- one (or more) loudspeaker(s); +- an audio codec; +- audio related controls; +- some means for identifying the speaker. + +The terminal must also include a control module to which these units will be connected, and a network interface unit. The other types of information require specific equipment detailed below. The terminal should include the equipment(s) for at least one media component besides audio. + +The equipment for handling multimedia documents includes one or more of the following functional units: + +- a microcomputer with a screen and optionally a printer; +- a still picture equipment with a camera or scanner, a screen and/or a printer; +- a telewriting equipment; +- a facsimile equipment. + +The basic equipment for video includes: + +- one (or more) camera(s); +- one (or more) screen(s); +- a video codec. + +When video is present, means must be provided for displaying the outgoing picture, either permanently or by substituting it on the screen to the incoming picture. If sound and/or image sources are locally switched, then an indication must be given of which one is sent out. + +NOTE – Testing of the outgoing picture: it should be possible for the user to put an off-line terminal into a self-test procedure, which includes the codec, in order to test and control the outgoing picture. + +Possible enhancements to the equipment in each location are for example: + +- voice switched or manually switched multiple microphones; +- multiple cameras performing some of the following functions: + - overall view of the room; + - partial views of the assembly; + - views of individual participants; +- the pictures from these cameras may be switched or combined in different windows; +- additional dedicated cameras; +- multiple screens, for instance for displaying side-by-side the different windows; +- zooming and panning; +- far-end camera control; +- various indications such as identification of the displayed location in a multipoint conference; + +- controls for conducting the conference, asking for the floor, etc.; +- auxiliary cameras for viewing a blackboard, objects, etc.; +- videotape recorder for displaying and sending any sequence of images, or recording the meeting; +- telewriting facility; +- pointer; +- multiple loudspeakers; +- equipment for encryption and decryption of user signals. + +## 4.6 Applications + +Some possible applications are indicated here as examples: + +- unconducted meeting between distant parties; +- formal conducted meeting between distant parties; +- panel discussion; +- elaboration of a document, with or without cooperative document handling; +- presentation of a report in a large company with subsequent discussion; +- presentation of an object (e.g. a new product) and comments on this; +- negotiation of a contract with possible private discussions or advice from invited experts; +- lecturing; +- distant education or training; +- tele-auction sale. + +## 4.7 Supplementary services + +For further study. + +# 5 Static aspects + +## 5.1 Service level + +### 5.1.1 General aspects of the service + +Throughout the communication phase, the communication task conferencing is used with the media component audio and one or more other media components. + +Retrieval of documents needed for the conference, either at the beginning or during the course of the conference may optionally be available; it uses the communication task receiving with one or more media components. + +Storage of the documents drafted during the conference in a server, and/or transmission of copies to users outside the conference may also be offered; it uses the communication task sending, with one or several media components; it may be used with the media components audio and possibly video for storing in a server or sending to other users the records of the meeting. + +The relationship between the three levels of the multimedia service reference model (service, communication tasks and media components) is shown in Figure 4 (see Recommendation F.700). + +### 5.1.2 Quality of service + +#### 5.1.2.1 General considerations + +The quality of service depends on the quality of each media component, of their association (e.g. synchronization of sound and image), of the control and procedures. It also depends on the combined quality of the various parts of the system and its environment. + +![A hierarchical diagram showing relationships between 'Multimedia conference services', three 'Communication tasks' (conferencing, receiving, sending), and six 'Media components' (audio, video, text, graphics, still Picture, other). It includes a legend for Main (solid), Secondary (dashed), and Optional (thin) relationships.](bd0b93e7a46ede276d0a3b79ac487bd9_img.jpg) + +The diagram illustrates the relationships between service level, communication tasks, and media components. At the top level is 'Multimedia conference services'. This connects to three communication tasks: 'Communication task conferencing', 'Communication task receiving', and 'Communication task sending'. Each of these tasks connects to six media components: 'Media component audio', 'Media component video', 'Media component text', 'Media component graphics', 'Media component still Picture', and 'Other media component'. + +Relationship types are defined by line styles: + +- Main relationship (mandatory):** Solid line +- Secondary relationship (one of them at least mandatory):** Dashed line +- Optional:** Thin line + +Specific connections: + +- Conferencing:** Main relationships to audio, video, and text; secondary relationships to graphics, still Picture, and other media component; optional relationship to receiving. +- Receiving:** Main relationships to audio, video, text, graphics, still Picture, and other media component; optional relationship to conferencing. +- Sending:** Main relationships to audio, video, text, graphics, still Picture, and other media component; optional relationship to conferencing. + +T0104730-95/d04 + +A hierarchical diagram showing relationships between 'Multimedia conference services', three 'Communication tasks' (conferencing, receiving, sending), and six 'Media components' (audio, video, text, graphics, still Picture, other). It includes a legend for Main (solid), Secondary (dashed), and Optional (thin) relationships. + +FIGURE 4/F.702 +**Relationships between service level, communication tasks and media components** + +Particular attention should be given to the MCUs. They support the main part of the complex multipoint procedures and are the crossroads of all signals. The degradation of the signals going through an MCU should thus be kept as low as possible, bearing in mind that it may build up in the case of cascaded MCUs. + +#### **5.1.2.2 Overall delay** + +The nominal overall delay is defined as the sum of the transmission delay and the characteristic delay of the conference terminals (see Note). When applicable, the characteristic delay of the MCU(s) must also be added. The subjective effect of this delay on the quality of service has to be taken into account because an excessive value may impair user acceptability. + +Under normal conditions the overall delay should not exceed 400 ms when video is not present, in conformance with Recommendation G.114 for the audiographic conference service or CDH without video. Another Recommendation is expected for audiovisual services. However, when no means are available to fulfill this condition, the call should not be rejected. + +NOTE – The characteristic delay of a terminal is the delay introduced by the terminal when the transmitted information contents for the various media is minimal; for the video, this means that there is no movement or only very small movements in the transmitted image, e.g. only lips and eyes of the users are moving. + +### **5.1.3 Security aspects** + +Adequate means should be provided to avoid unauthorized intrusion in the conference call, for instance by authentication or screening of new entrants. Whenever possible, a participant should be able to know the identities of all the others, except in specific cases such as teleteaching or telelecturing where this possibility may be limited to the conference controller. + +### **5.1.4 General charging principles** + +Three modes of charging should be offered: + +- charging to one of the participants; +- shared charging where each participant is charged according to his participation; +- charging to a third party. + +If charging is to the convener and he leaves the conference, other participants may continue without him, either with the same charging if the convener has accepted it beforehand, or charging to another participant who volunteers for it, or shared charging. + +Optionally other possibilities may be offered, such as: + +- multiple charging: several participants are charged, each for his own part and for the parties he invites; +- shared charging with invited parties: the part of the connection relevant to the invited party (or parties) is shared between the other parties, together with their own part. + +After disconnection of one or several parties during the course of a conference, it may happen that a simpler and cheaper configuration would be available; if the change of configuration is possible, then it should be proposed to the users. + +## **5.2 Communication task level** + +### **5.2.1 Communication tasks used for the service** + +Multimedia conference services are built around the communication task conferencing (see Annex B.2/F.700). Other tasks such as receiving and sending may also be used for specific purposes (see 5.1.1). + +### **5.2.2 Quality of service** + +#### **5.2.2.1 Differential delay between sound and image** + +When video is present, the differential delay between sound and image should be kept low enough for lip synchronism to be subjectively insured. + +## **5.3 Media component level** + +### **5.3.1 Media components used for the service** + +The media component audio (see Annex A.1/F.700) is set up at the beginning of the call and is usually present throughout, but it may be temporarily interrupted to increase the bit rate available for other media components. + +One other media component at least should also be present, permanently or not, i.e. one of the following components: + +- Video (A.2/F.700). +- Text (A.3/F.700). +- Graphics (A.4/F.700). +- Still pictures (A.5/F.700). +- Others (for further study). + +### **5.3.2 Quality of service** + +#### **5.3.2.1 Sound quality** + +The following two levels of sound quality are preferred: + +- the basic quality is 7 kHz bandwidth (conforming to Recommendation G.722) equivalent to present broadcast TV sound (level 1 of media component audio); +- the high quality is stereophonic and should be at least 7 kHz and preferably 15 kHz bandwidth (level 1 or 2 of media component audio). + +A lower quality, equivalent to that of the PCM telephony conforming to Recommendation G.711 (level 0 of media component audio) may also be used if there are difficulties or technical constraints in using the higher levels, or in intercommunication situations, or for short periods while priority is given to fast transfer of other types of information. + +For the quality levels of the media component audio, see A.1/F.700. + +Means for efficient echo control must be provided. When video is present, these means should have the capability to accommodate the large transmission delay that may be induced by the synchronization of sound and image (see 5.2.2). + +##### **5.3.2.2 Image quality for the video** + +When video is present, its quality must be sufficient for reproducing the fluid movements of a small group of people sitting at a table (see Note). Three grades of image quality may be offered: + +- basic videoconference quality: at least two persons with only small movements can be viewed; this is level 2 of the media component video; +- enhanced quality videoconference: the quality should be that of broadcast TV; this is level 3 of the media component video; +- high quality videoconference: the quality should be that of HD-TV; this is level 4 of the media component video. + +For the quality levels of the media component video, see A.2/F.700. + +A factor of quality related to video is the recovery time for picture build-up when the video source is changed; it must be kept low enough to avoid any disturbance to the meeting. Precise values are for further study. + +Complete definition and methods of evaluation of the quality of motion pictures require further studies. + +NOTE – When motion video is present, the smoothness of the movement in the displayed image depends on the ability of the system to convey rapid changes; the amount of information which has to be transmitted will of course increase with the affected area in the picture and the speed of the movement. It is generally considered that in a normal videoconference environment, only part of the picture will be moving and that any fast movement will be limited to a very small portion of the image. Some degradation of the picture quality such as jerkiness, blurring or other artifacts may be allowed during the brief and exceptional instants when these conditions are not met. + +# **6 Dynamic aspects** + +## **6.1 Activation phase** + +### **6.1.1 Provision, withdrawal** + +The multimedia conference services may be provided after prior arrangement with a service provider, or they may be generally available. + +### **6.1.2 Call establishment** + +Call establishment may be by reservation or on demand. + +### **6.1.3 Reservation** + +Reservation may be made from any multimedia terminal taking part in a conference. It may also be made from another terminal which could be for instance a simple telephone or videophone terminal, a computer with communication facilities, a videotex terminal or a fax. Reservation may also be made by a third party. + +To allow reservation through any of these terminals, three modes of reservation should be offered: + +- automatic reservation, through a computer with appropriate communication facilities or a multimedia terminal with equivalent capabilities; +- simplified automatic reservation, through audiotex or videotex; +- manual reservation, through an operator called from a telephone or other audiovisual terminal or by fax; the operator will then enter the reservation into the automatic system. + +In all three cases, the information required from the user should be identical and the procedures should be equivalent, and as far as possible independent of the type of multimedia conference. The conditions of reservation should also be the same (minimum delay between reservation and conference, conditions for cancellation or modification of the demand, conditions for early termination or prolongation, etc.). It is important for the user that the procedures from the different terminals and for the different types of multimedia conferences be harmonized as much as possible, so that a user will not have to learn the use of different procedures. + +An acknowledgement listing all the parameters of the reservation should be sent to the user whenever possible (i.e. if the user has a terminal suitable for this acquittal and known to the reservation server). + +The information required for reservation are: + +- date and time of the proposed conference; +- duration or time for ending the conference; +- type of multimedia conference and quality level; +- access numbers of participating terminals; +- in-call modifications, if offered by the network and MCU (additions, disconnections, splitting, change of type); +- additional functions required (any function not offered by the basic MCU); +- billing process (to one user, shared or other) and identification of the charged user(s). + +The following parameters will be defined in other Recommendations or agreed between Administrations: + +- minimum booking period; +- incrementation steps for defining the booking period; +- minimum notice time for booking; +- conditions and possible charges for cancelling or postponing the conference, or for changing information given at booking time; + +- conditions and charges for in-call modifications not defined when the reservation was made; +- conditions and charges for prolongation or early termination of the call. + +### **6.1.4 On demand** + +Three types of procedures may be offered for direct dialling: + +- 1) point-to-point calls for establishing direct point-to-point connections between terminals; +- 2) meet me conference calls where each participant calls the MCU at a pre-arranged time; +- 3) add-on conference calls where the convener calls the MCU and then the participants; alternately, the MCU may call the participants from a list transmitted by the convener or previously stored; a particular instance of this call procedure is the supplementary service preset conference, in which the user dials a specific abbreviated number causing the MCU to call a preset list of terminals. + +For the meet me conference, arrangements also have to be made for reservation of the MCU, or at least recognition of calling terminals by the MCU. Reservation of the MCU requires the same information as for complete reservation, or a subset of these (e.g. the number of participants may be needed but not their identity). Means for authentication are essential in order to prevent intrusion. + +## **6.2 Communication phase** + +### **6.2.1 Call set-up** + +In the case of reservation, the connections are established at the agreed time; if the terminals are already on, they will immediately enter the conference set-up phase; otherwise they will do so when they are switched on. + +In the on-demand establishment, the ordinary call set-up procedure is used in the first case (point-to-point configuration). + +Procedures for the other two configurations are left for further study. + +### **6.2.2 Addition of another party during the call** + +#### **6.2.2.1 Reservation case** + +In case of reservation, the additional connection is made at the scheduled time if defined at the time of booking, and set-up proceeds as above. An unscheduled addition may also be requested; it may be made by direct dialling if this is offered; otherwise the conditions are for further study. + +#### **6.2.2.2 On-demand case** + +During the course of the conference, a call can be made to another terminal in order to make it participate; this may be done by any party, but has to be accepted by the conference controller or by the party being charged for the conference; in case of shared billing, the new entrant must be advised that he will have to pay for his share. + +Conversely, a user external to the conference may ask for entry; this has to be accepted by the party being charged for the conference, or by the convener in case of shared billing. + +In both cases, all participants must be advised of the identity of the newcomer; the latter should be advised of the identities of all the parties already present and of the billing status. Addition of a new party is dependent on the availability of network and MCU resources. In case of denial, the caller should be advised of the reason. + +If the new terminal is not compatible with the type of the current conference, it may enter by one of the intercommunication or interworking processes described in clause 8, or the participants may be given the choice to change the type of their conference if the MCU supports it. + +### **6.2.3 Conference splitting** + +#### **6.2.3.1 Conference splitting in case of reservation** + +During the course of the conference, users may want to split into two or more groups. If this optional feature is supported by the MCU, it may be made at a preset time, in which case the users will be advised a few minutes beforehand; the time may also be left open, and it is then initiated by the convenor or the Chairman. + +Resuming the initial conference may be made at a preset time, or may be allowed at any time. In both cases the participants should receive some appropriate warning beforehand. + +Unscheduled splitting may also be offered, but is then dependent on the availability of network and MCU resources. + +Detailed procedures are for further study. + +##### **6.2.3.2 Conference splitting in case of on-demand establishment** + +If the MCU can support it, the possibility of splitting the conference may be offered to the users; this may, however, be conditioned by the availability of network and MCU resources; nevertheless, temporary exclusion of one or more participants so that the others may have a private conversation does not require any additional resources and should then always be granted. + +Resuming the initial conference will be allowed at any time. The participants should receive some audible warning. + +#### **6.2.4 Change of communication mode** + +Changes in user requirements arising during the course of the conference may require a change of communication mode, i.e. changes in the network connections or changes in the in-band channel that are transparent to the network. If the network and MCU support it, then in-call change of mode should be allowed when all the terminals have the capabilities for the new mode; this may be initiated by the controller or the convener. If it requires additional network resources, then it is subject to the availability of these resources, unless the conference has been set up by reservation and provisions have been made for the change when making the reservation. + +If some of the terminals do not have the capabilities required for the new mode, then the requesting party should be informed and given the choice between one of the following courses supported by the system: + +- a) the request may be abandoned; +- b) the change may be made using one of the intercommunication or interworking processes described in clause 8; +- c) if the MCU does not support terminals with secondary status, then the requested mode may be established between the terminals that have the required capabilities while the other terminals are temporarily excluded from the conference; +- d) alternately a less demanding mode may be proposed. + +Ideally this would be an automatic feature of the service that would allow fall-back to a predetermined or highest common level of capability unless the user makes an explicit choice. + +#### **6.2.5 Call prolongation in the case of reservation** + +In case of reservation, the convenor, or any participant in the case of shared billing, can ask for a prolongation of the call beyond the reserved time; this may be granted or not according to the availability of the network or the MCUs. Detailed procedures are for further study. + +#### **6.2.6 Terminal disconnection** + +Each participant may disconnect his terminal at any time. + +#### **6.2.7 Conference management procedures** + +#### **6.2.7.1 Conference set-up** + +Each terminal exchanges with the MCU all the information relevant to the operation of the conference that it is capable of handling; it may include the following: + +- terminal identification (access number, location, company, etc.); +- identification of participants (names, company, etc.); +- terminal capabilities. + +As soon as a terminal has completed conference set-up, it is considered active and should be sent an appropriate indication. All other active terminals should be informed of its entry in the conference, and it should receive a list of all the already active terminals (see 6.2.7.5 for exceptions to this rule). + +#### **6.2.7.2 Conference modes** + +There are two conference modes, conducted and unconducted. They are described in B.2/F.700. + +#### **6.2.7.3 Audio muting and video inhibition** + +Any participant may temporarily prevent his terminal from sending out audio or video signals. A suspended video should be replaced by an adequate notice. A terminal should provide a visual indication when its audio is muted or its video inhibited. + +#### **6.2.7.4 Checking the list of participants** + +Every participant may at any time request to check the identity of the other participating terminals. Except in special types of conferences (see 6.2.7.5), he should then receive a list of the terminals active in the conference at that moment. + +#### **6.2.7.5 Participants with special status** + +It may be possible to distinguish certain categories of participants and to give them different status, either permanently or temporarily. Examples are given hereafter. + +##### **Members and auditors** + +Members are able to take the floor at any time whereas auditors have to be given the floor by the Chairman. This feature can be used for instance if there is a panel whose members are allowed to speak freely, while in the audience only one auditor at a time is allowed to ask a question. Additionally, if the number of participants is large, mixing the voices of the panel members only will limit the addition of noise to a small number of sources and avoid unwanted disturbances. + +###### **Anonymous participants** + +In a public meeting, e.g. a lecture, a show or an auction sale, some or all participants may be allowed to remain anonymous, i.e. they will not be seen by other participants except the controller or the Chairman, and their names or other identification will not be made available either. + +#### **6.2.7.6 Videoconference with multiple pictures** + +Multiple pictures may be transmitted by various means. At the receiving end, they may be displayed on multiple screens or in windows on one screen. + +If the number of screens or windows is sufficient, then every participant can be viewed permanently in all locations and no switching is necessary. This is called continuous presence. In most cases however, the number of subpictures is smaller and a selection has to be made. The selection may be operated by the participants or automatically or a mixture of both. The criteria for automatic selection are an implementation matter (e.g. latest speakers, or Chairman and latest speakers, or members of a panel, etc.). However, the following rules should be applied: + +- preferably only one window should be changed at a time, or two subpictures interchanged (e.g. if one window is larger than the others); +- the number of subpictures should be limited in order to keep them large enough; +- when the MCU is using several windows, a terminal should not be allowed to split them into smaller windows; +- the commands, if available, should be kept simple to use and should not be a burden on the Chairman. + +Appendix I shows possible commands and procedures for controlling multiple subpictures. + +#### **6.2.7.7 Access to other services** + +###### **6.2.7.7.1 General** + +There are two ways of accessing other services: + +- access by one of the participants, who forwards the multimedia documents or information received to the other participants in the conference; this requires an additional access to the network, the appropriate rights of access to the service and possibly specific rights for forwarding documents to other users; the service may be public or private, e.g. retrieval of multimedia documents from a private server; +- access through the MCU, which then has to provide an access to the network, rights of access to the service (unless it can use the rights of one or several of the participants in the conference through some authentication procedure), and possibly conversion of protocols, of coding or of media. + +###### **6.2.7.7.2 Retrieval services** + +For further study. + +## **6.3 Termination phase** + +### **6.3.1 Call termination** + +Call termination may be initiated: + +- by the convener or controller; +- by the network or the MCU at the end of the reservation period; in this case participants should be given notice beforehand, allowing time for them to request for a prolongation; +- by the network when only one participant is left; before the call is terminated, he may be offered the possibility to make reservations for future conferences; +- by the network in the case of direct dialling, if resources are needed for setting up a reserved conference; in this case, the users should be given notice long enough in advance to leave them reasonable time for bringing their conference to a temporary end; optionally, they may be offered the choice of automatically resuming the conference (after prior notice) when resources will become available again. + +# 7 Service profiles + +## 7.1 Different types of multimedia conference services + +All multimedia conference services have in common the capability for transmission of sound with a good quality. They can be divided into different types of profiles according to the other kinds of information exchanged: + +- Audiographic conference, using various types of documents such as still pictures, graphics, texts, under different forms such as facsimile or files. +- Videoconference, with moving pictures and optionally all types of documents used in an audiographic conference. +- Collaborative Document Handling (CDH), using microcomputers with file transfer and joint editing capabilities. + +Although possibly not supplying the same quality level, the videophone conference service is also included as a special case of a multimedia conference service. This service allows communication between three or more videophones, and thus belongs to the services enabling group communication. The multipoint connection may be established through the same types of MCUs as for the videoconference service. + +## 7.2 List of service profiles + +Taking into account the different types of facilities that may be used for handling documents, and the various options that may be offered, the following profiles of the generic multimedia conference service are defined. As stated in Recommendation F.700, these profiles are references for the offering of services, but do not preclude any enhancement or additional functions. Additional profiles may be added to respond to evolving user needs. + +### 1) *Videoconference service* + +- Profile 1a: Basic videoconference: wideband audio, CIF video for limited movements; conference conductor facilities are optional. +- Profile 1b: Standard videoconference service: wideband audio, CIF video, conference conductor facilities. +- Profile 1c: Enhanced videoconference: wideband audio, TV quality video, conference conductor facilities (see Note). +- Profile 1d: High quality videoconference: stereophonic audio, HD-TV quality video, conference conductor facilities (see Note). + +### 2) *Audiographic conference service* + +This service may be offered with two levels of audio quality and with three types of exchanged data. The minimum audio quality is level A0, equivalent to PCM telephony (profiles 2a); the basic audio quality is level A1, wideband audio. The three types of exchanged data are still pictures from a camera or equivalent system (profiles 2a1 and 2b1), fax pictures with hard copy capability (profiles 2a2 and 2b2), and microcomputer files (profiles 2a3 and 2b3). + +- Profile 2a1: Basic audiographic conference with still pictures: basic audio, multipoint communication with still pictures, annotation and pointer. +- Profile 2a2: Basic audiographic conference with fax functions: basic audio, multipoint communication with fax. +- Profile 2a3: Basic audiographic conference with file transfer: basic audio, multipoint communication with file transfer. +- Profile 2b1: Standard audiographic conference with still pictures: wideband audio, multipoint communication with still pictures, annotation and pointer. +- Profile 2b2: Standard audiographic conference with fax functions: wideband audio, multipoint communication with fax. +- Profile 2b3: Standard audiographic conference with file transfer: wideband audio, multipoint communication with file transfer. + +### 3) *Collaborative Document Handling (CDH)* + +- Profile 3: Collaborative document handling: wideband audio, multipoint communication with file transfer and joint editing; optionally video (see Note). + +### 4) *Videophone conference service* + +- Profile 4a: Low bit-rate videophone conference: basic audio, minimum video. +- Profile 4b: Basic videophone conference: basic audio, basic video. + +NOTE – The values shown for the profiles 1c, 1d and 3 are provisional. + +| Service | Profile | Audio | Video | Text, graphics and/or still pictures | File transfer | Control functions | +|-------------|--------------------------------------------------------------------------|-------|-------|--------------------------------------|---------------|-------------------| +| VCS | P1a | A1 | V2* | O | O | O | +| | P1b | A1 | V2 | O | O | M | +| | P1c | A1 | V3 | O | O | M | +| | P1d | A2 | V4 | O | O | M | +| AGCS | P2a1/2 | A0 | – | 1M | O | M | +| | P2a3 | A0 | – | M | M | M | +| | P2b1/2 | A1 | – | 1M | O | M | +| | P2b3 | A1 | – | M | M | M | +| CDHS | P3 | A1 | O | O | M | M | +| VPCS | P4a | A0 | V0* | O | O | O | +| | P4b | A0 | V1* | O | O | O | +| A0, A1, ... | Minimum mandatory audio quality level | | | | | | +| V0, V1, ... | Minimum mandatory video quality level | | | | | | +| V0*, V1* | Minimum mandatory video quality level, with a limited amount of movement | | | | | | +| M | Mandatory | | | | | | +| 1M | One at least of the media components mandatory | | | | | | +| O | Optional | | | | | | + +Note that additional profiles can be obtained by appropriate combinations of the above profiles: + +- any combination of P2a1, P2a2 and P2a3 or P2b1, P2b2 and P2b3 provides an audiographic conference profile; +- P1a, P1b or P1c with any combination of P2b1, P2b2 or P2b3 provides a videoconference profile with data capability; +- P4 with any combination of P2a1, P2a2 or P2a3 provides a desktop videotelephone conference profile. + +The following table recapitulates the requirements for the various types of profiles in terms of media components and their minimal levels of quality. + +| Media components | P1-VCS | P2-AGCS | P3-CDH | P4-VPH | +|--------------------------------------|--------------------------------------------------------------------------------------------------|---------|--------|--------| +| Audio | A1 | A0 | A1 | A0 | +| Video | V2 | – | O | V1 (*) | +| Text, graphics and/or still pictures | O | M | O | O | +| Data | O | O | M | O | +| A0, A1, V0, V1, V2 | Are the minimal levels for audio and video | | | | +| M | Mandatory (i.e. it must be available but not necessarily used) | | | | +| O | Optional | | | | +| (*) | If the low bit-rate videotelephony was also included, then the minimal video quality would be V0 | | | | + +# **8 Interworking and intercommunication** + +## **8.1 General** + +Terminals with different characteristics and capabilities may be connected in a multimedia conference. Three ways of interworking or intercommunication may then be provided: + +- use of a common mode which every terminal in the call can handle; this will decrease the quality and functionalities to those of the terminal with the lowest quality level for each media component; however, communication is always possible because all terminals have at least the basic audio capability; +- use of their normal mode for the terminals supporting higher quality levels, with partial functionalities only for the terminals with lower quality levels, which are thus given a secondary status; for instance some terminals may have audio and video while others may have audio capability only; it may also happen that terminals have different and incompatible data capabilities that they will not be able to use, e.g. facsimile and still pictures; +- transcoding in the MCU so that each terminal receives and transmits the signals it is able to handle. + +## **8.2 Interworking/intercommunication between two terminals** + +This subclause describes various situations of interworking or intercommunication between two terminals. They are supposed to be either directly connected or through an MCU without transcoding capabilities. + +### **8.2.1 Different types of audiographic conference terminals** + +These terminals can communicate with audio and with their common graphic components. In the worst case they may have incompatible graphic components and will then communicate with sound only. + +### **8.2.2 Audiographic conference and videoconference terminals** + +If the videoconference terminal has no optional graphics components, then the communication can only use sound. If it has graphic components, then the situation is the same as above: those components that are common, if any, can be added to the common mode. + +### **8.2.3 Different types of videoconference terminals** + +The higher level terminal can usually fall back to the lower level video mode. If this is not the case, then the common mode includes only sound. + +If graphic components are present, then those that are common, if any, can be added to the common mode. + +### **8.2.4 CDH and other types of terminals** + +CDH terminals may be considered as videoconference terminals with data handling if they have video capability, and as audiographic conference terminals if they do not; their intercommunication process is therefore the same as for these terminals. + +## **8.3 Interworking/intercommunication between several terminals** + +### **8.3.1 Different types of multimedia conference terminals** + +There is often a choice between different modes. In some cases, the choice will be obvious and can be automatic, for instance when only one terminal has higher capabilities than the others. In other cases, choices may have to be made between a common mode with lower quality level and/or reduced functionalities, and higher quality with one or more terminals relegated to secondary status (if the MCU supports this). + +A default mode may be automatically provided, but whenever there is a choice, the final decision should be left to the users. + +The mode may also be changed during the course of the conference. For instance a basic videoconference terminal unable to handle multimedia documents may have to be temporarily excluded from the conference (or from the video medium) while the other terminals exchange documents. In an audiographic conference with different types of terminals, it may be decided at times to exchange files, giving secondary status to terminals that do not have this capability, and at other times to transmit facsimile documents, giving secondary status to other terminals that do not support this facility. It is a desirable feature that common mode arbitration be provided by the MCU on a default basis. + +### **8.3.2 Videophone and other multimedia conference terminals, multimedia conference and videophone terminals** + +#### **8.3.2.1 Audiographic conference and videophone terminals** + +These terminals can intercommunicate with sound only. Basic videophones have a lower quality level than audiographic conference terminals which will have to fall back to this lower level to establish communication. However, with broadband videophones, the situation may be the reverse. The ability to select one video picture in the video terminal or in the MCU for sending it as a still picture to the AGC terminal would be a desirable feature. + +#### **8.3.2.2 Videoconference and videophone terminals** + +If the video codecs have a common mode, then this can be used for intercommunication. However, with a basic videophone terminal, this may not be desirable because of the limited image quality, insufficient to show several participants simultaneously. The other possibility is then for the videoconference rooms to communicate between themselves in their usual mode, while the videophone terminal participates with sound only. + +#### **8.3.3 Addition of a telephone terminal** + +The capability for adding a telephone to a multimedia conference meeting should be offered. It will then participate with sound only, and possibly with any other terminal (e.g. facsimile, telewriting, etc.) that its user may have. + +#### **8.3.4 Communication with a facsimile terminal** + +The MCU may support the connection to a facsimile terminal. This entails the ability to extract the facsimile signals from the information received from other terminals in the conference in order to forward them to the facsimile terminal, and conversely to insert in the common data channel the information received from the facsimile terminal. + +# **Annex A** + +## **Multipoint Conference Units (MCUs)** + +(This annex forms an integral part of this Recommendation) + +### **A.1 General** + +In a multipoint conference between multimedia terminals, three or more terminals exchange audio, video and data signals, to which control and indication signals have to be added in order to ensure proper and harmonious progress of the conference. To provide multipoint communication among multiple terminals in the conference, the connections are usually established through one or several MCUs (Multipoint Conference Units): each terminal is connected to an MCU, and when several MCUs are present, connections are established between them. + +The MCUs have to perform the following functions: + +- a) controlling the call, i.e. establishing, modifying and releasing the connections, either directly or through the reservation system; +- b) controlling the conference, through control and indication signals exchanged with the terminals; +- c) handling, distributing and when necessary adapting (for instance transcoding) the signals transmitted by the terminals. + +The functional model of an MCU is shown in Figure A.1. It includes the following modules: + +- a control module performing the control functions; +- an audio module processing audio signals; +- a video module processing video signals; +- a data module processing text, still picture and graphics information; +- ports and interfaces. + +A port is a functional entity that handles all the signals to and from an external system, i.e. a terminal, another MCU or a node in the network performing some distributed functions (see A.5). + +Interfaces are the physical modules which adapt the signals entering or leaving the MCU. + +It is not mandatory for an MCU to include all these modules. Some MCUs may be dedicated to a specific service or to a limited set of service profiles. For instance an MCU dedicated to audiographic conferencing does not need a video module; an MCU dedicated to basic videotelephony may have no data module. Another case where the MCU may be simplified is when it is built for an environment in which some of its usual functions are not needed, for instance when some functions are provided by nodes in the network (see A.5). + +Note that this functional model applies to a single conference. An MCU is often able to support several independent conferences simultaneously. + +![Functional model of an MCU diagram showing internal modules (Control, Call control, Conference control, Audio, Video, Data) connected to multiple Ports (1, 2, ..., n) via Interfaces, which are then connected to a Network(s). Dashed lines represent internal control signals from the Control module to the other modules.](088921fa3f5a44c8551815122517eefd_img.jpg) + +The diagram illustrates the functional model of an MCU. On the left, a large box labeled 'MCU' contains several internal modules. At the top, a 'Control module' is connected to 'Call control' and 'Conference control'. Below these, 'Audio module', 'Video module', and 'Data module' are shown. Dashed lines represent internal control signals originating from the 'Control module' and connecting to each of these three modules. To the right of the internal modules, there are multiple 'Port' boxes labeled 'Port 1', 'Port 2', and 'Port n'. Solid lines connect each of the internal modules ('Call control', 'Conference control', 'Audio module', 'Video module', and 'Data module') to each of the ports. These ports are then connected to a vertical 'Interfaces' box. Finally, the 'Interfaces' box is connected to an oval labeled 'Network(s)'. + +----- Internal control signals + +T0104740-95/d05 + +Functional model of an MCU diagram showing internal modules (Control, Call control, Conference control, Audio, Video, Data) connected to multiple Ports (1, 2, ..., n) via Interfaces, which are then connected to a Network(s). Dashed lines represent internal control signals from the Control module to the other modules. + +FIGURE A.1/F.702 +**Functional model of an MCU** + +### A.2 Call control functions + +The call control functions vary with the method for call set-up. If the call set-up is only by reservation, then the connections are controlled by the reservation system, and there is no need for the MCU to take any particular action. However, the MCU may exchange information with the reservation system in order to: + +- confirm that all connections are correctly established; +- add or drop a terminal; +- change the bit rate of all the connections depending on the capabilities of the terminals and the requests from the call controller; +- reduce the bit rate of a terminal with secondary status; +- split the conference or merge a split conference, if this entails changes in the network resources (e.g. in the case of multiple MCUs); +- request or drop a connection to a server; +- request a prolongation of the conference; +- make a reservation for another conference or request information on already booked future conferences; +- provide or obtain charging information; +- put an end to the conference. + +Requests affecting the cost of the call have to be made by the call controller, and the MCU forwards them to the reservation system. The response to requests requiring increased resources is dependent upon the availability of those resources. + +In the case of a meet me conference by direct dialling, the situation is similar: no action on the network for call control is required from the MCU; it may however set up, modify or drop connections for the above-mentioned purposes, or call a reservation system concerning future conferences. + +In an add-on conference, after the first connection has been set up between the call controller and the MCU, the connections to the other terminals may be established by the call controller and then transferred to the MCU if the network supports this procedure. Alternately the connections may be established by the MCU from a list provided by the call controller. During the course of the conference, the same process may be repeated in order to add other terminals. Other actions for call control are the same as in a meet me conference. + +## A.3 Conference control functions + +#### Mandatory functions + +The mandatory functions are those necessary to support a non-conducted conference, and also the transmission of means for identifying the speaker. + +#### Optional functions + +Support of a conducted conference is optional but highly desirable. Other optional control functions are related to the allocation of bandwidth to the various media components, to individual choices by participants of the information they want to receive, to the contents of pictures with multiple windows, to indications on the identities of participants, to security features and to conference splitting. + +#### Description of the control functions + +The control functions vary according to the way each of the media components is handled. For instance if in each location the terminal permanently receives all sounds (usually by mixing them), all pictures (e.g. in several windows of a composite picture) and all other information on a common transmission medium, then no control function is mandatory, but optional functions may be offered. In the general case however, selections for audio and/or video have to be made and control functions are required. + +The main control functions are described in B.2/F.700, in particular the various control modes (unconducted and conducted) and conference splitting. An example of the control of multiple subpictures is given in Appendix I. + +#### Selection of the transmission mode + +Changes in user requirements arising during the course of the call may result in changes in the transmission mode including changes in the connections; then it is part of the call control functions described in A.2. When it only implies reallocation of bit rates inside the existing connections and/or changes in the coding algorithms, then the procedure is simpler because it can be entirely handled within the MCU. Before making the change, the MCU must check whether all terminals can support the requested mode; if this is not the case, then the MCU should notify it and ask for confirmation; the possible choices are: abandon the change, choose another mode compatible with the capabilities of all terminals (or of most of them), relegate some terminals to secondary status or temporarily exclude them from the conference; in those last two cases, all participants should be informed of the new status of the terminals involved (secondary status or temporary exclusion); it is only optional for an MCU to support these status. + +#### Identifications + +If the image of the speaker is always transmitted, then no other means for identification is mandatory. In most cases however the MCU will send for identification purposes a number it has arbitrarily allocated to each terminal. In the initiation phase, it may leave it to the participants or to their terminals to exchange information on their identities and locations and to correlate them with these arbitrary numbers; it may also request this information from the various terminals and broadcast them to everyone. + +#### Security features + +Security features may include screening of the entrants in the conference, authentication of participants or terminals, encryption of signals and exchange of encryption keys. + +#### Status of the participants + +If there are participants with special status (see 6.2.7.5) then this is controlled by the MCU, and it affects the way the signals from these participants are handled. + +## A.4 Handling the various media components + +The MCU handles the various media components as described in B.2.2.2/F.700. This includes: + +- mixing and/or switching the audio signals; +- switching and/or combining the video signals; +- distributing the other information types on common data channels and controlling these channels. + +## A.5 Distributed MCU functions + +As stated in 4.3, the term MCU has been used here indifferently for a unique equipment or for a system of distributed and interconnected equipment because they are functionally equivalent. Examples of configurations with single and multiple MCUs are shown in Figure 2. Each terminal is connected to one of the MCUs, and the MCUs exchange the appropriate signals in order to behave as a single entity, except for possible transmission delays and limitations in the available user signals due to the capabilities of the inter-MCU connections. + +Other configurations may be set up in networks offering broadcasting, selection and/or non-symmetrical connections. In addition to the three configurations described above, these networks may support configurations in which some functions of the MCUs are distributed in nodes of the network, i.e. points in the network capable of accomplishing functions such as switching, broadcasting, distributing, possibly multiplexing of signals. These nodes may be considered as carrying in a distributed way some simple functions of MCUs. An example of broadcasting and multiplexing in a node is illustrated in Figure A.2: the MCU transmits information to a distant node which broadcasts them to the neighbouring terminals connected to it; it multiplexes the signals received from these terminals on a single channel to the MCU. + +![Diagram of broadcasting and multiplexing in a node. An MCU on the left is connected to terminals A and B. It is also connected to a central Node via two parallel lines. The Node is connected to terminals C, D, and E. Arrows indicate signal flow: from A and B into the MCU, from the MCU into the Node, and from the Node out to C, D, and E. A return arrow is shown on the lower parallel line from the Node back to the MCU. The text 'T0104750-95/d06' is at the bottom right.](11edb7fcedf09ac6a817f8d7b8c61eec_img.jpg) + +Diagram of broadcasting and multiplexing in a node. An MCU on the left is connected to terminals A and B. It is also connected to a central Node via two parallel lines. The Node is connected to terminals C, D, and E. Arrows indicate signal flow: from A and B into the MCU, from the MCU into the Node, and from the Node out to C, D, and E. A return arrow is shown on the lower parallel line from the Node back to the MCU. The text 'T0104750-95/d06' is at the bottom right. + +FIGURE A.2/F.702 +**Broadcasting and multiplexing in a node** + +Another example of a node is the space segment of a satellite network. The satellite provides the multipoint function by broadcasting the signals it receives to all earth stations (Figure A.3). If each conference site is able to access a satellite earth station providing connection with the space segment, then all signals are available there. An MCU is needed on each site to select or combine the signals for each media component. However, if the terminals are able to handle multiple signals, e.g. by sending the audio signals to several loudspeakers and displaying the pictures on several screens, then a multichannel multipoint configuration is obtained (see 4.3) and MCUs are not required. + +![Diagram of a multipoint conference around a satellite network. A central 'Space Segment' (satellite) is shown at the top. Below it are three earth stations, each connected to an MCU. Zigzag lines represent radio signals between each earth station and the satellite. The text 'T0104760-95/d07' is at the bottom right.](dd5771673aececa53d42ece89218299d_img.jpg) + +Diagram of a multipoint conference around a satellite network. A central 'Space Segment' (satellite) is shown at the top. Below it are three earth stations, each connected to an MCU. Zigzag lines represent radio signals between each earth station and the satellite. The text 'T0104760-95/d07' is at the bottom right. + +FIGURE A.3/F.702 +**Multipoint conference around a satellite network** + +## **A.6 Ports and interfaces** + +The interfaces of the MCUs depend upon the network(s) to which they are connected and upon the configuration. An MCU may be connected to several networks. For instance an audiographic conference may be set up with the audio on the PSTN and other media components on a dedicated data network; in such a case each port is connected to two interfaces, one with the PSTN and the other with the data network. Another example of connection to different networks is a private MCU connected to local terminals through a private network and to distant terminals through a public network. + +In the first example above, there were two interfaces for each port. Conversely, the physical interfaces to the ISDN are usually primary rate accesses to which several 2 B or 6 B ports to different terminals or MCUs are jointly connected. + +## **A.7 Interworking/intercommunication** + +The various cases of interworking or intercommunication are presented in clause 8. + +Intercommunication between videoconference and videotelephony at the same bit rate is mandatory. It may exclude the data channels if both types of terminals do not have that capability. + +The possibility of joining a conference with a telephone terminal should be offered. + +All other cases of interworking or intercommunication are optional, in particular the capability of connecting terminals with secondary status and the capability of temporarily excluding a terminal while other terminals use a communication mode that it does not support. + +# **Appendix I** + +## **Control of multiple windows** + +(This appendix does not form an integral part of this Recommendation) + +### **I.1 General** + +Multiple windows offer many different possibilities and their control may thus be complex to operate. For the pictures broadcast to all terminals, it may therefore be done by the Chairman or left to the controller or to another participant (conference secretary). Some multipoint systems may also allow participants to individually request specific pictures to be sent to them. The commands should be kept as simple as possible and should be assisted by an automatic selection process based on detection of speech and on commands for taking or allocating the floor. A possible set of commands is shown hereafter. + +### **I.2 Starting, stopping and organizing multiple windowing** + +Commands should allow starting and stopping the multiple windows mode, and defining the number, dimensions and lay-out of the windows, preferably as standard partitions of the screen. This should not prevent implementers to display them on several different screens if they are available in the conference room. The minimum size of a window is dependent on the resolution of the picture. Some windows may be larger than others, but they are supposed to have the same format so that any subpicture may be displayed in any window. A window of a different format, e.g. double width or double height, could also be reserved to a specific terminal capable of generating this format. + +### **I.3 Selection of the windows' contents** + +Windows may be allocated through specific commands to participants nominally, according to a stable status (Chairman, lecturer, members of a panel, etc.) or to a transient status (present speaker, previous speaker, new entrant, etc.). + +It is necessary however to limit the number of changes in the picture composition, and to prevent the various elementary pictures from jumping simultaneously across the screen from one window to another; for instance this would happen each time some participant would take the floor if windows were allocated specifically to the last speakers. + +To avoid this, a list is defined, containing by default the latest speakers with priorities increasing in the order of taking the floor. Other participants may be inserted in the list by specific commands. Windows assigned to this list are considered equivalent; elementary pictures are allocated to these windows in the order of priorities; when a new elementary picture has to be displayed, it displaces the one with the lowest priority by taking its window; thus only one window is affected by the change (Figure I.2). + +Alternately, a specific window, for instance a window of larger size, may be allocated to the present speaker. Then if a participant already on the screen takes the floor, he will leave his window to the previous speaker and take the specific window (Figure I.2). + +If specific windows are also allocated to other participants (e.g. the Chairman or panel members), whose picture will take the larger window while they speak and then return to their original specific window, then three windows at most may be affected at a time (Figure I.3). + +Instead of being allocated to a given elementary picture, a window may be used for other purposes such as written announcements, graphics or for the time rotating display of all participants not present on the screen. + +A terminal may send a picture composed of several elementary pictures from different sources. This is not allowed while the MCU is itself transmitting composed pictures, because windows inside a window would be too small. However, two or more windows may be allocated to the same terminal if it is able to generate them and send them independently. + +#### Examples of picture composition + +Mode 1 - All windows equivalent, some participants $P_i$ have priority for having their pictures displayed, the remaining windows are allocated to the latest speakers $S_j$ (the present or latest non-priority speaker is $S_1$ , and the priority decreases with increasing rank $j$ ). + +Each priority participant is allocated a fixed window. Any other speaker is allocated a window when he takes the floor and keeps it until his priority has become the lowest so that he is replaced by a new speaker. Thus only one window may change at a time. Figure I.1 shows that: + +- when a priority participant takes the floor, there is no change in the picture composition and the ranks of all non-priority speakers are unchanged; +- when a non-priority speaker already on the screen takes the floor again, there is no change in the picture composition; he becomes $S_1$ and the ranks of other speakers increase by one unit if smaller than he had; ranks larger than he had are unchanged; +- when a participant not yet on the screen takes the floor, he takes the window of the former speaker with the lowest priority; he becomes $S_1$ and all non-priority previous speakers have their rank increased by one unit. + +### Key to Figures I.1 to I.3 + +Figures I.1 to I.3 show examples of the arrangement of windows on a screen and of the evolution of their contents. + +**P** denotes a participant who has been permanently allocated a specific window. **p** denotes (in Figures I.2 and I.3) a participant who has been allocated a priority for being on the screen, but in any available window. **S** denotes a non-priority participant who is on the screen because he is a previous speaker, but will drop out when his priority becomes too low. + +In the Figures, shading indicates a window that has changed, while the darker framing shows the window allocated to the present speaker. When the rank of a speaker **S** has changed, his previous rank is shown in parenthesis. Arrows show pictures moved into a window, or out of a window, or from one window to another. + +Mode 2 – A specific window is allocated to the participant who has the floor, while all other windows are considered equivalent. + +The pictures selected for display are the same as in mode 1, but they are moved so that the present speaker is always in the preferred window. If he was already on the screen, there is an exchange between the pictures of the present and former speakers; if the new speaker was not already on the screen, then the picture with the lowest priority is dropped and the picture of the former speaker takes its window. Thus two windows are affected by each change of speaker. This process is shown in Figure I.2. + +![](08dce7ad4c512fdf0c0cde60415fade6_img.jpg) + +| | | | +|----|----|----| +| P1 | P2 | P3 | +| S1 | S2 | S3 | +| S4 | S5 | S6 | + +| | | | +|----|----|----| +| P1 | P2 | P3 | +| S1 | S2 | S3 | +| S4 | S5 | S6 | + +| | | | +|---------------|---------------|---------------| +| P1 | P2 | P3 | +| S2
(ex S1) | S3
(ex S2) | S4
(ex S3) | +| S1
(ex S4) | S5 | S6 | + +| | | | +|---------------|---------------|---------------| +| P1 | P2 | P3 | +| S3
(ex S2) | S4
(ex S3) | S5
(ex S4) | +| S2
(ex S1) | S6
(ex S5) | new S1 | + +Initial state: +S1 has the floor + +P2 takes the floor: +no change on the screen + +S4 takes the floor: +again no change on the screen (S4 becomes the new S1) + +A participant not yet on the screen takes the floor (he becomes the new S1) + +T0105020-96/d08 + +FIGURE I.1/F.702 +**All windows equivalent** + +![](318886a86a1dcc59e1fc83db6f157c60_img.jpg) + +| | | | +|----|----|----| +| S1 | | P1 | +| | | P2 | +| S2 | S3 | S4 | + +| | | | +|----|----|----| +| P1 | | S1 | +| | | P2 | +| S2 | S3 | S4 | + +| | | | +|----|----|----| +| P2 | | S1 | +| | | P1 | +| S2 | S3 | S4 | + +| | | | +|---------------|---------------|---------------| +| S1
(ex S3) | | S2
(ex S1) | +| | | p1 | +| S3
(ex S2) | p2
(ex S1) | S4 | + +Initial state: +S1 has the floor + +P1 takes the floor + +P2 takes the floor + +S3 takes the floor (and becomes the new S1) + +| | | | +|---------------|----|---------------| +| S1
(new) | | S3
(ex S2) | +| | | p1 | +| S4
(ex S3) | p2 | S2
(ex S1) | + +| | | | +|---------------|----|---------------| +| S1
(ex S3) | | S2
(ex S1) | +| | | p1 | +| S4 | p2 | S3
(ex S2) | + +| | | | +|----|----|----| +| p1 | | S2 | +| | | S1 | +| S4 | p2 | S3 | + +| | | | +|-------------|----|---------------| +| S1
(new) | | S3
(ex S2) | +| | | S2
(ex S1) | +| p1 | p2 | S4
(ex S3) | + +A participant not yet on the screen takes the floor (he becomes the new S1) + +S3 takes the floor (he becomes the new S1) + +p1 takes the floor + +A participant not yet on the screen takes the floor (he becomes the new S1) + +T0105030-96/d09 + +FIGURE I.2/F.702 +**A specific window allocated to the participant who has the floor** + +Mode 3 – Specific windows are allocated to the participant who has the floor and to some other priority participants. + +Some of the priority participants, e.g. the Chairman, may be allocated fixed windows (Pi), while others, e.g. panel members, are allocated any window in a set (pk); the latter case decreases slightly the number of windows affected by changes of speakers, because after taking the floor those pictures may return to any free window in the set instead of a specific one. + +Again the pictures selected for display are the same as in modes 1 and 2, but the number of windows affected by each change of speaker is now two or three (see Figure I.3). + +![A 4x4 grid of diagrams showing window allocation states for participants S1, S2, S3, P1, p1, and p2. Each diagram shows a grid of windows with labels and arrows indicating changes in ownership or position. The grid is organized into four rows and four columns. Row 1: Initial state (S1 has floor), P1 takes the floor, S1 takes the floor again, P1 takes the floor again. Row 2: S3 takes the floor (and becomes the new S1), p1 takes the floor, p2 takes the floor, S1 takes the floor. Row 3: S3 takes the floor (and becomes the new S1), P1 takes the floor, p1 takes the floor, P1 takes the floor again. Row 4: A participant not yet on the screen takes the floor (he becomes the new S1), p2 takes the floor, A participant not yet on the screen takes the floor (he becomes the new S1), Another participant not yet on the screen takes the floor (and becomes the new S1).](7ae836e598020d937ed1478c2ef13025_img.jpg) + +Initial state: +S1 has the floor + +P1 takes the floor + +S1 takes the floor again + +P1 takes the floor again + +S3 takes the floor (and becomes the new S1) + +p1 takes the floor + +p2 takes the floor + +S1 takes the floor + +S3 takes the floor (and becomes the new S1) + +P1 takes the floor + +p1 takes the floor + +P1 takes the floor again + +A participant not yet on the screen takes the floor (he becomes the new S1) + +p2 takes the floor + +A participant not yet on the screen takes the floor (he becomes the new S1) + +Another participant not yet on the screen takes the floor (and becomes the new S1) + +A 4x4 grid of diagrams showing window allocation states for participants S1, S2, S3, P1, p1, and p2. Each diagram shows a grid of windows with labels and arrows indicating changes in ownership or position. The grid is organized into four rows and four columns. Row 1: Initial state (S1 has floor), P1 takes the floor, S1 takes the floor again, P1 takes the floor again. Row 2: S3 takes the floor (and becomes the new S1), p1 takes the floor, p2 takes the floor, S1 takes the floor. Row 3: S3 takes the floor (and becomes the new S1), P1 takes the floor, p1 takes the floor, P1 takes the floor again. Row 4: A participant not yet on the screen takes the floor (he becomes the new S1), p2 takes the floor, A participant not yet on the screen takes the floor (he becomes the new S1), Another participant not yet on the screen takes the floor (and becomes the new S1). + +T0105950-96/d10 + +FIGURE I.3/F.702 +Specific windows allocated to some participants and to the one who has the floor + +NOTE – A variant of this mode is to allocate windows permanently to the priority participants and, when they take the floor, to display them a second time in the speaker's window if it is larger; this reduces the number of windows affected by a change of speaker to two only. + +# ITU-T RECOMMENDATIONS SERIES + +- Series A Organization of the work of the ITU-T +- Series B Means of expression +- Series C General telecommunication statistics +- Series D General tariff principles +- Series E Telephone network and ISDN +- Series F Non-telephone telecommunication services** +- Series G Transmission systems and media +- Series H Transmission of non-telephone signals +- Series I Integrated services digital network +- Series J Transmission of sound-programme and television signals +- Series K Protection against interference +- Series L Construction, installation and protection of cables and other elements of outside plant +- Series M Maintenance: international transmission systems, telephone circuits, telegraphy, facsimile and leased circuits +- Series N Maintenance: international sound-programme and television transmission circuits +- Series O Specifications of measuring equipment +- Series P Telephone transmission quality +- Series Q Switching and signalling +- Series R Telegraph transmission +- Series S Telegraph services terminal equipment +- Series T Terminal equipments and protocols for telematic services +- Series U Telegraph switching +- Series V Data communication over the telephone network +- Series X Data networks and open system communication +- Series Z Programming languages \ No newline at end of file diff --git a/marked/F/T-REC-F.703-200011-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.703-200011-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..b18e5b07ba6a266813c0ce76e92f4e93c0d9fd4b --- /dev/null +++ b/marked/F/T-REC-F.703-200011-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c7a97195c2fd68ad39d82d19b808d1738dce3aa52e94325870a031b5a2ea7c01 +size 8275 diff --git a/marked/F/T-REC-F.703-200011-I_PDF-E/5e92d9e8e9ce204e405bff2367f88176_img.jpg b/marked/F/T-REC-F.703-200011-I_PDF-E/5e92d9e8e9ce204e405bff2367f88176_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ad22ddea2cc3a7d2421687bf9208240f1fc6abe2 --- /dev/null +++ b/marked/F/T-REC-F.703-200011-I_PDF-E/5e92d9e8e9ce204e405bff2367f88176_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4e2040847704268a47f7b4e8bac4e4450a9a17c54cebf0f7dfaa34278f9a3680 +size 10076 diff --git a/marked/F/T-REC-F.703-200011-I_PDF-E/af7916c89a458fdab6c3f443217388ae_img.jpg b/marked/F/T-REC-F.703-200011-I_PDF-E/af7916c89a458fdab6c3f443217388ae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..b91833d5e92e1c3eaaa5a756cb892d7456e05270 --- /dev/null +++ b/marked/F/T-REC-F.703-200011-I_PDF-E/af7916c89a458fdab6c3f443217388ae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:707df85d766a4e771c239ba615e0b8d536b10ae65afebd580c938bf1f0478776 +size 92953 diff --git a/marked/F/T-REC-F.703-200011-I_PDF-E/raw.md b/marked/F/T-REC-F.703-200011-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..ff3be4f0423e408d341e5ce0353d1a27a9a0e16c --- /dev/null +++ b/marked/F/T-REC-F.703-200011-I_PDF-E/raw.md @@ -0,0 +1,592 @@ + + +![ITU logo: A globe with a lightning bolt and the letters ITU.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with a lightning bolt and the letters ITU. + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.703** + +(11/2000) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Audiovisual services + +--- + +**Multimedia conversational services** + +ITU-T Recommendation F.703 + +(Formerly CCITT Recommendation) + +--- + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| AUDIOVISUAL SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# **Multimedia conversational services** + +## **Summary** + +The purpose of this ITU-T Recommendation is to define the Generic Multimedia Conversational Service and to describe their general features, regardless of the network environment in which these services are provided. These services allow conversational communication between two users in two different locations. The Generic Multimedia Conversational Services are one of the generic service types identified in ITU-T Recommendation F.700, and their description follows the methodology described therein. The user requirements for the various applications supported by the service are translated into generic service specifications, independent of the constraints of specific implementations. Thus this ITU-T Recommendation sets out the requirements that will satisfy the users' needs and allow proper intercommunication on an international basis of services offered by different providers and of equipments from different manufacturers. + +Various instances (profiles) of this generic service are considered here with a network independent view. Specific service descriptions for each network will be issued in other Recommendations. The detailed technical specifications of the terminal, network and protocol aspects for each of them are described in dedicated ITU-T Recommendations of the H.200, H.300, H.400 series. + +Along with the F.700 methodology, the description relies on the media components and the communication tasks described respectively in Annexes A and B of ITU-T Recommendation F.700 together with the middleware service elements in Annexes C for control and processing functions. These are service independent modular communication capabilities. + +## **Source** + +ITU-T Recommendation F.703 was prepared by ITU-T Study Group 16 (2001-2004) and was approved under the World Telecommunication Standardization Assembly Resolution 1 procedure on 17 November 2000. + +## FOREWORD + +ITU (International Telecommunication Union) is the United Nations Specialized Agency in the field of telecommunications. The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of the ITU. The ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics. + +The approval of Recommendations by the Members of the ITU-T is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +## INTELLECTUAL PROPERTY RIGHTS + +The ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. The ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, the ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementors are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database. + +© ITU 2001 + +All rights reserved. No part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and microfilm, without permission in writing from the ITU. + +## CONTENTS + +###### Page + +| | | | +|-------|-------------------------------------------------------------|---| +| 1 | Scope..... | 1 | +| 2 | Definition ..... | 1 | +| 3 | References and terminology..... | 1 | +| 3.1 | References..... | 1 | +| 3.2 | Terminology..... | 2 | +| 4 | Description..... | 3 | +| 4.1 | General description ..... | 3 | +| 4.2 | Functional model ..... | 3 | +| 4.3 | Configuration ..... | 3 | +| 4.4 | Terminal aspects ..... | 3 | +| 4.5 | Applications ..... | 4 | +| 4.6 | Supplementary services..... | 4 | +| 5 | Static aspects..... | 4 | +| 5.1 | Service level..... | 4 | +| 5.1.1 | General aspects of the service..... | 4 | +| 5.1.2 | Quality of service..... | 5 | +| 5.1.3 | Security aspects ..... | 5 | +| 5.1.4 | General charging principles..... | 5 | +| 5.2 | Communication task level ..... | 6 | +| 5.2.1 | Communication tasks used for the service ..... | 6 | +| 5.2.2 | Quality of service related to the communication tasks ..... | 6 | +| 5.3 | Media component level..... | 6 | +| 5.3.1 | Media components used for the service..... | 6 | +| 5.3.2 | Quality of service related to the media components..... | 6 | +| 6 | Dynamic aspects ..... | 7 | +| 6.1 | Activation phase ..... | 7 | +| 6.1.1 | Provision, withdrawal..... | 7 | +| 6.1.2 | Call establishment..... | 8 | +| 6.2 | Communication phase..... | 8 | +| 6.2.1 | Call set-up..... | 8 | +| 6.2.2 | Change of communication mode..... | 8 | +| 6.2.3 | Conference calls..... | 8 | +| 6.2.4 | Audio muting and video inhibition..... | 8 | +| 6.2.5 | Access to other services..... | 8 | +| 6.3 | Termination phase..... | 9 | + +| | | Page | +|-------|-------------------------------------------------------------|-------------| +| 7 | Service profiles ..... | 9 | +| 7.1 | Different types of Multimedia Conversational Services ..... | 9 | +| 7.2 | List of service profiles ..... | 9 | +| 8 | Interworking and intercommunication..... | 11 | +| 8.1 | General..... | 11 | +| 8.2 | Intercommunication with a telephone terminal ..... | 11 | +| 8.3 | Videophone and other multimedia conference terminals ..... | 11 | +| 8.3.1 | Audiographic conference and videophone terminals..... | 11 | +| 8.3.2 | Videoconference and videophone terminals..... | 11 | +| 8.3.3 | Communication with a facsimile terminal..... | 11 | +| 8.3.4 | Text telephone interworking..... | 11 | +| 8.3.5 | Total Conversation interworking..... | 11 | + +# Introduction + +The purpose of this ITU-T Recommendation is to define the Generic Multimedia Conversational Service and to describe their general features, regardless of the network environment in which these services are provided. These services allow conversational communication between two users in two different locations. The Generic Multimedia Conversational Services are one of the generic service types identified in ITU-T Recommendation F.700, and their description follows the methodology described therein. The user requirements for the various applications supported by the service are translated into generic service specifications, independent of the constraints of specific implementations. Thus this ITU-T Recommendation sets out the requirements that will satisfy the users' needs and allow proper intercommunication on an international basis of services offered by different providers and of equipments from different manufacturers. + +Various instances (profiles) of this generic service are considered here with a network independent view. Specific service descriptions for each network will be issued in other Recommendations. The detailed technical specifications of the terminal, network and protocol aspects for each of them are described in dedicated ITU-T Recommendations of the H.200, H.300, H.400 series. + +Along with the F.700 methodology, the description relies on the media components and the communication tasks described respectively in Annexes A and B of ITU-T Recommendation F.700. These are service independent modular communication capabilities. + + + +# **Multimedia conversational services** + +# **1 Scope** + +The purpose of this ITU-T Recommendation is to define the Generic Multimedia Conversational Service and to describe their general features, regardless of the network environment in which these services are provided. These services allow conversational communication between two users in two different locations. The Generic Multimedia Conversational Services are one of the generic service types identified in ITU-T Recommendation F.700, and their description follows the methodology described therein. The user requirements for the various applications supported by the service are translated into generic service specifications, independent of the constraints of specific implementations. Thus this ITU-T Recommendation sets out the requirements that will satisfy the users' needs and allow proper intercommunication on an international basis of services offered by different providers and of equipments from different manufacturers. + +Various instances (profiles) of this generic service are considered here with a network independent view. Specific service descriptions for each network will be issued in other Recommendations. The detailed technical specifications of the terminal, network and protocol aspects for each of them are described in dedicated ITU-T Recommendations of the H.200, H.300, H.400 series. + +Along with the F.700 methodology, the description relies on the media components and the communication tasks described respectively in Annexes A and B of ITU-T Recommendation F.700. These are service independent modular communication capabilities. + +# **2 Definition** + +The Multimedia Conversational Services provide real time transmission of voice together with motion video and/or various types of multimedia information between two users in two different locations. The information exchanged may include all information types. When moving pictures are present, their quality should be at least sufficient for the adequate representation of the fluid movements of a person displayed in head and shoulders view. + +NOTE – The smoothness of the movements in the reproduced picture is essentially dependent on the amount of motion with respect to the transfer rate and compression scheme of transmitted picture information. The above requirements are supposed to be met under such conditions where the amount of motion is limited or the throughput is high enough not to impair the received picture. Degradation may appear as increased blurring, jerkiness and/or various artefacts in the reproduced picture. + +The media components used are described in the Annexes A of ITU-T Recommendation F.700 on Audiovisual/Multimedia Services. Media component Audio (Annex A.1) is usually present, together with one or more of the media components Video (Annex A.2), Text (Annex A.3), Graphics (Annex A.4), Still Pictures (Annex A.5). + +# **3 References and terminology** + +## **3.1 References** + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; all users of this Recommendation are therefore encouraged to investigate the possibility of applying the + +most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. + +- ITU-T Recommendation F.700 (2000), *Framework Recommendation for audiovisual/multimedia services*. +- ITU-T Recommendation F.702 (1996), *Multimedia conference services*. +- ITU-T Recommendation G.114 (2000), *One-way transmission time*. +- CCITT Recommendation G.711 (1988), *Pulse code modulation (PCM) of voice frequencies*. +- CCITT Recommendation G.722 (1988), *7 kHz audio-coding within 64 kbit/s*. +- ITU-T Recommendation T.120 (1996), *Data protocols for multimedia conferencing*. +- ITU-T Recommendation T.140 (1998), *Protocol for multimedia application text conversation*. +- ITU-T H-series Recommendations – Supplement 1 (1999), *Application profile – Sign language and lip-reading real-time conversation using low bit-rate video communication*. + +## 3.2 Terminology + +**3.2.1 Collaborative Document Handling service (CDH):** A service that provides bidirectional transfer of data between two or more locations so that users are able to work on a common document, for drafting or amending it collectively. + +**3.2.2 communication modes:** A communication mode is defined by the various channels supporting the media used for the service. Changes in communication mode may occur during the course of the call, in order to set up or eliminate one of the media, or to change its quality level and thus the bit-rate allocation. It may be used for instance to temporarily add a channel for transmitting still pictures. + +**3.2.3 Multipoint Conference Unit (MCU):** Equipment that provides multipoint connections among three or more conference rooms. + +**3.2.4 muting:** Preventing sound to be transmitted from a terminal equipment. + +**3.2.5 videoconference service:** An audiovisual conference service providing bidirectional real-time transfer of voice and motion video between groups of users in two or more separate locations. Although the audio and motion video informations are the essential part of the service, other types of information, such as high resolution still pictures, text or graphics may also be exchanged. + +**3.2.6 videophone service:** An audiovisual conversational service providing bidirectional symmetric real-time transfer of voice and motion video between two locations. The minimum requirement is that under normal conditions the picture information transmitted is sufficient for the adequate representation of fluid movements of a person displayed in head and shoulders view. + +**3.2.7 total conversation service:** An audiovisual conversation service providing bidirectional symmetric real-time transfer of motion video, text and voice between users in two or more locations. + +**3.2.8 text telephone service:** An audiovisual conversation service providing bidirectional real time transfer of text and optionally audio between users in two locations. Audio may be transmitted alternating with text or simultaneously with text. + +# 4 Description + +## 4.1 General description + +A multimedia conversational service provides real time bidirectional communication via telecommunication networks between users in two different locations; it usually combines an audio facility with motion video of the users and/or transmission of multimedia informations; however the audio may not be present in some particular applications. The service is applicable to dedicated terminal equipments or to microcomputer based terminals. + +Multimedia Conversational Services are essentially built around the communication task Conversing described in Annex A.1/F.700. Other communication tasks (receiving, sending) may optionally also be used. The control functions are described in Annex C.2/F.700 – Middlewaren service element Conversation control. + +## 4.2 Functional model + +In a multimedia conversational service, two terminals exchange multimedia informations through a telecommunication network (Figure 1). + +![Functional model diagram showing Terminal A connected to a Network, which is connected to Terminal B.](5e92d9e8e9ce204e405bff2367f88176_img.jpg) + +``` +graph LR; TerminalA[Terminal A] --- Network((Network)); Network --- TerminalB[Terminal B]; +``` + +T1609370-01 + +Functional model diagram showing Terminal A connected to a Network, which is connected to Terminal B. + +**Figure 1/F.703 – Functional model** + +## 4.3 Configuration + +The basic configuration is point to point between two terminals communicating through a bidirectional connection. This connection is usually symmetrical, but in special cases the media components present in each direction may be different, or they may be the same but with different bit rates and quality of service. + +A conference call may be set up as a supplementary service in conformance with ITU-T Recommendation F.702, in which case the configuration is multipoint to multipoint; the terminals are then connected through a multipoint conference unit (MCU) that fulfils three functions: + +- a) managing the call, setting up and closing the connections; +- b) managing the conference, through control and indication signals exchanged with the terminals; +- c) handling the signals received and sent on each connection, switching, distributing, multiplexing and when necessary adapting and combining them as appropriate. + +## 4.4 Terminal aspects + +Audio is used for the service except in exceptional cases. Therefore in order to perform the basic functions necessary for multimedia conversational services, the terminal equipment should include the following units necessary for audio communication: + +- a microphone; +- one (or more) loud-speaker(s); +- an audio codec; +- optionally some audio related controls. + +The terminal must also include a network interface unit. The other types of information require specific equipments detailed below. The terminal should include the equipment(s) for at least one media component besides audio. + +The equipment for handling multimedia documents includes one or more of the following functional units: + +- a micro-computer with a screen and a modem; +- a still picture equipment with a camera or scanner, a screen and a modem; +- a telewriting equipment; +- optionally a printer. + +The basic equipment for video includes: + +- a camera; +- a screen; +- a video codec. + +When video is present, means must be provided for displaying the outgoing picture, either permanently or by substituting it on the screen to the incoming picture. + +NOTE – Testing of the outgoing picture: it should be possible for the user to put an off-line terminal into a self-test procedure, which includes the codec, in order to test and control the outgoing picture. + +## **4.5 Applications** + +Some possible applications are indicated here as examples: + +- various types of conversation between two distant parties, similar to a telephone conversation but enhanced with motion video, still pictures or text; +- conversation between two distant parties, where one of them at least has difficulty in hearing and uses another media, either because of a disability or because of a noisy environment; examples are conversation with sign language using a videophone, lip reading to help hearing impaired people, text conversation; +- elaboration of a document between two parties, with or without cooperative document handling. + +## **4.6 Supplementary services** + +For further study. + +# **5 Static aspects** + +## **5.1 Service level** + +### **5.1.1 General aspects of the service** + +Throughout the communication phase, the communication task Conversing is usually active with the media component audio and one or more other media components. + +Exchange of various types of documents may optionally be available, using the communication task Sending with one or more media components. + +The relationship between the three levels of the multimedia service reference model (service, communication tasks and media components) is shown on Figure 2. + +![Diagram showing relationships between a multimedia conversational service, communication tasks, and media components.](af7916c89a458fdab6c3f443217388ae_img.jpg) + +The diagram illustrates the relationships between a 'Multimedia conversational service' and its constituent parts. At the top, a box labeled 'Multimedia conversational service' is connected by a solid line (Main relationship) to two boxes below: 'Communication Task Conversing' and 'Communication Task Sending'. From 'Communication Task Conversing', a solid line connects to 'Media Component Audio' (oval), and dashed lines (Secondary relationship) connect to 'Media Component Video' (oval), 'Media Component Text' (oval), 'Media Component Graphics' (oval), 'Media Component Still Picture' (oval), and 'Other Media Component' (oval). From 'Communication Task Sending', a solid line connects to 'Media Component Still Picture' (oval), and dashed lines connect to 'Media Component Audio' (oval), 'Media Component Video' (oval), 'Media Component Text' (oval), 'Media Component Graphics' (oval), and 'Other Media Component' (oval). A legend at the bottom left defines the line types: solid for 'Main relationship (mandatory)', dashed for 'Secondary relationship (one of them at least mandatory)', and dotted for 'Optional'. The code 'T1607940-00' is in the bottom right. + +Diagram showing relationships between a multimedia conversational service, communication tasks, and media components. + +**Figure 2/F.703 – Relationships between service level, communication tasks and media components** + +### 5.1.2 Quality of service + +#### 5.1.2.1 General considerations + +The quality of service depends on the quality of each media component (see 5.3.1), of their association (e.g. synchronisation of sound and image), of the control and procedures. It also depends on the combined quality of the various parts of the system and its environment. + +#### 5.1.2.2 Overall delay + +The nominal overall delay is defined as the sum of the transmission delay and the characteristic delay of the terminals (see Note). The subjective effect of this delay on the quality of service has to be taken into account because an excessive value may impair user acceptability. + +Under normal conditions the overall delay should not exceed 400 ms, in conformance with ITU-T Recommendation G.114. Another Recommendation is expected for audiovisual services. However, when no means are available to fulfil this condition, the call should not be rejected. + +NOTE – The characteristic delay of a terminal is the delay introduced by the terminal when the transmitted information contents for the various media is minimal; for the video, this means that there is no movement or only very small movements in the transmitted image, e.g. only lips and eyes of the users are moving. + +### 5.1.3 Security aspects + +No specific means for security are usually required. They may however be included for particular applications. + +### 5.1.4 General charging principles + +The same charging options should be offered as for telephony. + +## **5.2 Communication task level** + +### **5.2.1 Communication tasks used for the service** + +Multimedia conversational services are built around the communication task Conversing (see Annex B.1/F.700). Other tasks such as Receiving and Sending may also be used for specific purposes. + +### **5.2.2 Quality of service related to the communication tasks** + +#### **Differential delay between sound and image** + +When video is present, the differential delay between sound and image should be kept low enough for lip synchronism to be subjectively insured. The differential delay should not be allowed to become more than 120 ms, and should preferably remain below 100 ms. ITU-T H-series Recommendations – Supplement 1 (*Application profile – Sign language and lip-reading real-time conversation using low bit-rate video communication*) reports that asynchronism under 100 ms is required. + +## **5.3 Media component level** + +### **5.3.1 Media components used for the service** + +The Media component Audio (see Annex A.1/F.700) is normally set up at the beginning of the call and is usually present throughout, but it may be temporarily interrupted to increase the bit-rate available for other media components. + +One other media component at least should also be present, permanently or not, i.e. one of the following components: + +- Video (Annex A.2/F.700); +- Text (Annex A.3/F.700); +- Graphics (Annex A.4/F.700); +- Still Pictures (Annex A.5/F.700); +- others (for further study). + +### **5.3.2 Quality of service related to the media components** + +#### **5.3.2.1 Sound quality** + +The following two levels of sound quality are preferred: + +- the basic quality is equivalent to that of the PCM telephony conforming to ITU-T Recommendation G.711 (level 0 of media component Audio); +- the high quality is 7 kHz bandwidth (conforming to ITU-T Recommendation G.722) equivalent to present broadcast TV sound (level 1 of media component Audio). + +For the quality levels of the media component Audio, see Annex A.1/F.700. + +Means for efficient echo control must be provided. When video is present, these means should have the capability to accommodate the large transmission delay that may be induced by the synchronisation of sound and image (see 5.2.2). + +#### **5.3.2.2 Image quality for the video** + +When video is present, its quality should be sufficient for reproducing the fluid movements of a person in head and shoulders view (see Note). Four grades of image quality may be offered: + +- minimum videophone quality: one person with only very small movements can be viewed with a limited resolution; this is level 0 of the media component Video and is applicable to low bit-rate connections; +- basic videophone quality: one person with only small movements can be viewed; this is level 1 of the media component Video; it allows limited language perception; +- enhanced videophone quality: at least 2 persons with only small movements can be viewed; this is level 2 of the media component Video; it is equivalent to the basic videoconference quality; it allows a good language perception level, a good usability for sign language and lip-reading, and a comfortable intercommunication with the videoconference service; +- high quality videophone: the quality should be that of broadcast TV and possibly that of HD-TV; this is level 3 or 4 of the media component Video. + +For the quality levels of the media component Video, see Annex A.2/F.700. + +A factor of quality related to video is the recovery time for picture build-up when the video source is changed; it is important to keep it low in the case of conference calls, but may also apply when two or more cameras are used alternately, e.g. one showing the user and another one for documents. + +Complete definition and methods of evaluation of the quality of motion pictures require further studies. + +NOTE – When motion video is present, the smoothness of the movement in the displayed image depends on the ability of the system to convey rapid changes; the amount of information which has to be transmitted will of course increase with the affected area in the picture and the speed of the movement. It is generally considered that in a normal conversation, only part of the picture will be moving and that any fast movement will be limited to a small portion of the image. Some degradation of the picture quality such as blurring or other artefacts may be allowed to occur when these conditions are not met. + +#### **5.3.2.3 Text Quality** + +Text conversation is supposed to be supported according to the principles of T.140 [ITU-T Recommendation T.140, *Protocol for multimedia application text conversation*]. + +Preferred display areas and other requirements can be found in ETSI ETR333 [Human Factors: Text Telephony, User requirements and proposals]. + +Text quality is measured in corrupted characters, dropped characters and characters replaced by the missing text marker [see ITU-T Recommendation T.140]. + +Another factor of quality is the ability to use a complete set of characters for various languages. + +For the quality levels of the media component Text, see Annex A.3/F.700. + +# **6 Dynamic aspects** + +## **6.1 Activation phase** + +### **6.1.1 Provision, withdrawal** + +The multimedia conversational services may be provided after prior arrangement with a service provider, or they may be generally available. + +### **6.1.2 Call establishment** + +Call establishment is on demand. Incoming calls shall be announced with means that is perceivable by the users, e.g. audible, visual or tactile alerting signals preferably selected by the user. + +Call progress shall be announced to the caller with visible and audible signals. + +## **6.2 Communication phase** + +### **6.2.1 Call set-up** + +The call is made according to the general procedure applicable to the network involved. If several modes are allowed, negotiation occurs between the terminals for the selection of the mode to be used. This is usually an automatic process relying on the general options of each user. When required, additional channels are established and aggregated to the initial channel. + +### **6.2.2 Change of communication mode** + +Changes in user requirements arising during the course of the communication may require a change of communication mode, i.e. changes in the network connections or changes in the in-band channel that are transparent to the network. If the network supports it, then in-call change of mode should be allowed when both terminals have the capabilities for the new mode. A change of mode that entails an increased cost for the communication may only be initiated from the charged end. + +### **6.2.3 Conference calls** + +A conference call is a supplementary service. It is established in conformance with ITU-T Recommendation F.702 and is made through a multipoint control unit (MCU). It uses the communication task Conferencing (Annex B.2/F.700), and the middleware service element Conference Control (Annex C.1/F.700). + +### **6.2.4 Audio muting and video inhibition** + +Any participant may temporarily prevent his terminal from sending out audio or video signals. A suspended video should be replaced by an adequate notice. A terminal should provide a visual indication when its audio is muted or its video inhibited. + +### **6.2.5 Access to other services** + +#### **6.2.5.1 General** + +Access to other services may be made by one of the participants, who forwards the multimedia documents or informations received to the other participant in the call; this requires an additional access to the network, the appropriate rights of access to the service and possibly specific rights for forwarding documents to other users; the service may be public or private, e.g. retrieval of multimedia documents from a private server. + +In the case of a conference call, access may also be made through the MCU, which then has to provide an access to the network, rights of access to the service (unless it can use the rights of one or several of the participants in the conference through some authentication procedure), and possibly conversion of protocols, of coding or of media. + +#### **6.2.5.2 Retrieval services** + +For further study. + +## 6.3 Termination phase + +Call termination may be initiated indifferently from the calling or from the called end. + +# 7 Service profiles + +## 7.1 Different types of Multimedia Conversational Services + +Multimedia conversational services can be divided into different types of profiles according to the various kinds of information exchanged and to their quality levels. They usually have in common the capability for transmission of sound. The different types are: + +- Videophone service, with audio and moving pictures and optionally various types of data; +- Voice and data services, with audio and various types of data; +- Text telephony, with real time text, optionally combined with audio; +- Total Conversation service, with moving pictures, real time text and audio; +- Collaborative Document Handling Service (CDH), with real time text, data and possibly graphics; this service is often used in multipoint configurations and is therefore described in ITU-T Recommendation F.702. + +## 7.2 List of service profiles + +The following profiles of the Generic Multimedia Conversational Services are defined. As stated in ITU-T Recommendation F.700, these profiles are references for the offering of services, but do not preclude any enhancement or additional functions. Additional profiles may be included in the future to respond to evolving user needs. + +### 1) *Videophone service:* + +- Profile 1a, Low bit-rate videophony: low bit-rate audio, QCIF video or less with limited movements capability; +- Profile 1b, Basic videophony: PCM telephony audio, QCIF video or CIF video with limited movements capability; +- Profile 1c, Enhanced videophony: wideband audio, CIF video. +- Profile 1d, High quality videophony: wideband audio, TV quality video or HD-TV quality video. + +### 2) *Voice and data services:* + +These services may be offered with two levels of audio quality and with three types of exchanged data. The minimum audio quality is level A0, low bit-rate audio (profiles 2a); the basic audio quality is level A1, equivalent to PCM telephony audio. The three types of exchanged data are still pictures from a camera or equivalent system (profiles 2a1, 2b1), text (profiles 2a2, 2b2), and microcomputer files (profiles 2a3, 2b3). + +- Profile 2a1, low bit-rate audio with still pictures; +- Profile 2a2, low bit-rate audio with text; +- Profile 2a3, low bit-rate audio with file transfer; +- Profile 2a1, basic audio with still pictures; +- Profile 2a2, basic audio with text; +- Profile 2a3, basic audio with file transfer. + +- 3) *Text Telephone Service:* +- Profile 3a, Usable text conversation, with text only; + - Profile 3b, Usable text conversation alternating between text and audio (Introduced here for historical reasons, to describe existing services. There is no desire to limit new implementations to the alternating text/audio profile); + - Profile 3c, Good text conversation with simultaneous usable audio. +- 4) *Total Conversation Service:* +- Profile 4a, Minimum total conversation, with usable audio, usable text and perceivable video; + - Profile 4b, Standard total conversation : with usable audio, good text and good motion optimised video. + +The following table recapitulates the requirements for the various types of profiles in terms of media components and their minimal levels of quality. + +| Service | Profile | Audio | Video | Text | Graphics, and/or Still Pictures | File transfer | +|---------|---------|-------|----------|------|---------------------------------|---------------| +| VPS | P1a | A0 | V0* | O | O | O | +| | P1b | A1 | V1* | O | O | O | +| | P1c | A2 | V2 | O | O | O | +| | P1d | A3 | V3 or V4 | O | O | O | +| VDS | P2a | A0 | O | O* | O* | O* | +| | P2b | A1 | O | O* | O* | O* | +| TTS | P3a | - | - | T1 | O | O | +| | P3b | A1* | - | T1* | O | O | +| | P3c | A1 | - | T2 | O | O | +| TCS | P4a | A1 | V0 | T2 | O | O | +| | P4b | A1 | V2 | T2 | O | O | + +A0, A1, ... minimum mandatory audio quality level +A1\* minimum mandatory audio quality level alternating with text +V0, V1, ... minimum mandatory video quality level +V0\*, V1\*, ... minimum mandatory video quality level, with a limited amount of movement +T0, T1 minimum mandatory text quality level +T0\* minimum mandatory text alternating with audio +O optional +O\* one at least of the media components mandatory + +NOTE – Telephony and text only service are monimedia services, but apart from that difference they share the same specification as the above services. + +# **8 Interworking and intercommunication** + +## **8.1 General** + +Terminals with different characteristics and capabilities may be connected in a multimedia conversation. They use a common mode that both terminals can handle. Any media that both terminals supports will be present with the lower of the two quality levels if these differ. + +## **8.2 Intercommunication with a telephone terminal** + +The communication is made with sound only. + +## **8.3 Videophone and other multimedia conference terminals** + +### **8.3.1 Audiographic conference and videophone terminals** + +These terminals can intercommunicate with sound only. Basic videophones have a lower quality level than audiographic conference terminals which will have to fall back to this lower level to establish communication. However, with broadband videophones, the situation may be the reverse. The ability to select one video picture in the video terminal (or in the MCU in the case of a multipoint call) for sending it as a still picture to the AGC terminal would be a desirable feature. + +### **8.3.2 Videoconference and videophone terminals** + +If the video codecs have a common mode, then this can be used for intercommunication. However, with a basic videophone terminal, this may not be desirable in a multipoint call because of the limited image quality, insufficient to show several participants simultaneously. The other possibility is then for the videoconference rooms to communicate between themselves in their usual mode, while the videophone terminal participates with sound only. + +### **8.3.3 Communication with a facsimile terminal** + +The communication is not possible unless the multimedia conversational terminal supports facsimile on its data channel. + +### **8.3.4 Text telephone interworking** + +It may be of interest for users of multimedia conversation terminals to be able to conduct text conversations with text telephones. This interworking may be possible either if the terminal has text conversation capabilities and a gateway for interworking with the text telephones is available, or the terminal has capabilities for direct interworking with the actual type of text telephone. + +### **8.3.5 Total Conversation interworking** + +Total Conversation terminals should be capable of interworking with Total Conversation terminals in other networks through gateways. They should also interwork with other multimedia terminals with a subset of their functionalities. + + + + + +# ITU-T RECOMMENDATIONS SERIES + +| | | +|-----------------|--------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of the ITU-T | +| Series B | Means of expression: definitions, symbols, classification | +| Series C | General telecommunication statistics | +| Series D | General tariff principles | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Construction, installation and protection of cables and other elements of outside plant | +| Series M | TMN and network maintenance: international transmission systems, telephone circuits, telegraphy, facsimile and leased circuits | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks and open system communications | +| Series Y | Global information infrastructure and Internet protocol aspects | +| Series Z | Languages and general software aspects for telecommunication systems | + +\*18874\* \ No newline at end of file diff --git a/marked/F/T-REC-F.71-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.71-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..64672153e39c851d9e16fdbc702e526c7ba7e8a7 --- /dev/null +++ b/marked/F/T-REC-F.71-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:454a9958ffe168868cb7d38a0eb24418dafe31a7a4245c992089b2316ac37d3e +size 7392 diff --git a/marked/F/T-REC-F.72-199610-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.72-199610-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3a58f68885c03bc2fb95bb409d68f7f55947475f --- /dev/null +++ b/marked/F/T-REC-F.72-199610-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:97adcbdebb3326a1af9817c0266d6c1cef088a8f97dc377b1e5fd01e721d14f7 +size 8269 diff --git a/marked/F/T-REC-F.72-199610-I_PDF-E/3121ebddccf183ca63bb9781be440a7e_img.jpg b/marked/F/T-REC-F.72-199610-I_PDF-E/3121ebddccf183ca63bb9781be440a7e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ea48a81924e2fbfbe8bd76647db776faab4c18be --- /dev/null +++ b/marked/F/T-REC-F.72-199610-I_PDF-E/3121ebddccf183ca63bb9781be440a7e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ea41160ecd6b81183d34ce0d444b1f9a17d307f114f23297b1a23d0cebe2fd87 +size 23663 diff --git a/marked/F/T-REC-F.72-199610-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg b/marked/F/T-REC-F.72-199610-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a9a75c758fe95eb3b0b421be354a5392599b21b1 --- /dev/null +++ b/marked/F/T-REC-F.72-199610-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7b598bc818c6ede417aa5d30ee1867bb614388e50f6044741239bf6e133a0aac +size 97386 diff --git a/marked/F/T-REC-F.72-199610-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg b/marked/F/T-REC-F.72-199610-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1b71275216af54f66d5a660ad60623670554bd30 --- /dev/null +++ b/marked/F/T-REC-F.72-199610-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c82531eee1a0abd755efc247f37819a77bc338d60a1ee9e5ac868b88f7fc8dae +size 113645 diff --git a/marked/F/T-REC-F.72-199610-I_PDF-E/990567efebf979be51f56d1150012c9d_img.jpg b/marked/F/T-REC-F.72-199610-I_PDF-E/990567efebf979be51f56d1150012c9d_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f40ec976dcad614e566f9a1e1909d7375978116f --- /dev/null +++ b/marked/F/T-REC-F.72-199610-I_PDF-E/990567efebf979be51f56d1150012c9d_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9f1cce0773c03941a0b1b74a7e360b7c7f9af2248d9e1add2857d670ffffae83 +size 35955 diff --git a/marked/F/T-REC-F.720-199208-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.720-199208-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..5724dddb0631c7b3da6cc1dc7f6ae93d978a90ab --- /dev/null +++ b/marked/F/T-REC-F.720-199208-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0b697f1f89b3584c606562ea134607091ac54e193693a197e87ada685013c885 +size 8277 diff --git a/marked/F/T-REC-F.720-199208-I_PDF-E/2fa4a1bf91d0f34e87c689fbc1211fe3_img.jpg b/marked/F/T-REC-F.720-199208-I_PDF-E/2fa4a1bf91d0f34e87c689fbc1211fe3_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..8c51923a5f3b550a8d600aadea73b93c69a22c8f --- /dev/null +++ b/marked/F/T-REC-F.720-199208-I_PDF-E/2fa4a1bf91d0f34e87c689fbc1211fe3_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5b1863d13cc9d50154fe5340518bdb0242a65248271c613629f272ef6e0bc850 +size 58942 diff --git a/marked/F/T-REC-F.720-199208-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.720-199208-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a964d818064f7dd9b197f2fb020277aaf3f7cb43 --- /dev/null +++ b/marked/F/T-REC-F.720-199208-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d3bd57b63b5340e52c6f854fcaf77e595b3cbc0b0e851454d6b2ff70b1d43959 +size 5900 diff --git a/marked/F/T-REC-F.721-199208-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.721-199208-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3ae9a090bc50643fe1ed50d32c36eac50b037f0d --- /dev/null +++ b/marked/F/T-REC-F.721-199208-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:836e63fb2ebf9ef2ed10f0ef430abaf50e506751b9a563f93e3d25eeaa7a61a6 +size 8184 diff --git a/marked/F/T-REC-F.721-199208-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg b/marked/F/T-REC-F.721-199208-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f38b858e4e2597e9d3fd4200768740a7e2b4401a --- /dev/null +++ b/marked/F/T-REC-F.721-199208-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7e1ae4acf8266a09deebb11bf7cdba5724242e3aeaf35d74f0da382f27cb7b5a +size 6002 diff --git a/marked/F/T-REC-F.721-199208-I_PDF-E/raw.md b/marked/F/T-REC-F.721-199208-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..3caacf01b18a894f7318da8b1f58e51a0f1d4890 --- /dev/null +++ b/marked/F/T-REC-F.721-199208-I_PDF-E/raw.md @@ -0,0 +1,456 @@ + + +![ITU logo: A globe with a lightning bolt and the letters ITU.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with a lightning bolt and the letters ITU. + +INTERNATIONAL TELECOMMUNICATION UNION + +**CCITT** + +**F.721** + +THE INTERNATIONAL +TELEGRAPH AND TELEPHONE +CONSULTATIVE COMMITTEE + +(08/92) + +**TELEMATIC, DATA TRANSMISSION, ISDN +BROADBAND, UNIVERSAL, PERSONAL +TELECOMMUNICATIONS AND +TELECONFERENCE SERVICES +OPERATIONS AND QUALITY OF SERVICE** + +--- + +**VIDEOTELEPHONY TELESERVICE FOR ISDN** + +![CCITT logo: A globe with the letters CCITT.](6ed175c791b5e156d9c98a8dbcc3318c_img.jpg) + +CCITT logo: A globe with the letters CCITT. + +**Recommendation F.721** + +--- + +# FOREWORD + +The CCITT (the International Telegraph and Telephone Consultative Committee) is a permanent organ of the International Telecommunication Union (ITU). CCITT is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The Plenary Assembly of CCITT which meets every four years, establishes the topics for study and approves Recommendations prepared by its Study Groups. The approval of Recommendations by the members of CCITT between Plenary Assemblies is covered by the procedure laid down in CCITT Resolution No. 2 (Melbourne, 1988). + +Recommendation F.721 was prepared by Study Group I and was approved under the Resolution No. 2 procedure on the 4th of August 1992. + +# --- CCITT NOTES + +- 1) In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized private operating agency. +- 2) A list of abbreviations used in this Recommendation can be found in Annex A. + +# **VIDEOTELEPHONY TELESERVICE FOR ISDN** + +(1992) + +The CCITT, + +*considering* + +- (a) that considerable efforts have been undertaken worldwide in order to develop videotelephone equipment based on rapid improvements in the quality of video codec algorithms; +- (b) that videotelephones of some manufacturers are already available on the market; +- (c) that first trials of videotelephony have been performed nationally as well as internationally; +- (d) that a number of countries intend to introduce the videotelephone service as soon as possible; +- (e) that the ISDN will be an appropriate network for providing the videotelephony teleservice; +- (f) that ISDN is offered in a number of countries, + +*recognizes* + +the need for a standardized international videotelephony teleservice which will guarantee the compatibility of videotelephones on a worldwide basis and therefore, + +*recommends* + +that the videotelephony teleservice, where implemented, respect the requirements stated in this Recommendation. + +# **1 Definition** + +The **videotelephony teleservice** is a symmetrical, bidirectional, realtime, audiovisual teleservice in which speech and moving pictures are interchanged by means of one or two B-channels, using 64 kbit/s circuit-mode connections in the ISDN. The picture information transmitted is sufficient for adequate representation of fluid movements of a person displayed in head and shoulders view. + +# **2 Description** + +## **2.1 General description** + +The videotelephony teleservice is defined as a fully standardized ISDN teleservice following the principles given in Recommendation I.210. + +Two cases can be identified for the videotelephony teleservice: + +*Case I:* Videotelephone based on using one circuit-mode 64 kbit/s connection. + +*Case II:* Videotelephone based on using two circuit-mode 64 kbit/s connections. + +For case I the 64 kbit/s connection carries both speech and video information; for case II the first connection either carries speech or both speech and some video information and the second connection carries video information. + +The basic videotelephony teleservice is characterized by the transmission of moving pictures displayed continuously in colour, simultaneously with the speech of the persons involved in the call (generally two in the case of a point-to-point connection). + +The speech quality of this teleservice must be at least as good as that applicable to the telephony teleservice in the 64 kbit/s ISDN based on bandwidths of 3.1 kHz or 7 kHz respectively. + +The videotelephony teleservice shall allow the communication between: + +- two users (e.g. terminals) in a point-to point configuration via the ISDN over one or two circuit mode 64 kbit/s connection(s); +- three or more users in a multipont configuration as invoked by some supplementary services. + +Videotelephone terminals must be capable of supporting the telephony teleservice. + +An essential feature of the service is that, besides videotelephony, it also provides to the user the possibility to communicate with other ISDN telephone or videotelephone terminals by using only the speech communication facility. It shall be possible to use videotelephone terminals to communicate with 3.1 kHz telephone terminals connected to the public switched telephone network (PSTN). + +## 2.2 Specific terminology + +*Fall-back:* Procedure performed either by the network or by the calling videotelephone terminals to establish calls to 3.1 kHz telephone terminals. + +*Call 1:* The first call invoked in the videotelephony teleservice. It identifies the first 64 kbit/s connection between the subscribers. The call is invoked for all the service cases. + +*Call 2:* The second call invoked in the videotelephony teleservice. It identifies the second 64 kbit/s connection between the two subscribers. The call is invoked for case II only ( $2 \times 64$ kbit/s). + +*Retention timer:* This timer specifies the amount of time that the network retains the call information of the original call upon encountering busy or being released. This timer is a network provider option. The value of this timer is greater than 15 seconds. + +*Videotelephone terminal:* A terminal that supports the videotelephony teleservice. + +*3.1 kHz telephone terminal:* A terminal that supports the telephony 3.1 kHz teleservice. + +*7 kHz telephone terminal:* A terminal that supports the telephony 7 kHz teleservice. + +## 3 Procedures + +## 3.1 Provision/withdrawal + +3.1.1 This teleservice may be provided after prior arrangement with the service provider or be generally available. + +3.1.2 As a service provider option, the videotelephony teleservice can be offered with several subscription options which apply separately to each ISDN number, all or a group of ISDN numbers on the interface. For each subscription option, only one value can be selected. + +| Subscription option | Value | +|---------------------------------------------------------------------------|--------------------------------------------------------------------------------------------| +| Maximum number of information channels available at the called user | $m$ ,
where $m$ is not greater than the number of information channels on the interface | +| Maximum number of total calls presented simultaneously at the called user | $n$ ,
where $n$ is not greater than the number of information channels on the interface | + +The called user is identified by either one ISDN number, all or a group of ISDN numbers on the interface. + +## 3.2 *Normal procedures* + +### 3.2.1 *Activation/deactivation/registration* + +Not applicable. + +### 3.2.2 *Invocation and operation* + +##### 3.2.2.1 *Originating the service (call request)* + +Call 1 shall be set up first. After this call has been accepted by the called user, call 2 can be originated if necessary. + +The characteristics of initial call set up for case I and case II shall be identical for call 1. + +Call 1 is devoted to multimedia information transfer (e.g. high quality speech, video and data). The transmission audio mode on the end-to-end digital path is defined according to Recommendations H.221 and H.242. + +Call 2 is devoted to video information transfer. The end-to-end path is framed according to Recommendation H.221. + +Call 2 is invoked by the calling terminal when call 1 is in the active phase and after end-to-end mode initialization procedure is completed. When the second connection is active, end-to-end alignment procedure occurs and the relative delay between the two connections is adjusted until complete synchronization is achieved according to Recommendation H.221. + +The in-band protocol shall be established according to Recommendation H.242. + +The videotelephony call shall be originated by the originating user activating the terminal, performing service selection (if applicable from the originating terminal) and terminating customer selection. During this process, the originating user shall be given the appropriate indications which refer to the state of the call. + +The call request procedure from the user's point of view must be available as an operation preferably similar to that for telephony even if two separated calls are established in the network. + +Audio tones provided to the user shall be as for the 3.1 kHz telephony teleservice. + +*Note* – Call 1 should be presented with an alerting phase. Call 2 should be associated to an automatic answer at the called interface. + +If call 2 cannot be completed due to e.g. remote access conditions, network congestion, call 1 can be maintained or released by the calling terminal according to intercommunication requirements. If call 1 is maintained, the user can re-attempt the establishment of call 2. + +##### 3.2.2.2 *Indications during call set-up and call acceptance (answer)* + +At the called side, the two calls are accepted after successful checking of compatibility information by the terminal(s) addressed. + +After initiating a call, the calling user shall receive an acknowledgement that the network is able to process the call. The called user shall receive an indication of the arrival of an incoming videotelephony call. The calling user shall also be given an indication that the call is being offered to the called user, when an indication is received by the network that the called user is being informed of this call. When the call reaches the called user and the connection is established, an indication shall be sent to the calling user. + +The acceptance of the videotelephone call by the terminating user (answer) causes the indication to be removed, and bidirectional communication paths to be provided. + +A called user has the control whether his picture is transmitted to the calling user. + +The called user may also provide other information, for use by the network in supplementary services provided to the other user (e.g. connected line identity). + +Note that in the case where a 3.1 kHz terminal is established first, the acceptance of the call is done according to normal telephony teleservice procedures. + +##### 3.2.2.3 *Terminating the call (call release)* + +A request to terminate the videotelephony teleservice may be generated by either of the users. If one user terminates the call, the other user is given an appropriate indication. + +In general, the release of a videotelephone call should be the same to the release of a telephone call; picture and sound are released simultaneously. + +##### 3.2.2.4 *Change of terminal communication mode* + +As a consequence of end-to-end integrity on videotelephony teleservice, 7 kHz telephony teleservice and some 3.1 kHz telephony teleservice calls (i.e. where no speech processing is performed by the network), it will be possible to use the B-channel protocols given in Recommendations G.725 and H.242. + +Depending on the terminal capabilities, it may be possible to change between the following communication modes according to Tables 2/H.320 and 3/H.320: + +- 3.1 kHz speech (Recommendation G.711); +- 7 kHz speech (Recommendation G.722); +- different videotelephone terminal modes. + +*Note 1* – The user may be required to establish additional calls in some cases. + +*Note 2* – As an option, in some circumstances establishment of a videotelephone call can be based on call set up as a 3.1 kHz telephone call and then change to a videotelephone call if a change of service by using an end-to-end procedure is possible. In the case where a 3.1 kHz telephone call is first established, on request of the calling user, the calling terminal will try to achieve framing and to exchange terminal capabilities on the existing B-channel. If it succeeds, this channel is used in the same way as a channel which is the result of a call 1 establishment. The change from the telephone call to a videotelephone call will not cause an interruption of the voice communication. If framing cannot be achieved on the existing B-channel, the calling user has to release the telephone call and to request for call 1 establishment. This change cannot be performed without an interruption of the existing communication. + +### 3.2.3 *Interrogation* + +Not applicable. + +### 3.3 *Exceptional procedures* + +#### 3.3.1 *Failure situations due to user error* + +- i) A user inputting an improper service request shall be given an appropriate failure indication by the network and the call set-up shall be ceased. +- ii) A user inputting a non-valid network number shall be given an appropriate failure indication by the network and the call set-up will be ceased. + +### 3.3.2 *Failure situations due to called user state* + +- i) A calling user attempting to establish a call to a user who is identified by the network to be busy [either network determined user busy (NDUB) or user determined user busy (UDUB)] shall be given an appropriate failure indication by the network and the call set-up shall be ceased. +- ii) A user attempting to establish a call to a user whose terminal equipment fails to respond, shall be given an appropriate failure indication by the network and the call set-up shall be ceased. + +- iii) On a call to a user whose terminal equipment has responded that the called user is being informed of the call, but has failed to answer within a defined period, the calling user attempting to establish the call shall be given an appropriate failure indication by the network and the call set-up shall be ceased. + +### 3.3.3 *Failure situation due to network conditions* + +A user attempting to establish a call but meeting call failure situations due to network conditions (e.g. congestion), shall be given an appropriate failure indication by the network. + +#### 3.3.4 *Failure situations due to called user state and/or network condition* + +A user attempting to establish a call but meeting call failure situations due to network conditions (e.g. congestion) or called user state (e.g. NDUB or UDUB), can have call information retained for the duration of the retention timer. + +### 3.3.5 *Decreasing of service quality due to fall-back* + +In the case of decreasing of service quality due to audio and/or visual fall-back, appropriate indication should be given to both users, even if problems occur in one direction only. + +# 4 **Network capabilities for charging** + +Charging capabilities are outside the scope of this Recommendation. + +## 5 **Intercommunication and interworking considerations** + +## 5.1 *Intercommunication/interworking with other terminals* + +Intercommunication with 3.1 kHz and 7 kHz ISDN terminals or interworking with PSTN shall be offered. + +### 5.1.1 *General principles* + +The videotelephony service shall include voice encoding according to Recommendation G.711. + +It may include additional speech encoding as an optional feature. + +The following fundamental requirements shall be fulfilled: + +- i) The user of a videotelephone terminal shall be able to establish calls to 3.1 kHz and 7 kHz telephone terminals (if the 7 kHz capability is supported) connected to the ISDN and to telephone terminals connected to the PSTN. Optionally, it should be able to reach other ISDN audiovisual terminals. +- ii) A videotelephone terminal shall be able to accept calls from 3.1 kHz and 7 kHz (if 7 kHz capability is supported) telephone terminals connected to the ISDN and from 3.1 kHz telephone terminals connected to the PSTN. Optionally, it should be able to accept calls from other ISDN audiovisual terminals. + +As an option, videotelephone terminals may be pre-programmed to receive incoming videotelephone calls only. This latter functions may be requested by users possessing, e.g. both a videotelephone terminal and a 3.1 kHz telephone terminal connected to the same access arrangement. + +### 5.1.2 *Fall back procedure* + +##### 5.1.2.1 *Fall back in the destination network* + +Fall-back to 3.1 kHz telephony shall be an inherent feature of videotelephony teleservice and shall be provided as a default procedure. + +The user shall be offered the possibility of indicating whether he requires interworking/fall-back to the 3.1 kHz telephony teleservice. A request for the videotelephony teleservice without fall-back (if indicated by the calling terminal) shall be possible. + +The following procedure shall apply: + +- If the calling user has indicated that fall-back is allowed, the network may offer the call to the called user at all videotelephone and 3.1 kHz telephone terminals if possible. The called user can accept the call either as a videotelephone or a 3.1 kHz telephone call at any terminal where the call is offered. + +*Note* – The called terminals may recognize the fall-back situation and indicate it to the user. + +- The calling user shall be informed of the resultant telecommunication service, i.e. the videotelephony or 3.1 kHz telephony teleservice. +- If no terminal accepts the call, this shall be indicated to the calling user. +- If a busy condition is met at the terminals, supplementary services, e.g. completion of calls to busy subscriber shall apply. + +*Note* – Echo cancellation will be disabled for videotelephone calls. If fall-back occurs, there is no current signalling mechanism for re-enabling the echo cancellers. + +- When fall-back is not implemented by the network (possible short-term situation), fall-back may be performed end-to-end by the calling videotelephone terminal by originating a 3.1 kHz telephone call. + +#### 5.1.2.2 *Fall-back when the ISDN does not offer the videotelephone teleservice* + +If the calling user has indicated that fall-back is allowed but the destination network does not support the videotelephone capabilities the calling user shall receive an indication that fall-back has occurred and an indication of the resultant telecommunication service. + +The called user shall be offered the incoming call as a telephony 3.1 kHz call. + +## 5.2 *Interworking with private ISDNs* + +If the called user is on private ISDN, the fall-back procedures will be performed by the private ISDN. + +The result of call presentation (videotelephony or 3.1 kHz telephony) within the private ISDN shall be indicated to the public ISDN. + +# 6 **Attributes/values** + +## 6.1 *Application of the attribute method* + +Depending on the case that applies the videotelephony teleservice description is based on invocation of one or two calls: call 1 and call 2 described according to the attribute method. + +## 6.2 *Low layer attributes* + +- a) Call 1 + - 1) Transfer mode: + - circuit. + - 2) Transfer rate: + - 64 kbit/s. + - 3) Transfer capability: + - 7 kHz. + +*Note* – Before 7 kHz audio bearer service is available, as an interim solution videotelephones should use “unrestricted digital information” as the transfer capability when calling other videotelephones. + +- 4) Structure: + - 8 kHz integrity. +- 5) Establishment of communication: + - demand. +- 6) Symmetry: + - bidirectional symmetric. +- 7) Configuration of communication: + - point-to-point, multipoint. + +Note that in the case where fall-back to the 3.1 kHz telephony teleservice occurs, values of 3.1 kHz telephony teleservice bearer capability apply. Also if optionally a 3.1 kHz telephone call is established first, attributes of 3.1 kHz telephony teleservice bearer capability apply (telephone call instead of call 1). + +#### b) Call 2 + +- 1) Transfer mode: + - circuit. +- 2) Transfer rate: + - 64 kbit/s. +- 3) Transfer capability: + - unrestricted digital information. +- 4) Structure: + - 8 kHz integrity. +- 5) Establishment of communication: + - demand. +- 6) Symmetry: + - bidirectional symmetric. +- 7) Configuration of communication: + - point-to-point, multipoint. + +## 6.3 *Access attributes* + +#### a) Call 1 + +- 8) Access channel and rate: + - D(16) or D(64) for signalling, B(64) for user information. +- 9.1 Signalling access protocol, layer 1: + - Recommendations I.430/I.431. +- 9.2 Signalling access protocol, layer 2: + - Recommendation Q.921. +- 9.3 Signalling access protocol, layer 3: + - Recommendation Q.931. +- 9.4 Information access protocol, layer 1: + - Recommendations H.221, G.711, G.722 (option), H.242, H.261, AV.254. +- 9.5 Information access protocol, layer 2. +- 9.6 Information access protocol, layer 3. + +Note that in the case where fall-back to the 3.1 kHz telephony teleservice occurs or if optionally a telephone call is established first, attribute 9.4 has the following values: Recommendations I.430/I.431, G.711. + +- b) Call 2 + - 8) Access channel and rate: + - D(16/64) for signalling, B(64) for user information. + - 9.1 Signalling access protocol, layer 1: + - Recommendations I.430/I.431. + - 9.2 Signalling access protocol, layer 2: + - Recommendation Q.921. + - 9.3 Signalling access protocol, layer 3: + - Recommendation Q.931. + - 9.4 Information access protocol, layer 1: + - Recommendations H.221, H.242, H.261, AV.254. + - 9.5 Information access protocol, layer 2. + - 9.6 Information access protocol, layer 3. + +## 6.4 *High layer attributes* + +- a) Call 1 + - 10) Type of user information: + - Speech (telephony), video, data, audiovisual (information). + - 11) Layer 4 protocol functions: + - Recommendation H.221. + - 12) Layer 5 protocol functions: + - Recommendation H.242. + - 13) Layer 6 protocol functions: + - Recommendations G.722 (option), G.711, H.261. + - 14) Layer 7 protocol functions. + +Note that in the case where fall-back to 3.1 kHz telephony teleservice occurs or, if optionally a 3.1 kHz telephone call is established first, the value of attribute 10) is “speech” and the value of attribute 13) is “Recommendation G.711”. + +- b) Call 2 + - 10) Type of user information: + - video. + - 11) Layer 4 protocol functions: + - Recommendation H.221. + - 12) Layer 5 protocol functions: + - Recommendation H.242. + - 13) Layer 6 protocol functions: + - Recommendation H.261. + - 14) Layer 7 protocol functions. + +## 6.5 *General attributes* + +# 7 **Supplementary services provided** + +Supplementary services should be considered as applicable to the videotelephone communication as a whole, even if two separated calls (i.e. call 1 and call 2) are established in the network. Restrictions identified should be applied by the videotelephone terminal. + +Application of telephony supplementary services to the videotelephony teleservice is described separately. + +*Note* – Only one set of addressing information i.e. ISDN number address should be allocated to a given videotelephone terminal and the same addressing information should always be used for call 1 and call 2 request. + +# **8 Quality of Service** + +## **8.1 *Synchronism of speech and lip movement (lip synchronism)*** + +No subjectively discernible difference in the delay of the speech and video signal. + +## **8.2 *Sound quality*** + +No significant difference compared to the speech quality used in the 64 kbit/s ISDN telephony services based on bandwidths of 3.1 kHz or 7 kHz. + +## **8.3 *Picture quality*** + +Optimization of picture quality is for further study including the need for adequate representation of fluid movements (see Note). + +*Note* – The urgent need to develop both objective and subjective quality parameters for the received motion picture is widely identified. + +## **8.4 *The overall delay*** + +The overall delay is defined to consist of transmission delay and the characteristic delay of a videotelephone terminal. A characteristic delay of a videotelephone terminal is the delay introduced by the terminal when only lips and eyes of the talking user are moving. + +The overall effect on quality by the delays introduced by video codecs and transmission facilities needs to be taken into account in the service. Increased delays may impair user acceptability. + +Maximum allowable delay including maximum number of satellite hops are left for further study. + +For the videotelephony teleservice case II, it is possible that one 64 kbit/s connection is routed via terrestrial path while the other is routed via satellite. In this case, the resynchronization is performed by the terminal. + +The Quality of Service is the same as if both channels were routed via satellite. + +# **9 Intercommunication/interworking possibilities** + +- 7 kHz telephony; +- 3.1 kHz ISDN telephony; +- 3.1 kHz PSTN telephony; +- other videotelephony modes; +- audiovisual services; +- others for further study. + +# **10 Operational and commercial aspects** + +For further study. + +# ANNEXE A + +(to Recommendation F.721) + +## **Alphabetical list of abbreviation used in this Recommendation** + +ISDN Integrated services digital network + +NDUB Network determined user busy + +PSTN Public switched telephone network + +UDUB User determined user busy \ No newline at end of file diff --git a/marked/F/T-REC-F.723-199607-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.723-199607-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3a58f68885c03bc2fb95bb409d68f7f55947475f --- /dev/null +++ b/marked/F/T-REC-F.723-199607-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:97adcbdebb3326a1af9817c0266d6c1cef088a8f97dc377b1e5fd01e721d14f7 +size 8269 diff --git a/marked/F/T-REC-F.723-199607-I_PDF-E/raw.md b/marked/F/T-REC-F.723-199607-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..77f56f6b52a2d2055cc43f736455b68b26eb5e3d --- /dev/null +++ b/marked/F/T-REC-F.723-199607-I_PDF-E/raw.md @@ -0,0 +1,370 @@ + + +![ITU logo: A globe with a lightning bolt and the letters ITU.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with a lightning bolt and the letters ITU. + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.723** + +(07/96) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Audiovisual services + +--- + +**Videophone service in the Public Switched +Telephone Network (PSTN)** + +ITU-T Recommendation F.723 + +(Previously «CCITT Recommendation») + +--- + +# ITU-T F-SERIES RECOMMENDATIONS NON-TELEPHONE TELECOMMUNICATION SERVICES + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| AUDIOVISUAL SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to ITU-T List of Recommendations.* + +# FOREWORD + +The ITU-T (Telecommunication Standardization Sector) is a permanent organ of the International Telecommunication Union (ITU). The ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Conference (WTSC), which meets every four years, establishes the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics. + +The approval of Recommendations by the Members of the ITU-T is covered by the procedure laid down in WTSC Resolution No. 1 (Helsinki, March 1-12, 1993). + +ITU-T Recommendation F.723 was prepared by ITU-T Study Group 1 (1993-1996) and was approved under the WTSC Resolution No. 1 procedure on the 19th of July 1996. + +--- + +## NOTE + +In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +© ITU 1996 + +All rights reserved. No part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and microfilm, without permission in writing from the ITU. + +# CONTENTS + +*Page* + +| | | | +|---------|-----------------------------------------------------|---| +| 1 | Introduction ..... | 1 | +| 2 | Normative references ..... | 1 | +| 3 | Definitions..... | 2 | +| 4 | Terminology ..... | 2 | +| 5 | Description ..... | 2 | +| 5.1 | General description ..... | 2 | +| 5.2 | Basic functionalities..... | 3 | +| 5.3 | Service categories ..... | 3 | +| 5.4 | Possible applications ..... | 3 | +| 5.5 | Optional features..... | 3 | +| 6 | Procedures ..... | 3 | +| 6.1 | Normal procedures..... | 4 | +| 7 | Network aspects ..... | 4 | +| 8 | Terminal aspects..... | 4 | +| 9 | Quality of service ..... | 4 | +| 9.1 | Picture quality ..... | 4 | +| 9.2 | Audio quality ..... | 4 | +| 9.3 | Overall delay..... | 5 | +| 10 | Interworking/intercommunication..... | 5 | +| 10.1 | Intercommunication with telephony ..... | 5 | +| 10.2 | Intercommunication with mobile videotelephony ..... | 5 | +| 10.3 | Intercommunication with ISDN videotelephony ..... | 5 | +| Annex A | – Attributes and values..... | 5 | +| A.1 | Low layer attributes ..... | 5 | +| A.2 | Access attributes ..... | 6 | +| A.3 | High layer attributes ..... | 6 | +| A.4 | General attributes..... | 6 | + +# VIDEOPHONE SERVICE IN THE PUBLIC SWITCHED TELEPHONE NETWORK (PSTN) + +(Geneva, 1996) + +# 1 Introduction + +This Recommendation contains the description and network specific service requirements for videophone services offered in the Public Switched Telephone Network (PSTN). The substance of this Recommendation complements the main body of the Supplement to Recommendation F.720, which deals with network independent service requirements for respective Low Bit Rate (LBR) videophone services provided in networks such as PSTN and digital mobile telecommunication networks across Low Bit Rate (LBR) channels. The difference between the service requirements in these two network domains stems from variations in access rates, mobility, robustness of digital wireless transmission and different terminal environments. In addition to network specific requirements, the network independent requirements for LBR videophone services and general requirements for all videophone services, included in Recommendation F.720, apply for the service as well. + +Due to bandwidth and technical restrictions, the quality of service is limited and may be inadequate for many applications, in particular in the professional domain. Hence, it is essential that users of the service can utilize the reduced network capabilities as efficiently as possible with a flexibility in the channel allocation between voice, video, image and data. + +This service can be used on a stand-alone basis or as part of a multimedia application. In the latter case, the same requirements apply. + +## 2 Normative references + +The following Recommendations and other references contain provisions which, through reference in the text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; all users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. + +- ITU-T Recommendation G.723.1 (1996), *Dual rate speech coder for multimedia communications transmitting at 5.3 and 6.3 kbit/s.* +- ITU-T Recommendation H.223 (1996), *Multiplexing protocol for low bit rate multimedia communication.* +- ITU-T Recommendation H.245 (1996), *Control protocol for multimedia communication.* +- ITU-T Recommendation H.263 (1996), *Video coding for low bit rate communication.* +- ITU-T Recommendation H.281 (1994), *A far-end camera control for videoconferences using H.224.* +- ITU-T Recommendation H.324 (1996), *Terminal for low bit rate conversational multimedia communication.* +- ITU-T Recommendation T.120 (1996), *Data protocols for multimedia conferencing.* +- ITU-T Recommendation T.126 (1996), *Multipoint still image and annotation protocol.* +- ITU-T Recommendation V.8 bis (1996), *Procedures for the identification and selection of common modes of operation between DCEs and between DTEs over the general switched telephone network and on leased point-to-point telephone-type circuits.* +- ITU-T Recommendation V.34 (1994), *A modem operating at data signalling rates of up to 28 800 bit/s for use on the general switched telephone network and on leased point-to-point 2-wire telephone-type circuits.* + +# 3 Definitions + +For the purposes of this Recommendation, the following definitions apply. + +**3.1 channel aggregation:** The capability to aggregate basic PSTN data connections in given discrete data rate increments to establish one higher bit rate data channel. + +**3.2 dynamic channel allocation:** The principle of allocating in a flexible way the available channel capacity between various types of data, comprising audio, video, control information, still pictures and other application specific data to maximize the quality of service. + +**3.3 fall-back1):** A procedure, initiated by either the videophone or the network that reduces the bit rate of the voiceband connection. + +**3.4 in-band signalling:** Signalling between videophone units that support user control functions, monitoring functions, user alerts or advisory information about the status of the videophone units or the network connection. + +**3.5 basic videophone:** A videophone containing mandatory features only and conforming to performance parameters contained in the Recommendations without any optional features nor any improved performance parameters. + +**3.6 enhanced videophone:** A videophone containing enhanced operational features and conforming with performance parameters defined as "optional" in the Recommendations. + +**3.7 motion rendition:** The ability of the receiving entity to reproduce motion from the transmitted video. + +## 4 Terminology + +For the purposes of this Recommendation, the following abbreviations are used. + +| | | +|-------|---------------------------------------| +| CIF | See Recommendation F.720 (Supplement) | +| PSTN | Public Switched Telephone Network | +| QCIF | See Recommendation F.720 (Supplement) | +| SQCIF | See Recommendation F.720 (Supplement) | + +## 5 Description + +## 5.1 General description + +The general service description characteristics to all LBR videophone services is included in the Supplement of the F.720 Recommendation. + +The service offers a real-time conversational two-way audiovisual end-to-end communication, comprising video, audio and optional2) in-band data transfer capabilities. As a rule, the audiovisual information is transferred along a single PSTN connection, based on LBR data channels. The aggregation of two PSTN connections for achieving increased overall capacity and QOS may be offered as an option (further studies are required). + +For the video the service should support the coding scheme defined by Recommendation H.263 and a spatial resolution conforming to QCIF and the terminal requirements covered by Recommendation H.324. + +--- + +1) Besides at set-up, fall-back may also occur under degraded network conditions through bit-rate reduction. + +2) As a terminal capability in-band data transfer is optional, but as a service functionality it is mandatory. + +### 5.2 Basic functionalities + +The basic functionalities, characteristic of videophone services in general as well as those of the LBR videophone services have to be supported. + +The basic PSTN speech communication facility, i.e. that of an analogue telephone, must be included in the terminal, so that the user shall be able to use the terminal as a normal telephone as well. + +Concerning the fall-back, a slow repetition rate video mode, manually controllable from the receiving end, must be supported. + +Dynamic channel allocation shall be provided as a mandatory system capability. Dynamic channel allocation is executed by means of a framing, synchronization and terminal negotiation mechanism in conformance to Recommendations H.245 and H.324 and other relevant ITU-T Recommendations. + +### 5.3 Service categories + +As a rule, the service employs a single PSTN connection. + +The possibility to aggregate two or more PSTN connections to set up a higher bit rate connection for improving the QOS (multilink operation) as well as the possibility to temporarily extend or reduce the bearer capabilities during the call without disconnecting the call (in-call modification) are for further study. + +### 5.4 Possible applications + +In the consumer/residential segment the envisaged applications are real-time human-to-human interaction, based on head-and-shoulders view, and remote surveillance and event monitoring e.g. for baby-sitting, security, as well as other non-conversational applications. + +In business/institutional segment the foreseen applications are remote expert consultation, requiring audiovisual support, remote surveillance and recognition, remote trouble-shooting, remote inspection and accessing videoconferences. + +The above applications may be offered on a stand-alone basis or as part of a value-added multimedia application requiring extended terminal capabilities. + +From the service point of view the basic videophone terminal has to support QCIF only for the motion video. Some terminals, i.e. enhanced videophones may support also SQCIF3) format or at least they are capable of receiving SQCIF video frames. + +### 5.5 Optional features + +The facilities and functionalities defined in Supplement 1 to Recommendation F.720 in general apply. + +Furthermore the possibility to access a multipoint videophone call has to be provided in accordance with T.120-Series protocols. For a possible still picture data transfer in multipoint calls, as well as for data transfer of telepointer and annotation, Recommendation T.126 applies. + +In case a far-end camera control is provided, Recommendation H.281 shall apply. + +# 6 Procedures + +The service is envisaged to address the consumer/residential market using low-cost stand-alone terminals with a minimum set of functionalities. + +--- + +3) According to the current Recommendation H.263, the capability of receiving SQCIF video frames is supported by all H.263 video codecs as a mandatory feature. + +### **6.1 Normal procedures** + +#### **6.1.1 Call procedures** + +The call set-up and call initialization procedures are defined in the Recommendation on network independent requirements for LBR videophone services (see Supplement to Recommendation F.720). + +When a videophone call has been originated and the called terminal is incompatible or unable to comply with the call request, the calling subscriber is given an appropriate indication. The call will progress automatically as a telephone call if the calling user accepts the call. + +## **7 Network aspects** + +The transfer rate achieved on a given PSTN line is strongly dependent on the transmission quality, i.e. crosstalk and noise level, which may have a substantial impact on the overall QOS with the consequence that operation on only the low repetition rate mode is possible. Therefore, maintaining appropriate signal/interference ratio in the local loop enabling the maximum data rate or a rate close to it is important to guarantee adequate overall QOS necessary for gaining user acceptance. + +The service can be offered also on leased point-to-point telephone type circuits. + +## **8 Terminal aspects** + +The service is envisaged to address the consumer/residential market using low-cost stand-alone terminals with a minimum set of functionalities. The terminals with enhanced functional capabilities may consist of PCs equipped with auxiliary plug-in units providing the needed terminal capabilities for PSTN videotelephony (see Supplement to Recommendation F.720). + +The bearer capabilities for transferring the digital audiovisual data over analogue PSTN lines are implemented using V.34 modems with bit rates up to 28.8 kbit/s and beyond4). + +## **9 Quality of service** + +The QOS is dependent on the quality of the obtained telephone line and networks involved, which may result in a reduced modem transfer rate and impaired audiovisual quality as well as delayed call set-up due to prolonged modem start-up phase. + +### **9.1 Picture quality** + +The picture quality depends on the available modem transfer rate, amount of movements in the picture and the selected channel rate. Increased amount of changes in the image content causes degraded performance of motion rendition. For further study5). + +### **9.2 Audio quality** + +See Recommendation F.720 (Supplement). + +--- + +4) The V.34 modems are envisaged to be upgraded in the near future resulting in an increased data transfer rate of 33.6 kbit/s. + +5) The exact required QOS is likely to depend on the applications and therefore further studies are required. + +### 9.3 Overall delay + +Under normal conditions, the maximum overall delay should not exceed the values applicable to international videophone calls6). + +## 10 Interworking/intercommunication + +### 10.1 Intercommunication with telephony + +Intercommunication with PSTN is mandatory and normally it takes place using the analogue speech communication facility of the videophone terminal. Furthermore digital speech communication is possible between two compatible videophone terminals. + +### 10.2 Intercommunication with mobile videotelephony + +Intercommunication with low bit rate mobile videophone services should be provided. Variations in the access rates of different mobile networks may require rate adaptation. This must be supported by terminals with the aid of the in-band signalling mechanism, comprising a standard set of rates. + +Intercommunication with mobile videophone services of higher service categories is carried out through a potential fall-back to the best common mode, supported by the two services involved. + +### 10.3 Intercommunication with ISDN videotelephony + +See Supplement to Recommendation F.720. + +## Annex A + +## Attributes and values + +(This annex forms an integral part of this Recommendation) + +### A.1 Low layer attributes + +| | Attributes | Values | +|---|--------------------------------|--------------------------------------------------------------------------------------------------------| +| 1 | Transfer mode | circuit | +| 2 | Transfer rate | maximum 28.8 kbit/s (and beyond), under degraded network conditions may be less | +| 3 | Transfer capability | 3.1 kHz audio for Rec. V.34 (videotelephony)
3.1 kHz speech for telephone mode (analogue telephony) | +| 4 | Structure | N/A | +| 5 | Establishment of communication | on demand | +| 6 | Symmetry | bidirectional symmetric | +| 7 | Configuration of call | point-to-point | + +6) For international telephone calls the current limit is 400 ms determined by Recommendation G.114. A new Recommendation is under preparation in ITU-T Study Group 12 which will be applicable for audiovisual services. + +### A.2 Access attributes + +| | Attributes | Values | +|-----|--------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| 8 | Access channel and rate | in videophone mode: 5.3/6.4 kbit/s speech and 23.5/22.4 kbit/s video, or optionally in data mode 28.8 kbit/s or in speech and data mode: 5.3/6.4 kbit/s speech and 23.5/22.4 kbit/s data | +| 9.1 | Signalling access protocol, layer 1 | Rec. V.8/V.8 bis | +| 9.2 | Signalling access protocol, layer 2 | Rec. V.8/V.8 bis | +| 9.3 | Signalling access protocol, layer 3 | | +| 9.4 | Information access protocol, layer 1 | Rec. H.223 | +| 9.5 | Information access protocol, layer 2 | Rec. H.245 | +| 9.6 | Information access protocol, layer 3 | | + +### A.3 High layer attributes + +| | Attributes | Values | +|----|----------------------------|-------------------------------------------------------------------------| +| 10 | Type of user information | audio and video and/or data or plain data | +| 11 | Layer 4 protocol functions | | +| 12 | Layer 5 protocol functions | | +| 13 | Layer 6 protocol functions | Rec. G.723.1 for audio
Rec. H.263 for video
T.120-Series for data | +| 14 | Layer 7 protocol functions | | + +### A.4 General attributes + +| | Attributes | Values | +|----|-----------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| 15 | Supplementary services provided | for further study | +| 16 | Quality of service | audio: 3.1 kHz telephony, toll quality speech
video
synchronization of audio and video: no subjectively discernible delay between speech and video or minimal audio delay (inserted speech delay disabled)
data: for further study | +| 17 | Intercommunication/interworking possibilities | with LBR videophone services in mobile networks
with ISDN videophone service: to the extent feasible with gateways in the network
with other audiovisual services (on telephony only)
with telephony
with other services: for further study | +| 18 | Operational and commercial aspects | for further study | + +# ITU-T RECOMMENDATIONS SERIES + +- Series A Organization of the work of the ITU-T +- Series B Means of expression +- Series C General telecommunication statistics +- Series D General tariff principles +- Series E Telephone network and ISDN +- Series F Non-telephone telecommunication services** +- Series G Transmission systems and media +- Series H Transmission of non-telephone signals +- Series I Integrated services digital network +- Series J Transmission of sound-programme and television signals +- Series K Protection against interference +- Series L Construction, installation and protection of cables and other elements of outside plant +- Series M Maintenance: international transmission systems, telephone circuits, telegraphy, facsimile and leased circuits +- Series N Maintenance: international sound-programme and television transmission circuits +- Series O Specifications of measuring equipment +- Series P Telephone transmission quality +- Series Q Switching and signalling +- Series R Telegraph transmission +- Series S Telegraph services terminal equipment +- Series T Terminal equipments and protocols for telematic services +- Series U Telegraph switching +- Series V Data communication over the telephone network +- Series X Data networks and open system communication +- Series Z Programming languages \ No newline at end of file diff --git a/marked/F/T-REC-F.724-200509-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg b/marked/F/T-REC-F.724-200509-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..36cf51f8ba07e44aa7a4b009f3ba08eb2b570b1c --- /dev/null +++ b/marked/F/T-REC-F.724-200509-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:11780faeee75ba39ab4ed9de67a9e67abec0d9ba3068ebec661dea70c16e29d3 +size 4338 diff --git a/marked/F/T-REC-F.724-200509-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.724-200509-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3a3fa8503cd5aa93518020022b72b563eac847bc --- /dev/null +++ b/marked/F/T-REC-F.724-200509-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:351b5f5fa48944e7eb1ccc036e47c22bfda38d7df2b3649a370f29e2afa097fe +size 48053 diff --git a/marked/F/T-REC-F.724-200509-I_PDF-E/raw.md b/marked/F/T-REC-F.724-200509-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..40f19af404668ad815f326bc7d271150ecc72866 --- /dev/null +++ b/marked/F/T-REC-F.724-200509-I_PDF-E/raw.md @@ -0,0 +1,549 @@ + + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.724** + +(09/2005) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Audiovisual services + +--- + +**Service description and requirements for +videotelephony services over IP networks** + +ITU-T Recommendation F.724 + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +## **TELEGRAPH SERVICE** + +| | | +|-----------------------------------------------------------------|-------------| +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | + +## **MOBILE SERVICE** + +| | | +|--------------------------------------------------------|-------------| +| Mobile services and multdestination satellite services | F.110–F.159 | +|--------------------------------------------------------|-------------| + +## **TELEMATIC SERVICES** + +| | | +|-------------------------------------------|-------------| +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | + +## **MESSAGE HANDLING SERVICES** + +F.400–F.499 + +## **DIRECTORY SERVICES** + +F.500–F.549 + +## **DOCUMENT COMMUNICATION** + +| | | +|--------------------------------------|-------------| +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | + +## **DATA TRANSMISSION SERVICES** + +F.600–F.699 + +## **AUDIOVISUAL SERVICES** + +**F.700–F.799** + +## **ISDN SERVICES** + +F.800–F.849 + +## **UNIVERSAL PERSONAL TELECOMMUNICATION** + +F.850–F.899 + +## **HUMAN FACTORS** + +F.900–F.999 + +*For further details, please refer to the list of ITU-T Recommendations.* + +# **Service description and requirements for videotelephony services over IP networks** + +## **Summary** + +The purpose of this Recommendation is to define the videotelephony services provided over IP networks. The services enable two users in different locations to have conversational communications over IP networks. The service description, function model, service profiles, application scenarios and requirements of the IP videotelephony services are provided. The IP videotelephony services are one class of the generic services identified in ITU-T Rec. F.720, and their description follows the methodology described in ITU-T Rec. F.701. The requirements for the IP videotelephony services are derived from the scenarios for different applications the said services can support. Therefore, the services requirements accommodate the demands of different kinds of users and assist in enabling intercommunications between the services provided by different telecom operators and/or different IP videotelephony systems. + +## **Source** + +ITU-T Recommendation F.724 was approved on 13 September 2005 by ITU-T Study Group 16 (2005-2008) under the ITU-T Recommendation A.8 procedure. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications. The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure e.g. interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementors are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database. + +© ITU 2005 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## CONTENTS + +| | Page | +|------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 2.1 Normative References ..... | 1 | +| 2.2 Informative References ..... | 1 | +| 3 Definitions ..... | 1 | +| 4 Conventions ..... | 2 | +| 5 Prose descriptions ..... | 2 | +| 6 Functional model and service profile ..... | 2 | +| 6.1 Functional model ..... | 2 | +| 6.2 Service model ..... | 3 | +| 6.3 Service profiles ..... | 4 | +| 7 Service scenarios ..... | 5 | +| 7.1 Business videotelephony service scenarios ..... | 5 | +| 7.2 Residential videotelephony service scenarios ..... | 7 | +| 8 IP videotelephony service requirements ..... | 7 | +| 8.1 User requirements..... | 7 | +| 8.2 Application requirements ..... | 8 | +| 8.3 Security requirements ..... | 8 | +| 8.4 Authentication and accounting requirements ..... | 8 | +| 8.5 Interworking and intercommunication requirements ..... | 9 | +| 8.6 QoS requirements ..... | 10 | + + + +# Service description and requirements for videotelephony services over IP networks + +# 1 Scope + +This Recommendation provides the service descriptions, function model, service profiles, application scenarios and requirements of the videotelephony services based on the IP network. The objective is to describe the service requirements, facilitate interoperability, and provide a guideline to the user, service provider, network carrier, and administrator as well for deploying videotelephony services over IP networks. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +## 2.1 Normative References + +None. + +## 2.2 Informative References + +- ITU-T Recommendation F.700 (2000), *Framework Recommendation for multimedia services*. +- ITU-T Recommendation F.701 (2000), *Guideline Recommendation for identifying multimedia service requirements*. +- ITU-T Recommendation F.703 (2000), *Multimedia conversational services*. +- ITU-T Recommendation F.720 (1992), *Videotelephony services – General*. +- ITU-T Recommendation Y.1541 (2002), *Network performance objectives for IP-based services*. +- ITU-T H-series Recommendations – Supplement 1 (1999), *Sign language and lip-reading real time conversation using low bit-rate video communication*. + +# 3 Definitions + +This Recommendation defines the following terms: + +**3.1 basic videotelephony service:** A videotelephony service supporting basic voice, video, real-time text and optional multimedia data or control facility. + +**3.2 enhanced videotelephony service:** A videotelephony service supporting advanced voice, video, real-time text and at least one type of multimedia data and/or control facility. + +# 4 Conventions + +In this Recommendation the following conventions are used: + +- "shall" indicates a mandatory requirement. +- "should" indicates a suggested but optional course of action. +- "may" indicates an optional course of action rather than a recommendation that something take place. + +# 5 Prose descriptions + +The IP videotelephony service provides real-time end-to-end bidirectional communications between two subscribers in different locations on the IP network by means of voice, video, real-time text and other forms of multimedia data and/or control facility. + +The way to place a call on IP videotelephony services is the same as that for conventional telephony services in the sense that the subscriber dials a number, or enters another type of identifier, to initiate a call. In addition to videotelephony calls, the subscriber can also use the videotelephony services in other applications. + +A subscriber of IP videotelephony services can be located in any place covered by IP networks, e.g., office buildings, meeting rooms, hotels, residences, telephone booths on the street or even on board a transport vehicle. + +There are two types of IP videotelephony calls: + +- Point-to-point calls; +- Multiparty calls (utilizing devices for audio, video and text mixing, e.g., the Multipoint Control Unit (MCU)). + +There are two major types of IP videotelephony terminals: + +- Videotelephony sets; +- Softphones running on computers. + +Other types of videotelephony terminals, such as PSTN videophones, ISDN videophones or even conventional phones and mobile phones, can communicate with the IP videophones. These terminals may have limited capabilities regarding sending or receiving various types of information in videotelephony calls, but they can at least intercommunicate with IP videophones in a voice-only mode. + +# 6 Functional model and service profile + +## 6.1 Functional model + +The functional model of IP videotelephony services is shown in Figure 1. IP videotelephony terminals exchange voice, video, real-time text and multimedia data in a point-to-point or multipoint way over IP networks. They can also intercommunicate with other videotelephony or conventional telephony terminals via interworking units. The call control unit processes call signalling and controls sessions, and the Authentication, Authorization and Accounting (AAA) unit performs the function of user, authentication, authorization and accounting. + +![Functional model of IP videotelephony services diagram showing connections between terminals, call control, AAA, and interworking.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +The diagram illustrates the functional model of IP videotelephony services. It features four rectangular boxes representing terminals: 'IP videotelephony terminal A' (top left), 'IP videotelephony terminal B' (top right), 'IP videotelephony terminal C' (bottom left), and 'other videotelephony or telephony terminal' (bottom right). Two oval shapes represent functional blocks: 'Call control' (top center) and 'Interworking' (bottom center). A third oval labeled 'AAA' is positioned to the right of 'Call control'. Solid lines represent 'User information' paths: from terminal A to terminal B, and from terminal C through the 'Interworking' block to the 'other videotelephony or telephony terminal'. Dashed lines represent 'Call information' paths: from terminal A to 'Call control', from terminal C to 'Call control', and from 'Call control' to terminal B. Dotted lines represent 'AAA information' paths: from 'Call control' to 'AAA', and from 'AAA' to terminal B. A legend at the bottom left defines the line types: solid for 'User information', dashed for 'Call information', and dotted for 'AAA information'. The label 'F.724\_F01' is in the bottom right corner. + +Functional model of IP videotelephony services diagram showing connections between terminals, call control, AAA, and interworking. + +**Figure 1/F.724 – Functional model of IP videotelephony services** + +## 6.2 Service model + +### 6.2.1 Service model: the functional view + +There are two models for IP videotelephony services from the functional view. Other models may exist or can be developed. However, they are a subject beyond the scope of this Recommendation and are thus for further study. + +#### 1) Basic videotelephony services + +They refer to those IP videotelephony services that support the mandatory basic features. Only the voice, video and real-time text of conversational parties are transmitted between them. + +#### 2) Enhanced videotelephony services + +They refer to those IP videotelephony services that support the enhanced optional features (i.e., other forms of multimedia data and/or remote control function). In addition to voice, video and real-time text, multimedia data like still pictures, pre-recorded audio/video clips, text messages, and collaboration-related data such as white-boarding information can be transmitted between videophones. Other optional features should include convening of a conference and control of a conference, support for far-end camera control and other remote control. + +### 6.2.2 Service model: the usage view + +From the view of use environment, there are two types of IP videotelephony services. + +#### 1) Residential videotelephony services + +For residential videotelephony service users, the following applications should be supported: + +- Face-to-face conversation with audio, video, and real-time text; +- Face-to-face conversation with simultaneous transfer of pictographical data such as picture, pre-recorded video clips and files of other kinds; +- Remote video surveillance for home security inspection and unmanned baby-sitting for kids, etc; +- Emergency calls in audio, video and text. + +#### 2) Business videotelephony services + +For business service users, the following applications should be supported: + +- Face-to-face conversation with audio, video, and real-time text; +- Face-to-face conversation with simultaneous transfer of pictographical data such as images, documents and files of other kinds; + +- Remote video surveillance; +- Remote consultation; +- Remote diagnosis in telemedicine; +- Participation in videoconferencing; +- Emergency calls in audio, video and text. + +## 6.3 Service profiles + +While all videotelephony services have the common capability for transmission of audio, video and real-time text, they can be divided into different types of profiles according to the quality level of audio and video, and other multimedia data exchanged. + +An IP videotelephony service may be offered with two levels of audio quality, three levels of video quality, one level of real-time text, five types of exchanged data and two types of control facility. The basic audio quality is level A0, equivalent to 3.4 kHz PCM telephony; the enhanced audio quality is level A1, equivalent to 7 kHz or 14 kHz wideband audio. The three levels of video are level V1 for QCIF video, level V2 for CIF video and level V3 for SDTV video. The level of real-time text is T2, good conversational text. The five types of exchanged multimedia data are still pictures, video clips, text messages, file transfer and joint editing. The two types of control facility are remote control and conference conductor. + +Taking into account the factors mentioned above, the following profiles of videotelephony service are defined. The service profile descriptions are not intended to impose a particular and detailed way of offering the services, but to illustrate the approach to the profile definition. + +- Profile a: Basic videotelephony service: basic audio, QCIF or CIF video, real-time text, optional multimedia data or control facility; +- Profile b: Enhanced Basic videotelephony service: wideband audio, CIF video, real-time text, optional multimedia data or control facility; +- Profile c: Enhanced videotelephony service: wideband audio, CIF or SDTV video, real-time text, multimedia data and/or control facility. + +The above defined service profiles are summarized in Table 1. + +**Table 1/F.724 – Service profiles** + +| Profile | Audio | Video | Text | Still pictures, pre-recorded video clips, text messages, file transfer and/or joint editing | Control functions | +|------------|------------------------------------------------|----------|------|---------------------------------------------------------------------------------------------|-------------------| +| Pa | A0 | V1 or V2 | T2 | O | O | +| Pb | A1 | V2 | T2 | O | O | +| Pc | A1 | V2 or V3 | T2 | 1M | 1M | +| A0, A1 | Minimum mandatory audio quality level | | | | | +| V1, V2, V3 | Minimum mandatory video quality level | | | | | +| T2 | Mandatory text level | | | | | +| 1M | At least one of the media components mandatory | | | | | +| O | Optional | | | | | + +These profiles ensure at least a minimum level of communication. Conformance to a profile ensures intercommunication with other terminals of the same profile. A terminal or a service function unit may conform to one or several profile(s), and may have capabilities beyond those embodied in the profile(s). + +# **7 Service scenarios** + +This clause describes typical service scenarios to illustrate the videotelephony service and to derive its technical requirements. + +## **7.1 Business videotelephony service scenarios** + +### **7.1.1 Business call** + +The branch offices of Company A are geographically distributed all over its country and Company A has equipped all its branch offices with videophones for routine work-related communications. + +Step 1: Jacky, head of the branch office in regional division X, encounters some problems with his sales and he calls Hans, director of sales at headquarters, using videophone, to seek help and guidance and to discuss possible solutions with him. + +Step 2: Hans takes Jacky's video call. Because Jacky and Hans are old friends, they exchange some greetings (e.g., "you look healthy", "I like your tie", etc.). Jacky reports the problems to Hans, in the hope that the sales strategies can be adjusted. They then discuss different possibilities and methods to solve these problems under the existing sales strategies. Hans says he will start collecting information on the current status of the branch offices in the other regional divisions and then convene a meeting to discuss these problems if they are found to be common. The two end the call. + +Step 3: Hans makes video calls with other branch offices to talk about their execution of sales strategies, and finds that similar problems to those he has encountered also exist in these branch offices. So he decides to convene a meeting to discuss these problems. + +Step 4: Hans' secretary reserves a meeting through the videoconferencing service system and specifies that the videoconferencing terminal be placed in the headquarters meeting room, and the videophones in the branch offices of the four regional divisions, as the meeting participants. + +Step 5: At the scheduled start of the meeting, the system calls the reserved terminals. Hans does not use the videophone on his desk this time; instead, he goes to the meeting room to participate in the meeting. After the meeting is established, Hans successfully requests to be the chairman and begins to chair the meeting. + +Step 6: Hans first asks Jacky to report on the problems he has encountered. When Jacky begins his talk, the image of Jacky appears on the screens of all the meeting participants' terminals. After Jacky finishes his speech, Allan, head of branch office Y, requests to make an intervention. So Hans grants Allan the floor and switches the image on the screens of all the meeting participants' terminals to the image of Allan. + +Step 7: After Allan finishes speaking, Hans suggests a free discussion and sets the picture mode to the four-in-one continuous presence mode so that everyone can see everyone else in the ensuing discussion. + +Step 8: At the end of the discussion, Hans makes a summary and the image switches back to the image of Hans. Hans first expresses appreciation for the efforts and achievements of the various sales offices, then says it is possible to reasonably adjust the sales strategies based on the different opinions given, and schedules a date to release the new sales strategies. Hans then declares the end of the meeting and the participants say goodbye to one another. After all exit, the meeting automatically terminates. + +### **7.1.2 Business trip** + +Hans often goes on business trips. To ensure that he does not miss any important calls, he sets his office phone calls to be forwarded to his mobile phone prior to leaving on a business trip. Hans has recently become a subscriber to 3G services so he can use his new 3G cellular phone to take forwarded video calls. + +Step 1: Hans gets off the plane and takes a taxi to the hotel, when a videotelephony call from Jacky comes through. Hans takes the video call with his 3G mobile phone in the taxi. Jacky requests information on an important sales project and the two have a discussion on this project. + +Step 2: Hans' mobile phone displays a low battery warning, so Hans chooses to switch off the video function and leaves only the voice in use. + +Step 3: After the conversation is over, Jacky hangs up. The videotelephony service provider charges the call between Jacky and Hans (the call involves both video conversation and voice-only conversation) according to the corresponding rates. + +### **7.1.3 Customer service** + +Peter has bought a new digital product from Company B. However, the product malfunctions when Peter tries to use it. So Peter dials Company B's free call service number. What is special is that the number is a videophone number. + +Step 1: Peter uses his home videophone to place the call and a receptionist of Company B takes the call. + +Step 2: Peter explains the problem to the receptionist and demonstrates the malfunction in front of the camera of the videophone. Since this problem is not on the FAQ list, the receptionist cannot provide an immediate answer and says she needs to consult the technical support engineer. She asks Peter to hold on whilst she consults one of the technical support engineers for advice. In order to make Peter feel comfortable while he is waiting, the receptionist plays a video clip of Company B via the videophone to Peter who can view the clip on his videophone screen. + +Step 3: The receptionist then comes back with a solution and explains the problem was caused by the fact that Peter did not do the settings correctly. The receptionist then demonstrates to Peter how to set the product using video and the problem is solved. + +Step 4: Peter thanks the receptionist and the receptionist also thanks Peter for purchasing one of Company B's products. Peter then hangs up. + +### **7.1.4 Call between a deaf and blind person and a service centre** + +Sally is deaf and blind. She was born deaf but later became blind, so she learned sign language very well and still likes to express herself in sign language when her conversation partner understands it. She has learnt to read Braille to be able to communicate in text. + +Step 1: Sally calls with her videophone equipped with Braille display, to ask her service centre about a leaflet she has received. + +Step 2: The service centre is staffed with signing personnel. Marco answers this call, recognizes his customer Sally, and types the greeting "Hi Sally, this is Marco, how can I help you?" + +Step 3: Sally reads the response on the Braille display, and then signs in front of the camera: "I received a leaflet that I would like you to tell me about. I am placing it under my document camera so you can look at it." + +Step 4: Marco types: "OK, I will take a look." And he activates the supplementary video channel and zooms in on the leaflet, reads it and types to Sally. "It says that there is a discussion meeting in your building tonight about the opportunity to build a balcony on to your apartment. 7 P.M. in the coffee shop." + +Step 5: Sally signs: "Wonderful, thanks and bye." + +Step 6: Marco types: "Cheers, bye" and hangs up. + +## **7.2 Residential videotelephony service scenarios** + +### **7.2.1 Family call** + +John is in charge of sales in a big multinational company and he often goes on business trips. His son Tom is in primary school and his daughter Jessie is in kindergarten. Every time John goes for a business trip, John misses his family very much and he always uses the videophone in his hotel room to make video calls home. + +Step 1: This time John finds that the hotel where he stays does not offer a room videophone. So he goes to the commerce centre of the hotel to use the videophone there. John arrives at the commerce centre and dials the number of his home videophone. The call is connected. + +Step 2: John's wife Lisa answers the call. After they talk for a while, their children come to talk with John. Tom shows the model aeroplane he has made in his handicraft class before the camera while Jessie sings a song for her daddy. + +Step 3: John transmits the scenic photos he has taken with his digital camera during his journey via the USB port of the videophone to Lisa, Tom and Jessie. While looking at the photos displayed on the screen, comments are made on the beautiful photos and John promises that he will return with Lisa, Tom and Jessie to go for sightseeing trip in the summer vacation. John types the exact name of the place in the text area so that Lisa can make a note of it and start looking for booking opportunities. Tom, Jessie and Lisa then are very happy. + +Step 4: John talks with Lisa for a while again and then the call ends. After hanging up, John pays the commerce centre for his call. + +### **7.2.2 Monitoring** + +John takes all his family on vacation and hopes that he can keep an eye on his house whilst they are away. Thanks to his home videophone which provides monitoring functionality, it can be done. + +Step 1: Before leaving home, John enables the monitoring function on his videophone and sets an authentication password. + +Step 2: During his vacation, John uses the hotel videophone to call his home videophone. He keys in the password and his home videophone automatically starts transmitting the video captured from the camera mounted on the ceiling of the living room. + +Step 3: John then switches to the camera in his garage using remote control and then to the cameras in his kids' bedrooms and his backyard in succession. + +Step 4: After John gets back home from vacation, he cancels the monitoring settings of his home videophone. + +# **8 IP videotelephony service requirements** + +## **8.1 User requirements** + +Basic requirements: + +- Ability to exchange real-time video, audio and real-time text; +- Ability to select audiovisual mode or voice-only mode; +- Ability to make videotelephony calls at any place covered by IP networks; +- Ability to make videotelephony calls in handset mode and handsfree mode; + +- Ability to make videotelephony calls by people with hearing or speech disabilities. + +Enhanced requirements: + +- Ability to exchange multimedia data including still images, live and pre-recorded video clips, and collaboration data such as whiteboard; +- Ability to implement remote control; +- Ability to join and conduct a videoconference. + +## **8.2 Application requirements** + +Basic requirements: + +- Voice and video switching processing; +- Allow for various access means, such as xDSL, Ethernet, WLAN, GSM and 3G etc., and videotelephony service provider should support at least one of them; +- Support for interworking between different videotelephony systems or networks through gateways; +- Support for subscriber management and numbering. Use of E.164 numbering plan is mandatory; +- Support for PSTN-like dialling modes. A keypad should be implemented on the videophone; +- Support for audio arrangement of handset function and handsfree function; +- Support for entry and display of real-time text. The specific method for text entry (e.g., keypad, integral or detachable keyboard, touch screen, verbal recognition) is beyond the scope of this Recommendation. The specific method for text display (e.g., video screen, Braille, verbal) is beyond the scope of this Recommendation. + +Enhanced requirements: + +- Support for dynamic creation and termination of video streams; +- Support for fallback from audiovisual mode to voice-only mode; +- Support for upgrading from voice-only mode to audiovisual mode; +- Support for real-time multimedia data exchanging, such as still pictures, live and pre-recorded video clips, text messages, and collaboration data; +- Support for other types of dialling modes, such as use of aliases. + +## **8.3 Security requirements** + +The security of IP videotelephony calls should be guaranteed. There are three levels of security: + +- Subscriber authentication and authorization; +- Call security; +- Security of media streams. + +## **8.4 Authentication and accounting requirements** + +The subscriber authentication of IP videotelephony services is used to ensure that only legal subscribers can have access to IP videotelephony services and accurate accounting should be implemented for the IP videotelephony calls made by subscribers. + +## **8.5 Interworking and intercommunication requirements** + +Three types of interworking or intercommunication are related to IP videotelephony service: + +- Interworking and intercommunication between terminals with different capability sets; +- Interworking and intercommunication between terminals in different networks (PSTN, ISDN, 3G etc.); +- Interworking and intercommunication between different IP videotelephony systems. + +Transcoding or bit-rate conversion in the interworking unit may be needed so that each terminal receives and transmits the signals it is able to handle. + +### **8.5.1 Terminals with different capabilities** + +Terminals may have different characteristics and capabilities, thus conforming to different profiles that the videotelephony service provider may offer. When they intercommunicate with each other, a common mode of the profiles will be used. This will adapt the service quality and functionalities to those of the terminal with the lowest quality level for each media component; however, communication is always possible because all terminals conform to the common basic profile. + +### **8.5.2 Terminals in different networks** + +IP videotelephony service needs to intercommunicate and interwork with the videotelephony service in other (non-IP) networks. In addition, the interworking between IP videotelephony calls and conventional telephony calls should be guaranteed. + +- Intercommunication and interworking between IP videotelephony terminal and PSTN/ISDN/3G, etc. videotelephony terminal; +- Intercommunication and interworking between IP videotelephony terminal and PSTN/ISDN/mobile phone. + +The requirements of this type of intercommunication include: + +- Audio transcoding or bit-rate conversion; +- Video transcoding or bit-rate conversion; +- Real-time text transcoding; +- Data transcoding or bit-rate conversion; +- Call control signalling conversion. + +### **8.5.3 Different IP videotelephony systems** + +There may be many IP videotelephony systems, such as H.323-based and SIP-based videotelephony systems. Intercommunication is needed if the terminals are located in different service systems. The requirements of this type of intercommunication include: + +- Audio/video/text/data transcoding or bit-rate conversion; +- Authentication between different systems; +- Accounting between different systems; +- Call control signalling conversion; +- Subscriber resource sharing and security. + +## **8.6 QoS requirements** + +The QoS of IP videotelephony calls should be guaranteed. Since the major media elements of videotelephony calls are voice, video and text, it is necessary to guarantee clear voice, clear head-and-shoulder image, continuous and smooth video of certain motion levels and text with good performance. In enhanced videotelephony services, the quality of multimedia data should also be guaranteed. + +To guarantee the QoS of IP videotelephony services, IP networks should provide QoS guarantee to support the bidirectional real-time service. + +### **8.6.1 Audio quality** + +IP videotelephony should support basic audio (3.4 kHz) and wideband audio (7 kHz or 14 kHz). + +IP videotelephony should be capable of performing acoustic echo-cancellation. + +IP videotelephony should have error resilience mechanisms to recover from packet loss. + +For a videotelephone with the audio arrangement of handset or handsfree function, the sensitivity and loudness rating should be guaranteed. + +### **8.6.2 Video quality** + +IP videotelephony should be able to provide a smooth video depending on the application. + +IP videotelephony should be able to provide reliable video colours. + +IP videotelephony should have error resilience mechanisms to recover from packet loss. + +### **8.6.3 Text quality** + +IP videotelephony should support good text quality for real-time conversation. The presentation should be smooth, covering any jerkiness caused by transmission in blocks. The delay between each character entry and its display should be low so that the experience of a direct conversation is maintained. The reliability should be good, so that transmission errors are much more rare than typing errors, and indicated to the users. + +### **8.6.4 Lip synchronization** + +IP videotelephony should be capable of performing lip-synchronization so that there is no humanly perceptible asynchronism between audio and video. + +### **8.6.5 Overall delay** + +The overall delay comprises two parts: network transmission delay and delay due to processing on IP videophone terminals. The latter is caused by the codec on terminals to perform encoding and decoding. + +The overall delay for IP videotelephony should be within specified limits since any delay greater than this threshold will cause unacceptable degradation in QoS. + +### **8.6.6 Network transmission quality** + +In order to provide videotelephony services over an IP network, the IP network should be able to provide end-to-end QoS guarantee. The required QoS has different aspects such as low delays, low jitters, and low packet loss. The required network transmission quality for the IP videotelephony service should be defined according to ITU-T Rec. Y.1541. + + + +# SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|---------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | General tariff principles | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects and next-generation networks | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.731-199707-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.731-199707-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..5d816e651f263a9b060dc411f00346e36d71d245 --- /dev/null +++ b/marked/F/T-REC-F.731-199707-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:eb291899a5433182a3ae7faaf556a2802a1499f2d0cee3c05eba56e561489341 +size 8176 diff --git a/marked/F/T-REC-F.732-199610-I_PDF-E/raw.md b/marked/F/T-REC-F.732-199610-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..5eaf85bb000c1f6f36490fb09a4f9cdd7e2cc5fa --- /dev/null +++ b/marked/F/T-REC-F.732-199610-I_PDF-E/raw.md @@ -0,0 +1,957 @@ + + +![ITU logo](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +The logo of the International Telecommunication Union (ITU) features the letters 'ITU' in a bold, sans-serif font, superimposed on a stylized globe with intersecting lines. + +ITU logo + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.732** + +(10/96) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Audiovisual services + +--- + +**Multimedia conference services in the B-ISDN** + +ITU-T Recommendation F.732 + +(Previously CCITT Recommendation) + +--- + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1–F.109 | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | F.110–F.159 | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | F.160–F.399 | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | F.550–F.599 | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| AUDIOVISUAL SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to ITU-T List of Recommendations.* + +# FOREWORD + +The ITU-T (Telecommunication Standardization Sector) is a permanent organ of the International Telecommunication Union (ITU). The ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Conference (WTSC), which meets every four years, establishes the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics. + +The approval of Recommendations by the Members of the ITU-T is covered by the procedure laid down in WTSC Resolution No. 1 (Helsinki, March 1-12, 1993). + +ITU-T Recommendation F.732 was prepared by ITU-T Study Group 1 (1993-1996) and was approved by the WTSC (Geneva, October 9-18, 1996). + +--- + +## NOTE + +In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +© ITU 1997 + +All rights reserved. No part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and microfilm, without permission in writing from the ITU. + +# CONTENTS + +| | | Page | +|-----|---------------------------------------------------------------|-------------| +| 1 | Scope..... | 1 | +| 2 | Definition ..... | 1 | +| 3 | Terminology ..... | 1 | +| 4 | Description ..... | 1 | +| 4.1 | General description..... | 1 | +| 4.2 | Functional model ..... | 1 | +| 4.3 | Configuration..... | 1 | +| 4.4 | Roles of the participants ..... | 3 | +| 4.5 | Terminal aspects ..... | 3 | +| 4.6 | Applications..... | 3 | +| 4.7 | Supplementary services ..... | 3 | +| 5 | Static aspects ..... | 3 | +| 5.1 | Service level ..... | 3 | +| 5.2 | Communication task level..... | 4 | +| 5.3 | Media component level..... | 4 | +| 6 | Dynamic aspects..... | 4 | +| 6.1 | Activation phase ..... | 4 | +| 6.2 | Communication phase..... | 4 | +| 6.3 | Termination phase ..... | 5 | +| 6.4 | Call termination ..... | 5 | +| 7 | Service profiles ..... | 5 | +| 7.1 | Different types of multimedia conference services..... | 5 | +| 7.2 | List of service profiles ..... | 5 | +| 8 | Interworking/intercommunication..... | 7 | +| 8.1 | Interworking between narrow-band and broadband terminals..... | 7 | +| 8.2 | Access type 7 ..... | 12 | +| 8.3 | Hybrid MCU..... | 15 | +| 8.4 | Interworking with LANs..... | 15 | +| 8.5 | Intercommunication between services ..... | 18 | +| 9 | Attribute table and values..... | 18 | +| 9.1 | Information transfer high layer attributes..... | 18 | +| 9.2 | Information transfer low layer attributes..... | 19 | +| 9.3 | Access attributes ..... | 19 | +| | Annex A – References..... | 20 | + +# **SUMMARY** + +This Recommendation encompasses a broad range of related services, including the videoconference service, and also the audiographic conference service, the videophone conference service and CDH (Collaborative Document Handling). It is part of a general framework for Recommendations on multimedia services. + +This Recommendation refers to Recommendation F.702 (Multimedia conference services) and contains only those parts that require complements or choices between options in order to apply it specifically to B-ISDN and to take full advantage of its enhanced capabilities. + +Because in the coming years the broadband conference rooms will have to interwork with the existing narrow-band systems, a large part of this Recommendation is devoted to the various possible configurations for interworking (see clause 8). Thus, when B-ISDN is interconnected with 64 kbit/s-based ISDN, it will also have to provide the services which are currently provided by this network, and insure interworking with existing narrow-band terminals. + + + +# MULTIMEDIA CONFERENCE SERVICES IN THE B-ISDN + +(Geneva, 1996) + +# 1 Scope + +This Recommendation describes the multimedia conference services supported by a B-ISDN. It relies on Recommendation F.702 for the main body of the description, and only gives the specific complements relevant to the B-ISDN that derive from its capabilities and its limitations. Thus, when reading this Recommendation, it is necessary to refer back to Recommendation F.702. The numbers of the clauses are those of F.702, and thus they do not form a continuous sequence because only a few clauses need a specific complement for the B-ISDN. In addition, there is a new clause 9 on the attributes for the various profiles. Because interworking with the N-ISDN is a major requirement and will probably be a very frequent way of setting up conferences in the next few years, this is described in great detail in clause 8. + +# 2 Definition + +See Recommendation F.702. + +# 3 Terminology + +See Recommendation F.702. + +# 4 Description + +## 4.1 General description + +See Recommendation F.702. + +## 4.2 Functional model + +See Recommendation F.702. + +## 4.3 Configuration + +### 4.3.1 Configuration at the network access + +The following figures describe the possible basic scenarios corresponding to a multimedia conference offered at the network access. + +#### 4.3.1.1 Access type 1 + +![Diagram showing a B-TE connected to a B-ISDN via a line labeled S_B = T_B.](fb139b01fa81528410fe1d28fbb60486_img.jpg) + +The diagram illustrates a configuration where a B-TE (Basic Terminal Equipment) is connected to a B-ISDN (Broadband Integrated Services Digital Network). A horizontal line connects the two components, with the label $S_B = T_B$ positioned above it. Below the B-ISDN box, there is a small reference code: T0105600-96/d01. + +Diagram showing a B-TE connected to a B-ISDN via a line labeled S\_B = T\_B. + +Figure 1/F.732 – Configuration + +![Diagram of the U-plane for F.732. On the left, four stacked boxes represent B/H ch, AAL 1, ATM, and PHY. Arrows point from these boxes to a central stack of two boxes labeled ATM and PHY. To the right of the central stack is another stack of two boxes labeled ATM and PHY. A horizontal line connects the B/H ch box to the right side of the diagram.](3121ebddccf183ca63bb9781be440a7e_img.jpg) + +Diagram of the U-plane for F.732. On the left, four stacked boxes represent B/H ch, AAL 1, ATM, and PHY. Arrows point from these boxes to a central stack of two boxes labeled ATM and PHY. To the right of the central stack is another stack of two boxes labeled ATM and PHY. A horizontal line connects the B/H ch box to the right side of the diagram. + +T0105610-96/d02 + +Figure 2/F.732 – U-plane + +![Diagram of the C-plane for F.732. On the left, six stacked boxes represent Q.2931, Q.2130, Q.2110, AAL 5, ATM, and PHY. Arrows point from these boxes to a central stack of six boxes with the same labels. To the right of the central stack is a stack of six boxes labeled B-ISUP, MTP3, Q.2140, Q.2110, AAL 5, ATM, and PHY.](d0abac95583b52a3b35f74a215567334_img.jpg) + +Diagram of the C-plane for F.732. On the left, six stacked boxes represent Q.2931, Q.2130, Q.2110, AAL 5, ATM, and PHY. Arrows point from these boxes to a central stack of six boxes with the same labels. To the right of the central stack is a stack of six boxes labeled B-ISUP, MTP3, Q.2140, Q.2110, AAL 5, ATM, and PHY. + +T0105620-96/d03 + +Figure 3/F.732 – C-Plane + +#### 4.3.1.2 Access type 2 + +![Diagram of the configuration for access type 2. It shows three main blocks: B-TE, B-NT2, and B-ISDN. B-TE is connected to B-NT2 via a line labeled S_B. B-NT2 is connected to B-ISDN via a line labeled T_B.](07b17a620c75522d53916a11e12d1bff_img.jpg) + +Diagram of the configuration for access type 2. It shows three main blocks: B-TE, B-NT2, and B-ISDN. B-TE is connected to B-NT2 via a line labeled S\_B. B-NT2 is connected to B-ISDN via a line labeled T\_B. + +T0105630-96/d04 + +Figure 4/F.732 – Configuration + +![Diagram of the U-plane for access type 2. On the left, four stacked boxes represent B/H ch, AAL 1, ATM, and PHY. Arrows point from these boxes to a first stack of three boxes labeled AAL 1, ATM, and PHY. From this first stack, arrows point to a second stack of three boxes labeled AAL 1, ATM, and PHY. From the second stack, arrows point to a third stack of two boxes labeled ATM and PHY. To the right of the third stack is another stack of two boxes labeled ATM and PHY. A horizontal line connects the B/H ch box to the right side of the diagram.](7affafe7362a2d2d072e9d4bf515f0bb_img.jpg) + +Diagram of the U-plane for access type 2. On the left, four stacked boxes represent B/H ch, AAL 1, ATM, and PHY. Arrows point from these boxes to a first stack of three boxes labeled AAL 1, ATM, and PHY. From this first stack, arrows point to a second stack of three boxes labeled AAL 1, ATM, and PHY. From the second stack, arrows point to a third stack of two boxes labeled ATM and PHY. To the right of the third stack is another stack of two boxes labeled ATM and PHY. A horizontal line connects the B/H ch box to the right side of the diagram. + +T0105640-96/d05 + +Figure 5/F.732 – U-plane + +![Diagram of the C-Plane protocol stack showing layers Q.2931, Q.2130, Q.2110, AAL 5, ATM, and PHY across four entities. The first three entities have a full stack, while the fourth entity is missing the Q.2130 layer. To the right, a separate stack lists B-ISUP, MTP3, Q.2140, Q.2110, AAL 5, ATM, and PHY.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +The diagram illustrates the C-Plane protocol stack across four entities. Each entity is represented by a vertical column of protocol layers. The first three entities have identical stacks: Q.2931, Q.2130, Q.2110, AAL 5, ATM, and PHY. The fourth entity's stack is missing the Q.2130 layer, consisting of Q.2931, Q.2110, AAL 5, ATM, and PHY. To the right of the main diagram, a separate vertical stack lists the following layers from top to bottom: B-ISUP, MTP3, Q.2140, Q.2110, AAL 5, ATM, and PHY. Horizontal double-headed arrows connect corresponding layers between adjacent entities. + +Diagram of the C-Plane protocol stack showing layers Q.2931, Q.2130, Q.2110, AAL 5, ATM, and PHY across four entities. The first three entities have a full stack, while the fourth entity is missing the Q.2130 layer. To the right, a separate stack lists B-ISUP, MTP3, Q.2140, Q.2110, AAL 5, ATM, and PHY. + +T0105650-96/d06 + +**Figure 6/F.732 – C-Plane** + +## 4.4 Roles of the participants + +See Recommendation F.702. + +## 4.5 Terminal aspects + +Two types of terminals may be used on the B-ISDN: those that are specifically designed for the B-ISDN and those that are adapted from the narrow-band terminals. When video is present (videoconference, videophone and some CDH terminals), the first kind of terminals conform to Recommendation H.310, while the second kind conforms to Recommendation H.321. + +### Audio + +The normal audio mode for conferencing services is wideband audio conforming to Recommendation G.722. A G.711 codec should also be present for interworking purposes. G.728 coding may also be used in interworking or fall-back situations, or temporarily to leave a larger bandwidth to the video or to the data subchannels, e.g. while showing an object or transmitting a document. + +The use of higher quality audio codecs is for further study. + +### Video (when applicable) + +Video codecs shall conform to Recommendation H.262 or H.261; H.261 codecs shall always be present for interworking purposes. + +## 4.6 Applications + +See Recommendation F.702. + +## 4.7 Supplementary services + +For further study. + +# 5 Static aspects + +## 5.1 Service level + +### 5.1.1 General aspects of the service + +See Recommendation F.702. + +### 5.1.2 Quality of service + +See Recommendation F.702. + +### **5.1.3 Security aspects** + +See Recommendation F.702. + +### **5.1.4 General charging principles** + +See Recommendation F.702. + +## **5.2 Communication task level** + +### **5.2.1 Communication tasks used for the service** + +See Recommendation F.702. + +### **5.2.2 Quality of service** + +See Recommendation F.702. + +## **5.3 Media component level** + +### **5.3.1 Media components used for the service** + +See Recommendation F.702. + +### **5.3.2 Quality of service** + +See Recommendation F.702. + +# **6 Dynamic aspects** + +## **6.1 Activation phase** + +See Recommendation F.702. + +## **6.2 Communication phase** + +### **6.2.1 Call set-up** + +See Recommendation F.702. + +### **6.2.2 Addition of another party during the call** + +See Recommendation F.702. + +### **6.2.3 Conference splitting** + +See Recommendation F.702. + +### **6.2.4 Change of communication mode** + +See Recommendation F.702. + +### **6.2.5 Call prolongation in the case of reservation** + +See Recommendation F.702. + +### **6.2.6 Call disconnection** + +See Recommendation F.702. + +### **6.2.7 Conference management procedures** + +The procedures for multipoint broadband conference services are those of the T.120-Series of Recommendations. Audio and video control use the T.126 procedure. Other media components should be transmitted on the common data subchannel supporting the T.120-Series protocol and are controlled through that channel. + +Interworking with narrow-band terminals requires the use of the procedures described in Recommendation H.242 and Recommendation H.243 for multipoint aspects. + +## 6.3 Termination phase + +## 6.4 Call termination + +See Recommendation F.702. + +## 7 Service profiles + +## 7.1 Different types of multimedia conference services + +See Recommendation F.702. + +## 7.2 List of service profiles + +The following Media Components (MCs) are defined for the broadband conferences services: + +- MC 1: audio; +- MC 2: video; +- MC 3: synchronous signalling; +- MC 4: data. + +B-ISDN is able to support all profiles of multimedia conference services. Thus, the following service profiles may be offered over a B-ISDN network: + +### 1) *Videoconference services* + +MC 1: mandatory, MC 2: mandatory, MC 3: mandatory, MC 4: optional + +Profile 1a – Basic videoconference service, with a transfer rate of 128 kbit/s; terminals shall conform to Recommendations H.310 and H.321, or to H.321 only. + +Profile 1b – Standard videoconference service, with a transfer rate of 384 kbit/s; terminals shall conform to Recommendation H.310 and H.321, or to H.321 only. + +Profile 1c – Enhanced videoconference service; for further study. + +Profile 1d – High quality videoconference service; for further study. + +### 2) *Audiographic conference services* + +MC 1: mandatory, MC 2: not applicable, MC 3: mandatory, MC 4: mandatory + +Profile 2 – All audiographic conference services use wideband audio conforming to Recommendation G.722 and a data subchannel supporting the T.120-Series protocol. + +### 3) *Cooperative document handling services* + +MC 1: optional, MC 2: optional, MC 3: mandatory, MC 4: mandatory + +Profile 3 – All cooperative document handling services have audio to one or more of the Recommendations G.711, G.722, G.728; they have a data subchannel supporting the T.120-Series protocol, and in particular T.127 for file transfer; other data functions are for further study; video conforming to H.261 and/or H.262 is optional. + +### 4) *Videophone conference supplementary services* + +MC 1: mandatory, MC 2: mandatory, MC 3: optional, MC 4: optional + +Profile 4b – Basic videophone conference service; this service applies to basic videophone terminals with a transfer rate of 128 kbit/s, and does not require any specific enhancement to the terminals; these terminals conform to Recommendations H.310 and H.321, or to H.321 only; they have G.728 and/or G.722 audio codecs (and G.711 for interworking purposes); they have H.261 and optionally H.262 video codecs; the profile is the same as Profile 1a except that G.722 coding is only optional instead of mandatory. + +Figure 7 shows the support of audio, video, data and signalling over an ATM connection. + +![Functional organization diagram showing two identical systems. Each system takes inputs for Audio (G.711, G.722, G.728), Video (H.261, H.262), and Synchronous signalling into an H.222.1 block. Inside H.222.1, the first three inputs go to H.222.0 (Program and Transport streams), and the fourth goes to Network adaptation. These then connect to an AAL block (AAL 1 for the first three, AAL 5 for the fourth), which contains CS (I.363) and SAR (I.363) components. Finally, they connect to an ATM I.363 block. Data (T.120-Series) is shown as a separate input line.](007b053fe94a8348f75128a584503fd0_img.jpg) + +The diagram illustrates the functional organization of two identical systems. Each system processes the following inputs: + +- Audio:** G.711, G.722, G.728 +- Video:** H.261, H.262 +- Synchronous signalling** +- Data:** T.120-Series + +The first three inputs (Audio, Video, and Synchronous signalling) are processed by an **H.222.1** block. Inside this block, they are first handled by **H.222.0**, which generates a **Program stream** and a **Transport stream**. These streams then pass through **Network adaptation**. The output of the Network adaptation is then processed by an **AAL 1** block, which contains **CS 1 I.363** and **SAR 1 I.363** components. The output of AAL 1 is then sent to an **ATM I.363** block. + +The **Data (T.120-Series)** input is processed by an **AAL 5** block, which contains **CS 5 I.363** and **SAR 5 I.363** components. The output of AAL 5 is also sent to the **ATM I.363** block. + +The diagram is symmetrical, showing two identical processing paths side-by-side. + +Functional organization diagram showing two identical systems. Each system takes inputs for Audio (G.711, G.722, G.728), Video (H.261, H.262), and Synchronous signalling into an H.222.1 block. Inside H.222.1, the first three inputs go to H.222.0 (Program and Transport streams), and the fourth goes to Network adaptation. These then connect to an AAL block (AAL 1 for the first three, AAL 5 for the fourth), which contains CS (I.363) and SAR (I.363) components. Finally, they connect to an ATM I.363 block. Data (T.120-Series) is shown as a separate input line. + +T0105660-96/d07 + +Figure 7/F.732 – Functional organization + +# 8 Interworking/intercommunication + +## 8.1 Interworking between narrow-band and broadband terminals + +### 8.1.1 Configurations + +#### 8.1.1.1 Reference configuration + +The reference configuration illustrates the various possible connections between terminals and networks. + +![Figure 8/F.732 – Scenarios for interworking](410562339ce067fdc6fa41940c118658_img.jpg) + +The diagram shows a central connection between 'B-ISDN capabilities' and 'N-ISDN capabilities' via 'IWF 6'. On the left, 'B-ISDN capabilities' is connected to 'B-TE' and 'B-NT2' (which is further connected to 'B-TE'). It is also connected to 'IWF 1', 'IWF 2', and 'IWF 3'. 'IWF 1' connects to 'N-TE'. 'IWF 2' connects to 'N-NT2' which connects to 'N-TE'. 'IWF 3' connects to 'N-TE'. On the right, 'N-ISDN capabilities' connects to 'N-TE' and 'N-NT2' (which connects to 'N-TE'). A shaded box labeled 'For further study' contains 'B-TE', 'B-NT2', 'B-TE', 'IWF 4', and 'IWF 5' connected to 'N-ISDN capabilities'. + +B-TE is a future broadband terminal +B-NT2 is a broadband NT2 +N-TE is a narrow-band terminal +N-NT2 is a narrow-band NT2 +IWF i ( $1 \leq i \leq 6$ ) corresponds to the $i^{\text{th}}$ specific group of interworking functions + +T0105670-96/d08 + +Figure 8/F.732 – Scenarios for interworking + +**Figure 8/F.732 – Scenarios for interworking** + +#### 8.1.1.2 Standardized configurations + +These configurations are extracted from I.413 and I.580. + +#### 8.1.1.3 Practical configuration + +The general practical configuration is based on the standardized configurations for interworking, but also reflects the fact that a B-ISDN does not only support broadband capabilities but also 64 kbit/s-based ISDN capabilities. It implies that all 64 kbit/s-based ISDN services are integrated in B-ISDN with a broadband signalling protocol. The interworking uses the broadband signalling protocol and a B-NT2 is necessarily present to connect the N-TEs (64 kbit/s-based TEs) to the B-ISDN. Interworking between broadband services and 64 kbit/s-based ISDN services is performed by call control mapping. Several kinds of signalling interworking are possible depending on the network architecture and the communication configuration: + +- 1) UNI signalling interworking between DSS 2 and DSS 1 (e.g. case of an IWF between an N-TE and a B-NT2); +- 2) NNI signalling interworking between B-ISUP and N-ISUP (e.g. case of an IWF between the B-ISDN and N-ISDN public networks); +- 3) UNI/NNI signalling interworking between DSS 1 and B-ISUP (e.g. case of an N-TE or N-NT2 attached to an integrated broadband/narrow-band local exchange, the IWF being done in the exchange); +- 4) UNI/NNI signalling interworking between DSS 2 and N-ISUP (e.g. case of a B-NT2 or B-TE attached to an integrated broadband/narrow-band local exchange, the IWF being done in the exchange). + +NOTE – The support of N-ISDN supplementary services in B-ISDN is required. + +![Figure 9/F.732 – Standardized configurations for interworking. The diagram shows the connectivity between B-TE (Broadband Terminal Equipment) and N-TE (Narrowband Terminal Equipment) through B-ISDN and N-ISDN networks using Interworking Functions (IWF).](b9ecbc3baefab13719e000faa6e0c7eb_img.jpg) + +``` + +graph LR + subgraph B-TE_Group [B-TE] + B-TE1 -- S_B --- B-NT2_1[B-NT2] -- T_B --- B-ISDN + B-TE2 -- R_B --- B-TA_1[B-TA] -- S_B --- B-NT2_2[B-NT2] -- T_B --- B-ISDN + TE2_1[TE2] -- R --- B-TA_2[B-TA] -- S --- B-NT2_3[B-NT2] -- T_B --- B-ISDN + end + + subgraph N-TE_Group_Left [N-TE] + TE1_1[TE1] -- S --- B-NT2_4[B-NT2] -- T_B --- B-ISDN + TE2_2[TE2] -- R --- TA_1[TA] -- S --- B-NT2_5[B-NT2] -- T_B --- B-ISDN + end + + B-ISDN -- IWF_6 --- N-ISDN[N-ISDN 64 kbit/s based] + B-ISDN -- T_B --- IWF_2_1[IWF 2] + B-ISDN -- T_B --- IWF_2_2[IWF 2] + + N-ISDN -- T --- NT2_1[NT2] -- S --- TE1_2[TE1] + N-ISDN -- T --- NT2_2[NT2] -- S --- TA_2[TA] -- R --- TE2_3[TE2] + IWF_2_1 -- T --- NT2_3[NT2] -- S --- TE1_3[TE1] + IWF_2_2 -- T --- NT2_4[NT2] -- S --- TA_3[TA] -- R --- TE2_4[TE2] + + subgraph N-TE_Group_Right [N-TE] + TE1_2 + TA_2 + TE2_3 + TE1_3 + TA_3 + TE2_4 + end + +``` + +The diagram illustrates interworking configurations between B-ISDN and N-ISDN. On the left, B-TE and N-TE configurations connect to a B-ISDN cloud. The B-ISDN cloud connects to an N-ISDN (64 kbit/s based) cloud via IWF 6. It also connects directly to IWF 2 units. On the right, N-TE configurations connect to the N-ISDN cloud and IWF 2 units via NT2 and TA components. Reference points include R, S, T, RB, SB, and TB. + +Figure 9/F.732 – Standardized configurations for interworking. The diagram shows the connectivity between B-TE (Broadband Terminal Equipment) and N-TE (Narrowband Terminal Equipment) through B-ISDN and N-ISDN networks using Interworking Functions (IWF). + +T0105680-96/d09 + +NOTE – NT2 may be empty (S = T) and B-NT2 may also be empty (SB = TB). If NT2 is empty, relevant IWF is IWF 3 instead of IWF 2. + +**Figure 9/F.732 – Standardized configurations for interworking** + +### 8.1.2 Scenarios at the network access + +The following figures describe the possible basic scenarios at the network access for a multimedia conference. There are seven possible types of accesses, depending on whether the terminal is broadband or narrow-band, whether it accesses a broadband or narrow-band network, and whether it is connected directly or through an NT2. + +#### 8.1.2.1 Access for a broadband terminal connected to the B-ISDN + +Access types 1 and 2 are used in this case and have already been described in 3.3.1. + +NOTE – The mapping of IWF (to be understood as generic) and TA is for further study. + +#### 8.1.2.2 Access type 3 + +This is an access for a narrow-band terminal to the B-ISDN. + +![Figure 10/F.732 – Configuration diagram showing N-TE connected to IWF 3 via S = T, and IWF 3 connected to B-ISDN via T_B.](bffdddb47fced140f8d17fdc2a29f592_img.jpg) + +The diagram shows three main components connected in a linear fashion from left to right: an 'N-TE' (Narrow-band Terminal Equipment) block, an 'IWF 3' (Interface Function) block, and a 'B-ISDN' (Broadband ISDN) block. The connection between N-TE and IWF 3 is labeled 'S = T'. The connection between IWF 3 and B-ISDN is labeled 'T\_B'. A small code 'T0105690-96/d10' is located below the B-ISDN block. + +Figure 10/F.732 – Configuration diagram showing N-TE connected to IWF 3 via S = T, and IWF 3 connected to B-ISDN via T\_B. + +Figure 10/F.732 – Configuration + +![Figure 11/F.732 – U-Plane diagram showing data flow from B/H ch through PHY, AAL 1, ATM, and PHY to another ATM and PHY block.](a7c51c18111139f9aca2805114108565_img.jpg) + +This diagram illustrates the U-plane data flow. On the left, a 'B/H ch' (Broadband/Hybrid channel) is connected to a 'PHY' (Physical) layer. This is followed by another 'PHY' layer which connects to a stack containing 'AAL 1', 'ATM', and 'PHY'. This stack is connected to a third 'ATM' and 'PHY' block. A small code 'T0105700-96/d11' is located below the rightmost block. + +Figure 11/F.732 – U-Plane diagram showing data flow from B/H ch through PHY, AAL 1, ATM, and PHY to another ATM and PHY block. + +Figure 11/F.732 – U-Plane + +![Figure 12/F.732 – C-Plane diagram showing multiple signaling channels (Q.931, Q.921, D ch) being processed through various layers (Q.931, Q.921, Q.2931, Q.2130, Q.2110, AAL 5, ATM, PHY) to reach network protocols like B-ISUP, MTP3, Q.2140, Q.2110, AAL 5, ATM, and PHY.](3f1987804d7d78bc3b3bc560e974280a_img.jpg) + +This diagram shows the C-plane signaling paths. On the left, four input channels are listed: 'Q.931', 'Q.921', 'D ch', and 'PHY'. Arrows from these inputs point to a central processing block. Inside this block, the signals are routed through various layers: 'Q.931' and 'Q.921' lead to a 'Q.2931' layer; 'Q.921' also leads to a 'Q.2130' and 'Q.2110' layer; 'D ch' leads to an 'ATM' and 'PHY' layer. These processed signals then exit the central block and are mapped to a final stack of protocols on the right: 'B-ISUP', 'MTP3', 'Q.2140', 'Q.2110', 'AAL 5', 'ATM', and 'PHY'. A small code 'T0105710-96/d12' is located below the final stack. + +Figure 12/F.732 – C-Plane diagram showing multiple signaling channels (Q.931, Q.921, D ch) being processed through various layers (Q.931, Q.921, Q.2931, Q.2130, Q.2110, AAL 5, ATM, PHY) to reach network protocols like B-ISUP, MTP3, Q.2140, Q.2110, AAL 5, ATM, and PHY. + +Figure 12/F.732 – C-Plane + +#### 8.1.2.3 Access type 4 + +This is another access for a narrow-band terminal to the B-ISDN. + +![Figure 13/F.732 – Configuration diagram showing a linear flow: N-TE connected to N-NT2 via interface S, N-NT2 connected to IWF 2 via interface T, and IWF 2 connected to B-ISDN via interface T_B.](0f985b39edc1d52ba3600c438bc8f0a5_img.jpg) + +Figure 13/F.732 – Configuration diagram showing a linear flow: N-TE connected to N-NT2 via interface S, N-NT2 connected to IWF 2 via interface T, and IWF 2 connected to B-ISDN via interface T\_B. + +T0105720-96/d13 + +Figure 13/F.732 – Configuration + +![Figure 14/F.732 – U-Plane diagram showing data flow through multiple B/H channels and PHY layers. It includes internal components like AAL 1, ATM, and PHY, leading to an external ATM/PHY interface.](a83ba9e3e2c1e21dd69953a7b09e45b4_img.jpg) + +Figure 14/F.732 – U-Plane diagram showing data flow through multiple B/H channels and PHY layers. It includes internal components like AAL 1, ATM, and PHY, leading to an external ATM/PHY interface. + +T0105730-96/d14 + +Figure 14/F.732 – U-Plane + +![Figure 15/F.732 – C-Plane diagram showing signaling paths for various protocols including Q.931, Q.921, D ch, Q.2931, Q.2130, Q.2110, AAL 5, ATM, and PHY. It maps these to higher-level protocols like B-ISUP, MTP3, Q.2140, and Q.2110.](b90144cfbb81a2d610d920240fda689d_img.jpg) + +Figure 15/F.732 – C-Plane diagram showing signaling paths for various protocols including Q.931, Q.921, D ch, Q.2931, Q.2130, Q.2110, AAL 5, ATM, and PHY. It maps these to higher-level protocols like B-ISUP, MTP3, Q.2140, and Q.2110. + +T0105740-96/d15 + +Figure 15/F.732 – C-Plane + +#### 8.1.2.4 Access type 5 + +This is another access for a narrow-band terminal to the N-ISDN. + +![Figure 16/F.732 – Configuration diagram showing a linear flow: N-TE connected to N-NT2 via interface S, and N-NT2 connected to a 64 kbit/s based ISDN via interface T.](2de714cedbbdd36f901f71bafa78f75a_img.jpg) + +Figure 16/F.732 – Configuration diagram showing a linear flow: N-TE connected to N-NT2 via interface S, and N-NT2 connected to a 64 kbit/s based ISDN via interface T. + +T0105750-96/d16 + +Figure 16/F.732 – Configuration + +![Diagram of the U-plane for F.732. It shows a sequence of components: B/H ch, PHY, B/H ch, PHY, B/H ch, PHY, B/H ch, PHY, and a final B/H ch, PHY block. Arrows indicate bidirectional data flow between adjacent components. Reference T0105760-96/d17 is shown at the bottom right.](8fbdfc3d17fb1dae7b2d8f5a287fa9fc_img.jpg) + +T0105760-96/d17 + +Diagram of the U-plane for F.732. It shows a sequence of components: B/H ch, PHY, B/H ch, PHY, B/H ch, PHY, B/H ch, PHY, and a final B/H ch, PHY block. Arrows indicate bidirectional data flow between adjacent components. Reference T0105760-96/d17 is shown at the bottom right. + +**Figure 17/F.732 – U-Plane** + +![Diagram of the C-plane for F.732. It shows a sequence of components: Q.931, Q.921, D ch, PHY, Q.931, Q.921, D ch, PHY, Q.931, Q.921, D ch, PHY, Q.931, Q.921, D ch, PHY, and a final stack of N-ISUP, MTP3, MTP2, MTP1. Arrows indicate bidirectional data flow between adjacent components. Reference T0105770-96/d18 is shown at the bottom right.](2cde062fd82833415971a8bd1a2cafab_img.jpg) + +T0105770-96/d18 + +Diagram of the C-plane for F.732. It shows a sequence of components: Q.931, Q.921, D ch, PHY, Q.931, Q.921, D ch, PHY, Q.931, Q.921, D ch, PHY, Q.931, Q.921, D ch, PHY, and a final stack of N-ISUP, MTP3, MTP2, MTP1. Arrows indicate bidirectional data flow between adjacent components. Reference T0105770-96/d18 is shown at the bottom right. + +**Figure 18/F.732 – C-plane** + +#### **8.1.2.5 Access type 6** + +This is still another access for a narrow-band terminal to the B-ISDN. + +![Diagram of the configuration for access type 6. It shows a sequence of components: N-TE, S, IWF 1, S_B, B-NT2, T_B, and B-ISDN. Arrows indicate bidirectional data flow between adjacent components. Reference T0105780-96/d19 is shown at the bottom right.](21ad58fee90f2be50708ff541d225507_img.jpg) + +T0105780-96/d19 + +Diagram of the configuration for access type 6. It shows a sequence of components: N-TE, S, IWF 1, S\_B, B-NT2, T\_B, and B-ISDN. Arrows indicate bidirectional data flow between adjacent components. Reference T0105780-96/d19 is shown at the bottom right. + +**Figure 19/F.732 – Configuration** + +![Diagram of the U-plane for access type 6. It shows a sequence of components: B/H ch, PHY, B/H ch, PHY, AAL 1, ATM, PHY, AAL 1, ATM, PHY, AAL 1, ATM, PHY, ATM, PHY, and a final ATM, PHY block. Arrows indicate bidirectional data flow between adjacent components. Reference T0105790-96/d20 is shown at the bottom right.](43837b056625d3d6ce615e4c02f163bb_img.jpg) + +T0105790-96/d20 + +Diagram of the U-plane for access type 6. It shows a sequence of components: B/H ch, PHY, B/H ch, PHY, AAL 1, ATM, PHY, AAL 1, ATM, PHY, AAL 1, ATM, PHY, ATM, PHY, and a final ATM, PHY block. Arrows indicate bidirectional data flow between adjacent components. Reference T0105790-96/d20 is shown at the bottom right. + +**Figure 20/F.732 – U-Plane** + +![Protocol stack diagram for the C-Plane showing layers Q.931, Q.921, D ch, PHY, Q.2931, Q.2130, Q.2110, AAL 5, ATM, and PHY across multiple nodes. A reference stack on the right lists B-ISUP, MTP3, Q.2140, Q.2110, AAL 5, ATM, and PHY.](76b0cd79baaedd942af4dc42f2e764b8_img.jpg) + +| | | | | | | | | | | +|-------|-------|-------|--------|-------|--------|--------|-------|--------|-----------------------------------------------------------| +| Q.931 | <---> | Q.931 | Q.2931 | <---> | Q.2931 | Q.2931 | <---> | Q.2931 | B-ISUP
MTP3
Q.2140
Q.2110
AAL 5
ATM
PHY | +| Q.921 | <---> | Q.921 | Q.2130 | <---> | Q.2130 | Q.2130 | <---> | Q.2130 | | +| | | | Q.2110 | <---> | Q.2110 | Q.2110 | <---> | Q.2110 | | +| | | | AAL 5 | <---> | AAL 5 | AAL 5 | <---> | AAL 5 | | +| D ch | <---> | D ch | ATM | <---> | ATM | ATM | <---> | ATM | | +| PHY | <---> | PHY | PHY | <---> | PHY | PHY | <---> | PHY | | + +Protocol stack diagram for the C-Plane showing layers Q.931, Q.921, D ch, PHY, Q.2931, Q.2130, Q.2110, AAL 5, ATM, and PHY across multiple nodes. A reference stack on the right lists B-ISUP, MTP3, Q.2140, Q.2110, AAL 5, ATM, and PHY. + +T0105800-96/d21 + +Figure 21/F.732 – C-Plane + +## 8.2 Access type 7 + +This is an access for a narrow-band terminal to the N-ISDN. + +![Configuration diagram showing an N-TE connected to a 64 kbit/s-based ISDN via an S = T interface.](9cd90f495b95ad2116ff780248c26d95_img.jpg) + +N-TE ------------------ S = T ------------------ 64 kbit/s-based ISDN + +Configuration diagram showing an N-TE connected to a 64 kbit/s-based ISDN via an S = T interface. + +T0105810-96/d22 + +Figure 22/F.732 – Configuration + +![U-Plane diagram showing B/H ch and PHY layers between an N-TE and a 64 kbit/s-based ISDN. A reference stack on the right shows B/H ch and PHY.](51db757d054ce1ce83c436a3578b56ca_img.jpg) + +| | | | | +|--------|----------------------------------------------------------------------|--------|---------------| +| B/H ch | <------------------------------------------------------------------> | B/H ch | B/H ch
PHY | +| PHY | <------------------------------------------------------------------> | PHY | | + +U-Plane diagram showing B/H ch and PHY layers between an N-TE and a 64 kbit/s-based ISDN. A reference stack on the right shows B/H ch and PHY. + +T0105820-96/d23 + +Figure 23/F.732 – U-Plane + +![Protocol stack diagram for the C-Plane showing layers Q.931, Q.921, D ch, PHY, and a reference stack on the right with N-ISUP, MTP3, MTP2, and MTP1.](e928f4874ed492d3ad4c6fa2d29aedbc_img.jpg) + +| | | | | +|-------|----------------------------------------------------------------------|-------|----------------| +| Q.931 | <------------------------------------------------------------------> | Q.931 | N-ISUP
MTP3 | +| Q.921 | <------------------------------------------------------------------> | Q.921 | | +| D ch | <------------------------------------------------------------------> | D ch | MTP2 | +| PHY | <------------------------------------------------------------------> | PHY | MTP1 | + +Protocol stack diagram for the C-Plane showing layers Q.931, Q.921, D ch, PHY, and a reference stack on the right with N-ISUP, MTP3, MTP2, and MTP1. + +T0105830-96/d24 + +Figure 24/F.732 – C-Plane + +### 8.2.1 Scenarios for interworking between B-ISDN and N-ISDN + +![Figure 25/F.732 – Configuration: A block diagram showing B-ISDN connected to IWF 6, which is connected to 64 kbit/s-based ISDN.](7f25db95ce3916c0e09803b861a2f7bc_img.jpg) + +Figure 25/F.732 – Configuration: A block diagram showing B-ISDN connected to IWF 6, which is connected to 64 kbit/s-based ISDN. + +T0105840-96/d25 + +Figure 25/F.732 – Configuration + +![Figure 26/F.732 – U-Plane: A detailed protocol stack diagram for the U-plane. It shows multiple ATM/PHY units on the left connected to a central B/H ch and PHY block. This central block is connected to another B/H ch and PHY block, which is then connected to a third B/H ch and PHY block on the right.](10781f43062bf3e9601a1e086710556c_img.jpg) + +Figure 26/F.732 – U-Plane: A detailed protocol stack diagram for the U-plane. It shows multiple ATM/PHY units on the left connected to a central B/H ch and PHY block. This central block is connected to another B/H ch and PHY block, which is then connected to a third B/H ch and PHY block on the right. + +T0105850-96/d26 + +Figure 26/F.732 – U-Plane + +![Figure 27/F.732 – C-plane if N-ISDN <-> B-ISDN <-> N-ISDN: A detailed protocol stack diagram for the C-plane. It shows a sequence of stacks: B-ISUP/MTP3/Q.2140/Q.2110/AAL 5/ATM/PHY on the left, followed by a B-ISUP/MTP3/Q.2140/Q.2110/AAL 5/ATM/PHY stack, then a central N-ISUP/MTP3/MTP2/MTP1 stack, followed by another N-ISUP/MTP3/MTP2/MTP1 stack, and finally an N-ISUP/MTP3/MTP2/MTP1 stack on the right.](12de9b926df0384ec07702671827c9cd_img.jpg) + +Figure 27/F.732 – C-plane if N-ISDN <-> B-ISDN <-> N-ISDN: A detailed protocol stack diagram for the C-plane. It shows a sequence of stacks: B-ISUP/MTP3/Q.2140/Q.2110/AAL 5/ATM/PHY on the left, followed by a B-ISUP/MTP3/Q.2140/Q.2110/AAL 5/ATM/PHY stack, then a central N-ISUP/MTP3/MTP2/MTP1 stack, followed by another N-ISUP/MTP3/MTP2/MTP1 stack, and finally an N-ISUP/MTP3/MTP2/MTP1 stack on the right. + +T0105860-96/d27 + +Figure 27/F.732 – C-plane if N-ISDN <-> B-ISDN <-> N-ISDN + +![Figure 28/F.732 – C-plane if B-TE <-> B-ISDN <-> N-ISDN <-> N-TE: A detailed protocol stack diagram for the C-plane. It shows a sequence of stacks: Q.2931/Q.2130/Q.2110/AAL 5/ATM/PHY on the left, followed by a B-ISUP/MTP3/Q.2140/Q.2110/AAL 5/ATM/PHY stack, then a central N-ISUP/MTP3/MTP2/MTP1 stack, followed by another N-ISUP/MTP3/MTP2/MTP1 stack, and finally a Q.931/Q.921/D ch/PHY stack on the right.](327ba94498e3381cf08eb41e3fd3d77f_img.jpg) + +Figure 28/F.732 – C-plane if B-TE <-> B-ISDN <-> N-ISDN <-> N-TE: A detailed protocol stack diagram for the C-plane. It shows a sequence of stacks: Q.2931/Q.2130/Q.2110/AAL 5/ATM/PHY on the left, followed by a B-ISUP/MTP3/Q.2140/Q.2110/AAL 5/ATM/PHY stack, then a central N-ISUP/MTP3/MTP2/MTP1 stack, followed by another N-ISUP/MTP3/MTP2/MTP1 stack, and finally a Q.931/Q.921/D ch/PHY stack on the right. + +T0105870-96/d28 + +Figure 28/F.732 – C-plane if B-TE <-> B-ISDN <-> N-ISDN <-> N-TE + +#### 8.2.1.1 Point-to-point scenarios + +Any end-to-end point-to-point scenario is described by connecting two among all the relevant above figures. + +#### 8.2.1.2 Multipoint scenarios + +##### 8.2.1.2.1 General multipoint considerations + +There are three kinds of scenarios which may possibly support a multipoint multimedia conference: + +- 1) Every terminal establishes an end-to-end (unidirectional) connection to each of the other terminals. In this case, the interworking is covered by the paragraphs dealing with the point-to-point interworking. + +![Diagram of a multipoint scenario using point-to-point connections. It shows a stack of components: TE, TA, B-NT2, and B-ISDN. The TE is connected to the TA via an S interface. The TA is connected to the B-NT2 via an S_B interface. The B-NT2 is connected to the B-ISDN via a T_B interface. The B-ISDN is connected to an IWF (Interworking Function) via a T_B interface. The IWF is connected to an N-ISDN via a T interface. The N-ISDN is connected to an N-TE via a T interface. The diagram also shows a B-TE connected to the B-ISDN via a T_B interface. A legend indicates that double-headed arrows represent 'Two reverse ATM connections', single-headed arrows represent 'One H/B circuit (bidirectional)', and a solid black circle represents a 'Mapping entity of two reverse ATM connections onto a bidirectional B or vice versa'. A note states that IWF 1 corresponds to TA. The diagram is labeled T0105880-96/d29.](523ab7b925beb555f88b2e1e1336974f_img.jpg) + +Two reverse ATM connections + +One H/B circuit (bidirectional) + +Mapping entity of two reverse ATM connections onto a bidirectional B or vice versa + +NOTE – IWF 1 corresponds to TA. + +Diagram of a multipoint scenario using point-to-point connections. It shows a stack of components: TE, TA, B-NT2, and B-ISDN. The TE is connected to the TA via an S interface. The TA is connected to the B-NT2 via an S\_B interface. The B-NT2 is connected to the B-ISDN via a T\_B interface. The B-ISDN is connected to an IWF (Interworking Function) via a T\_B interface. The IWF is connected to an N-ISDN via a T interface. The N-ISDN is connected to an N-TE via a T interface. The diagram also shows a B-TE connected to the B-ISDN via a T\_B interface. A legend indicates that double-headed arrows represent 'Two reverse ATM connections', single-headed arrows represent 'One H/B circuit (bidirectional)', and a solid black circle represents a 'Mapping entity of two reverse ATM connections onto a bidirectional B or vice versa'. A note states that IWF 1 corresponds to TA. The diagram is labeled T0105880-96/d29. + +**Figure 29/F.732 – Example of a multipoint scenario using point-to-point connections** + +It should be noted that the B/H-channels are always bidirectional and the ATM connections are unidirectional. Thus, an interworking constraint is to map one onto the other and vice versa and to determine the associated protocols which are for further study. + +- 2) Every terminal establishes an end-to-end point-to-multipoint (unidirectional) connection to all other terminals. + +It should be noted that an N-TE is not able to establish a point-to-multipoint connection and if there are N terminals, each terminal should have N – 1 decoders and one coder. In the case of the N-TE it does not seem a viable solution. Consequently, such scenarios including an N-TE are not considered realistic. + +The way for establishing a multipoint connection in the B-ISDN from an endpoint in the N-ISDN is not defined. Thus, the multipoint aspects of the connection have to be treated in the B-ISDN and the MCU has to be directly connected to the B-ISDN. + +For the same reason as in the first case, this scenario is for further study. + +- 3) Every terminal establishes a couple of end-to-end point-to-point opposite (unidirectional) connections between itself and an MCU. + +![Diagram of a multipoint scenario using point-to-point connections. It shows a central N-ISDN block connected to two B-ISDN blocks via IWF (Interworking Function) blocks. Each B-ISDN block is connected to a B-TE (Basic Terminal Equipment) block. The diagram illustrates various connections: two opposite ATM connections (double-headed arrows), one H/B circuit (bidirectional) (single-headed arrow), mapping entities (small black dots), and multipoint functions (large black dots). The label T_B is used to identify specific connection points. A reference code T0105890-96/d30 is present in the bottom right of the diagram area.](5b8a756d9a71c35f17db8bcb90b438a3_img.jpg) + +←→ Two opposite ATM connections + → One H/B circuit (bidirectional) + ● Mapping entity of two opposite ATM connections onto a bidirectional B or vice versa + ● Multipoint functions + +T0105890-96/d30 + +Diagram of a multipoint scenario using point-to-point connections. It shows a central N-ISDN block connected to two B-ISDN blocks via IWF (Interworking Function) blocks. Each B-ISDN block is connected to a B-TE (Basic Terminal Equipment) block. The diagram illustrates various connections: two opposite ATM connections (double-headed arrows), one H/B circuit (bidirectional) (single-headed arrow), mapping entities (small black dots), and multipoint functions (large black dots). The label T\_B is used to identify specific connection points. A reference code T0105890-96/d30 is present in the bottom right of the diagram area. + +NOTE – IWF is IWF 6. + +**Figure 30/F.732 – Example of a multipoint scenario using point-to-point connections** + +For the time being, the initial scenario should be based on an MCU whose accesses are narrow-band. In a second phase, all MCU accesses should be broadband. Offering narrow-band and broadband access in the same MCU needs further considerations. + +##### 8.2.1.2.2 MCU connected to the N-ISDN + +##### 8.2.1.2.3 MCU connected to the B-ISDN + +## 8.3 Hybrid MCU + +The hybrid MCU has access to both the B-ISDN and N-ISDN, thus bridging the two networks for setting up a multimedia conference with terminals connected to each network. + +## 8.4 Interworking with LANs + +The same types of configurations and scenarios apply to interworking with LANs. If a terminal connected to an LAN supports one or more N-ISDN compatible modes (possibly through a terminal adapter), then it can be connected to an N-ISDN compatible terminal on the B-ISDN network; the gateway between the LAN and the B-ISDN ensures compatibility and continuity of the channels used on both networks. + +For instance, for terminals with video capability, conformance of a B-ISDN terminal with Recommendation H.321 ensures compatibility with LAN terminals conforming to Recommendation H.322 (for a LAN with guaranteed bandwidth) or to Recommendation H.323 (for an LAN without guaranteed bandwidth). + +![Diagram of a multipoint scenario using point-to-point connections. A central B-ISDN block is connected to an MCU at the bottom via three lines. Above the B-ISDN block is a B-NT2, which is connected to a TA (labeled S_B) and then to a TE (labeled S). Two B-TE blocks are connected to the B-ISDN block via lines labeled T_B. Reference T0105900-96/d31 is shown.](e69b9188aa2c14ec6b21c83f711fef65_img.jpg) + +Diagram of a multipoint scenario using point-to-point connections. A central B-ISDN block is connected to an MCU at the bottom via three lines. Above the B-ISDN block is a B-NT2, which is connected to a TA (labeled S\_B) and then to a TE (labeled S). Two B-TE blocks are connected to the B-ISDN block via lines labeled T\_B. Reference T0105900-96/d31 is shown. + +↔ Two opposite point-to-point connections + +NOTE – IWF 1 corresponds to TA. + +**Figure 31/F.732 – Example of multipoint scenario using point-to-point connections** + +![Diagram of a scenario including a narrow-band MCU. A central B-ISDN block is connected to an MCU at the bottom via three lines. Above the B-ISDN block is a B-NT2, which is connected to a TA (labeled S_B) and then to a TE (labeled S). To the left, an N-TE is connected to an N-ISDN block, which is connected to an IWF block. This IWF block is connected to the B-ISDN block. Another IWF block is connected to the B-ISDN block and to an N-ISDN block on the right. The N-ISDN block on the right is connected to an N-TE. Reference T0105910-96/d32 is shown.](09955ff8214ffb6947951fc0f60eb6ab_img.jpg) + +Diagram of a scenario including a narrow-band MCU. A central B-ISDN block is connected to an MCU at the bottom via three lines. Above the B-ISDN block is a B-NT2, which is connected to a TA (labeled S\_B) and then to a TE (labeled S). To the left, an N-TE is connected to an N-ISDN block, which is connected to an IWF block. This IWF block is connected to the B-ISDN block. Another IWF block is connected to the B-ISDN block and to an N-ISDN block on the right. The N-ISDN block on the right is connected to an N-TE. Reference T0105910-96/d32 is shown. + +NOTE – IWF 1 corresponds to TA and IWF is IWF 6. + +**Figure 32/F.732 – Example of a scenario including a narrow-band MCU** + +![Diagram of a scenario including a broadband MCU. It shows three main vertical paths. The left path consists of TE, S, TA, S_B, B-NT2, and T_B connected to a B-ISDN block, which is also connected to a B-TE at T_B. The middle path consists of N-TE connected to T, which is connected to a B-ISDN block. The right path consists of TE, S, TA, S_B, B-NT2, and T_B connected to a B-ISDN block, which is connected to an MCU. All three B-ISDN blocks are interconnected by a horizontal bus. Two IWF (Interface) blocks are shown on this bus. A B-TE at T_B is also connected to the MCU.](c914f51f4427bc672dd0526cfc90ebe9_img.jpg) + +Diagram of a scenario including a broadband MCU. It shows three main vertical paths. The left path consists of TE, S, TA, S\_B, B-NT2, and T\_B connected to a B-ISDN block, which is also connected to a B-TE at T\_B. The middle path consists of N-TE connected to T, which is connected to a B-ISDN block. The right path consists of TE, S, TA, S\_B, B-NT2, and T\_B connected to a B-ISDN block, which is connected to an MCU. All three B-ISDN blocks are interconnected by a horizontal bus. Two IWF (Interface) blocks are shown on this bus. A B-TE at T\_B is also connected to the MCU. + +T0105920-96/d33 + +NOTE – IWF 1 corresponds to TA and IWF is IWF 6. + +**Figure 33/F.732 – Scenario including a broadband MCU** + +![Diagram of a scenario including a hybrid MCU. It shows three main vertical paths. The left path consists of N-TE connected to T, which is connected to an N-ISDN block, which is connected to an IWF block, which is connected to a B-ISDN block. The middle path consists of N-TE connected to T, which is connected to an N-ISDN block. The right path consists of B-TE connected to T_B, which is connected to a B-ISDN block, which is connected to an IWF block, which is connected to a B-ISDN block. All three B-ISDN blocks are interconnected by a horizontal bus. Two IWF (Interface) blocks are shown on this bus. A B-TE at T_B is connected to the MCU via a 'Broadband access' path. The MCU is also connected to the N-ISDN block via a 'Narrow-band access' path. A dotted box labeled 'Same geographical zone' contains the B-TE, B-NT2, TA, S, and TE components of the right path.](844077b3034f0030b404207db0ad76b4_img.jpg) + +Diagram of a scenario including a hybrid MCU. It shows three main vertical paths. The left path consists of N-TE connected to T, which is connected to an N-ISDN block, which is connected to an IWF block, which is connected to a B-ISDN block. The middle path consists of N-TE connected to T, which is connected to an N-ISDN block. The right path consists of B-TE connected to T\_B, which is connected to a B-ISDN block, which is connected to an IWF block, which is connected to a B-ISDN block. All three B-ISDN blocks are interconnected by a horizontal bus. Two IWF (Interface) blocks are shown on this bus. A B-TE at T\_B is connected to the MCU via a 'Broadband access' path. The MCU is also connected to the N-ISDN block via a 'Narrow-band access' path. A dotted box labeled 'Same geographical zone' contains the B-TE, B-NT2, TA, S, and TE components of the right path. + +T0105930-96/d34 + +NOTE – IWF 1 corresponds to TA and IWF is IWF 6. + +**Figure 34/F.732 – Scenario including a hybrid MCU** + +## 8.5 Intercommunication between services + +All terminals have the media component audio. The basic common audio mode is G.722. However, some videophone terminals (and possibly some CDH terminals) may not have that capability; in that case G.711 is used, or alternately G.728 if all terminals in the conference have that capability. + +For the media component video, all terminals which support it use H.262 coding, or H.261 for interworking with narrow-band systems. They will always be able to communicate with this coding algorithm, by aligning the bit rate of the video subchannel on the lowest bit rate. If the resulting quality is not sufficient, then a higher bit rate may be chosen, with some terminals relegated to secondary status and participating with sound (and possibly data) only. + +For the other media components, all terminals using them have a common data subchannel with the T.120 protocol. The bit rate of the subchannel will be aligned on the terminal with the lowest bit rate capability. However, this does not imply that all terminals will have a common media component. For instance, a group of participants may be able to exchange facsimile while another group (overlapping or not) may exchange still pictures. + +# 9 Attribute table and values + +## 9.1 Information transfer high layer attributes + +**Table 1/F.732 – Information transfer high layer attributes for the broadband conference services** + +| | | | +|------|-----------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------| +| 1 | Name of the service | Broadband conference | +| 2 | Service type | May correspond to each of the different service profiles (see 7.2) | +| 3 | Connection mode | Connection oriented | +| 4 | Media Component(s) MC(s) | MC 1: audio
MC 2: video
MC 3: synchronous signalling
MC 4: data
Applicability of MC to service profile is specified in 7.2 | +| 5 | Establishment of the service | Subscription
Use on: either demand or reserved | +| 6 | Configuration of the service | Point-to-point
Multipoint | +| 7 | Supplementary services provided | For further study | +| 8 | Quality of service | For further study | +| 9 | Interworking capabilities | With narrow-band videoconference and narrow-band videotelephony | +| 10 | Operational and commercial aspects | Outside of the scope | +| 11 | Protocol stack above ATM | | +| 11.1 | ATM Adaptation layer | MC 1: AAL 1
MC 2: AAL 1
MC 3: AAL 1
MC 4: AAL 1 or AAL 5 | +| 11.2 | Information access protocols
– Layer 2 above AAL | H.222.0, H.222.1 | + +**Table 1/F.732 – Information transfer high layer attributes for the broadband conference services (concluded)** + +| | | | +|------|-----------------------------------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| 11.3 | Information access protocols
– Layer 3 above AAL | H.245 | +| 11.4 | Information access protocols
– Layer 4 | For further study | +| 11.5 | Information access protocols
– Layer 5 | For further study | +| 11.6 | Information access protocols
– Layer 6 | G.711, G.722 or G.728 for audio
H.261 or H.262 for video (profiles 1, 4, optional for profile 3)
T.120-Series for data (profiles 2, 3, optional for profiles 1, 4) | +| 11.7 | Information access protocols
– Layer 7 | For further study | + +## 9.2 Information transfer low layer attributes + +**Table 2/F.732 – Information transfer low layer attributes for the broadband conference service** + +| | | | +|-----|--------------------------------------|---------------------------------------| +| 1 | Information transfer mode | ATM cell | +| 2 | Topology | | +| 2.1 | Symmetry of the communication | Bidirectional symmetric | +| 2.2 | Configuration of the ATM connections | Point-to-point or point-to-multipoint | +| 3 | Structure of the ATM connections | Cell sequence integrity | + +## 9.3 Access attributes + +**Table 3/F.732 – Access attributes for the broadband conference services** + +| | | | +|-----|--------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------| +| 1 | Access attributes for user information | | +| 1.1 | Type of connections supporting the MC(s) | VC 1 for the set of audio, video, data and synchronous signalling
VC 1 for the set of audio, video and synchronous signalling and
VC 2 for data | +| 1.2 | ATM transfer capability of the ATM connections | VC 1: DBR
VC 2: DBR, SBR, ABR, ABT-DT, ABT-IT | +| 1.3 | Source traffic descriptor of the ATM connections | VC 1: The values are for further study taking I.371 into account
VC 2: The values are for further study taking I.371 into account | +| 1.4 | QOS class of the ATM connections | VC 1: The values are for further study taking I.356 into account
VC 2: The values are for further study taking I.356 into account | + +**Table 3/F.732 – Access attributes for the broadband conference services (concluded)** + +| | | | +|-----|---------------------------------------------------------|--------------------------------------------------------------------------------| +| 2 | Access attributes for signalling | | +| 2.1 | Type of signalling connection | VC | +| 2.2 | ATM transfer capability of ATM connection | DBR | +| 2.3 | Source traffic descriptor of ATM connection | Values are for further study | +| 2.4 | QOS class of ATM connection | Class 1 | +| 3 | Access attributes for management | | +| 3.1 | Type of management connection | VC | +| 3.2 | ATM transfer capability of ATM connection | DBR | +| 3.3 | Source traffic descriptor of ATM connection | Values are for further study | +| 3.4 | QOS class of ATM connection | Class 1 | +| 4 | Access attributes for fast allocation of resources | For further study | +| 5 | Access protocols | | +| 5.1 | Access protocol – Physical layer | See Rec. I.432 | +| 5.2 | Access protocol – ATM layer | See Recs. I.150, I.361 | +| 5.3 | Signalling access protocol – ATM Adaptation Layer (AAL) | Recommendations of the SAAL series Q.2100, Q.2110 and Q.2130 | +| 5.4 | Signalling access protocol – Layer 3 (above AAL) | Recommendations of the Q.2900 (DSS 2) series Q.2931, Q.2961, Q.2962 and Q.2963 | +| 5.5 | Management access protocol – ATM Adaptation Layer (AAL) | For further study | +| 5.6 | Management access protocol – Layer 3 (above AAL) | For further study | +| 5.7 | Access protocol for fast allocation of resources | For further study | + +# Annex A + +# References + +- ITU-T Recommendation F.702 (1996), *Multimedia conference services*. +- CCITT Recommendation G.711 (1988), *Pulse code modulation (PCM) of voice frequencies*. +- CCITT Recommendation G.722 (1988), *7 kHz audio-coding within 64 kbit/s*. +- CCITT Recommendation G.728 (1992), *Coding of speech at 16 kbit/s using low-delay code excited linear prediction*. + +- ITU-T Recommendation H.222.0 (1996), *Information technology – Generic coding of moving pictures and associated audio information: Systems*. +- ITU-T Recommendation H.222.1 (1996), *Multimedia multiplex and synchronization for audiovisual communication in ATM environments*. +- ITU-T Recommendation H.245 (1996), *Control protocol for multimedia communication*. +- ITU-T Recommendation H.261 (1993), *Video codec for audiovisual services at $p \times 64$ kbit/s*. +- ITU-T Recommendation H.262 (1995), *Information technology – Generic coding of moving pictures and associated audio information: video*. +- ITU-T Recommendation H.310 (1996), *Broadband and audiovisual communication systems and terminals*. +- ITU-T Recommendation H.321 (1996), *Adaptation of H.320 visual telephone terminals to B-ISDN environments*. +- ITU-T Recommendation H.322 (1996), *Visual telephone systems and terminal equipment for local area networks which provide a guaranteed quality of service*. +- ITU-T Recommendation H.323 (1996), *Visual telephone systems and equipment for local area networks which provide a non-guaranteed quality of service*. +- ITU-T Recommendation I.150 (1995), *B-ISDN asynchronous transfer mode functional characteristics*. +- ITU-T Recommendation I.356 (1996), *B-ISDN ATM layer cell transfer performance*. +- ITU-T Recommendation I.361 (1995), *B-ISDN ATM layer specification*. +- ITU-T Recommendations I.363.1 (1996), I.363.3 (1996) and I.363.5 (1996), *B-ISDN ATM Adaptation Layer specification*. +- ITU-T Recommendation I.371 (1996), *Traffic control and congestion control in B-ISDN*. +- ITU-T Recommendation I.413 (1993), *B-ISDN user-network interface*. +- ITU-T Recommendation I.580 (1995), *General arrangements for interworking between B-ISDN and 64 kbit/s based ISDN*. +- ITU-T Recommendation Q.921 (1993), *ISDN user-network interface – Data link layer specification*. +- ITU-T Recommendation Q.931 (1993), *ISDN user-network interface layer 3 specification for basic call control*. +- ITU-T Recommendation Q.2100 (1994), *B-ISDN Signalling ATM Adaptation Layer (SAAL) overview description*. +- ITU-T Recommendation Q.2110 (1994), *B-ISDN ATM adaptation layer – Service Specific Connection Oriented Protocol (SSCOP)*. +- ITU-T Recommendation Q.2130 (1994), *B-ISDN Signalling ATM adaptation layer – Service Specific Coordination Function for support of Signalling at the User-Network Interface (SSCF at UNI)*. +- ITU-T Recommendation Q.2140 (1995), *B-ISDN Signalling ATM Adaptation Layer – Service Specific Coordination Function for support of Signalling at the Network Node Interface (SSCF at NNI)*. +- ITU-T Recommendation Q.2931 (1995), *Digital Subscriber Signalling System No. 2 – User-Network Interface (UNI) layer 3 specification for basic call/connection control*. +- ITU-T Recommendations Q.2961.1 (1995), and Q.2961.2 (1997), *Digital Subscriber Signalling System No. 2 – Additional traffic parameters*. +- ITU-T Recommendation Q.2962 (1996), *Digital subscriber Signalling System No. 2 – Connection characteristics negotiation during call/connection establishment phase*. +- ITU-T Recommendation Q.2963.1 (1996), *Digital subscriber Signalling System No. 2 – Connection modification: Peak cell rate modification by the connection owner*. +- ITU-T Recommendation T.127 (1995). *Multipoint binary file transfer protocol*. + + + +# ITU-T RECOMMENDATIONS SERIES + +- Series A Organization of the work of the ITU-T +- Series B Means of expression: definitions, symbols, classification +- Series C General telecommunication statistics +- Series D General tariff principles +- Series E Overall network operation, telephone service, service operation and human factors +- Series F Non-telephone telecommunication services** +- Series G Transmission systems and media, digital systems and networks +- Series H Audiovisual and multimedia systems +- Series I Integrated services digital network +- Series J Transmission of television, sound programme and other multimedia signals +- Series K Protection against interference +- Series L Construction, installation and protection of cables and other elements of outside plant +- Series M Maintenance: international transmission systems, telephone circuits, telegraphy, facsimile and leased circuits +- Series N Maintenance: international sound programme and television transmission circuits +- Series O Specifications of measuring equipment +- Series P Telephone transmission quality, telephone installations, local line networks +- Series Q Switching and signalling +- Series R Telegraph transmission +- Series S Telegraph services terminal equipment +- Series T Terminals for telematic services +- Series U Telegraph switching +- Series V Data communication over the telephone network +- Series X Data networks and open system communication +- Series Z Programming languages \ No newline at end of file diff --git a/marked/F/T-REC-F.733-200509-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg b/marked/F/T-REC-F.733-200509-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..36cf51f8ba07e44aa7a4b009f3ba08eb2b570b1c --- /dev/null +++ b/marked/F/T-REC-F.733-200509-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:11780faeee75ba39ab4ed9de67a9e67abec0d9ba3068ebec661dea70c16e29d3 +size 4338 diff --git a/marked/F/T-REC-F.733-200509-I_PDF-E/b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg b/marked/F/T-REC-F.733-200509-I_PDF-E/b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3f5666bae063d5483652a0ebdefa474745f37094 --- /dev/null +++ b/marked/F/T-REC-F.733-200509-I_PDF-E/b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a7d09841d1f105bcde93015b79dbbfdfe6ddb9140cbdd2962433db7a8ba37478 +size 66412 diff --git a/marked/F/T-REC-F.735.1-202008-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.735.1-202008-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..22dc7ff0fb30f976cc1e576c45d2b7e65fa5ddc7 --- /dev/null +++ b/marked/F/T-REC-F.735.1-202008-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aa14933e776f7d0be8b2d60b81cbb6c1ee49deb5080e0281da05296f48189055 +size 6859 diff --git a/marked/F/T-REC-F.735.1-202008-I_PDF-E/dbe553cf16dd14073b89a8263a428664_img.jpg b/marked/F/T-REC-F.735.1-202008-I_PDF-E/dbe553cf16dd14073b89a8263a428664_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..b60bc553aec9ec98499ca4221bf07c7cab84b92e --- /dev/null +++ b/marked/F/T-REC-F.735.1-202008-I_PDF-E/dbe553cf16dd14073b89a8263a428664_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ebdc58140cec8358d9352f2ce55764ef174ce1ce78b42fc7b8c612e453a57465 +size 93002 diff --git a/marked/F/T-REC-F.735.1-202008-I_PDF-E/raw.md b/marked/F/T-REC-F.735.1-202008-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..e232e65bfe87ae3f625c7bb9ee6f45bbcc7c2a11 --- /dev/null +++ b/marked/F/T-REC-F.735.1-202008-I_PDF-E/raw.md @@ -0,0 +1,319 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.735.1** + +(08/2020) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +# --- **Requirements for software-defined cameras** + +Recommendation ITU-T F.735.1 + +**ITU-T** + +![ITU logo: A globe with a red lightning bolt and the text 'ITU International Telecommunication Union'.](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a stylized globe with a red lightning bolt striking across it. To the right of the globe, the text "ITU" is written in a bold, blue font, and below it, "International Telecommunication Union" is written in a smaller, blue font. + +ITU logo: A globe with a red lightning bolt and the text 'ITU International Telecommunication Union'. + +## ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.735.1 + +# Requirements for software-defined cameras + +## Summary + +With the development of hardware and intelligent algorithms, various intelligent algorithms are developed for a large number of scenarios. In order to meet the various scenarios and requirements, cameras supporting algorithms on-demand deployment and online upgrade without service interruption are needed. A software-defined camera is a kind of intelligent premises unit (IPU), see Recommendation ITU-T F.743.1, which provides a technical approach to decoupling hardware and software and to support algorithms on-demand deployment, online upgrade without service interruption and continuous online learning to adapt to various scenarios. This Recommendation defines the typical scenarios, functional architecture and requirements of such software-defined cameras. + +Recommendation ITU-T F.735.1 can facilitate a cooperation between camera manufacturers, customers and intelligent algorithm providers. The aim is to define the basic requirements and functional architecture for software-defined cameras. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.735.1 | 2020-08-13 | 16 | 11.1002/1000/14323 | + +## Keywords + +Software-defined camera. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2020 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|--------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Typical scenarios ..... | 2 | +| 6.1 Algorithms on-demand deployment and online upgrade ..... | 2 | +| 6.2 Multi-algorithms parallel computing..... | 3 | +| 6.3 Online learning ability ..... | 3 | +| 7 Functional architecture ..... | 3 | +| 8 Requirements for software-defined camera ..... | 4 | +| 8.1 User requirements..... | 4 | +| 8.2 Service requirements ..... | 4 | +| Bibliography..... | 5 | + + + +# Recommendation ITU-T F.735.1 + +# Requirements for software-defined cameras + +# 1 Scope + +Currently, with the development of hardware and intelligent algorithms, various intelligent algorithms are developed for a large number of scenarios. In order to meet the various scenarios and requirements, cameras supporting algorithms on-demand deployment and online upgrade without service interruption are needed. A software-defined camera is a kind of intelligent premises unit (IPU), see [ITU-T F.743.1], which provides a technical approach to decoupling hardware and software and to support algorithms on-demand deployment, online upgrade without service interruption and continuous online learning to adapt to various scenarios. This Recommendation defines the typical scenarios, functional architecture and requirements of such software-defined cameras. + +The scope of this Recommendation includes: + +- 1) typical scenarios of software-defined cameras; +- 2) the functional architecture of software-defined cameras; +- 3) the requirements of software-defined cameras. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T F.743.1] Recommendation ITU-T F.743.1 (2015), *Requirements for intelligent visual surveillance*. + +[ITU-T H.626] Recommendation ITU-T H.626V2 (2019), *Architectural requirements for visual surveillance*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following term defined elsewhere: + +**3.1.1 traffic/pedestrian flow analysis** [ITU-T F.743.1]: Motion analysis and feature classification of the video of moving target(s) within the specified single or multiple video monitoring area. The pedestrian flow data can be detected accurately. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 human body attribute extraction**: Detection of the human body in video and extraction of its characteristics. These characteristics may include gender, age, clothes type, clothes colour, hairstyle and backpack, etc. + +**3.2.2 software-defined camera:** A software-defined camera is a kind of IPU (see [ITU-T F.743.1]), which provides a technical approach to decoupling hardware and software and to support algorithms' on-demand deployment, online upgrade without service interruption and continuous self-adaptive learning to adapt to various scenarios. + +**3.2.3 vehicle recognition:** Detection of the car in video and extraction of its characteristics. These characteristics may include brand, type, colour, etc. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-----|---------------------------------| +| AI | Artificial Intelligence | +| API | Application Programme Interface | +| APP | Application | +| IPU | Intelligent Premises Unit | +| ISP | Image Signal Processing | +| PIV | Premises Intelligent Video | +| PU | Premises Unit | +| SDC | Software-Defined Camera | + +# 5 Conventions + +In this Recommendation: + +The keywords "is required to" and "shall" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. + +The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement needs not be present to claim conformance. + +The keywords "can optionally" and "may" indicate an optional requirement which is permissible, without implying any sense of being recommended. These terms are not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +# 6 Typical scenarios + +This clause describes typical scenario examples illustrating the definition and work mechanism of software-defined cameras (SDCs). + +## 6.1 Algorithms on-demand deployment and online upgrade + +Software-defined cameras can support various scenarios through multi-algorithms on-demand deployment and upgrade. In the traditional surveillance systems, if the administrator wants to add a new feature or upgrade an algorithm, they have to restart the camera system and stop the surveillance service. In SDC systems, the administrator can deploy new algorithms through sending a deployment or upgrade command to software-defined cameras, then these cameras will apply this new feature without the system restarting and other service interruption. + +## 6.2 Multi-algorithms parallel computing + +The SDC supports multi-algorithm parallel computing, which means the camera can run multiple algorithms simultaneously, i.e., vehicle recognition, traffic/pedestrian flow analysis, human body attribute extraction, etc. The combination of multi-algorithms can perform as a specific mode. Users can switch between different modes to adapt different scenarios. + +## 6.3 Online learning ability + +The SDC supports continuous online learning. The SDC can automatically collect, clean and label the video or image data captured in the actual scene as a training set. It then use this training set to train the original training model which is already embedded internally, and adjusts the parameters of the model to complete final optimization of the model, making the learning model more suitable for the actual scene, as well as improving the accuracy of the intelligent analysis result. + +# 7 Functional architecture + +Figure 7-1 shows the overall functional architecture of an SDC. This architecture contains three layers, the SDC hardware layer, the SDC OS layer and the application layer. The descriptions of the three layers are as follows. The application here is related to the software which is used in the camera to complete video business and intelligent analysis. + +![Functional architecture of SDC diagram](dbe553cf16dd14073b89a8263a428664_img.jpg) + +The diagram illustrates the functional architecture of an SDC, organized into three main layers: + +- Application Layer:** The top layer, containing "Intelligent App 1 ... Intelligent App N" and "Non-intelligent App 1 ... Non-intelligent App N". A vertical "APP manager" block is positioned to the right of the application boxes. +- SDC OS Layer:** The middle layer, which includes: + - "Basic hardware service API" and "Common software service API" at the top. + - "Basic hardware function module" (containing "Video", "Codec", "ISP", and "...") and "Common software function module" (containing "API gateway", "Security", "Log", and "...") below the APIs. + - A "Virtualization layer" below the function modules. + - The "Kernel (Hardware and AI driver)" at the bottom of the OS layer. +- SDC hardware Layer:** The bottom layer, consisting of the "Hardware/Chip set". + +The diagram is labeled F.735.1(20)\_F7-1 in the bottom right corner. + +Functional architecture of SDC diagram + +F.735.1(20)\_F7-1 + +**Figure 7-1 – Functional architecture of SDC** + +SDC hardware layer: This layer provides the corresponding hardware resource and AI chips for the upper layer. + +SDC OS layer: This layer is composed of the basic hardware service API, the common software service API, the common software function module, the basic hardware function module, the virtualization layer, the APP manager and the kernel. The description of each module is as follows. + +The kernel is the driver for hardware and AI chips. + +The basic hardware function module is a function module for supporting the basic hardware capabilities of a camera which contains video, ISP, crypto, codec, PTZ, etc. + +The basic hardware service API is the service API of the basic hardware function module. + +The common software function module is used to provide the common software capabilities of a camera which include API gateway interface, event management, security, log, etc. + +The common software service API is used to provide service interfaces and works as an API gateway for the camera; all the result intelligent APPs are provided through this common software service API. + +The APP manager is a container manager which supports APP life cycle management, installation, activation, deactivation, uninstallation, upgrade, license management, etc. + +The virtualization layer is used to complete the APP resources isolation and support the online upgrade and deployment without any service interruption; the major technology used here is the container technology. + +The application layer is composed of intelligent APPs and non-intelligent APPs. The intelligent APPs include APPs such as face recognition, license plate recognition, etc., the non-intelligent APPs could be APPs for video streaming, transforming, etc. + +# **8 Requirements for software-defined camera** + +The SDC is a kind of IPU; there are two types of requirement for an IPU (refer to clause 6 in [ITU-T F.743.1]), one is the requirement of the premises unit (PU) (refer to clause 6.2.8 in [ITU-T H.626V2]) including video/audio acquisition, image snapshot, recording/playback, event and alarm, PTZ control, etc. The other one is the requirements of PIV which can be referred to clause 6 in [ITU-T F.743.1]. This clause focuses on those requirements which belong to software-defined technology of cameras, and not those requirements which belong to the IPU. + +## **8.1 User requirements** + +USR-001: The SDC is required to support users to view the information of the deployed applications, including name, version, status, license and other information of the application. + +USR-002: The SDC is required to support users to manage applications, including installation, activation, deactivation, upgrade and uninstallation of the application in the SDC. + +## **8.2 Service requirements** + +SRV-001: The SDC is required to support applications on-demand deployment and online upgrade without service interruption. + +SRV-002: The SDC is required to support the life cycle management of applications, including installation, uninstallation, and activation, deactivation, upgrade and license management. + +SRV-003: The SDC is required to support APP license verification and algorithm model file encryption to protect the intellectual property of the APP. + +SRV-004: The SDC is recommended to support sharing intelligent analysis services with other PUs. In order to complete this service sharing, the SDC shall support video acquisition from other PUs, decoding the video, completing the intelligent analysis and sending the analysis results to the CU. + +SRV-005: The SDC is recommended to embed a training model inside, and support automatic collection, cleaning and labelling the video or image data captured in the actual scene as a training set, and using these training sets to train the model and adjust the model parameter to complete the final optimization of the model. + +# Bibliography + +- [b-ITU-T H.Sup.1] Supplement 1 to the ITU-T H-series Recommendations (1999), *Application profile – Sign language and lip-reading real-time conversation using low bit rate video communication.* + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.74-199208-I_PDF-E/raw.md b/marked/F/T-REC-F.74-199208-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..1f55f16f4aa22e33e3a737549ea1530a6f5c5e8a --- /dev/null +++ b/marked/F/T-REC-F.74-199208-I_PDF-E/raw.md @@ -0,0 +1,111 @@ + + +![ITU logo: A globe with a lightning bolt and the letters ITU.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with a lightning bolt and the letters ITU. + +INTERNATIONAL TELECOMMUNICATION UNION + +**CCITT** + +**F.74** + +THE INTERNATIONAL +TELEGRAPH AND TELEPHONE +CONSULTATIVE COMMITTEE + +(08/92) + +**TELEGRAPH AND MOBILE SERVICES +OPERATIONS AND QUALITY OF SERVICE** + +--- + +**INTERMEDIATE STORAGE DEVICES +ACCESSED FROM THE INTERNATIONAL +TELEX SERVICE USING SINGLE STAGE +SELECTION – ANSWERBACK FORMAT** + +**Recommendation F.74** + +--- + +![CCITT logo: A globe with the letters CCITT.](84a1d09fb489061482111515543b60dc_img.jpg) + +CCITT logo: A globe with the letters CCITT. + +Geneva, 1992 + +# FOREWORD + +The CCITT (the International Telegraph and Telephone Consultative Committee) is a permanent organ of the International Telecommunication Union (ITU). CCITT is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The Plenary Assembly of CCITT which meets every four years, establishes the topics for study and approves Recommendations prepared by its Study Groups. The approval of Recommendations by the members of CCITT between Plenary Assemblies is covered by the procedure laid down in CCITT Resolution No. 2 (Melbourne, 1988). + +Recommendation F.74 was revised by Study Group I and was approved under the Resolution No. 2 procedure on the 4th of August 1992. + +--- + +# CCITT NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication Administration and a recognized private operating agency. + +# **INTERMEDIATE STORAGE DEVICES ACCESSED FROM THE INTERNATIONAL TELEX SERVICE USING SINGLE STAGE SELECTION – ANSWERBACK FORMAT** + +*(revised 1992)* + +# **1 Introduction** + +1.1 There is an increasing trend for mailbox and other intermediate storage devices to be connected accessed from the international telex service, with the ability to receive telex messages. + +1.2 Such devices are allocated network addresses that are part of the telex numbering plan thereby allowing customers of the international telex service to communicate with them using normal telex procedures. + +1.3 The message could remain in the storage device for some time awaiting forward transmission to or retrieval by the intended recipient. + +1.4 The sender of the message (i.e. a telex customer) may be unaware that the addressee is a storage device, and will assume that, providing the answerback is present and correct at both start and end of the local record, the recipient will receive the message without any positive action on his part. + +1.5 Such storage devices should be required to accept all calls delivered correctly by the international telex network, and to react promptly to ensure that ineffective usage of the international telex network is minimized. + +# **2 Definitions** + +## **2.1 intermediate storage device** + +An intermediate storage device is defined as any device which provides the storage of messages originating in the international telex service prior to onward transmission to or retrieval by the intended recipient. + +# **3 Scope** + +3.1 The provisions of this Recommendation apply to intermediate storage devices connected to a national telex network identified by individual telex addresses which are part of the national telex numbering plan. This Recommendation recognizes that currently implemented intermediate storage devices may not conform to these provisions. However, there would be advantages if existing equipment complied with the requirements of this Recommendation where possible, but that in any case new equipment should comply in order to maintain the quality of the international telex service. + +3.2 The provisions of this Recommendation have particular application within the interworking scenarios described in the F.80- and U.200-Series Recommendations. + +# **4 Call establishment to an intermediate storage device** + +4.1 Incoming calls to an intermediate storage device shall be handled in accordance with normal telex procedures, that is: + +- i) return of call connect in accordance with U-Series Recommendations; +- ii) return of answerback formatted in accordance with Figure 1/F.74 in response to a received **WRU** signal; +- iii) return of a clear confirmation signal in response to a received clear. + +# 5 Format of answerbacks for intermediate storage devices + +5.1 The answerbacks of intermediate storage devices shall be formatted in accordance with the rules given in Recommendation F.60; by in addition, shall contain combination 22 in figures (= sign) separating the alpha and figure parts of the answerback. The use of “=” in this way should be confined to the unique purpose of identifying such intermediate storage devices. + +- Figure-shift or (if required by the network) letter-shift +- Carriage-return +- Line-feed +- National telex number of the intermediate storage device or (if letter-shift is fitted in the first position) figure-shift followed by the national telex number of the intermediate storage device +- “=” Combination No. 22 (see Notes 1 and 2) +- Letter-shift +- Space (optional) +- Letters indicating as explicitly as possible the name of the individual customer +- Space +- One or two letters of the telex network identification code listed in Recommendation F.69 +- Letter shift (if required by the network) + +*Note 1* – It should be noted that some Administrations participating in the international telex service use combination 22 in the figure case (=) as the initial printing character of the answerback to indicate a bilingual terminal in its default Latin mode. + +*Note 2* – In networks which have in use answerback codes which are arranged in an order other than the preferred in Recommendation F.60, § 3.4.3, combination 22 in the figure case (=) (identifying an intermediate storage device) should be inserted immediately following the customer's number. + +FIGURE 1/F.74 + +### Answerback format \ No newline at end of file diff --git a/marked/F/T-REC-F.740-199308-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.740-199308-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..26d56ec823674592ecfa2511a33f2d20928ab1d6 --- /dev/null +++ b/marked/F/T-REC-F.740-199308-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c1994bce95d7014fedac3f2643fc5367febda657267df102b15e46470c9f73f8 +size 8195 diff --git a/marked/F/T-REC-F.740-199308-I_PDF-E/8e14350b4b669119a3bdfca7869110ca_img.jpg b/marked/F/T-REC-F.740-199308-I_PDF-E/8e14350b4b669119a3bdfca7869110ca_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..38b5c228d089237cdb84b59eae11b11f0139aab5 --- /dev/null +++ b/marked/F/T-REC-F.740-199308-I_PDF-E/8e14350b4b669119a3bdfca7869110ca_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:88c687aa4d59061dfdd91b57d96bf8bfebcb9af0871176bbd3b246f4d7dae98a +size 67509 diff --git a/marked/F/T-REC-F.740-199308-I_PDF-E/af7916c89a458fdab6c3f443217388ae_img.jpg b/marked/F/T-REC-F.740-199308-I_PDF-E/af7916c89a458fdab6c3f443217388ae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a39a8f22987bded7c2c3d2348688e1d8b4e3d7ca --- /dev/null +++ b/marked/F/T-REC-F.740-199308-I_PDF-E/af7916c89a458fdab6c3f443217388ae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8dfe78190d15dcc2dbef35e5433639f0488540c3d8d1f66bbad32a784a84a035 +size 62396 diff --git a/marked/F/T-REC-F.740-199308-I_PDF-E/e23c9577e177bf077bac8951a775dd90_img.jpg b/marked/F/T-REC-F.740-199308-I_PDF-E/e23c9577e177bf077bac8951a775dd90_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..8bfd9e5b729f2365f4106c74a36f5dc80363a33f --- /dev/null +++ b/marked/F/T-REC-F.740-199308-I_PDF-E/e23c9577e177bf077bac8951a775dd90_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:249962d5cfe8f6ab2db73b4c9eedd776ff35760a7842e766418136ad728a8280 +size 28120 diff --git a/marked/F/T-REC-F.740.1-201911-I_PDF-E/2ee59e629035d641140e55f4d215b0d7_img.jpg b/marked/F/T-REC-F.740.1-201911-I_PDF-E/2ee59e629035d641140e55f4d215b0d7_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..93b9c2128d0f78369397a2059731e0bd37b3cb6f --- /dev/null +++ b/marked/F/T-REC-F.740.1-201911-I_PDF-E/2ee59e629035d641140e55f4d215b0d7_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:924b89c95f2db970eaffbde2bea3144d99502bd946bfbaf2e8471adbb560a860 +size 81562 diff --git a/marked/F/T-REC-F.740.1-201911-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.740.1-201911-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9002ee68e2b9ebf63d27d39ba8c73bdafb0e6be6 --- /dev/null +++ b/marked/F/T-REC-F.740.1-201911-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8df701536c25ab8246d2868de3c20c8a4e25168bb52c5eb9078278ab1b5b4fd8 +size 6859 diff --git a/marked/F/T-REC-F.740.1-201911-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg b/marked/F/T-REC-F.740.1-201911-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..b3f2d1f66f078e073af2c829c3ccbf79e69f99c1 --- /dev/null +++ b/marked/F/T-REC-F.740.1-201911-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:363aa88ebdc0206c41852c8d3d2e5fc4f9afd14f8789b6b624ff75a543ca46e3 +size 85054 diff --git a/marked/F/T-REC-F.740.10-202503-I_PDF-E/562f471e8153729557e6a4ee6343c32c_img.jpg b/marked/F/T-REC-F.740.10-202503-I_PDF-E/562f471e8153729557e6a4ee6343c32c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ee5686925280ec09bec57e6d83ecfcb8ace88ea6 --- /dev/null +++ b/marked/F/T-REC-F.740.10-202503-I_PDF-E/562f471e8153729557e6a4ee6343c32c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4632c69e5e83b9ad63ad5a77780b41c83c412649b6dc9f3a73ef95a3ac49d900 +size 51732 diff --git a/marked/F/T-REC-F.740.10-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.740.10-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ac295b1b0c7f14b1c5a33ef6682b1b017c812477 --- /dev/null +++ b/marked/F/T-REC-F.740.10-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37aed4fc114a39843438175326e0b7e1d09a9d6075723d82fea32040fd560ee4 +size 7192 diff --git a/marked/F/T-REC-F.740.10-202503-I_PDF-E/cfda9df1319e04207eb28bcefd1dab7b_img.jpg b/marked/F/T-REC-F.740.10-202503-I_PDF-E/cfda9df1319e04207eb28bcefd1dab7b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..8be55116b182fdf5888051e6c197bb2be23e8ce4 --- /dev/null +++ b/marked/F/T-REC-F.740.10-202503-I_PDF-E/cfda9df1319e04207eb28bcefd1dab7b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:63641d216490c7c9a044dfd8e84e03628bd0227e5c3354e0829a974f3610c6ec +size 35779 diff --git a/marked/F/T-REC-F.740.10-202503-I_PDF-E/raw.md b/marked/F/T-REC-F.740.10-202503-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..12fc8c8607baaa5ea250793cf111cee352c10f67 --- /dev/null +++ b/marked/F/T-REC-F.740.10-202503-I_PDF-E/raw.md @@ -0,0 +1,401 @@ + + +# Recommendation + +# **ITU-T F.740.10 (03/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +## --- **Requirements and procedure for cultural data annotation** + +![ITU logo](84a1d09fb489061482111515543b60dc_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.740.10 + +# Requirements and procedure for cultural data annotation + +## Summary + +In recent times, a large number of cultural resources have been digitized. During the utilization of these resources, the industry has found that the extent to which they are used often depends on whether they have detailed annotations. Annotation involves many research areas in the field of information and communication technology (ICT) such as computer vision, natural language processing and big data. Annotated resources are more easily retrievable, can be better incorporated into knowledge graphs, and are more comprehensible to machines, thus resulting in higher usage rates. Therefore, annotating cultural data is a prerequisite for its full utilization. + +Recommendation ITU-T F.740.10 aims to standardize the annotation of cultural data, ensuring that it is conducted in a manner that is both efficient and accurate. By establishing a clear procedure and its requirements, the document seeks to promote data exchange and enhance the overall utility of cultural data in various applications, and provide users with a better experience of using cultural data. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.740.10 | 2025-03-16 | 21 | 11.1002/1000/16223 | + +## Keywords + +Annotation, cultural data. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2025 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|---------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Overview ..... | 2 | +| 7 Semantic annotation model..... | 3 | +| 7.1 Core tags..... | 3 | +| 7.2 Professional tags..... | 3 | +| 7.3 Public tags ..... | 3 | +| 8 Annotation procedure ..... | 4 | +| 9 Requirements for annotation activities ..... | 4 | +| 9.1 Requirements for data collection activity..... | 4 | +| 9.2 Requirements for data processing and screening activity ..... | 5 | +| 9.3 Requirements for annotation scheme design activity..... | 6 | +| 9.4 Requirements for annotation tool selection activity ..... | 6 | +| 9.5 Requirements for annotator recruitment activity..... | 6 | +| 9.6 Requirements for annotation result output activity ..... | 7 | +| 9.7 Requirements for annotation procedure management activity ..... | 9 | +| Bibliography..... | 10 | + + + +## Requirements and procedure for cultural data annotation + +# 1 Scope + +This Recommendation specifies a procedure for cultural data annotation, which includes seven activities: data collection activity, data processing and screening activity, annotation scheme design activity, annotation tool selection activity, annotator recruitment activity, annotation result output activity and annotation procedure management activity. This document specifies requirements that should be fulfilled at each activity. + +This Recommendation targets cultural data existing in forms such as text, images, audio, video and 3D models. It carries out operations such as classification, recognition, segmentation and boundary delineation on its inherent constituent elements, and performs semantic annotations to form specialized datasets. These datasets are used for machine learning, semantic searches and data visualization related to cultural applications. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +# 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 annotate** [b-ISO 24612-1]: Process of adding linguistic information to electronic representation of data. + +**3.1.2 annotation** [b-ISO 24612-1]: Linguistic information added to electronic representation of data, independent of its representation. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 cultural data:** Information and statistical data used to describe, analyse and understand cultural phenomena, behaviours and products. + +**3.2.2 cultural element:** The fundamental functional unit of culture and which possesses a unique cultural connotation, composability, diversity and uncertainty. + +**3.2.3 semantic tag:** Semantic tags are based on the cultural attribute space and represent the essential attributes of cultural data in a multidimensional form. + +NOTE – To this end, defining semantic tagging is the process of annotating the concept categories, attributes, instances and other metadata of target resources. A typical semantic annotation process involves inputting the document to be annotated and the ontology into the annotation system. The annotation system obtains semantic + +information from the ontology and adds semantic information to the target document, ultimately outputting the annotated document. + +# **4 Abbreviations and acronyms** + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|--------------------------------| +| AVI | Audio Video Interleave | +| FLAC | Free Lossless Audio Codec | +| JSON | JavaScript Object Notation | +| MP4 | Moving Picture Experts Group-4 | +| PCM | Pulse-Code Modulation | +| PNG | Portable Network Graphic | +| RGB | Red, Green, Blue | +| TIF | Tag Image File Format | +| WAV | Waveform Audio File Format | + +# **5 Conventions** + +In this Recommendation: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted, if conformance to this Recommendation is to be claimed. +- The keywords "is recommended to" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. + +# **6 Overview** + +World culture has a long and profound history, with a diverse range of cultural relics and rich connotations. The development of digitization and artificial intelligence poses higher demands on the standards and procedure for cultural data annotation. There is an urgent need to address the management, sharing and interconnectivity of cultural data from different regions. Through artificial intelligence and digital technology the deeper meanings within cultural data can be unearthed, and a professional knowledge base for annotations can be established. + +In order to facilitate the circulation of cultural data, it is necessary to annotate this data as much as possible. Data annotation is also a key link for most deep learning algorithms to operate effectively. Annotation technology involves many research areas such as computer vision, natural language processing, pattern recognition and knowledge graphs. From the perspective of requirements for digital preservation and sharing of cultural relics, a large amount of existing cultural relic data needs to refine and organize related concepts and relationships to achieve the purpose of knowledge sharing and reuse. Therefore, it is necessary to mine the potential concepts in cultural relic neglect from a semantic perspective, so that the semantics of cultural data can be understood by both humans and computers. This will realize the semantic retrieval of cultural relic data and knowledge graphs, improve the ability and efficiency of computer retrieval models, provide technical support for the visualization of cultural data, and provide a more convenient way of experiencing cultural transmission and inheritance. + +# 7 Semantic annotation model + +To assist various roles in achieving more efficient annotation, it is recommended to use a three-tier semantic annotation model based on the "Core Tag – Professional Tag – Public Tag", as shown in Figure 1. + +![Figure 1: Illustration of the three-tier semantic annotation model. The diagram shows three concentric circles representing the tiers: Core tags (innermost), Professional tags (middle), and Public tags (outermost). The Core tags are divided into three segments: Advanced tags {Type, location, class, ...}, Extended tags {Origin, meaning, style, ...}, and Basic tags {Name, ethnicity, religion, ...}. Lines connect these segments to their respective domain attributes: Core domain attributes {Name, type, meaning, ...}, Professional domain attributes {Expert, field, resource, ...}, and Public domain attributes {Tag set, user group, source, and utility degree, ...}. The label F.740.10(25) is at the bottom right.](562f471e8153729557e6a4ee6343c32c_img.jpg) + +The diagram illustrates the three-tier semantic annotation model as three concentric circles. The innermost circle is blue and labeled "Core tags" at the top. It is divided into three segments: "Advanced tags {Type, location, class, ...}" on the left, "Extended tags {Origin, meaning, style, ...}" on the right, and "Basic tags {Name, ethnicity, religion, ...}" at the bottom. The middle circle is green and labeled "Professional tags" at the bottom. The outermost circle is orange and labeled "Public tags" at the bottom. To the right of the circles, three labels with lines pointing to the corresponding tiers are: "Core domain attributes {Name, type, meaning, ...}" pointing to the inner circle, "Professional domain attributes {Expert, field, resource, ...}" pointing to the middle circle, and "Public domain attributes {Tag set, user group, source, and utility degree, ...}" pointing to the outer circle. The label F.740.10(25) is at the bottom right of the diagram. + +Figure 1: Illustration of the three-tier semantic annotation model. The diagram shows three concentric circles representing the tiers: Core tags (innermost), Professional tags (middle), and Public tags (outermost). The Core tags are divided into three segments: Advanced tags {Type, location, class, ...}, Extended tags {Origin, meaning, style, ...}, and Basic tags {Name, ethnicity, religion, ...}. Lines connect these segments to their respective domain attributes: Core domain attributes {Name, type, meaning, ...}, Professional domain attributes {Expert, field, resource, ...}, and Public domain attributes {Tag set, user group, source, and utility degree, ...}. The label F.740.10(25) is at the bottom right. + +Figure 1 – Illustration of the three-tier semantic annotation model + +### 7.1 Core tags + +The core tags reflect the connotative description of tags, which includes fields such as time, origin, meaning, class, language, religion, style, region, name, type, location, ethnicity, etc. + +Based on the ownership and professionalism of the tag fields, the core tags are further divided into basic tags, advanced tags and extended tags. + +Basic tags refer to the essential tags that are indispensable during tag collection, describing the fundamental attributes of the annotated entity, including name, ethnicity, religion and region. + +Advanced tags are related tags that build upon basic tags to provide explanations or detailed descriptions, including type, location, time and class. + +Extended tags are a series of tags that supplement the annotated tags, specifically including origin, meaning and style. + +### 7.2 Professional tags + +Professional tags require domain-specific knowledge; therefore, they are generated through a stage of expert review and validation, and they include (but are not limited to): expert, field, scenario and resource. + +### 7.3 Public tags + +Public tags refer to attributes provided by the general public; they include, but are not limited to, tag set, user group, source and level of usage. Among these, the level of usage is calculated based on specific tagged objects (resources) and integrates multiple sets of evaluation indicators. This approach can better reflect the importance of tags in collaborative tagging than simple frequency. + +# 8 Annotation procedure + +The annotation procedure, carried out by the annotation team, mainly includes seven activities: data collection activity, data processing and screening activity, annotation scheme design activity, annotation tool selection activity, annotator recruitment activity, annotation result output activity and annotation procedure management activity. The specific content and interrelationships of each activity are as shown in Figure 2. + +![Flowchart of the annotation procedure showing seven activities in a linear sequence with a feedback loop from the final activity back to the first.](cfda9df1319e04207eb28bcefd1dab7b_img.jpg) + +``` +graph LR; A[Data collection activity] --> B[Data processing and screening activity]; B --> C[Annotation scheme design activity]; C --> D[Annotation tool selection activity]; D --> E[Annotator recruitment activity]; E --> F[Annotation result output activity]; F --> G[Annotation procedure management activity]; G --> A; +``` + +F.740.10(25) + +Flowchart of the annotation procedure showing seven activities in a linear sequence with a feedback loop from the final activity back to the first. + +**Figure 2 – Content and interrelationships of each annotation activity** + +**Data collection activity:** The annotation team collects cultural data, including images, audio, videos and text. + +**Data processing and screening activity:** The annotation team processes and filters the collected data, including deduplication, categorization, editing, etc. + +**Annotation scheme design activity:** The annotation team designs reasonable annotation methods and schemes based on the characteristics and requirements of cultural data, determining annotation elements and rules. + +**Annotation tool selection activity:** The annotation team chooses suitable annotation tools according to the requirements of the annotation scheme, such as annotation platforms and software. + +**Annotator recruitment activity:** The annotation team recruits suitable annotators according to the scale and difficulty of the annotation work, requiring annotators to possess professional knowledge and experience in related fields. + +**Annotation result output activity:** The annotation team produces annotation results in a standard format as output for use in subsequent programs or systems, such as machine learning and natural language processing. + +**Annotation procedure management activity:** The annotation team controls the quality of the annotation procedure management, including supervising the work status and results of annotators, promptly detecting and correcting annotation errors, and providing data analysis for improving annotation quality. + +# 9 Requirements for annotation activities + +### 9.1 Requirements for data collection activity + +Cultural data is recommended to include, but is not limited to, text, image, audio, video and 3D model data. The cultural data to be annotated and collected by the annotation team is recommended to not be less than the following requirements: + +- a) Image format cultural data is recommended to use lossless image formats, such as TIF, PNG, etc. +- b) Image resolution is required to be at least 72 dpi, RGB colour mode is recommended to be greater than 8-bit or 16-bit colour depth, and size is recommended to be at least $256 \times 256$ pixels. Images are recommended to ensure the integrity of the appearance, shape and + +structure of the object, with clear texture, balanced colours, distortion-free cultural elements, and accurate descriptions of distinctive content, such as text, lines, colours, shapes and techniques. + +- c) The position of the text object in an image is recommended to be precisely defined by specifying the coordinates of their bounding box's four vertices. Text format cultural data is recommended to not have meaningless spaces in paragraphs and recommended to be in text or JSON format. +- d) Audio format cultural data is recommended to use lossless audio formats, such as WAV, FLAC, etc., with a sampling rate of at least 44.1 kHz, PCM encoding and 24-bit or higher bit depth. Audio files are recommended to have a minimum duration of 3 seconds, ensure the integrity of the original sound, clear acoustic characteristics, balanced frequencies, distortion-free cultural elements, and accurate representations of distinctive content, such as speech, musical instruments, environmental sounds and recording techniques. +- e) Video format cultural data is recommended to meet the following requirements: It is recommended to use universal video formats (either compressed or uncompressed) such as MP4, AVI, etc. The integrity of the video is recommended to be maintained, and the video files are recommended to have a minimum duration of 3 seconds. The minimum video frame resolution is recommended to not be less than $320 \times 240$ , and the frame rate is recommended to not be less than 24 fps; the data is recommended to be stored using RGB colour mode and to use 8-bit or 16-bit colour storage; the video is recommended to fully record the complete actions of the characters or the complete appearance, shape and structure of the objects, with clear images, balanced colours and no distortion of the cultural elements' perspective. It is recommended to accurately and clearly record the iconic content such as actions and appearance. Videos are recommended to have basic metadata and audio. +- f) 3D model format cultural data is recommended to meet the following requirements: It is recommended to use high-precision 3D formats, such as point clouds and meshes. For point cloud data, the point cloud density is recommended to be at least 100 points per square metre, or the distance between points is recommended to be less than 0.1 metres; for mesh data, the number of faces is recommended to be more than 50 000. 3D models are recommended to ensure the integrity of the object's appearance, shape, structure, featuring clear textures, high-definition mappings and high-precision models, as well as accurate descriptions of unique content such as points, lines, meshes, shapes, normal vectors and the mapping between points and textures. In addition, when collecting cultural data, the model's reference coordinate system is recommended to be defined, which is very important for subsequent positioning, stitching, displaying and analysing. If the cultural relic has complex internal structures or multilayer constructions, the relationships between layers and the positions of independent components are recommended to be identified. + +## 9.2 Requirements for data processing and screening activity + +Data processing and screening activity carried out by the annotation team is recommended to meet the following requirements: + +- a) Data processing: The annotation team is recommended to perform desensitization, deduplication, normalization and formatting on the collected cultural data to ensure data tidiness, completeness and operability, making it suitable for subsequent annotation work. +- b) Data screening: The annotation team is recommended to filter the processed data based on the annotation scheme and quality requirements, removing unrelated or non-compliant data, leaving representative data that fully covers cultural information characteristics, reducing the difficulty and complexity of the annotation work. + +## **9.3 Requirements for annotation scheme design activity** + +Annotation design schemes carried out by the annotation team is recommended to meet the following requirements: + +- a) Annotation purpose and basis: Define the purpose and requirements of annotation, explain the basis and target of annotation. +- b) Annotation object and scope: Determine the object and scope of annotation, such as a particular category of cultural resources, a specific period, etc. +- c) Annotation elements and rules: Design annotation elements, rules and methods, defining the content, format and standards of annotation, such as tags, keywords, properties, etc. +- d) Annotation quality assessment: Develop annotation quality assessment indicators and standards, considering the impact of various factors on annotation results, such as annotator factors, data quality, annotation rules, etc. +- e) Annotation management: Plan the annotation procedure, establish quantification and management mechanisms for annotation workload, clarify tasks, time and responsible personnel for each step. +- f) Annotation implementation plan: Develop corresponding annotation implementation plans and schedules based on the specific requirements and actual conditions of the annotation scheme. + +### **9.4 Requirements for annotation tool selection activity** + +The selection of annotation tools carried out by the annotation team is recommended to meet the following requirements: + +- a) The annotation team is recommended to choose suitable annotation tools and software according to the annotation requirements and the characteristics of the annotation object. It is recommended to first have domain experts construct core category labels, and then use algorithms or models to iteratively optimize these labels. +- b) Evaluate whether the functions and features of annotation tools meet specified annotation requirements and standards, such as whether they support multi-person collaboration, multiple data formats, data visualization analysis functions, etc. +- c) Test and verify selected annotation tools, comparing their advantages and disadvantages and finding the most suitable annotation tool as much as possible. +- d) Determine the necessary preparations before using the annotation tools, such as installation, configuration, calibration, etc. +- e) The annotation tools are recommended to provide permission management capabilities, including account creation, authorization management, permission approval and role configuration. +- f) Train and guide annotators on the selected annotation tools to familiarize and master the usage. +- g) Develop corresponding countermeasures and solutions for possible problems and abnormal situations in the annotation procedure. +- h) Manage and track the version and upgrades of annotation tools to ensure their stability and reliability. +- i) Pay attention to copyright and licensing issues of annotation tools to ensure legal use. + +## **9.5 Requirements for annotator recruitment activity** + +Annotator recruitment carried out by the annotation team is recommended to include the following requirements: + +- a) The annotation team is recommended to determine the number and professional background of annotators according to the requirements of the annotation object and task, specifying the necessary skills and qualities. +- b) Develop annotator recruitment plans, including recruitment announcements, methods and channels. +- c) In the recruitment announcement, clarify the time, scope and compensation of annotation tasks, and require applicants to submit relevant supporting materials or complete specific test tasks. +- d) Conduct review of qualifications and interview tests for applicants, selecting suitable annotators. +- e) Determine the hiring method and work responsibility letter for annotators, clarifying their rights, obligations and work requirements. +- f) Train and guide annotators to familiarize them with annotation requirements, rules, tools and processes. +- g) Manage and assess annotators' performance, detecting and addressing problems and deficiencies during the annotation procedure in time. +- h) Establish confidentiality mechanisms and responsibility systems for annotators, strictly protecting the security and confidentiality of cultural data. + +## **9.6 Requirements for annotation result output activity** + +Annotation result output carried out by the annotation team is recommended to meet the following requirements: + +- a) Requirement for text format data annotation output + - i) The result output after text annotation is recommended to match the actual content of the text. + - ii) The results of text annotation obtained by the same algorithm performing the same operation on the same text are recommended to be stable. + - iii) Except when the text undergoes substantial changes, the text part of the output results is recommended to remain consistent with its state before the change. + - iv) The evaluation method for annotated data is recommended to balance subjective perception with objective metrics. Subjective evaluation indicators include: content accuracy, method stability, interaction experience, system durability, etc. Objective evaluation uses mathematical models to provide quantitative values for text annotation quality, serving as a primary tool for assessing the performance of annotation models. Many metrics cannot be directly observed by the human eye; instead, they are calculated through quantitative indicators to evaluate the algorithm's strengths and weaknesses and used as feedback for further optimizing the algorithm model. +- b) Requirement for image format data annotation output + - i) The result output after image annotation is recommended to match the actual content of the image. + - ii) The results of image annotation obtained by the same algorithm performing the same operation on the same image are recommended to be stable. + - iii) Except for substantial changes in the image content, when other changes occur to the image, the output results are recommended to remain consistent with those before the change. + - iv) The evaluation method for annotated data is recommended to balance subjective perception with objective metrics. Subjective evaluation indicators include: content + +accuracy, method stability, interaction experience and system durability, among others. Objective evaluation uses mathematical models to quantify the quality of pattern annotations, serving as a primary tool for assessing the performance of annotation models. Many of these metrics cannot be directly observed by the human eye; instead, they are calculated through quantitative indicators to evaluate the strengths and weaknesses of algorithms, and are used as feedback for further optimizing algorithm models. + +- c) Requirement for audio format data annotation output + - i) The result output after audio annotation is recommended to match the actual content of the audio. + - ii) The results of audio annotation obtained by the same algorithm performing the same operation on the same audio are recommended to be stable. + - iii) Except for substantial changes in the audio content, when other changes occur to the audio (editing, volume, quality), the output results are recommended to remain consistent with those before the change. + - iv) The evaluation method for annotated data is recommended to balance subjective perception with objective metrics. Subjective evaluation indicators include content accuracy, method stability, interaction experience, system durability, etc. Objective evaluation uses mathematical models to provide quantitative values for the quality of annotations related to the audio's region, genre and style, serving as a primary tool for assessing annotation model performance. Many metrics cannot be directly observed by the human ear; instead, they are calculated through quantitative indicators to evaluate the algorithm's strengths and weaknesses and used as feedback for further optimizing the algorithm model. +- d) Requirement for video format data annotation output + - i) The result output after video annotation is recommended to match the actual content of the video. + - ii) The results of video annotation obtained by the same algorithm performing the same operation on the same video and its frames are recommended to be stable. + - iii) Except for substantial changes in the video content, when other changes occur to the video, the output results are recommended to remain consistent with those before the change. + - iv) The evaluation method for annotated data is recommended to balance subjective perception with objective metrics. Subjective evaluation indicators include content accuracy, method stability, interaction experience, system durability, etc. Objective evaluation uses mathematical models to provide quantitative values for video annotation quality, serving as a primary tool for assessing annotation model performance. Many metrics cannot be directly observed by the human eye; instead, they are calculated through quantitative indicators to evaluate the algorithm's strengths and weaknesses and used as feedback for further optimizing the algorithm model. +- e) Requirement for 3D model format data annotation output + - i) The result output after 3D model annotation is recommended to match with the actual content of the 3D model. + - ii) Except for 3D model reconstruction or deformation processing, when other changes occur to the 3D model, the output results are recommended to remain consistent with those before the change. + - iii) For complex cultural relics with multiple layers, the relationships between layers and the positions of independent components are recommended to be identified. The types of + +materials used in the relics, such as bronze, ceramics, wood and stone, are recommended to also be indicated. + +- iv) The evaluation method for annotated data is recommended to assess the discrepancy between real data and measured data. Measured data is recommended to be obtained by averaging multiple measurements to achieve the final measurement result. Given the significant potential error in subjective human estimates, which cannot be directly observed by the human eye, it is recommended to use high-precision measuring tools to measure real data in the real world for obtaining measurements. + +### **9.7 Requirements for annotation procedure management activity** + +The annotation procedure management activity carried out by the annotation team is recommended to include the following requirements: + +- a) The annotation team is recommended to develop an annotation work quality management plan, clarify annotation quality objectives and standards, and develop corresponding quality management procedures and methods. +- b) Define the responsibilities and duties of annotation personnel, clarify the quality control points and quality inspection points of each step ensuring the controllability and reliability of annotation quality. +- c) Develop quality standards and assessment methods for annotation data, classify, screen and correct annotation data, promptly detecting and correcting annotation errors and problems. The evaluation of annotation results includes subjective and objective methods, and it is recommended to combine both types. Subjective evaluation methods assess annotation results based on human perception. It is recommended to use selected indicators and score the results manually using a scoring sheet. After scoring, a weighted average ranking is applied (weights are determined based on annotation needs and range from 0 to 1). The formula used is the average after removing the lowest and highest scores. For objective evaluation, it is suggested to use mathematical models to provide a quantified value of the annotation quality of patterns. Objective evaluation serves as the primary tool for assessing the performance of annotation models. Many indicators used are not directly observable by the human eye. Here, quantitative indicators are calculated to determine the strengths and weaknesses of algorithms and are used as feedback to further optimize the algorithm models. +- d) Determine the quality checks and quality assurance measures required during the annotation procedure, such as multi-person verification, repeated annotations, mutual checks, etc. +- e) Establish a dedicated quality assurance department staffed with experienced quality assurance personnel, independent of the annotation project manager, ensuring the independence and objectivity of annotation quality. +- f) Record and track problems and abnormal situations discovered during the annotation procedure, developing corresponding solutions to prevent similar problems from recurring. +- g) Statistically analyse the results of the annotation procedure, evaluating the stability and reliability of annotation quality, providing a robust data basis for improving annotation quality. + +## Bibliography + +- [b-ISO 24612-1] ISO 24612-1:2012, *Language resource management – Linguistic annotation framework (LAF)*. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.740.11-202503-I_PDF-E/04f51626e2e10a16e3eb2c4b33cb2742_img.jpg b/marked/F/T-REC-F.740.11-202503-I_PDF-E/04f51626e2e10a16e3eb2c4b33cb2742_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..e3691d63a49952542223807fa048791c326e48de --- /dev/null +++ b/marked/F/T-REC-F.740.11-202503-I_PDF-E/04f51626e2e10a16e3eb2c4b33cb2742_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6138f36d7460311db84e804b17867828aea9ea86d931a5d8d9ca15434285d333 +size 53571 diff --git a/marked/F/T-REC-F.740.11-202503-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/F/T-REC-F.740.11-202503-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ac295b1b0c7f14b1c5a33ef6682b1b017c812477 --- /dev/null +++ b/marked/F/T-REC-F.740.11-202503-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37aed4fc114a39843438175326e0b7e1d09a9d6075723d82fea32040fd560ee4 +size 7192 diff --git a/marked/F/T-REC-F.740.11-202503-I_PDF-E/2eb23c2210154279f8013a1594fbcc5a_img.jpg b/marked/F/T-REC-F.740.11-202503-I_PDF-E/2eb23c2210154279f8013a1594fbcc5a_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..e4bd5ccbcaaa3746def86445ea6bf0407c1c2ac7 --- /dev/null +++ b/marked/F/T-REC-F.740.11-202503-I_PDF-E/2eb23c2210154279f8013a1594fbcc5a_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:33f58eabd7f9ec26317e26df0ea2b9d65b1658f358cc28566baa4880fd866719 +size 44430 diff --git a/marked/F/T-REC-F.740.11-202503-I_PDF-E/8e14350b4b669119a3bdfca7869110ca_img.jpg b/marked/F/T-REC-F.740.11-202503-I_PDF-E/8e14350b4b669119a3bdfca7869110ca_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9525aa2a3e5e9800849835f7d61aa13173f1e642 --- /dev/null +++ b/marked/F/T-REC-F.740.11-202503-I_PDF-E/8e14350b4b669119a3bdfca7869110ca_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4cff011fd28d9ff8c869fd4c311d59fc61d33c23872f9a504cee794a998b384b +size 61827 diff --git a/marked/F/T-REC-F.740.11-202503-I_PDF-E/a24e89a6fe9bb70c83f8bf5202baba95_img.jpg b/marked/F/T-REC-F.740.11-202503-I_PDF-E/a24e89a6fe9bb70c83f8bf5202baba95_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..bea6b0dd55f72d25221f703fb818936a027d0431 --- /dev/null +++ b/marked/F/T-REC-F.740.11-202503-I_PDF-E/a24e89a6fe9bb70c83f8bf5202baba95_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9b60e32b63c086036064a8c5985f49f20178d2873dd7f64a6fb25d4799c6ed77 +size 71522 diff --git a/marked/F/T-REC-F.740.11-202503-I_PDF-E/raw.md b/marked/F/T-REC-F.740.11-202503-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..8fd69b107ec1afe4a98da8911e1042c5fc86fc59 --- /dev/null +++ b/marked/F/T-REC-F.740.11-202503-I_PDF-E/raw.md @@ -0,0 +1,688 @@ + + +# Recommendation **ITU-T F.740.11 (03/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +## --- **Requirements and framework of cloud-based augmented reality systems** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.740.11 + +# Requirements and framework of cloud-based augmented reality systems + +## Summary + +Based on cloud computing capabilities and massive data collaborative processing, cloud-based augmented reality systems can effectively support large-scale physical scene recognition and digital processing, provide more flexible virtual-scene integration and support multiperson interactive scenarios. Recommendation ITU-T F.740.11 does not change cloud computing's functions, but rather presents a basic concept of cloud-based augmented reality and specifies requirements and a framework of cloud-based augmented reality systems. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.740.11 | 2025-03-16 | 21 | 11.1002/1000/16224 | + +## Keywords + +Augmented reality, cloud-based augmented reality systems, framework, requirements. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2025 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|------------|------------------------------------------------------------------------------|----| +| 1 | Scope ..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 1 | +| 4 | Abbreviations and acronyms ..... | 1 | +| 5 | Conventions ..... | 2 | +| 6 | Overview ..... | 2 | +| 7 | Modes of cloud-based AR systems..... | 3 | +| 7.1 | Cloud-based AR with the integration of cloud and terminal capabilities ..... | 3 | +| 7.2 | Cloud-based AR relying on cloud processing capabilities..... | 5 | +| 8 | Framework of cloud-based augmented reality systems..... | 5 | +| 9 | Requirements of cloud-based augmented reality systems ..... | 6 | +| 9.1 | General requirements..... | 6 | +| 9.2 | Content layer requirements..... | 7 | +| 9.3 | Control layer requirements ..... | 7 | +| 9.4 | Resource layer requirements ..... | 10 | +| 9.5 | Terminal layer requirements..... | 10 | +| 9.6 | Network layer requirements ..... | 11 | +| 9.7 | OA&M requirements..... | 11 | +| 9.8 | Security requirements..... | 12 | +| Appendix I | – Use cases..... | 13 | +| I.1 | Cloud-based AR medical assistance..... | 13 | +| I.2 | Cloud-based AR navigation ..... | 14 | +| I.3 | Cloud-based AR industrial maintenance ..... | 15 | +| | Bibliography..... | 17 | + + + +# Recommendation ITU-T F.740.11 + +## Requirements and framework of cloud-based augmented reality systems + +# 1 Scope + +This Recommendation specifies requirements and a framework of cloud-based augmented reality (AR) systems. The scope of this Recommendation includes: + +- Overview; +- Modes of cloud-based augmented reality systems; +- Framework of cloud-based augmented reality systems; +- Requirements of cloud-based augmented reality systems. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T G.1036] Recommendation ITU-T G.1036 (2022), *Quality of experience influencing factors for augmented reality services*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 augmented reality** [b-ITU-T J.301]: A type of mixed reality where graphical elements are integrated into the real world in order to enhance user experience and enrich information. + +**3.1.2 augmented content** [b-ITU-T J.301]: A binary object, such as 2D images, 3D animated models or audio/video streaming files, to be augmented into a predefined augmentation region. + +**3.1.3 application** [b-ITU-T Y.101]: A structured set of capabilities, which provide value-added functionality supported by one or more services. + +## 3.2 Terms defined in this Recommendation + +None. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|----|-------------------------| +| 2D | Two Dimensional | +| 3D | Three Dimensional | +| AI | Artificial Intelligence | +| AR | Augmented Reality | + +| | | +|------|------------------------------------------| +| AVI | Audio Video Interleaved | +| CDN | Content Delivery Network | +| CPU | Central Processing Unit | +| DoF | Degree of Freedom | +| E2E | End-to-End | +| FOV | Field of View | +| FTTH | Fibre To The Home | +| GPU | Graphic Processing Unit | +| NIC | Network Interface Card | +| OA&M | Operation, Administration and Management | +| SDK | Software Development Kit | +| SLAM | Simultaneous Localization And Mapping | +| WDM | Wavelength Division Multiplexing | + +# 5 Conventions + +In this Recommendation: + +- The keywords "**is required**" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The keywords "**is recommended**" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement needs not be present to claim conformance. +- The keywords "**can optionally**" indicate an optional requirement which is permissible, without implying any sense of being recommended. These terms are not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +# 6 Overview + +Cloud-based AR and terminal-based AR are two different operation modes of AR. Terminal-based AR is the AR operation mode which adopts local content data storage and locally available running resources of the terminal to provide AR services. Terminal-based AR does not involve the retrieval and scheduling of content data and running resources in the cloud. The operation of terminal-based AR could be severely restrained by local running resource limitations, such as terminal computing power, battery and data storage. + +Cloud-based AR, based on the super computing power of the cloud, completes complex real-scene understanding, 3D modelling, multi-terminal data sharing, cloud-edge collaboration and other processing, builds a continuous combination of point cloud map and real-world coordinates, realizes real-time and updated 3D digital world model for real-world scanning, and provides the terminal with the ability to index and integrate virtual information. + +Cloud-based AR requires the collaborative provision of services based on various systems, collectively referred to as cloud-based AR systems. The E2E service of the cloud-based AR system refers to the subsystems running on the cloud, which distributes cloud-based AR services to terminals by relying on network capabilities and obtains remote operation instructions from a terminal. The cloud-based AR subsystems running on the cloud include a content system, control system, resource system, operation, administration and management (OA&M) system and security system. + +Cloud-based AR systems require the real-time collection of multiterminal collaboration, the low latency transmission of network uplink, the capabilities of cloud-edge collaboration and the integration capabilities of various multimedia processing, the efficient distribution of services/content and the control and coordination of various capabilities. + +Compared with the terminal-based AR mode, cloud-based AR systems have the characteristics of low terminal capacity requirements, multiperson interactive sharing and integrated AI processing capabilities. + +- The low terminal capacity requirements are mainly manifested in the completion of scene understanding, three-dimensional modelling and other processing through the cloud, and the use of multimedia streaming delivery and other modes, which can reduce the processing pressure of the terminal on logic operations and graphics rendering, thereby reducing the intensity of terminal operations and extending the battery life of terminals. +- Multiperson interactive sharing is mainly manifested in the digital processing of super-large scenes and multiperson data sharing. For the processing of ultralarge scenarios, based on the cloud-edge collaborative processing mechanism, the multiterminal crowdsourcing collection mode is adopted to obtain a wide range of map information and realize the digital processing of ultralarge outdoor scenes through intelligent understanding, resource aggregation and repositioning in the cloud. For multiperson data sharing, based on the cloud-unified large-scale real-world digital processing results and super virtual world construction capabilities, it provides high-end AR multiperson interactive services for terminals with basic functions such as display. +- The integrated AI processing capabilities, and first of all, the AI model training, need the super-scale computing power support of cloud response, and accurate recognition of real scenes also needs cloud capabilities. Therefore, the cloud-based AR can effectively integrate AI and other capabilities and improve the quality of AR user experience. + +# 7 Modes of cloud-based AR systems + +Cloud-based AR, based on the different tasks undertaken by the cloud and terminal, can be divided into two modes: cloud-based AR with the integration of cloud and terminal capabilities and cloud-based AR relying on cloud processing capabilities. + +## 7.1 Cloud-based AR with the integration of cloud and terminal capabilities + +Cloud-based AR with the integration of cloud and terminal capabilities is based on AR terminals with certain local rendering processing capabilities or real-scene digital correction capabilities; the cloud of the mode is used to complete the relevant digital processing processes such as real-scene spatial positioning, plane detection, coordinate binding and map construction in advance or real time, to form a cloud-based real-scene digital resource library. Subsequently, by reusing the real-scene digital resource library, cloud-based AR services are realized that integrate multiperson cloud and terminal integration. + +According to whether the virtual-scene rendering processing is carried out through the cloud, it can be subdivided into: the cloud-based AR mode which is based on real-scene digital results sharing by cloud and virtual-scene processing / virtual-real fusion output by terminal; and the cloud-based AR mode which is based on virtual-scene digital result delivery by cloud and virtual-real fusion output by terminal. + +### 7.1.1 Cloud-based AR mode based on real-scene digital results sharing by cloud and virtual-scene processing / virtual-real fusion output by terminal + +In the cloud-based AR mode which is based on real-scene digital results sharing by cloud and virtual-scene processing / virtual-real fusion output by terminal, as a cloud-based collaborative AR + +implementation method, the terminal's location information is collected, including the relevant positioning coordinates or the distance and angular between the terminal and base station, and the relevant location information is sent to the cloud management node, so that the management node in the cloud determines the target edge computing node that matches the terminal location information according to the principle of proximity, and forwards the terminal location information to the target edge computing node. The target edge computing node determines the target real-scene digitization result that matches the terminal location information and returns the target real-scene digitization result to the terminal. + +After obtaining the corresponding target real-scene digitization result, the terminal corrects the position and posture of the target real-scene digitalization result based on the position and posture information of the local sensor, obtains the corrected target real-scene digitization result and determines the reconstructed coordinate data of the real-scene map of the corrected target real-scene digitalization result. At the same time, based on the terminal location information, the corresponding virtual-scene material is extracted and the real-scene coordinate data of the corrected target real-scene digitization result is determined. Based on the real-scene coordinate data, the virtual-scene is rendered graphically to obtain and generate the virtual-scene result. Then the virtual-scene result and the corrected target real-scene digital results are layered overlay and fused to obtain and output the cloud AR virtual-real fusion results. + +If the terminal location information changes, the updated terminal location information is collected. If the updated terminal location information meets the pre-set difference conditions, the updated terminal location information is sent to the cloud, the cloud real-scene digital results are called and delivered, and the terminal completes the virtual-scene rendering and outputs the cloud AR virtual-real fusion result. + +### **7.1.2 Cloud-based AR mode based on virtual-scene digital result delivery by cloud and virtual-real fusion output by terminal** + +In the cloud-based AR mode which is based on virtual-scene digital result delivery by cloud and virtual-real fusion output by terminal, as an augmented reality implementation method, the cloud and terminal are used to generate optimized real-scene digital results, and the virtual-scene is processed by the cloud and is executed by the cloud server as the computing body. + +The cloud server determines the target real-scene based on the augmented reality request sent by the user terminal and verifies the target real-scene to determine whether there is information related to the target real-scene in the resource library of the cloud resource pool. If the target real-scene is verified, it obtains the real-time coordinate information of the target real-scene and the graphic directional data including the position, orientation and preliminary distance related to the target real-scene. + +Finally, the cloud server obtains the virtual-scene data from the resource library that matches the real-scene coordinate information and graphic directional data, and sends the virtual-scene data to the user terminal, so that the user terminal can combine the virtual-scene data with the target real-scene to output the virtual-reality fusion result. + +Constructing a virtual-scene by interacting with user terminals includes the following steps. The cloud server verifies the access permissions of the user terminal based on the verification request sent by the user terminal. After the access permission is verified, the cloud server queries whether the virtualization accuracy of the corresponding virtual-scene data meets the pre-set conditions in the resource library, based on the real-scene coordinate information reported by the user terminal. If the virtualization accuracy meets the pre-set conditions, the data is synchronized with the user terminal to obtain the initial real-scene digitization results generated by the SLAM capability of the user terminal calling its own real-time positioning and map construction. The cloud server utilizes the initial digitalization results of the real-scene, pre-set public data sets and artificial intelligence recognition results of the real-scene to construct multi-source data, and generates optimized real- + +scene digitalization results with higher accuracy using multi-source data. The cloud server utilizes the optimized digital results of the real-scene, to construct the virtual-scene and stores the corresponding virtual-scene data in the resource library. + +## 7.2 Cloud-based AR relying on cloud processing capabilities + +The mode of cloud-based AR entirely relies on cloud processing capabilities. The AR terminal uploads the current real-scene, completes the corresponding real-scene digitization in real time by the cloud, and then realizes cloud-based virtual-scene rendering, virtual-scene and real-scene digitization fusion output by calling the virtual-scene materials related to the cloud resource pool. Then the virtual-scene and real-scene digitization fusion output is sent to the terminal, and the terminal displays the cloud AR virtual-real fusion results. + +# 8 Framework of cloud-based augmented reality systems + +In [ITU-T G.1036], the reference framework for digital representation of cultural relics and artworks using augmented reality includes AR cloud creator, AR cloud management platform and mobile device. + +The digital representation of cultural relics and artworks using augmented reality is mainly to establish a digital communication mechanism between visitors and physical cultural relics/artworks, and there is no cross-regional large-scale multiperson interaction, nor is there more complex cloud processing and multiple edge computing collaboration, such as in multiperson cross-regional cloud-based AR navigation or cloud-based AR tourism. + +In order to widely adapt to cloud-based AR service scenarios, the end-to-end cloud-based AR system composition needs to be refined. Cloud-based AR systems consist of content layer, control layer, resource layer, network layer, terminal layer, security, OA&M and terminal layer. The content layer, control layer, resource layer, network layer, security, and OA&M of cloud-based AR systems run in the cloud. + +![Figure 8-1: Framework for cloud-based AR systems. The diagram shows a layered architecture. On the left, a dashed box labeled 'Run in cloud' contains three stacked layers: Content layer (top), Control layer (middle), and Resource layer (bottom). To the left of these layers are two boxes: 'Cloud-based AR OA&M' and 'Security'. The Control layer contains three sub-components: 'Cloud-based AR capability scheduling', 'Cloud-based AR capability integration', and 'Cloud-based AR capability distribution'. The Resource layer contains three sub-components: 'Computational resources', 'Running environment', and 'Resources collaboration'. To the right of the 'Run in cloud' box is the 'Network layer', which is connected to the 'Terminal layer' on the far right. The label 'F.740.11(25)' is at the bottom right.](8e14350b4b669119a3bdfca7869110ca_img.jpg) + +The diagram illustrates the framework for cloud-based AR systems. It is divided into two main parts: the cloud-based system and the terminal system, connected by a network layer. The cloud-based system, enclosed in a dashed box labeled 'Run in cloud', consists of three stacked layers: Content layer (top), Control layer (middle), and Resource layer (bottom). To the left of these layers are two additional components: Cloud-based AR OA&M and Security. The Control layer contains three sub-components: Cloud-based AR capability scheduling, Cloud-based AR capability integration, and Cloud-based AR capability distribution. The Resource layer contains three sub-components: Computational resources, Running environment, and Resources collaboration. The Network layer is positioned between the cloud-based system and the Terminal layer. + +Figure 8-1: Framework for cloud-based AR systems. The diagram shows a layered architecture. On the left, a dashed box labeled 'Run in cloud' contains three stacked layers: Content layer (top), Control layer (middle), and Resource layer (bottom). To the left of these layers are two boxes: 'Cloud-based AR OA&M' and 'Security'. The Control layer contains three sub-components: 'Cloud-based AR capability scheduling', 'Cloud-based AR capability integration', and 'Cloud-based AR capability distribution'. The Resource layer contains three sub-components: 'Computational resources', 'Running environment', and 'Resources collaboration'. To the right of the 'Run in cloud' box is the 'Network layer', which is connected to the 'Terminal layer' on the far right. The label 'F.740.11(25)' is at the bottom right. + +Figure 8-1 – Framework for cloud-based AR systems + +The content layer mainly provides a variety of cloud-based AR application content. Based on the application presentation mode, it can be divided into cloud-based AR games, cloud-based AR videos, cloud-based AR cartoons and cloud-based AR music. Based on the perspective of industry application, cloud-based AR content can be divided into cloud-based AR education, cloud-based AR social networking, cloud-based AR entertainment, cloud-based AR tourism and cloud-based AR industrial manufacturing. + +The control layer mainly provides cloud-based AR capability scheduling, cloud-based AR capability integration and cloud-based AR capability distribution. + +- Cloud-based AR capability scheduling: provides various cloud-based AR capability docking, coordination and other functions, including channel docking, resource scheduling, service allocation, operation mapping and location collaboration. +- Cloud-based AR capability integration: provide integrated services of various technical capabilities related to cloud-based AR, including real-scene understanding, 3D modelling, point cloud map, manipulation conversion, location service, AI processing and multiterminal data sharing. +- Cloud-based AR capability distribution: provides cloud-based AR capability distribution from resource pool to cloud-based AR users or terminals. + +The resource layer mainly provides computational resources, running environment and resources collaboration. + +The network layer provides the communication capabilities for cloud-based AR, such as QoS service, uplink services and network optimization. + +The terminal layer provides the cloud-based AR built-in services based on different types of terminals, including cloud-based AR services clients/APPs and plug-ins/SDK. It also provides service presentation on the terminals and collects manipulations of the terminals to return the control layer. It also provides the function of crowdsourcing collection about the real world. The terminal layer covers different types of AR terminals, such as optical see-through, video see-through and spatial augmented reality. + +The OA&M provides centralized operating, management and maintenance functions of cloud-based AR, such as the managements of content, user, technology integration, operation, resource, network capability docking and terminal. + +The security of cloud-based AR provides multidimensional security mechanisms, such as the securities of content, service, data and cloud resource pool, network and user experience. + +# **9 Requirements of cloud-based augmented reality systems** + +## **9.1 General requirements** + +### **9.1.1 Virtual-scene quality requirements** + +AQ-01: The output virtual-scene is required to have 2D audio-video processing capabilities and the ability to carry colour or transparent background information. + +AQ-02: The output virtual-scene is recommended to have 3D audio-video processing capabilities and the ability to carry colour and transparent background information. + +AQ-03: The output virtual-scene is recommended to achieve the same refresh rate as the AR terminal display. + +### **9.1.2 Interaction requirements** + +IR-01: It is required to achieve smooth human-computer interaction, and at least one common control method is required to be supported, such as handle, joystick, gesture or voice. + +IR-02: It is required to meet the 3DOF interaction capability and it is recommended to support the 6DOF interaction capability. + +### **9.1.3 Compatibility requirements** + +OR-01: It is required to support mainstream AR platform related content. + +OR-02: It is recommended to achieve compatibility with various types of AR terminals. + +### **9.1.4 Stability requirements** + +BR-01: It is required to provide a fault isolation mechanism between different functions. + +BR-02: It is required to ensure that short-term failures of related servers and other equipment do not cause alerts and service data. + +## **9.2 Content layer requirements** + +CL-01: The content of cloud-based AR is required to support the porting of all or part of the tasks such as reality scene digitization, virtual-scene rendering and data streaming of processing results to the cloud to achieve cloud-based execution. + +CL-02: The content of cloud-based AR is required to support the ability to capture and execute remote control instructions. + +CL-03: The content of cloud-based AR is required to have the ability to connect with the control layer and manage according to unified scheduling to complete distribution to various edge nodes and related AR terminals. + +CL-04: The content of cloud-based AR is required to support data synchronization, storage and on-demand calling, such as data distribution and scheduling to edge nodes and cloud AR terminals. + +CL-05: The content of cloud-based AR is required to connect with the cloud control layer and resource layer, and completing the docking of authentication, cloud operation, data synchronization and other capabilities. + +CL-06: The content of cloud-based AR is recommended to have the ability to flexibly adjust the screen output quality based on cloud and terminal processing capabilities. + +CL-07: The content of cloud-based AR is recommended to have a variety of flexible operation guidelines in order to improve user experience. + +CL-08: The content of cloud-based AR is recommended to have the ability to adjust the position, angle and lighting of virtual scenes according to changes in real-scene position, angle and lighting. + +## **9.3 Control layer requirements** + +### **9.3.1 Cloud-based AR capability scheduling** + +#### **9.3.1.1 Service node deployment and scheduling** + +ND-01: It is required to support the access of multiple service nodes and have the ability to flexibly expand according to the increase of service-bearing requirements. + +ND-02: It is required to have the ability to allocate and invoke service nodes in accordance with the principles of proximity, capability matching and load balancing. + +ND-03: It is required to support data synchronization and update between different service nodes and seamless service migration. + +#### **9.3.1.2 Server resource scheduling** + +RR-01: Within the service node, the server capacity is required to be flexibly expanded according to the increase in service-bearing requirements. + +RR-02: Within the service node, it is required to have the ability to allocate and call server resources according to the principles of capability matching and load balancing. + +RR-03: It is required to support data synchronization, update and seamless service migration between different servers in the same service node. + +#### **9.3.1.3 Running environment scheduling** + +EC-01: Among different servers, it is required to have the ability to flexibly expand the cloud-based AR operating environment according to the increase in service carrying demands. + +EC-02: In the server, according to the principle of capability matching, it is required to be assigned a running environment to match cloud-based AR users, such as the entire physical host, service virtual machine, container and service process. + +EC-03: It is required to support data synchronization, update and seamless migration between different servers in the same service node. + +#### **9.3.1.4 Service capability scheduling** + +CS-01: It is required to have the ability to adapt to cloud-based AR requests in order to call services such as authentication, billing, distribution and deployment in accordance with the resource allocation principle of load balancing. + +CS-02: It is required to have the ability to flexibly expand the software and hardware environment according to the bearing requirements of different functions. + +CS-03: It is recommended to support data synchronization, update and seamless migration between different servers in different service nodes and within the same service node. + +#### **9.3.1.5 Content resource scheduling** + +RS-01: It is required to have the ability to schedule cloud-based AR content related to different service nodes or different servers in the same node based on cloud-based AR requests. + +RS-02: It is required to have the ability to schedule corresponding software and hardware environments and to support the operation of cloud-based AR content based on cloud-based AR requests. + +RS-03: It is required to have the ability to support progress files related to cloud-based AR content and to complete data synchronization, update and seamless migration according to user requirements. + +#### **9.3.2 Cloud-based AR capability interaction** + +##### **9.3.2.1 Service construction** + +SC-01: It is required to have the ability to build cloud-based AR technology services based on cloud and edge nodes. + +SC-02: It is required to have the ability to build cloud-based AR running environment for users. + +##### **9.3.2.2 Operating environment requirements** + +OE-01: It is required to have support for mainstream AR interface specifications. + +OE-02: It is required to have the ability to support multiset parallel hosting, such as virtual machines, containers and multiprocess parallelism. + +OE-03: It is required to have the ability to support the isolation of operating environments. + +OE-04: It is required to support flexibly adjusting the output of cloud-based AR services in resolution, frame rate and field of view (FOV). + +OE-05: It is required to support the ability to seamlessly migrate cloud-based AR services based on different operating environments. + +OE-06: It is required to support the ability to call cloud-based AR content progress files based on the operating environment. + +#### **9.3.2.3 Capability requirements** + +CR-01: It is recommended to have the ability to support graphics rendering and video encoding. + +CR-02: It is required to have the ability for the seamless migration of cloud-based AR capabilities. + +CR-03: It is required to have the ability to support the digital processing of real scenes, including spatial coordinate anchoring, plane detection and point cloud map construction. + +CR-04: It is required to have the ability to complete real-time tracking and update the processing or call the appropriate digital processing results for the location information or real-scene data in real time for the cloud-based AR terminal. + +CR-05: For cloud-based AR relying on a cloud processing capabilities mode, it is required to have fusion based on real-scene digital results. + +CR-06: For cloud-based AR relying on a cloud processing capabilities mode, it is required to have the ability to render virtual scene based on the real-scene digital result and fuse the sound effect to complete the virtual-real output. + +CR-07: For the cloud-based AR mode based on virtual-scene digital result delivery by cloud and virtual-real fusion output by terminal, it is required to have the ability to provide cloud virtual-scene output data encoding and streaming processing and to deliver to the terminal. + +CR-08: For the cloud-based AR mode based on real-scene digital results sharing by cloud and virtual-scene processing / virtual-real fusion output by terminal, it is required to provide the ability to share and update reality digital processing resources. + +CR-09: It is recommended to have the ability to recognize real objects and to be able to accurately identify and subsequently digital process based on AI capabilities. + +#### **9.3.2.4 Support docking** + +SB-01: It is required to have the ability to connect with cloud-based AR operation services and provide cloud AR technology integration services. + +SB-02: It is required to have functions such as docking with cloud-based AR terminals, obtaining location information and control instructions. + +SB-03: It is required to have the ability to connect with OA&M to realize the management functions related to technical integration. + +#### **9.3.3 Cloud-based AR capability distribution** + +#### **9.3.3.1 Content distribution** + +CD-01: It is required to provide abilities for cloud-based AR content download or online operation. + +CD-02: It is required to provide CDN functions for cloud AR-related audio and video services as needed. + +CD-03: It is required to provide storage capabilities matched to cloud-based AR content distribution. + +CD-04: It is required to have the function of distributing AR content installation files and progress files to multiple service nodes and different servers of the same service node. + +#### **9.3.3.2 Resource distribution** + +RD-01: It is required to provide multiterminal and multi-edge node distribution functions with real-scene digital processing results. + +RD-02: It is required to provide a multiterminal distribution function of virtual-scene digital results completed in the cloud. + +RD-03: It is required to provide a cloud-based AR material distribution function to multi-edge multiterminal and nodes. + +#### **9.3.3.3 Technical competency distribution** + +TD-01: It is required to have the ability to distribute the technical functions and updated versions related for cloud-based AR to each service node and different servers of the same service node on demand, so as to complete the deployment and update of related functions. + +TD-02: It is required to have the ability to distribute the technical functions and updated versions related for cloud-based AR to each terminal on demand, and to implement the deployment of cloud-based AR services based on terminals. + +## **9.4 Resource layer requirements** + +### **9.4.1 Computational resources** + +PR-01: It is required to provide general-purpose hardware resources such as physical or virtual CPUs, GPUs, storage and NICs. + +PR-02: It is required to provide a common software system including a secure operating system, hardware drivers and a common graphics interface. + +PR-03: It is required to provide general functions such as configuration, expansion, migration and management of cloud resource pools. + +### **9.4.2 Running environment** + +RE-01: It is required to provide compatible mechanisms for concurrent reuse of virtual machine, container or application virtualization and other software and hardware resources. + +RE-02: It is required to provide secure and reliable virtualization, one-click migration and secure data synchronization mechanisms for virtual machines, containers or applications. + +RE-03: For large-scale digital processing, it is recommended to have the ability to collaborate across different service nodes as well as the ability to share and organize data. + +### **9.4.3 Resources collaboration** + +RC-01: It is required to provide data storage and data backup. + +RC-02: It is recommended to have remote storage and other capabilities to improve the high-speed sharing of data between different service nodes. + +## **9.5 Terminal layer requirements** + +### **9.5.1 Hardware** + +HW-01: It is required to use general-purpose CPU architectures, with complete computing, rendering, storage and other processing capabilities, which is suitable for cloud-based AR services. + +HW-02: It is required to provide real-time collection capabilities for 3DOF and related data. + +HW-03: It is required to provide effective controls, such as touch screens, gestures and sounds. + +HW-04: It is required to provide precise spatial positioning capability. + +HW-05: It is required to provide high-definition reality capture capability via built-in or external camera. + +HW-06: It is required to provide network access capability. + +HW-07: It is recommended to provide real-time collection capabilities for 6DOF and related data. + +### **9.5.2 Software** + +SW-01: It is required to support interconnection with the control layer to implement functions such as cloud-based AR functions, resource and content distribution, location information upload, receiving cloud-based AR service media streams and uploading control commands. + +SW-02: It is required to have the ability to upload real-time collected real scenes to the cloud. + +SW-03: It is recommended to have the ability to combine rendering output and sound effects of virtual-scene materials based on reality digital results and virtual-scene logic. + +SW-04: It is recommended to have the ability to locally store data such as real-scene digital results and virtual-scene materials. + +SW-05: It is required to complete the processing virtual scene based on local output or cloud delivery to fusion real-scene output. + +SW-06: AI prediction capability, which can provide real-scene recognition capabilities, can optionally be provided. + +SW-07: It is recommended to have real-scene digital processing capabilities such as plane detection, spatial recognition and coordinate anchoring. + +SW-08: It is required to adjust the coordinate mapping relationship of real-time digital processing according to the angle, position and distance of the collection, so as to lay the foundation for the virtual-real fusion output. + +## **9.6 Network layer requirements** + +CN-01: The uplink bandwidth of cloud-based AR terminals is required to be able to upload real-time location information and changes related to location information, such as orientation, angle and distance. + +CN-02: The uplink bandwidth for cloud-based AR terminals is recommended to meet the requirements for real-time upload of real-time shooting real-scene data to the cloud. + +CN-03: The downstream bandwidth for cloud-based AR terminals is required to meet the requirements for fusing and delivering virtual-scene rendering results with audio data. + +CN-04: It is required to provide a redundant packet transmission mechanism and a low packet loss rate, with the ability to stabilize data transmission. + +CN-05: Fibre to the home (FTTH) access technology is recommended for the use of wired access for cloud-based AR terminals. + +CN-06: It is recommended to have deterministic network access guarantees such as bandwidth and latency for cloud-based AR services. + +CN-07: It is recommended to have network transmission optimization mechanisms such as link path selection and network acceleration. + +CN-08: It is recommended to use fibre or wavelength division multiplexing (WDM) direct connection between cloud-based AR service nodes to reduce the delay of multiperson remote interaction. + +## **9.7 OA&M requirements** + +CO-01: It is required to provide the ability for cloud-based AR operation management based on cloud basic resource pools or general-purpose servers. + +CO-02: It is required to provide the ability to manage cloud resource pools related to the resource layer. + +CO-03: It is required to have interface management capabilities that provide various functions of cloud-based AR, including the ability to import and publish cloud-based AR. + +CO-04: It is required to have management capabilities such as access, authentication and removal for cloud-based AR capability interaction, cloud-based AR capability scheduling and cloud-based AR capability distribution. + +CO-05: It is required to complete user authentication, service billing, data analysis, operation and maintenance capabilities. + +CO-06: It is required to have the ability to connect with cloud-based AR terminals, complete user authentication, content synchronization and billing functions. + +CO-07: It is recommended to have built-in cloud-based AR terminals in the form of SDKs or APIs, and have operation/maintenance capabilities. + +## **9.8 Security requirements** + +SR-01: It is required to have the ability to divide security domains based on function or service nodes and to deploy firewalls, intrusion detection, anomalous traffic detection and filtering. + +SR-02: It is required to have antivirus capabilities and to process (logical) security assurance, including access restrictions, security audits and other capabilities. + +SR-03: It is required to provide copyright protection capabilities for cloud-based AR content to avoid problems such as recompilation and piracy. + +SR-04: It is required to have the ability to ensure service security through user management, permission management, log management and other means. + +SR-05: It is required to have security protection capabilities for metadata, user access data, business operation data and use encryption technology to encrypt important data transmitted on the Internet. + +# Appendix I + +## Use cases + +(This appendix does not form an integral part of this Recommendation.) + +Cloud-based augmented reality (AR) has the characteristics of adding colourful virtual-scene based on real-scene digitization and can achieve enhanced effects such as auxiliary information, interactive attributes and hidden attributes of real-scene. + +In order to fully demonstrate the features of cloud-based AR, cloud-based AR medical assistance, cloud-based AR navigation and cloud-based AR industrial maintenance are selected as typical cases of cloud-based AR mode, representing cloud-based AR relying on a cloud processing capabilities mode, cloud-based AR mode which is based on real-scene digital results sharing by cloud and virtual-scene processing / virtual-real fusion output by terminal, cloud-based AR mode which is based on virtual-scene digital result delivery by cloud and virtual-real fusion output by terminal. + +## I.1 Cloud-based AR medical assistance + +Cloud-based AR medical assistance uses AR technology to virtually annotate and display patients' lesions, and at the same time, through telemedicine experts, to complete virtual content such as disease diagnosis and a surgical program, so as to effectively improve the level of medical treatment and support medical training for students. + +Cloud-based AR medical assistance involves more complex etiological analysis and high-definition three-dimensional display, and at the same time needs to increase the assistance of telemedicine experts. Cloud-based AR medical assistance involves the accurate analysis and labelling of user lesions and three-dimensional display of surgery, which requires relatively powerful computing, rendering and AI computing power as support, and the terminal side cannot meet the requirements of computing power and flexibility, so it is necessary to complete the digital processing of lesions based on the cloud, and carry out flexible annotation, expert guidance and suggestions based on the digital processing results, three-dimensional display and simulation of surgical program. + +As a typical application of cloud-based AR relying on the cloud processing capabilities mode, the processing process of cloud-based AR medical assistance is shown in Figure I.1. + +![Diagram of the cloud-based AR medical assistance process showing the flow between a sick person, camera, terminal, cloud, and medical experts.](04f51626e2e10a16e3eb2c4b33cb2742_img.jpg) + +The diagram illustrates the cloud-based AR medical assistance process. It features a 'Sick person' on the left, a 'Camera' above them, and a 'Display device' (Terminal) below. A 'Cloud' is in the center, containing icons for a server, database, and AR headset. 'Medical experts' are on the right. The process is numbered 1 to 6: 1. Real-scene collection (Camera to Cloud), 2. Real-scene upload (Camera to Cloud), 3. Real-scene digital processing, virtual-scene rendering and virtual-real fusion (Cloud), 4. Input of guidance from medical experts (Medical experts to Cloud), 5. Distribution of virtual-real fusion results (Cloud to Terminal), and 6. Display of virtual-real fusion results (Terminal). Above the cloud, icons represent 'Real-scene digital processing', '3D annotation and model rendering', and 'Surgical program'. A reference code 'F.740.11(25)' is in the bottom right. + +Diagram of the cloud-based AR medical assistance process showing the flow between a sick person, camera, terminal, cloud, and medical experts. + +**Figure I.1 – Cloud-based AR medical assistance process** + +The cloud-based AR medical assistance mainly consists of two parts, terminal and cloud, of which the cloud part corresponds to the resource layer, control layer and QA&M. The terminal mainly includes the patient, the camera that collects the real scene and the corresponding display terminal, and the cloud mainly provides real-scene digital processing, 3D annotation, virtual-scene model + +rendering and surgical plan injection. Among them, the cloud also needs to support the input of medical experts' guidance. The overall process of cloud AR medical assistance is as follows: + +- Real-scene collection: Real-scene collection is carried out for patient-related lesions. +- Real-scene upload: The terminal uploads the real-scene results to the cloud. +- Cloud processing: The cloud dispatches computing resources at the resource layer through the control layer to complete the digital processing of real-scene, virtual-scene rendering and virtual-real fusion output. +- Expert opinion input: The cloud completes the input of relevant guidance of medical experts by connecting with medical experts; +- Processing result delivery: The cloud delivers the relevant virtual-real fusion results to the terminal side. +- Result display: The terminal completes the display of the corresponding virtual-real fusion results. + +### I.2 Cloud-based AR navigation + +Cloud-based AR navigation relies on a wide range of real-scene digital results and adds virtual guidance information according to the user's relevant travel purposes, because the real-scene digital results have multiuser sharing attributes and the addition of virtual guidance information does not require special computing power support, so it can be used as a typical application of the cloud-based AR mode which is based on real-scene digital results sharing by cloud and virtual-scene processing/virtual-real fusion output by terminal. + +Cloud-based AR navigation needs to be based on the digital processing of roads with massive traffic , transportation facilities, vehicles and surrounding buildings, and the digital processing results of this part of the real scene have the characteristics of shared use by multiple people, which is suitable for unified real-scene digital data aggregation, data distribution and data update operations through the cloud. At the same time, the virtual-scene rendering part related to cloud-based AR navigation has its own personalized characteristics for different users, and its rendering difficulty is not large, so it is more suitable to realize related virtual-scene processing and virtual-reality fusion output on the terminal side. + +The overall process of cloud-based AR navigation is shown in Figure I.2. + +![Diagram of the cloud-based AR navigation process showing interactions between a Terminal and a Cloud.](2eb23c2210154279f8013a1594fbcc5a_img.jpg) + +The diagram illustrates the cloud-based AR navigation process. On the left, a 'Terminal' icon (a building) is shown. On the right, a 'Cloud' icon contains three sub-components: 'Real-scene digital processing' (represented by a server rack), 'Navigation scheme' (represented by a lightning bolt), and 'Real-scene digital processing and navigation logic generation' (represented by a database, a server, and a camera). The process flow is as follows: 1. Acquisition of real-scene and location information (Terminal to Cloud); 2. Upload location information and key frames (Terminal to Cloud); 3. Real-scene digital processing and navigation logic generation (Cloud internal); 4. Real-scene digital results and navigation logic distribution (Cloud to Terminal); 5. Virtual scene rendering and virtual real fusion output (Terminal internal). The label 'F.740.11(25)' is in the bottom right corner. + +Diagram of the cloud-based AR navigation process showing interactions between a Terminal and a Cloud. + +**Figure I.2 – Cloud-based AR navigation process** + +The overall process of cloud AR navigation is as follows: + +- Real-scene and location information collection: The terminal completes the corresponding real-scene and the current location information collection. +- Upload location information and key frames: The terminal uploads the location information and real-scene key frames to the cloud. + +- Cloud processing: The cloud dispatches computing resources at the resource layer through the control layer to complete the digital processing of the real scene and the generation of navigation logic. +- Processing result delivery: The cloud delivers the relevant real-scene digital results and navigation logic to the terminal side. +- Terminal processing and display: The terminal completes the rendering of the virtual scene and the output of virtual-real fusion. + +## I.3 Cloud-based AR industrial maintenance + +The cloud-based AR industrial maintenance refers to the use of cloud-based AR capabilities to realize virtual enhanced dismantling of industrial equipment and amplification of fault points, so as to improve the fault location accuracy of on-site maintenance personnel and equipment maintenance efficiency. Because cloud-based AR industrial maintenance requires professional knowledge background and rich maintenance experience as its foundation, it is impossible to directly complete the real-scene logic related to professional maintenance based on the terminal side and at the same time, it is necessary to correct and adjust the relevant virtual-scene rendering points, lines, surfaces and stereoscopic imaging through specific fault situations during the maintenance process, and also at the same time, the virtual-scene rendering needs to be placed in the cloud for processing. + +Based on the characteristics of cloud-based AR industrial maintenance mentioned above, cloud-based AR industrial maintenance is a typical application of the cloud-based AR mode which is based on virtual-scene digital result delivery by cloud and virtual-real fusion output by terminal. + +The overall process of cloud-based AR industrial maintenance is shown in Figure I.3. + +![Figure I.3 – Cloud-based AR industrial maintenance process. The diagram illustrates a feedback loop between a 'Terminal' (represented by a headset icon) and a 'Cloud' (represented by a cloud icon with server symbols). The process consists of eight numbered steps: 1. Real-world capture of equipment (Terminal to Cloud); 2. Upload the device type and model (Terminal to Cloud); 3. Distribute the real-scene digital results of the equipment (Cloud to Terminal); 4. Real-scene digital comparison and analysis (Terminal); 5. Upload the results of the digital comparison of the real-scene of the equipment (Terminal to Cloud); 6. Digital calibration of equipment, maintenance virtual-scene rendering (Cloud); 7. The corrected equipment digitization results and maintenance simulation plan are issued (Cloud to Terminal); 8. Fusion output of equipment digitization and maintenance simulation solutions (Terminal). Above the Cloud icon, three labels with icons are shown: 'Real-scene digital processing' (starburst icon), 'Maintenance virtual-scene rendering' (dashed rectangle icon), and 'Maintenance solutions' (lightning bolt icon). The diagram is labeled 'F.740.11(25)' in the bottom right corner.](a24e89a6fe9bb70c83f8bf5202baba95_img.jpg) + +Figure I.3 – Cloud-based AR industrial maintenance process. The diagram illustrates a feedback loop between a 'Terminal' (represented by a headset icon) and a 'Cloud' (represented by a cloud icon with server symbols). The process consists of eight numbered steps: 1. Real-world capture of equipment (Terminal to Cloud); 2. Upload the device type and model (Terminal to Cloud); 3. Distribute the real-scene digital results of the equipment (Cloud to Terminal); 4. Real-scene digital comparison and analysis (Terminal); 5. Upload the results of the digital comparison of the real-scene of the equipment (Terminal to Cloud); 6. Digital calibration of equipment, maintenance virtual-scene rendering (Cloud); 7. The corrected equipment digitization results and maintenance simulation plan are issued (Cloud to Terminal); 8. Fusion output of equipment digitization and maintenance simulation solutions (Terminal). Above the Cloud icon, three labels with icons are shown: 'Real-scene digital processing' (starburst icon), 'Maintenance virtual-scene rendering' (dashed rectangle icon), and 'Maintenance solutions' (lightning bolt icon). The diagram is labeled 'F.740.11(25)' in the bottom right corner. + +**Figure I.3 – Cloud-based AR industrial maintenance process** + +The cloud-based AR industrial maintenance has the universality of equipment parts, and at the same time has the difference of its faults, and the corresponding maintenance plans are also very different, including the terminal collection of equipment real-scene, reporting equipment type and model, comparison and analysis of the real-scene digital results, and uploading the corresponding analysis and comparison results to the cloud, by obtaining the maintenance plan of virtual rendering issued by the cloud, to complete the localized final output; the cloud needs to complete the equipment real-scene digital correction and complete the maintenance virtual-scene rendering according to the maintenance plan. + +The overall process of cloud-based AR industrial maintenance is as follows: + +- Obtain the real scene of the equipment: The terminal obtains the real scene of the equipment through its built-in or external camera. + +- Upload equipment type and model: The terminal uploads the equipment type and model to the cloud. +- Delivery out digital results of the real-scene: The cloud distributes the digital processing results according to the relevant types and models of the equipment. +- Matching and analysis of real-scene digital results: After the terminal obtains the digital results of the equipment delivered by the cloud, it matches them with the current real-scene equipment and conducts corresponding comparison and analysis. +- Update the comparison results of the real-scene digital results of the equipment: The terminal reports the comparison results to the cloud, so that the cloud can locate the relevant equipment faults. +- Equipment digitization result correction and maintenance virtual-scene rendering: The cloud completes the equipment digital result correction and maintenance virtual-scene rendering processing according to the terminal upload result. +- Cloud processing result issuance: The cloud will deliver the calibrated equipment digitization results and maintenance simulation scheme to the terminal side. +- Terminal output: the terminal will obtain the equipment real-scene digital results and maintenance simulation scheme virtual-real fusion output. + +## Bibliography + +- [b-ITU-T J.301] Recommendation ITU-T J.301 (2014), *Requirements for augmented reality smart television systems.* +- [b-ITU-T Y.101] Recommendation ITU-T Y.101 (2000), *Global Information Infrastructure terminology: Terms and definitions.* + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.740.12-202512-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg b/marked/F/T-REC-F.740.12-202512-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c1629394c02f697f60ae954ad36bf274fa8082cb --- /dev/null +++ b/marked/F/T-REC-F.740.12-202512-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3763acd7a0ed56aa30f3f90468c1fb0cb54573eb9045bca3eddf8976b4f0f41d +size 7219 diff --git a/marked/F/T-REC-F.740.12-202512-I_PDF-E/b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg b/marked/F/T-REC-F.740.12-202512-I_PDF-E/b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3648e2fd48831fe870d8d88eda97bed949d227c5 --- /dev/null +++ b/marked/F/T-REC-F.740.12-202512-I_PDF-E/b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6fb0b3f3445295f4e8dd54475ebfe9d39da2b27cd0e8264d65c6bbb3c5ec08de +size 112444 diff --git a/marked/F/T-REC-F.740.12-202512-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg b/marked/F/T-REC-F.740.12-202512-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..4badcddf7bdefd9c3eb831803a2babf108aecfd4 --- /dev/null +++ b/marked/F/T-REC-F.740.12-202512-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:87bd35acc32089dd738e335db03786bc628e9622bb63ed9c89fc3fc3cc28d9bd +size 23758 diff --git a/marked/F/T-REC-F.740.12-202512-I_PDF-E/daa4a6fa7e2ba1954258f86b4928eb32_img.jpg b/marked/F/T-REC-F.740.12-202512-I_PDF-E/daa4a6fa7e2ba1954258f86b4928eb32_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..8a316e9a1de965de2832bd3ce63487c00b39dc08 --- /dev/null +++ b/marked/F/T-REC-F.740.12-202512-I_PDF-E/daa4a6fa7e2ba1954258f86b4928eb32_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aae2e371f3d93288a5221a85a68c0bf94c67c3ce363862b68fe8130614cab60b +size 24159 diff --git a/marked/F/T-REC-F.740.12-202512-I_PDF-E/f4fdd410cdb84df81274da55721e56fb_img.jpg b/marked/F/T-REC-F.740.12-202512-I_PDF-E/f4fdd410cdb84df81274da55721e56fb_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..bdefb0dc692538ef27a9bb4b5f4ed552b2149b94 --- /dev/null +++ b/marked/F/T-REC-F.740.12-202512-I_PDF-E/f4fdd410cdb84df81274da55721e56fb_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:56bcff40062051a5bd53115044695b5a81652da4abd1cfda529cac3a1635cd40 +size 29362 diff --git a/marked/F/T-REC-F.740.6-202309-I_PDF-E/raw.md b/marked/F/T-REC-F.740.6-202309-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..81d281e12e6fc59ade80c05bfb5f503bddec19bf --- /dev/null +++ b/marked/F/T-REC-F.740.6-202309-I_PDF-E/raw.md @@ -0,0 +1,531 @@ + + +# Recommendation**ITU-T F.740.6 (09/2023)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +### **Reference framework and requirements for Internet protocol multimedia subsystem early media and extension service system** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with white lines representing latitude and longitude, and the letters 'ITU' in a bold, blue, sans-serif font overlaid on the globe. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.740.6 + +# Reference framework and requirements for Internet protocol multimedia subsystem early media and extension service system + +## Summary + +Recommendation ITU-T F.740.6 specifies the requirements of the Internet protocol multimedia subsystem (IMS) early media and extension service (EMES) system. It defines the reference framework and requirements for network, user equipment (UE), and security of the system. It also describes several typical application scenarios including: + +- playing the early media before the called party alerting; +- playing the early media during the called party alerting; +- playing the early media when the call is failed; +- playing the media after the call is answered; +- playing the media during the release of the call; +- interactive scenarios. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.740.6 | 2023-09-13 | 16 | 11.1002/1000/15616 | + +## Keywords + +Customized alerting tones, customized ringing signal, early media, early media and extension service, EMES, IMS, ring back tone, ringing tone. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2023 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Overview ..... | 2 | +| 7 A reference framework of EMES system..... | 3 | +| 8 Requirements for EMES service provision ..... | 4 | +| 8.1 Requirements for the network ..... | 4 | +| 8.2 Requirements for the UE ..... | 5 | +| 8.3 Security requirements..... | 6 | +| Appendix I – EMES scenarios ..... | 8 | +| I.1 Playing scenarios ..... | 8 | +| I.2 Interactive scenarios ..... | 13 | +| Bibliography..... | 14 | + + + +# Recommendation ITU-T F.740.6 + +## Reference framework and requirements for Internet protocol multimedia subsystem early media and extension service system + +# 1 Scope + +This Recommendation aims to spread the early media and extension service (EMES) and help users access cultural contents with non-discriminatory treatment through short videos played in the service. The Recommendation defines the reference framework and network requirements, user equipment (UE) requirements, and security requirements of the EMES system. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T H.264] Recommendation ITU-T H.264 (2021), *Advanced video coding for generic audiovisual services*. + +[ITU-T H.265] Recommendation ITU-T H.265 (2023), *High efficiency video coding*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 IP multimedia subsystem** [b-3GPP TS 23.002]: The IP multimedia subsystem comprises all core network elements for provision of IP multimedia services comprising audio, video, text, chat, etc., and a combination of them delivered over the packet switched domain. + +**3.1.2 early media** [b-IETF RFC 3960]: Early media refers to media (e.g., audio and video) that is exchanged before a particular session is accepted by the called user. + +**3.1.3 precondition** [b-IETF RFC 3312]: A precondition is a set of constraints about the session which are introduced in the offer. The recipient of the offer generates an answer, but does not alert the user or otherwise proceed with session establishment. That only occurs when the preconditions are met. This can be known through a local event (such as a confirmation of a resource reservation), or through a new offer sent by the caller. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following term: + +**3.2.1 early media and extension service (EMES)**: A value-added service of call service, which is based on the additional media services related to the process of call service, such as early media and announcement media. EMES takes customized media content play and/or presents as the main service form, and introduces an interactive function, information service and other extended services in the process of media content play. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|--------|--------------------------------------------| +| AMR-WB | Adaptive Multi-Rate Wideband | +| CAT | Customized Alerting Tones | +| CRS | Customized Ringing Signal | +| EMEP | Early Media and Extension service Platform | +| EMES | Early Media and Extension Service | +| EVS | Enhance Voice Services | +| HD | High Definition | +| HPLMN | Home Public Land Mobile Network | +| IMS | IP Multimedia Subsystem | +| SDP | Session Description Protocol | +| SIM | Subscriber Identity Module | +| SIP | Session Initiation Protocol | +| UE | User Equipment | +| VPLMN | Visited Public Land Mobile Network | +| XR | Extended Reality | + +# 5 Conventions + +In this Recommendation: + +The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted if the conformance to this document is to be claimed. + +The keywords "is recommended" indicate a requirement which is recommended but not absolutely required. Thus, this requirement need not be present to claim conformance. + +The keywords "is not recommended" indicate a requirement which is not recommended but which is not specifically prohibited. Thus, conformance with this specification can still be claimed even if this requirement is present. + +In clause 8, mandatory requirements are labelled starting with R-, and recommended (i.e., optional) requirements are labelled starting with RR-. + +# 6 Overview + +Service innovation is the key to the competitiveness of telecom carriers. In some countries, the early media and extension service (EMES) serves as one of the service innovations based on 4G/5G. The customized audio ring-back tone as a traditional type of the EMES is popular because it plays the customized music when the calling party is waiting for the call answer. The customized video ring-back tone, as one of new types of the EMES, changes the early media during the called party alerting when the calling users can enjoy a short video, rather than just audio, while waiting for the answer to the call. The world witnessed a rapid growth in EMES and as of the end of April 2023, it has over 740 million subscribers and is supported by more than 1 billion smart phones. There is then a good opportunity to build telecom carriers' short video ecosystem based on the service. + +The EMES is one of the digital culture related services with the advanced digital multimedia technologies, IP multimedia subsystem (IMS) and 4G/5G technologies, etc. The EMES supports the + +creation, dissemination and representation of digital culture products. The popular contents of short video played in the EMES are digital cultural contents, such as music, film, drama, animation, etc. The EMES is usually applied to cultural dissemination scenarios, such as public service propagation, cultural tourism promotion, popular music video spreading, film and television drama sharing, weather forecasts, epidemic notices, etc. It is conducive to meeting the people's demand for digital cultural contents, promoting the production of digital culture contents, and facilitating the development of digital culture industry and ecology. + +It supports the exchange and sharing of diverse cultures worldwide via short videos containing digital culture contents played in IMS calls when the EMES are applied. Short videos played in the EMES can help users access cultural contents without discrimination. It also supports subscribers of the EMES allowing them to share their culture and learn from others, which, in the end, may perhaps make the world a better place. + +A number of telecom carriers in Asia, Europe and Africa have launched customized video ring-back tone service operations or plan to deploy this service. + +This Recommendation defines network requirements, UE requirements, and security requirements of the EMES system, which will help telecom carriers provide EMES faster and help to achieve interoperability among different telecom carriers. + +The development and deployment of 5G and other new technologies are expected to introduce new features to the EMES, such as the requirements to support HD video and XR video, and interactive videos are also described in this Recommendation. + +# 7 A reference framework of EMES system + +The EMES is generally based on the IP multimedia subsystem (IMS) services and multimedia content services and ecology. The EMES service system includes the service management subsystem and service provision subsystem, which work together to provide this service. The service management subsystem includes business management, content management, user management, operation management and access service server. The EMES service provision subsystem includes the early media and extension service platform (EMEP) media resource server, EMEP application server, IMS network and user equipment (UE). The requirements for the service provision subsystem are defined in clause 8 and the requirements for the service management subsystem are out of the scope of the Recommendation. + +A reference framework of EMES system is shown in Figure 1. + +![Figure 1 – Reference framework of EMES system. The diagram illustrates the architecture of the EMES system, divided into two main subsystems: Service management subsystem and Service provision subsystem. The Service management subsystem (top) includes an Access service server, User management server, Business management server, Content management server, and Operation management server. The Service provision subsystem (bottom) includes The calling party UE, The calling party IMS, The calling party EMEP (containing EMEP application server and EMEP media resource server), The called party EMEP (containing EMEP application server and EMEP media resource server), The called party IMS, and The called party UE. The calling party UE is connected to The calling party IMS, which is connected to The calling party EMEP. The calling party EMEP is connected to The called party EMEP, which is connected to The called party IMS, which is connected to The called party UE.](a234352dfaccdc24745c88eef7724cc6_img.jpg) + +The diagram shows the reference framework of the EMES system. It is structured as follows: + +- Service management subsystem** (top section): + - Access service server (top level) + - Four servers below it: User management server, Business management server, Content management server, and Operation management server. +- Service provision subsystem** (bottom section): + - On the left: The calling party UE connected to The calling party IMS. + - The calling party IMS connected to The calling party EMEP. + - The calling party EMEP contains two components: EMEP application server and EMEP media resource server. + - The calling party EMEP is connected to The called party EMEP. + - The called party EMEP also contains two components: EMEP application server and EMEP media resource server. + - The called party EMEP is connected to The called party IMS. + - The called party IMS is connected to The called party UE on the right. + +Figure 1 – Reference framework of EMES system. The diagram illustrates the architecture of the EMES system, divided into two main subsystems: Service management subsystem and Service provision subsystem. The Service management subsystem (top) includes an Access service server, User management server, Business management server, Content management server, and Operation management server. The Service provision subsystem (bottom) includes The calling party UE, The calling party IMS, The calling party EMEP (containing EMEP application server and EMEP media resource server), The called party EMEP (containing EMEP application server and EMEP media resource server), The called party IMS, and The called party UE. The calling party UE is connected to The calling party IMS, which is connected to The calling party EMEP. The calling party EMEP is connected to The called party EMEP, which is connected to The called party IMS, which is connected to The called party UE. + +Figure 1 – Reference framework of EMES system + +## **Service management subsystem:** + +- User management server: It provides user data management, and user service management functions including user authentication, user order, user content, user settings, etc. +- Business management server: It provides basic business management functions including service management, product management, tariff management, cash settlement management, etc., as well as business support management including process management, capability integration, etc. +- Content management server: It provides content copyright operation management functions including copyright content introduction, release, library management, etc., as well as application content operation and management functions including application content production, review, transcoding, release, library management, etc. +- Operation management server: It provides basic revenue management including billing and settlement, accounting revenue and expenditure, etc., marketing management including resource and activity management, and operation and maintenance management functions, etc. +- Access service server: It provides various user service portals, user service apps and other personnel access service functions, as well as various system access service functions. + +## **Service provision subsystem:** + +- EMEP application server: It provides basic service functions including IMS network signalling access, IMS call session management, IMS call flow control, out-of-band receiving and sending number processing, etc., and EMES service processing functions including service management subsystem interface, service data management, service scenario parsing, service scenario logic processing, service scenario flow processing, media negotiation control, and media play control, etc. +- EMEP media resource function: It provides basic service functions including IMS network media interface, media capability management, session management interface, media negotiation management, media play control interface, media file play, media stream transmission, media interaction service, and in-band receiving and sending number processing, etc., and EMES service processing functions including service management subsystem interface and media file management. +- IMS network: It provides basic service functions including IMS user terminal registration, IMS call flow control, public media negotiating and transcoding, media data anchoring and forwarding, network boundary protocol conversion, etc., and EMES service processing functions including service subscribing and triggering, service scenario status indication, service scenario media negotiation and play and media interaction service. +- The calling and the called party UE: Provide basic service functions including mobile/fixed network access, IMS network registration, IMS audio and video call, etc., and EMES service processing functions including service capability indication, service signalling procedure processing, service media processing, service media display, and terminal user interface interaction. + +# **8 Requirements for EMES service provision** + +## **8.1 Requirements for the network** + +### **8.1.1 Requirements for the EMEP** + +- R-NER-001: The EMEP is required to support related signalling procedures, media negotiations and media playing. + +- R-NER-002: The EMEP is required to support the audio codec AMR-WB and the video codec H.264 specified in [ITU-T H.264]. +- RR-NER-003: The EMEP is recommended to support the audio codec EVS and the video codec ITU-T H.265 specified in [ITU-T H.265]. +- RR-NER-004: The EMEP is recommended to support HD video, XR video and interactive media in a certain network condition for the voice and video call over IMS. +- R-NER-005: The EMEP is required to support call state processing and service provision on the EMES. +- RR-NER-006: The EMEP is recommended to support the provision of the EMES to the UE that supports the EMES when the precondition mechanism is not supported on the peer UE. +- RR-NER-007: The EMEP is recommended to support the processing and transmission of messages related to interactive functions. +- RR-NER-008: The EMEP is recommended to support a data channel for applying multimedia media including image, text, etc., and for applying interactive functions. + +### **8.1.2 Requirements for the IMS** + +- R-NER-008: The IMS is required to support triggering the EMES and transferring related parameters to the EMEP. +- R-NER-009: The IMS is required to support providing the call state so that the EMEP can provide the EMES which is applicable to the corresponding state. +- R-NER-010: The IMS is required to support precondition mechanism which can ensure the video experience for the EMES in IMS call. +- R-NER-011: The IMS is required to support the audio codec AMR-WB and the video codec ITU-T H.264 specified in [ITU-T H.264]. +- RR-NER-012: The IMS is recommended to support the audio codec EVS and the video codec ITU-T H.265 specified in [ITU-T H.265]. +- RR-NER-013: The IMS is recommended to support the EMES across IMS networks of different telecom carriers. +- RR-NER-014: The IMS is recommended to support roaming, i.e., if a user is in VPLMN, the EMEP in both HPLMN and VPLMN can provide EMES to the user based on telecom carriers' policy. +- R-NER-015: The IMS is required to support processing and forwarding of related media lines and header fields. +- RR-NER-016: The IMS is recommended to support transparent transmission of messages related to interactive functions, in cross-domain interworking scenarios. +- RR-NER-017: The IMS is recommended to support data channel for applying multimedia media such as image, text, etc., and for applying interactive functions. + +## **8.2 Requirements for the UE** + +### **8.2.1 Requirements for signalling processing** + +- R-UER-001: The UE is required to support related signalling procedures, media negotiations between the UE and other entities of IMS for EMES. +- RR-UER-002: The UE is recommended to support the processing and transmission of messages related to interactive functions. + +### **8.2.2 Requirements for media play** + +- R-UER-003: The UE is required to support playing the audio and/or video media for the EMES. +- RR-UER-004: The UE is recommended to support playing the image/text/other types of media for the EMES. +- R-UER-005: The UE is required to support the audio codec AMR-WB and the video codec ITU-T H.264 specified in [ITU-T H.264]. +- RR-UER-006: The UE is recommended to support the audio codec enhance voice services (EVS) and the video codec ITU-T H.265 specified in [ITU-T H.265]. + +### **8.2.3 Requirements for media player** + +- R-UER-007: The UE is required to support video media display for the EMES, such as landscape, portrait and full-screen display, as well as landscape and portrait content display. +- R-UER-008: The UE is required to support video media play of the EMES in non-video call without starting local camera. +- RR-UER-009: The UE is recommended to support video media play of the EMES in video call without displaying local camera image. +- RR-UER-010: The UE is recommended to support the display of call status-related text of prompts. +- R-UER-011: The UE is required to play the EMES media without negative affect on answering calls. + +### **8.2.4 Requirements for interaction** + +- RR-UER-012: The UE is recommended to support the interactive functions in all playing scenarios of the EMES, including displaying buttons such as "Like" and "Copy" in the planned area of the video window and reporting the user's operations on the relevant buttons to the network side. +- RR-UER-013: The UE is recommended to support the display of the corresponding animation effect in the video window according to the user's operations and the messages related to interactive functions. +- RR-UER-014: The UE is recommended to support a data channel for applying multimedia such as image, text, etc., and for applying interactive functions. + +### **8.2.5 Requirements for capability indication** + +- R-UER-015: The UE is required to support carrying the capability indication such as a video media feature tag in signalling messages when the specific capability is supported. + +## **8.3 Security requirements** + +- R-SER-001: It is required that the EMES contents be played to the user without negatively affecting the general call service. +- RR-SER-002: It is not recommended that the EMES contents be played to users that do not allow the EMES. +- R-SER-003: It is required for the EMEP to support authority security management, provide users with an authentication mechanism to ensure user authority and prevent unauthorized user access. +- R-SER-004: It is required for EMEP to ensure system security and have some necessary anti-risk capabilities such as preventing virus attacks. + +- R-SER-005: It is required for the EMEP to use hierarchical access control capabilities for man-machine interface commands, restrict unauthorized access and error access, and ensure data security and integrity. +- R-SER-006: It is required that the various operation and maintenance functions of the EMES system cannot endanger the security and integrity of the information stored and the information being transmitted. + +# Appendix I + +## EMES scenarios + +(This appendix does not form an integral part of this Recommendation.) + +This clause describes the typical application scenarios of IMS EMES. + +## I.1 Playing scenarios + +#### I.1.1 Playing the early media before the called party alerting + +Applying video/image/text/other types of media (excluding audio media in general) to the calling party before the called party alerting. + +NOTE – The relationships among EMES content, EMES tone content and media are shown in Figure I.1-1. It is noteworthy that the video EMES tone content generally contains video media without audio media, but can also contain both audio and video media based on the telecom carrier's policy, and the image EMES tone content contains image media, and so on. + +![Diagram showing the relationship between EMES content, EMES tone content, and media for playing scenario I.1.1. At the top, '(Customized) EMES content' branches into four boxes: '(Customized) Video EMES content', '(Customized) Image EMES content', '(Customized) Text EMES content', and 'Other type of media EMES content'. Below these are four boxes for 'tone content': 'Video EMES tone content', 'Image EMES tone content', 'Text EMES tone content', and 'Other type of media EMES tone content'. At the bottom are five media types in ovals: 'Audio media', 'Video media', 'Image media', 'Text media', and 'Other type of media'. Solid arrows show mandatory inclusion (e.g., Video media to Video EMES tone content), dashed arrows show optional inclusion (e.g., Video media to Video EMES content), and solid arrows with an 'X' show prohibitive inclusion (e.g., Audio media to Video EMES tone content).](f6e8acf9f931452d01688d311b5c0364_img.jpg) + +The diagram illustrates the relationships between different types of EMES content, tone content, and media. At the top level, '(Customized) EMES content' is shown in a box, which branches out via solid arrows to four boxes: '(Customized) Video EMES content', '(Customized) Image EMES content', '(Customized) Text EMES content', and 'Other type of media EMES content'. Below these, four boxes represent 'tone content': 'Video EMES tone content', 'Image EMES tone content', 'Text EMES tone content', and 'Other type of media EMES tone content'. Solid arrows point from each of the top-level EMES content boxes to their corresponding tone content boxes. Dashed arrows point from each of the top-level EMES content boxes to all four tone content boxes. At the bottom level, five ovals represent media types: 'Audio media', 'Video media', 'Image media', 'Text media', and 'Other type of media'. Solid arrows point from 'Video media', 'Image media', 'Text media', and 'Other type of media' to their corresponding tone content boxes. A solid arrow with an 'X' points from 'Audio media' to 'Video EMES tone content'. A legend at the bottom indicates: solid arrow for 'Mandatorily included in', dashed arrow for 'Optionally included in', and solid arrow with an 'X' for 'Prohibitively included in, in general'. The label 'F.740.6(23)' is in the bottom right corner. + +Diagram showing the relationship between EMES content, EMES tone content, and media for playing scenario I.1.1. At the top, '(Customized) EMES content' branches into four boxes: '(Customized) Video EMES content', '(Customized) Image EMES content', '(Customized) Text EMES content', and 'Other type of media EMES content'. Below these are four boxes for 'tone content': 'Video EMES tone content', 'Image EMES tone content', 'Text EMES tone content', and 'Other type of media EMES tone content'. At the bottom are five media types in ovals: 'Audio media', 'Video media', 'Image media', 'Text media', and 'Other type of media'. Solid arrows show mandatory inclusion (e.g., Video media to Video EMES tone content), dashed arrows show optional inclusion (e.g., Video media to Video EMES content), and solid arrows with an 'X' show prohibitive inclusion (e.g., Audio media to Video EMES tone content). + +**Figure I.1-1 – Relationship among EMES content, EMES tone content and media (for the playing scenario I.1.1)** + +Before the called party alerts, the customized EMES contents are played to the calling party. In parallel, some prompt texts such as "Calling..." are shown above the EMES contents for notifying the calling party of the current call state. + +When the called party alerts, the customized EMES contents may be stopped and switched to the customized alerting tones (CAT) contents, or the customized video EMES tone contents which contains audio and video media, the customized image/text/other types of media EMES tone contents, may continue to be played as CAT contents based on the telecom carrier's policy. + +Including 2 sub-scenarios: + +- The EMES contents played to the calling party are customized by the called party. +- The EMES contents played to the calling party are customized by the calling party. + +When the called party has customized EMES contents, the telecom carrier needs to set the priority of the EMES contents customized by the calling and the called party based on the telecom carrier's policy. + +#### I.1.2 Playing the early media during the called party alerting + +Including 2 sub-scenarios: + +- a) Applying audio/video ring tone to the calling party during the called party's alerting. + +NOTE 1 – If the customized audio/video ring back tone is played to the calling party, we can call it CAT. + +NOTE 2 – The relationship among CAT content, CAT tone content and media is shown in Figure I.1-2. + +![Diagram showing the relationship between CAT content, CAT tone content, and media. At the top, 'CAT content (Customized ring back tone content)' is shown. Below it, 'Audio CAT content (Customized audio ring back tone content)' and 'Video CAT content (Customized video ring back tone content)' are shown, both connected to the top CAT content by solid lines. Below these, five boxes represent 'tone content': 'Audio CAT tone content', 'Video CAT tone content', 'Image CAT tone content', 'Text CAT tone content', and 'Other type of media CAT tone content'. Solid lines connect 'Audio CAT tone content' to 'Audio CAT content' and 'Video CAT tone content' to 'Video CAT content'. Dashed lines connect 'Image CAT tone content', 'Text CAT tone content', and 'Other type of media CAT tone content' to both 'Audio CAT content' and 'Video CAT content'. At the bottom, five media types are shown in ovals: 'Audio media', 'Video media', 'Image media', 'Text media', and 'Other type of media'. Solid lines connect 'Audio media' to 'Audio CAT tone content' and 'Video media' to 'Video CAT tone content'. Dashed lines connect 'Image media', 'Text media', and 'Other type of media' to 'Image CAT tone content', 'Text CAT tone content', and 'Other type of media CAT tone content' respectively. A legend at the bottom indicates 'Mandatorily included in' with a solid line and 'Optionally included in' with a dashed line. The diagram is labeled F.740.6(23).](bffdddb47fced140f8d17fdc2a29f592_img.jpg) + +Diagram showing the relationship between CAT content, CAT tone content, and media. At the top, 'CAT content (Customized ring back tone content)' is shown. Below it, 'Audio CAT content (Customized audio ring back tone content)' and 'Video CAT content (Customized video ring back tone content)' are shown, both connected to the top CAT content by solid lines. Below these, five boxes represent 'tone content': 'Audio CAT tone content', 'Video CAT tone content', 'Image CAT tone content', 'Text CAT tone content', and 'Other type of media CAT tone content'. Solid lines connect 'Audio CAT tone content' to 'Audio CAT content' and 'Video CAT tone content' to 'Video CAT content'. Dashed lines connect 'Image CAT tone content', 'Text CAT tone content', and 'Other type of media CAT tone content' to both 'Audio CAT content' and 'Video CAT content'. At the bottom, five media types are shown in ovals: 'Audio media', 'Video media', 'Image media', 'Text media', and 'Other type of media'. Solid lines connect 'Audio media' to 'Audio CAT tone content' and 'Video media' to 'Video CAT tone content'. Dashed lines connect 'Image media', 'Text media', and 'Other type of media' to 'Image CAT tone content', 'Text CAT tone content', and 'Other type of media CAT tone content' respectively. A legend at the bottom indicates 'Mandatorily included in' with a solid line and 'Optionally included in' with a dashed line. The diagram is labeled F.740.6(23). + +**Figure I.1-2 – Relationship for CAT content, CAT tone content and media** + +When the called party alerts, the CAT contents are played to the calling party. In parallel, some prompt texts such as "The called party alerting" or "Call is waiting" are shown above the CAT contents for notifying the calling party of the current call state. + +Also including 2 sub-scenarios: + +- i) The CAT contents played to the calling party are customized by the called party. + +When the called party has no customization or cannot play the CAT contents customized by the called party for other reasons, the CAT contents can also be provided by the telecom carrier, such as news, weather, entertainment, and other information, etc. + +When the called party is in a call and the call waiting function is triggered, the CAT contents or common call waiting announcement tone is played to the calling party based on the telecom carrier's policy. + +- ii) The CAT contents played to the calling party are customized by the calling party. + +When the calling party has no customization or cannot play the CAT contents customized by the calling party for other reasons, the CAT contents can also be provided by the telecom carrier, such as news, weather, entertainment, and other information, etc. + +When the called party is in a call and the call waiting function is triggered, the CAT contents or common call waiting announcement tone is played to the calling party based on the telecom carrier's policy. + +When the called party has customized CAT contents, the telecom carrier needs to set the priority of the CAT contents customized by the calling and the called party based on the telecom carrier's policy. + +- b) Applying video/image/text/other types of media ringing tone to the called party in place of or in conjunction with local ring for alerting. + +NOTE 3 – If the customized video/image/text/other types ringing tone played to the called party, we can call it CRS. + +NOTE 4 – The relationship among customized ringing signal (CRS) content, CRS tone content and media is shown in Figure I.1-3. It is noteworthy that the video CRS tone content generally contains audio and video media but can also contain video media without audio media, and the image CRS tone content contains image media, and so on. + +![Diagram showing the relationship between CRS content, CRS tone content, and media. At the top is 'CRS content (Customized ringing tone content)'. Below it are four boxes: 'Video CRS content (Customized video ringing tone content)', 'Image CRS content (Customized image ringing tone content)', 'Text CRS content (Customized text ringing tone content)', and 'Other type of media CRS content'. Below these are four boxes: 'Video CRS tone content', 'Image CRS tone content', 'Text CRS tone content', and 'Other type of media CRS tone content'. At the bottom are five media types: 'Audio media', 'Video media', 'Image media', 'Text media', and 'Other type of media'. Solid arrows show mandatory inclusions: Audio and Video media to Video CRS tone content; Image media to Image CRS tone content; Text media to Text CRS tone content; Other type of media to Other type of media CRS tone content. Dashed arrows show optional inclusions: Audio media to Video, Image, and Text CRS content; Video media to Video, Image, and Text CRS content; Image media to Video, Text, and Other type of media CRS content; Text media to Video, Image, and Other type of media CRS content; Other type of media to Video, Image, and Text CRS content.](b90144cfbb81a2d610d920240fda689d_img.jpg) + +Diagram showing the relationship between CRS content, CRS tone content, and media. At the top is 'CRS content (Customized ringing tone content)'. Below it are four boxes: 'Video CRS content (Customized video ringing tone content)', 'Image CRS content (Customized image ringing tone content)', 'Text CRS content (Customized text ringing tone content)', and 'Other type of media CRS content'. Below these are four boxes: 'Video CRS tone content', 'Image CRS tone content', 'Text CRS tone content', and 'Other type of media CRS tone content'. At the bottom are five media types: 'Audio media', 'Video media', 'Image media', 'Text media', and 'Other type of media'. Solid arrows show mandatory inclusions: Audio and Video media to Video CRS tone content; Image media to Image CRS tone content; Text media to Text CRS tone content; Other type of media to Other type of media CRS tone content. Dashed arrows show optional inclusions: Audio media to Video, Image, and Text CRS content; Video media to Video, Image, and Text CRS content; Image media to Video, Text, and Other type of media CRS content; Text media to Video, Image, and Other type of media CRS content; Other type of media to Video, Image, and Text CRS content. + +**Figure I.1-3 – Relationship for CRS content, CRS tone content and media** + +When one number of the dual SIM dual active UE is in a call and the other number receives a new call, the common call waiting alerting tone rather than the CRS contents is played to the called party, to avoid excessive impact on the current call. + +Also including 2 sub-scenarios: + +- i) The CRS contents played to the called party are customized by the called party. +The video/image/text/other types of media CAT tone contents can be played to both the calling party and the called party, to enhance the called party's awareness of the CAT of the called party. +- ii) The CRS contents played to the called party are customized by the calling party, to share the identity information of the calling party with the called party. + +When the called party has customized CRS contents, the telecom carrier needs to set the priority of the video CRS contents customized by the calling and the called party based on the telecom carrier's policy. + +#### I.1.3 Playing the early media when the call is failed + +Applying audio/video/image/text/other types of media to the calling party in place of or in conjunction with traditional audio announcement tone when call is rejected or cancelled. + +NOTE – The relationship for EMES content, EMES tone content and media is shown in Figure I.1-4. It is noteworthy that the audio EMES tone content contains audio media, the video EMES tone content generally contains audio and video media but can also contain video media and no audio media based on the telecom carrier's policy, and the image EMES tone content contains image media, and so on. + +![Figure I.1-4: Relationship for EMES content, EMES tone content and media. The diagram shows three levels of hierarchy. At the top is '(Customized) EMES content'. Below it are five boxes: '(Customized) Audio EMES content', '(Customized) Video EMES content', '(Customized) Image EMES content', '(Customized) Text EMES content', and 'Other type of media EMES content'. Below these are five boxes for tone content: 'Audio EMES tone content', 'Video EMES tone content', 'Image EMES tone content', 'Text EMES tone content', and 'Other type of media EMES tone content'. At the bottom are five media types in ovals: 'Audio media', 'Video media', 'Image media', 'Text media', and 'Other type of media'. Solid arrows show mandatory inclusions: Audio media to Audio tone, Video media to Video tone, Image media to Image tone, Text media to Text tone, and Other media to Other tone. Dashed arrows show optional inclusions: Audio media to all other tone contents, Video media to Image, Text, and Other tone contents, Image media to Text and Other tone contents, and Text media to Other tone content.](33a8f3f01dfa8bce75d23017855a13c5_img.jpg) + +Figure I.1-4: Relationship for EMES content, EMES tone content and media. The diagram shows three levels of hierarchy. At the top is '(Customized) EMES content'. Below it are five boxes: '(Customized) Audio EMES content', '(Customized) Video EMES content', '(Customized) Image EMES content', '(Customized) Text EMES content', and 'Other type of media EMES content'. Below these are five boxes for tone content: 'Audio EMES tone content', 'Video EMES tone content', 'Image EMES tone content', 'Text EMES tone content', and 'Other type of media EMES tone content'. At the bottom are five media types in ovals: 'Audio media', 'Video media', 'Image media', 'Text media', and 'Other type of media'. Solid arrows show mandatory inclusions: Audio media to Audio tone, Video media to Video tone, Image media to Image tone, Text media to Text tone, and Other media to Other tone. Dashed arrows show optional inclusions: Audio media to all other tone contents, Video media to Image, Text, and Other tone contents, Image media to Text and Other tone contents, and Text media to Other tone content. + +**Figure I.1-4 – Relationship for EMES content, EMES tone content and media (for the playing scenario I.1.3, I.1.4, and I.1.5)** + +Including 2 sub-scenarios: + +- The EMES contents played to the calling party are customized by the called party. + +When the call is in one of in the following states: the called party's UE powered off, the called party unreachable (out of service area), the called party busy, the called party's number not in service, etc, the EMES contents (information shall be conveyed accurately and without ambiguity) play in place of or in conjunction with the general network prompt tone to enrich network perception. In parallel, some prompt texts such as "the called party + +power off" related to the states are shown above the EMES contents for notifying the calling party the current call state. + +- The EMES contents played to the calling party are customized by the calling party. + +When the call is in one of the following states: the called party's UE powered off, the called party unreachable (out of service area), the called party busy, the called party's number not in service, unallocated (unassigned) number, the calling party's number not in service, etc, the EMES contents (information shall be conveyed accurately and without ambiguity) play in place of or in conjunction with the general network prompt tone to enrich network perception. In parallel, some prompt texts such as "the called party power off" related to the states are shown above the EMES contents for notifying the calling party the current call state. + +When the called party has customized the EMES contents, the telecom carrier needs to set the priority of the EMES contents customized by the calling and the called party based on the telecom carrier's policy. + +#### **I.1.4 Playing the media after the call is answered** + +Applying audio/video/image/text/other types of media to the participants during the established call, including: + +NOTE 1 – Same as the Note in clause I.1.3. + +NOTE 2 – When video in the EMES and video of the call occur simultaneously, they can be displayed via different layouts and in windows without conflict. + +- Play the EMES contents (to the specific participants) customized by the operating participant or the operated participant, the subscriber or served user, the called or calling party, etc., when a call is held, muted, EMES is activated, etc. +- Play the EMES contents (to the specific participants) customized by the participant doing the operation or the participants being operated on, subscriber or served user, called or calling party, etc., in parallel with audio conversation, when EMES is activated, etc. +- Resume playing at a breakpoint. The calling party can continue to play the audio/video/image/text/other types of media CAT tone contents customized by the called or calling party that was interrupted due to off-hook. + +Use case: + +- 1) User A calls user B. + - 2) During alerting, a video CAT tone content is played to user A. + - 3) When user B accepts the call with audio call, the video CAT tone content may be continued to be played to user A with/without audio media based on the telecom carrier's policy and user's configuration, and maybe allow user A to close the video window of the CAT. +- Resume playing at a breakpoint. The called party can continue to play the video/image/text/other types of media CRS tone contents customized by the called or calling party that was interrupted due to off-hook. + +#### **I.1.5 Playing the media during the release of the call** + +Applying audio/video/image/text/other types of media to the calling or called party during the release of the call. + +NOTE – Same as the Note in clause I.1.3. + +- Play the EMES contents (to the calling or called party) customized by the called or calling party during the release of the call. + +- Resume playing at a breakpoint, the calling party can continue to play the audio/video/image/text/other types of media CAT tone contents customized by the called or calling party that was interrupted due to off-hook/hang-up, and the calling party can continue to play the audio/video/image/text/other types of media EMES tone contents in the call that was interrupted due to hang-up. + +Use case: + +- 1) User A calls user B. +- 2) During alerting, a video CAT tone content is played to user A. +- 3) When user B answers the call, the video CAT tone content may continue to be played to user A with/without audio media based on the telecom carrier's policy and user's configuration, and maybe allow user A to close the video window of the CAT. +- 4) If user A does not close the video window of the CAT, when user B hangs up the call, the CAT video may continue to play to user A. + +- Resume playing at a breakpoint The called party can continue to play the video/image/text/other types of media CRS tone contents customized by the called or calling party that was interrupted due to off-hook/hang-up, and the called party can continue to play the audio/video/image/text/other types of media EMES tone contents in the call that was interrupted due to hang-up. + +## **I.2     Interactive scenarios** + +Applying interactive functions in the above playing scenarios, such as users performing social operations (e.g., "Like", "Dislike", "Comment", etc.), service setting operations (e.g., "Copy", "Get reward", "Get more" for entering other service portals, etc.), etc. + +Wherein, the interactive functions include the calling or called user directly operates hotspot areas on the terminal screen, such as clicking a button or double-clicking the screen, to initiate interactive operations with the EMEP while watching EMES content. + +Use case: + +- 1) A customer uses a smartphone to call a business. +- 2) During the ringing stage, a customized video of the business is played to the customer. +- 3) While the video is playing, a menu of interactive actions represented by some icons or buttons is displayed as an overlay on the top of the video window, enabling the customer to select the menu. + +The user preference is expressed by the menu selection, for instance the customer can select to become a member of the business or to navigate the business website. + +# Bibliography + +- [b-3GPP TS 23.002] 3GPP TS 23.002, *Network architecture*. + +- [b-IETF RFC 3960] IETF RFC 3960 (2004), *Early Media and Ringing Tone Generation in the Session Initiation Protocol (SIP)*. +- [b-IETF RFC 3312] IETF RFC 3312 (2002), *Integration of Resource Management and Session Initiation Protocol (SIP)*. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.740.7-202309-I_PDF-E/raw.md b/marked/F/T-REC-F.740.7-202309-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..2ff30c8bdaec8ff0a7d88ed2b912b4736e4088d4 --- /dev/null +++ b/marked/F/T-REC-F.740.7-202309-I_PDF-E/raw.md @@ -0,0 +1,418 @@ + + +# Recommendation **ITU-T F.740.7 (09/2023)** + +SERIES F: Non-telephone telecommunication services +Multimedia services + +--- + +## **Reference architecture and requirements for a mobile terminal computational photography system** + +![ITU logo](390120de4fe440c42fea8154fcaad334_img.jpg) + +The logo of the International Telecommunication Union (ITU), featuring the letters 'ITU' in blue inside a circle with a globe-like design. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.740.7 + +# Reference architecture and requirements for a mobile terminal computational photography system + +## Summary + +Recommendation ITU-T F.740.7 specifies the reference architecture for a mobile terminal computational photography system, which includes three functional blocks, namely the optical imaging block, image processing block and application block. This Recommendation also describes requirements for the mobile terminal computational photography system. + +This Recommendation can facilitate cooperation between manufacturers, and enhance developing efficiency, so as to provide end users with better photography experiences. The aim is to define the reference architecture and requirements for the mobile terminal photography system. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.740.7 | 2023-09-13 | 16 | 11.1002/1000/15617 | + +## Keywords + +Computational photography, mobile terminal, MTCPS, reference architecture, requirements. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2023 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-----------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 1 | +| 5 Conventions ..... | 2 | +| 6 Overview ..... | 2 | +| 7 Reference architecture for MTCPS ..... | 3 | +| 7.1 Optical imaging block ..... | 4 | +| 7.2 Image processing block ..... | 4 | +| 7.3 Application block ..... | 5 | +| 8 Requirements for MTCPS ..... | 5 | +| 8.1 General requirements..... | 6 | +| 8.2 Service requirements ..... | 6 | +| 8.3 Security requirements..... | 7 | +| 8.4 Scalability requirements ..... | 7 | +| 8.5 Reliability requirements ..... | 7 | +| Appendix I – Typical application scenarios of MTCPS ..... | 8 | +| I.1 Mobile terminal photography scenarios ..... | 8 | +| I.2 Education application scenarios ..... | 8 | +| I.3 Entertainment application scenarios..... | 8 | +| I.4 Telecommuting scenarios..... | 8 | +| I.5 Robot application scenarios..... | 9 | +| I.6 Augmented reality scenarios ..... | 9 | +| Bibliography..... | 10 | + + + +# Recommendation ITU-T F.740.7 + +## Reference architecture and requirements for a mobile terminal computational photography system + +# 1 Scope + +This Recommendation specifies the reference architecture for a mobile terminal computational photography system, which includes an optical imaging block, image processing block and application block, as well as a description of each functional block. This Recommendation also describes requirements for the mobile terminal computational photography system. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 algorithm** [b-ISO/IEC 11557]: A set of rules for transforming the logical representation of data. + +**3.1.2 application** [b-ITU-T Y.101]: A structured set of capabilities, which provide value-added functionality supported by one or more services. + +**3.1.3 mobile terminal** [b-ITU-T X.1121]: An entity that has wireless network access function and connects a mobile network for data communication with application servers or other mobile terminals. + +**3.1.4 service** [b-ITU-T Y.101]: A structure set of capabilities intended to support applications. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following term: + +**3.2.1 mobile terminal computational photography system (MTCPS)**: A system with optical devices and computing units, which realizes computational photography relevant applications through the cooperation of hardware and software on a mobile terminal. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +3D 3-Dimensional + +API Application Programming Interface + +| | | +|-------|--------------------------------------------------| +| AR | Augmented Reality | +| HDR | High Dynamic Range | +| ISP | Image Signal Processor | +| MTCPS | Mobile Terminal Computational Photography System | +| OIS | Optical Image Stabilization | +| SDK | Software Development Kit | + +# 5 Conventions + +In this Recommendation: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement needs not be present to claim conformance. + +# 6 Overview + +With the rapid development of imaging and computing technology on mobile terminals such as smart phones, digital photography has become ubiquitous in the field of digital culture, which also promotes the progress of computational photography. The mobile terminal computational photography system (MTCPS) has transformed digital culture by making photography accessible, influencing visual communication through social media, and democratizing creativity while introducing new visual aesthetics. First, the accessibility and ubiquity of MTCPS has enabled widespread participation in photography, contributing to a culture of image or video creation and sharing. Second, MTCPS could integrate with social media platforms, and transform visual communication within digital culture, by allowing users to process photography contents instantly, and to convey thoughts and emotions. Finally, MTCPS has democratized creativity, enabling individuals to express themselves artistically without specialized equipment or technical expertise. The capabilities of MTCPS have also introduced new visual aesthetics, shaping the visual language and artistic practices of digital culture. + +In the process of optical imaging on mobile terminals, the components used for imaging are limited by size and power consumption and the hardware for receiving optical signals and controlling the hardware is quite different from the traditional photography equipment. Under some complicated circumstances, problems may arise when converting the optical signals into digital images via the hardware imaging system, such as severe image noise in low-light environments, narrow field of view, frustrating effects under backlight conditions, and low quality of distant imaging. These problems may become a hindrance for day-to-day photography of common mobile users and for some specific industry image applications. + +Mobile terminal computational photography system (MTCPS) refers to a system that creates image applications through imaging control, image computing and processing, and that combines artificial intelligence (AI) and mobile Internet technology to reconstruct scenes, light and objects relying on limited optical devices and computing units on the mobile terminal. MTCPS could use different calculation function units on a mobile terminal hardware for digital image signal sensing, computer vision analysis, semantic segmentation, image understanding, and synthetic transformation, which break the limitations of the traditional imaging model to increase and extend the capability of data acquisition in the traditional camera. + +This Recommendation specifies a reference architecture and details of each functional block, including the optical imaging block, the image software block and the application block. Through + +the cooperation of the hardware and software, the MTCPS can process various image data types and satisfy the applications in typical scenarios such as photography, education, entertainment, robotics, and augmented reality. + +In order to improve the development efficiency and quality, and facilitate industrial collaboration, the basic requirements for MTCPS, which include availability, reliability, credibility and scalability, are also described in this Recommendation. + +# 7 Reference architecture for MTCPS + +A reference architecture of MTCPS is shown as Figure 1, which includes an optical imaging block, an image processing block, and an application block. + +![Figure 1 – Reference architecture for MTCPS. The diagram shows three stacked blocks: Application block, Image processing block, and Optical imaging block. The Application block contains various apps like Panorama, 3D avatar, AR navigation, and Slow motion video. The Image processing block contains an Imaging control unit, a Computational photography processing unit (with service interface, services, and engines), and a Basic image processing unit (ISP). The Optical imaging block contains a Camera module (with various lenses) and an Imaging auxiliary module (with Flash, Focusing, etc.).](7a0db9703b68b3d06cdaeefc084c0006_img.jpg) + +The diagram illustrates the reference architecture for MTCPS, organized into three main horizontal blocks separated by dashed lines: + +- Application block:** Contains a list of applications: Panorama, 3D avatar, AR navigation, Slow motion video, Multiple exposure fusion HDR, and Other applications ... +- Image processing block:** Contains: + - Imaging control unit:** A vertical component on the left. + - Computational photography processing unit:** On the right, containing: + - Computational photography service interface:** The top layer. + - Computational photography services:** Below the interface, containing Service 1, Service 2, and Service N. + - Computational photography engines:** Below the services, containing Scene perception engine, Image analysis engine, Image understanding engine, Image reconstruction engine, and Other process engines ... + - Computational photography service manager:** Located between the Imaging control unit and the Computational photography processing unit. + - Basic image processing unit (ISP):** Located at the bottom of the Image processing block. +- Optical imaging block:** Contains: + - Camera module:** On the left, containing Wide angle camera, Ultra wide angle camera, Telephoto camera, Macro camera, and Other components ... + - Imaging auxiliary module:** On the right, containing Flash, Focusing, Irradiance sensor, Colour temperature, and Other components ... + +Figure 1 – Reference architecture for MTCPS. The diagram shows three stacked blocks: Application block, Image processing block, and Optical imaging block. The Application block contains various apps like Panorama, 3D avatar, AR navigation, and Slow motion video. The Image processing block contains an Imaging control unit, a Computational photography processing unit (with service interface, services, and engines), and a Basic image processing unit (ISP). The Optical imaging block contains a Camera module (with various lenses) and an Imaging auxiliary module (with Flash, Focusing, etc.). + +F.740.7(23) + +Figure 1 – Reference architecture for MTCPS + +When photography applications within the application block are activated by a user, the relevant imaging control instructions will be launched, and the camera module in the optical imaging block will generate and output image data stream, which will be processed by the basic image and computational photography processing unit in the image processing block. Finally, the processed image data will be transmitted back to the application block. + +The imaging control instructions include not only the control instructions issued to the optical imaging block (via the imaging control unit) to trigger optical imaging, but also the image processing instructions issued to the basic image processing unit and computational photography processing unit. + +The image processing indicates a series of processing works that the image data undergoes, after being generated from the optical sensor, and finally fed back to the applications. + +## 7.1 Optical imaging block + +A complete optical imaging block includes the following parts: + +- **Camera module.** This module is a necessity for an imaging block, which usually includes but is not confined to sensors, lens, focus actuators, and optical image stabilization (OIS) devices, constituting a camera. It is worth noting that the camera module may contain one or more cameras. With the support of the computational photography processing unit, multi-camera collaboration becomes an important feature that distinguishes MTCPS from other digital camera systems. +- **Imaging auxiliary module.** This module includes components related to the following aspects: flash, focusing, colour temperature and irradiance metering, etc. + +## 7.2 Image processing block + +The image processing block includes the following three parts: an imaging control unit, a basic image processing unit and a computational photography processing unit. + +- **Imaging control unit.** While translating an optical signal to a digital image signal, the imaging control unit will control the camera hardware according to the real-time statistical information of the image. This unit includes the process of focusing, exposure and flash by controlling lens, sensor and flashlight. +- **Basic image processing unit.** The basic image processing unit mainly realizes the common and basic image processing for digital image signal output from the hardware imaging system. This unit usually includes format conversion, dimension transformation (such as resizing or cropping), bad point correction, noise reduction, highlight suppression, backlight compensation, colour enhancement, lens shadow correction, and content encoding, etc. +- **Computational photography processing unit.** The computational photography processing unit mainly carries out complex image processing towards the single or multiple ways of the digital image signal. The computational photography processing unit is the core of the MTCPS functional module. MTCPS not only supports multi-lens, multi-sensor or multi-spectral cameras to generate better image quality, but also includes several image processing engines to implement the more advanced computer vision algorithms and deep learning algorithms. These technologies obtain more accurate image information and provide the adaptive function combination for different capture and processing demands. For example, MTCPS could provide specialized effects like 3D avatar, AI moon mode or multi-camera images fusion. + +MTCPS ensures flexibility and scalability in development, through the structured interface and internal modules in the computational photography processing unit, providing + +convenience for development on the one hand, and ease of management on the other. The computational photography processing unit comprises the following entities: + +- **Computational photography service interface.** This interface is the entrance of control instructions and image data between the computational photography processing unit and the peripheral units. It also provides an interface for development (via SDK or API) and management, and hides complicated details that users may not need to know. +- **Computational photography service manager.** The computational photography service manager forwards control instructions to one or more related computational photography engines, organizing and scheduling the computational photography algorithms distributed within each engine to run dynamically in the form of threads or processes. +- **Computational photography engines.** There are mainly four computational photography engines, namely the scene perception engine, the image analysis engine, the image understanding engine and the image reconstruction engine. Each computational photography engine contains a variety of related computational photography algorithms, providing corresponding image processing abilities, and to some extent, extending the capabilities of digital cameras in photography. +- **Scene perception engine** identifies and extracts useful scene information such as atmospheric situation, irradiation conditions, colour temperature, depth range, camera or subject motion status, etc., for subsequent processing by other engines. +- **Image analysis engine** analyses image and relevant information to perform various computer vision tasks, for instance, object detection, recognition and tracking, image segmentation, etc. +- **Image understanding engine** is responsible for integrating the low-level image information and high-level semantic information from other engines to perform photographic scene understanding and decision-making, for instance on, events happening in the scene, behaviour of subject, focus setting of the tracking object, best shutter timing, etc. +- **Image reconstruction engine** reconstructs the scene, light and the subject being photographed, by integrating the information and decision from other engines, presenting a higher quality or more favourable image to the end user. + +## 7.3 Application block + +The application block, abstracted as the one part of the reference architecture with which users interact directly, plays an important role in MTCPS. First, the application block contains a variety of applications, such as dual-camera bokeh, panorama, slow motion video, 3D avatar, AI moonshot, multiple camera fusion, and so on, to meet demands in various application scenarios. In addition, the application block also provides application management, content display, resource storage, network interconnection, social sharing and other common capabilities of mobile terminals. + +# 8 Requirements for MTCPS + +There are mainly five types of requirements for MTCPS, which are related to general requirements (including use, development and manufacture), service, security, scalability, and reliability, as described below. + +## **8.1 General requirements** + +### **8.1.1 Application user requirements** + +USR-001: MTCPS is recommended to support users to view the information of the deployed applications, including name, version, status, license and other information of the application. + +USR-002: MTCPS is recommended to support users to manage applications, including installation, activation, deactivation, upgrade and uninstallation of the application in the system. + +USR-003: MTCPS is recommended to support users to manage image contents, including storing, removing, transmitting and other operations. + +### **8.1.2 Developer requirements** + +DEV-001: MTCPS is recommended to provide a software development kit (SDK) or application programming interface (API), as well as an auditing and deploying mechanism for developing image application and image processing software. + +DEV-002: MTCPS is recommended to provide mechanisms to restrict developers from making changes to infrastructure of the hardware and software system, to prevent corruption or functional failure. + +### **8.1.3 Manufacturer requirements** + +MAN-001: MTCPS is recommended to provide mechanisms to allow manufacturers making changes to infrastructure of the hardware and software system, to support operations including addition, deletion, modification, replacement, or providing of customized functional modules. + +## **8.2 Service requirements** + +### **8.2.1 Requirements for computational photography imaging control services** + +SRV-001: MTCPS is recommended to provide computational photography requests parsing service for image applications, to translate the abstract photography requests to specific imaging control commands. + +SRV-002: MTCPS is recommended to provide imaging control services. The service controls the cameras and related auxiliary modules by receiving and executing imaging control commands. + +### **8.2.2 Requirements for computational photography image processing services** + +SRV-003: MTCPS is recommended to provide basic image processing services, such as automatic focusing, automatic exposure, automatic white balance, basal colour correction, noise reduction, etc. + +SRV-004: MTCPS is recommended to provide computational photography related image processing services, such as scene depth detection, object detection and recognition, semantic classification, semantic segmentation, multi-frame noise reduction, multi-frame fusion, etc. + +SRV-005: MTCPS is recommended to adopt image data transmission mechanisms to ensure efficient and reliable transmissions within and among the optical imaging block, image processing block and application block. + +### **8.2.3 Requirements for algorithm services in computational photography engines** + +SRV-006: MTCPS is recommended to provide application services for computational photography algorithm modules, including but not limited to query, selection, and combination, etc. + +SRV-007: MTCPS is recommended to provide management services for computational photography algorithm modules, including but not limited to loading, scheduling, unloading, and communication, etc. + +SRV-008: MTCPS is recommended to provide update services for computational photography algorithm modules. + +## **8.3 Security requirements** + +### **8.3.1 System security requirements** + +SEC-001: MTCPS is required to provide security assurance during the system operation to avoid health or safety hazards caused by hardware or software failures. + +SEC-002: MTCPS is recommended to provide security mechanisms to prevent malicious software from affecting the system operation. + +#### **8.3.2 Content security requirements** + +SEC-003: MTCPS is recommended to provide security mechanisms to protect end-user's privacy from unauthorized access. + +SEC-004: MTCPS is recommended to provide security mechanisms to ensure the integrity and copyrights of the image application software, image processing software, and end-users' private content from data corruption or unauthorized use. + +## **8.4 Scalability requirements** + +SCA-001: MTCPS is required to support the scalability of hardware devices, image processing software and image application software. + +## **8.5 Reliability requirements** + +REL-001: MTCPS is recommended to pass the system operation stability test to ensure the normal operation of the system under heavy load for a period of time. + +REL-002: MTCPS is recommended to support system suspension, sleeping or entering low power consumption mode for a period of time, and could be quickly switched to normal mode when necessary. + +# Appendix I + +## Typical application scenarios of MTCPS + +(This appendix does not form an integral part of this Recommendation.) + +Typical application scenarios of MTCPS are described in this appendix. + +## I.1 Mobile terminal photography scenarios + +People can use MTCPS-based terminals (such as smart phones) to obtain higher-quality photography effects. It is a common situation to take photos in dark and backlit environments. With the denoising, brightening, and deblurring functions provided by MTCPS, users can get clearer, brighter and wider dynamic ranges photos, even under less ideal lighting conditions. + +People can also use MTCPS-based terminals to achieve various photography effects, such as colour style filters, lighting effects, portrait beautification, panoramic stitching, etc., thereby enhancing the photography experience. + +### I.2 Education application scenarios + +Students can use MTCPS-based terminals to learn unfamiliar vocabularies independently. After taking pictures of the given text materials, words or sentences, from which information will be extracted via the text detection and recognition functions. Together with the voice synthesis and pronunciation functions (maybe from 3rd party applications), students can learn new vocabularies easier. + +Teachers can use MTCPS-based terminals to achieve rapid marking and correcting of students' assignments or examination submissions. After taking pictures of the submission (e.g., examination paper), and using the detection and recognition functions to locate and identify the contents submitted by students, inspection could be completed automatically, thereby reducing the teachers' burden. + +Terminals with MTCPS could also be used to perform non-contact photographing and scanning of a specific object, and then perform the three-dimensional reconstruction of the object to build a three-dimensional model that can be further displayed and printed for teaching purposes. + +## I.3 Entertainment application scenarios + +People can use MTCPS-based terminals to take pictures, and generate interesting photographic effects, such as cartooning and stylizing scenes or characters. The user could also add, delete and modify the background and foreground objects in the scene. After that, the user may share the edited content to others through the network transmission function. + +People can also use MTCPS-based terminals for posture and motion detection to achieve more interesting interactions. Through the detection of the key points, the skeleton information could be extracted to estimate the posture and movement of a certain person, which can be used to control other software programs or obtain feedback (visually or acoustically) to realize more interesting human-computer or interpersonal interactions. + +## I.4 Telecommuting scenarios + +More and more people have to work from home because of the pandemic. MTCPS-based terminals could provide background blur and virtual background functions to protect people's home privacy. Furthermore, relying on the portrait segmentation capability provided by MTCPS-based terminals, the video conference or remote meeting can integrate the speaker and the presentation content + +within the same display window to perform a more professional presentation and create a better meeting experience. + +### **I.5 Robot application scenarios** + +MTCPS can also be built into robots to realize scene perception and navigation. Firstly, features and landmarks of a certain target could be detected by a monocular or multi-view vision system. Secondly, taking advantage of above information, a scene depth calculation and 3D reconstruction system can easily identify obstacles and rebuild surrounding environment. Finally, after a series of complicated calculations using the information of obstacles, robot positioning and navigation can become possible. + +### **I.6 Augmented reality scenarios** + +MTCPS-based terminals can also be used in augmented reality (AR) related applications. Unlike traditional two-dimensional image capturing techniques, MTCPS can reconstruct the 3D information and superimpose virtual information of certain objects, providing an immersive experience when capturing on mobile terminals. + +## Bibliography + +- [b-ITU-T X.1121] Recommendation ITU-T X.1121 (2004), *Framework of security technologies for mobile end-to-end data communications.* +- [b-ITU-T Y.101] Recommendation ITU-T Y.101 (2000), *Global Information Infrastructure terminology: Terms and definitions.* +- [b-ISO/IEC 11557] ISO/IEC 11557 (1992), *Information technology – 3,81 mm wide magnetic tape cartridge for information interchange – Helical scan recording – DDS-DC format using 60 m and 90 m length tapes.* + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.740.8-202406-I_PDF-E/042733dc5e8e7f5f30b60adba3266cde_img.jpg b/marked/F/T-REC-F.740.8-202406-I_PDF-E/042733dc5e8e7f5f30b60adba3266cde_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..6b6a63bf938655e857186e38a83515b2a01cb3a9 --- /dev/null +++ b/marked/F/T-REC-F.740.8-202406-I_PDF-E/042733dc5e8e7f5f30b60adba3266cde_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:391ecc0bed0f28346739db92ae6f77cf0b02b406323fcec3fb165a5f3b70f7ca +size 103282 diff --git a/marked/F/T-REC-F.740.8-202406-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/F/T-REC-F.740.8-202406-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ac295b1b0c7f14b1c5a33ef6682b1b017c812477 --- /dev/null +++ b/marked/F/T-REC-F.740.8-202406-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37aed4fc114a39843438175326e0b7e1d09a9d6075723d82fea32040fd560ee4 +size 7192 diff --git a/marked/F/T-REC-F.740.8-202406-I_PDF-E/raw.md b/marked/F/T-REC-F.740.8-202406-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..fb61a80c2927134bca084faa60a05b3b1452ef98 --- /dev/null +++ b/marked/F/T-REC-F.740.8-202406-I_PDF-E/raw.md @@ -0,0 +1,499 @@ + + +# Recommendation **ITU-T F.740.8 (06/2024)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +## **Requirements and architecture for live virtual tour systems using panoramic video and augmented reality** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.740.8 + +# Requirements and architecture for live virtual tour systems using panoramic video and augmented reality + +## Summary + +Recommendation ITU-T F.740.8 defines a live virtual tour system using panoramic video and augmented reality. By using panoramic video and augmented reality technologies, the live virtual tour system can enable network users to experience the remote tour by watching the immersive multimedia content and using the related interactive functions. This Recommendation describes the requirements and architecture for the live virtual tour system using panoramic video and augmented reality. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.740.8 | 2024-06-13 | 16 | 11.1002/1000/15904 | + +## Keywords + +Architecture, augmented reality, live virtual tour system, panoramic video, requirements. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2024 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|--------------|-----------------------------------------------------------------------------------------------------------|----| +| 1 | Scope ..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 1 | +| 4 | Abbreviations and acronyms ..... | 1 | +| 5 | Conventions ..... | 2 | +| 6 | Requirements for live virtual tour system using panoramic video and augmented reality ..... | 2 | +| 6.1 | User requirements..... | 2 | +| 6.2 | Service requirements ..... | 3 | +| 6.3 | Security requirements..... | 4 | +| 6.4 | Quality of service (QoS) and quality of experience (QoE) requirements ..... | 5 | +| 6.5 | Terminal requirements..... | 5 | +| 7 | Architecture for live virtual tour system using panoramic video and augmented reality ..... | 5 | +| 7.1 | An overview of architecture for live virtual tour system using panoramic video and augmented reality..... | 5 | +| 7.2 | Functional modules ..... | 6 | +| 7.3 | Reference points ..... | 8 | +| Appendix I – | Use cases for LVTS using PV and AR ..... | 10 | +| I.1 | Live tour guide through PV ..... | 10 | +| I.2 | Roaming of multiple PV scenes ..... | 10 | +| I.3 | Live tour guide through PV with AR effects..... | 10 | +| | Bibliography..... | 11 | + + + +# Recommendation ITU-T F.740.8 + +# Requirements and architecture for live virtual tour systems using panoramic video and augmented reality + +# 1 Scope + +This Recommendation specifies the requirements and architecture for live virtual tour systems using panoramic video and augmented reality including: + +- 1) Requirements for live virtual tour systems using panoramic video and augmented reality; +- 2) Architecture for live virtual tour systems using panoramic video and augmented reality; +- 3) Reference points for live virtual tour systems using panoramic video and augmented reality. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 augmented reality** [b-ITU-T J.301]: A type of mixed reality where graphical elements are integrated into the real world in order to enhance user experience and enrich information. + +**3.1.2 reference point** [b-ITU-T Y.2012]: A conceptual point at the conjunction of two non-overlapping functional entities that can be used to identify the type of information passing between these functional entities. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 panoramic video**: A kind of video shot by using an omnidirectional camera or a collection of cameras and formed by recording each frame of view in every direction at the same time. + +**3.2.2 live virtual tour system**: An online tour system which enables network users to experience the remote tour by watching the multimedia contents and using the related interactive functions. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|--------------------------| +| AR | Augmented Reality | +| FOV | Field of View | +| LVTS | Live Virtual Tour System | + +| | | +|-----|-----------------------| +| PV | Panoramic Video | +| QoE | Quality of Experience | +| QoS | Quality of Service | + +# 5 Conventions + +In this Recommendation: + +The keywords "is required" indicate a requirement which must be strictly followed and from which no deviation is permitted if the conformance to this document is to be claimed. + +The keywords "is recommended" indicate a requirement which is recommended but not absolutely required. Thus, this requirement need not be present to claim conformance. + +Requirements are identified using the following conventions: + +- Requirement number xx in clause n is of the form R-n-xx; +- Recommended requirement number yy in clause n is of the form RR-n-yy; +- Requirement number xx in subclause n.m is of the form R-n.m-xx; +- Recommended requirement number yy in subclause n.m is of the form RR-n.m-yy. + +# 6 Requirements for live virtual tour system using panoramic video and augmented reality + +## 6.1 User requirements + +Live virtual tour system (LVTS) users can be divided into two roles, virtual tour content publisher, and audience. The users of different roles participate in the different parts of the system, so they have the different requirements. The LVTS content publishers are responsible for creating and managing the virtual tour content, and the LVTS audiences can watch the virtual tour content and interact with the live LVTS. + +### 6.1.1 LVTS content publisher requirements + +- R-6.1-01: The system is required to support the registration and de-registration of the LVTS content publisher through the interface provided by the system. +- R-6.1-02: The system is required to support the LVTS content publisher to login and logout from the system conveniently. +- R-6.1-03: The system is required to support the LVTS content publisher to view and modify the personal information, including name, password, identity information, etc. +- R-6.1-04: The system is required to support the LVTS content publisher to manage the live panoramic video (PV) channels, including addition, deletion, browsing, retrieval, audit, and ban. And the system is also required to support the LVTS content publisher to configure the basic information of live PV channels, such as name, introduction, start time, end time, etc. +- R-6.1-05: The system is required to support the LVTS content publisher to broadcast the live PV through a created live PV channel by using the PV capturing and transmission devices, such as hand-held camera or fixed PV acquisition equipment. +- R-6.1-06: The system is required to support the LVTS content publisher to manage the augmented reality (AR) elements of the live PV channels and PV. Specifically, the LVTS content publisher can create, modify and delete the AR elements in the live PV channels and PV. +- RR-6.1-01: The system is recommended to support the LVTS content publisher to manage the PV on demand, including addition, deletion, browsing, retrieval, audit, and ban. And the + +system is also recommended to support the LVTS content publisher to configure the basic information of PV on demand, such as name, introduction, publishing time, end time, etc. + +- RR-6.1-02: The system is recommended to support the LVTS content publisher to get the real-time and historical monitoring information of the live PV channels and PV on demand through the interfaces provided by the system, including number of online audiences, total number of views, etc. +- RR-6.1-03: The system is recommended to support the LVTS content publisher to manage the media resources of the live PV channels and PV. Specifically, the LVTS content publisher can upload, modify and delete the media resources used in the live PV channels and PV. +- RR-6.1-04: The system is recommended to support the LVTS content publisher to manage the route of auto guiding function, including addition, deletion, modification and preview. +- RR-6.1-05: The system is recommended to support the LVTS content publisher to control the authority for watching the virtual tour PV with AR elements. + +### **6.1.2 LVTS audience requirements** + +- R-6.1-07: The system is required to support the registration and de-registration of the LVTS audience through the interface provided by the system. +- R-6.1-08: The system is required to support the LVTS audience to login and logout from the system conveniently. +- R-6.1-09: The system is required to support the LVTS audience to watch the live PV or PV on demand with AR elements through different devices, such as PC, large screen devices, mobile devices, etc. LVTS audience can freely adjust the viewing angle of the PV. +- RR-6.1-06: The system is recommended to support the LVTS audience to have basic control over the live PV or PV, such as playing or pausing PV, turning sound on or off, entering or exiting full screen, etc. +- RR-6.1-07: The system is recommended to support the LVTS audience to send the comments for the virtual tour PV with AR elements. And users can see the other comments sent by the other LVTS audiences. +- RR-6.1-08: The system is recommended to support the LVTS audience to interact with the PV with AR elements. For example, the LVTS audience can click the AR hotspot label to trigger the corresponding events embedded in the label, such as popping up image, video, web page, etc. +- RR-6.1-09: The system is recommended to support the LVTS audience to change the PV scenes. The LVTS audience can switch to another PV scene according to their personal preferences by clicking the scene label. +- RR-6.1-10: The system is recommended to support the LVTS audience to use the auto guiding function. The system can automatically switch the PV viewing angle according to the scheduled route configured by the LVTS content publisher and trigger the corresponding AR hotspot label. + +## **6.2 Service requirements** + +- R-6.2-01: The system is required to support the registration, de-registration, login, logout, and information modification of the users through the interface provided by the system. +- R-6.2-02: The system is required to support to the management of the live PV channels. +- R-6.2-03: The system is required to support the live PV distribution from the video capturing devices to the audience viewing devices. +- R-6.2-04: The system is required to support the PV rendering to transfer the PV streaming to the PV that can be played smoothly in the screen. + +- R-6.2-05: The system is required to support the storage of PV configuration file, AR elements and other related media resources. +- RR-6.2-01: The system is recommended to support the management of the PV on demand. +- RR-6.2-02: The system is recommended to support the PV distribution from the servers where the PV is stored to the audience viewing devices. +- RR-6.2-03: The system is recommended to support the transcoding of the PV, and the PV transcoding templates can be configured. +- RR-6.2-04: The system is recommended to support the storage and playback of the live PV stream. +- RR-6.2-05: The system is recommended to support the monitoring information collection and display of the live video channels and PV on demand, such as number of online audiences, total number of views, etc. +- RR-6.2-06: The system is recommended to support the AR interaction in PV. The interactive AR elements are superimposed on a specific location in the PV frame, and the user can click on these interactive AR elements to trigger the corresponding event. +- RR-6.2-07: The system is recommended to support the auto guiding function. The system can automatically switch the viewing angle and trigger the related interactive AR elements according to the pre-set rules. +- RR-6.2-08: The system is recommended to support the panoramic roaming function. The users can switch the PV scenes according to their personal preferences by clicking the scene label. +- RR-6.2-09: The system is recommended to support the instant message interaction for the live PV channel and the comment for the PV on demand. +- RR-6.2-10: The system is recommended to support the message management for the live PV channel and the comment management for the PV on demand. + +## 6.3 Security requirements + +### 6.3.1 Access security requirements + +- R-6.3-01: The system is required to provide the mechanisms for authentication and authorization. It is required to provide the different interfaces for the users of different roles. Only the authorized users are allowed to access the corresponding part of system and use the related function. +- R-6.3-02: The system is required to forbid an unauthorized user to publish the PV through the system or send messages through the system. +- RR-6.3-01: The system is recommended to provide the mechanisms for viewing condition certification. The audience should be checked whether they can watch the PV according to the viewing conditions. + +### 6.3.2 Content security requirements + +- RR-6.3-02: The system is recommended to provide the security mechanisms for PV and AR elements, such as video encryption, AR element configuration file, etc., to ensure the integrity and secrecy of the media data. +- RR-6.3-03: The system is recommended to provide the mechanisms to protect the copyrights of the PV and AR elements. +- RR-6.3-04: The system is recommended to provide the logo addition function for the PV to maintain the copyright of the initiator of PV content. + +- RR-6.3-05: The system is recommended to provide the detection and prevention function for the sensitive elements of the PV content. + +### **6.3.3 System security requirements** + +- R-6.3-03: The system is required to protect the user privacy. The password and other private information should be encrypted in the system. +- R-6.3-04: The system is required to have a certain ability to resist the vicious attacks. +- R-6.3-05: The system is required to have the capability of disaster recovery. +- RR-6.3-06: The system is recommended to provide a troubleshooting mechanism. + +## **6.4 Quality of service (QoS) and quality of experience (QoE) requirements** + +- R-6.4-01: The system is required to respond rapidly to the user's operation. For example, when users drag the PV to adjust the angle of view, the PV will change the perspective instantly and correctly. +- R-6.4-02: The system is required to provide the reasonable PV rendering method to ensure that the panoramic can be played correctly. +- RR-6.4-01: The system is recommended to implement the QoS policies on the application layer, the transport layer or both. In the case of network delay, jitter, sequence error and packet loss, the system can recover and repair the damage, and adjust the PV flow according to the changes in network bandwidth. +- RR-6.4-02: The system is recommended to provide the high-quality PV transmission. When the PV is played, the pictures and the voice are required to be clear and fluent, and the multimedia is required to be synchronized. +- RR-6.4-03: The system is recommended to support the low end-to-end delay transmission of PV from the PV source device to the audience viewing device. +- RR-6.4-04: The system is recommended to support the PV frame rate with no lower than 30 frames/s. +- RR-6.4-05: The system is recommended to ensure that the AR elements should be set reasonably. + +## **6.5 Terminal requirements** + +- RR-6.5-01: The system is recommended to support the various PV capturing devices, including hand-held panoramic camera, panoramic camera connected to mobile phone, fixed PV acquisition equipment, etc. +- RR-6.5-02: The system is recommended to support the various audience devices to access the system to watch the PV, such as personal computer, mobile phone, large screen, etc. + +# **7 Architecture for live virtual tour system using panoramic video and augmented reality** + +This clause defines the functional architecture for LVTS using panoramic video (PV) and augmented reality (AR), the functional modules and reference points for the architecture. + +## **7.1 An overview of architecture for live virtual tour system using panoramic video and augmented reality** + +The goal of LVTS using PV and AR is to build a scenic spot oriented live streaming system with interesting interactive functions. Visitors can watch the live panoramic views of the scenic area anytime and anywhere. AR effects are applied to improve the interactivity of the system. There are two roles of the system users: content publishers and audiences. The publishers can publish PV + +contents, manage the accounts and edit the system functions. The audiences can see the panoramic live content and experience the interactive functions with different kinds of terminal devices. Therefore, the system needs a publisher module to create the panoramic live video stream and an audience module to display the panoramic content for online tourists. Also, the system needs a management module to manage the whole system. What's more, the system needs a module to process and deliver the video streams and a data storage module to store the necessary information and associated data of the system. + +Based on the requirements and the system analysis, the architecture for the LVTS using PV and AR is shown in Figure 7-1. + +![Figure 7-1 – Architecture for LVTS using PV and AR. The diagram shows five main modules: Publisher module, Management module, Video delivery module, Data storage module, and Audience module. The Publisher module contains three units: Panoramic video acquisition unit, Stream publishing unit, and User authentication unit. The Management module contains four units: Media stream management unit, Media resource management unit, AR editing unit, Account management unit, and System management unit. The Video delivery module contains two units: Video distribution unit and Video transcoding unit. The Data storage module contains three units: Video storage unit, AR element storage unit, and Configuration file storage unit. The Audience module contains four units: Panoramic video display unit, AR element rendering unit, Service control unit, and User authentication unit. Arrows labeled C1 through C8 indicate the flow of data and control between these modules and units. C1 is between the Panoramic video acquisition unit and the Media stream management unit. C2 is between the Stream publishing unit and the Video distribution unit. C3 is between the Management module and the Video delivery module. C4 is between the Management module and the Data storage module. C5 is between the Video distribution unit and the AR element storage unit. C6 is a long arrow at the bottom from the Video delivery module to the User authentication unit in the Audience module. C7 is between the Account management unit and the Panoramic video display unit. C8 is between the AR element storage unit and the Service control unit. A label F.740.8(24) is in the bottom right corner.](042733dc5e8e7f5f30b60adba3266cde_img.jpg) + +Figure 7-1 – Architecture for LVTS using PV and AR. The diagram shows five main modules: Publisher module, Management module, Video delivery module, Data storage module, and Audience module. The Publisher module contains three units: Panoramic video acquisition unit, Stream publishing unit, and User authentication unit. The Management module contains four units: Media stream management unit, Media resource management unit, AR editing unit, Account management unit, and System management unit. The Video delivery module contains two units: Video distribution unit and Video transcoding unit. The Data storage module contains three units: Video storage unit, AR element storage unit, and Configuration file storage unit. The Audience module contains four units: Panoramic video display unit, AR element rendering unit, Service control unit, and User authentication unit. Arrows labeled C1 through C8 indicate the flow of data and control between these modules and units. C1 is between the Panoramic video acquisition unit and the Media stream management unit. C2 is between the Stream publishing unit and the Video distribution unit. C3 is between the Management module and the Video delivery module. C4 is between the Management module and the Data storage module. C5 is between the Video distribution unit and the AR element storage unit. C6 is a long arrow at the bottom from the Video delivery module to the User authentication unit in the Audience module. C7 is between the Account management unit and the Panoramic video display unit. C8 is between the AR element storage unit and the Service control unit. A label F.740.8(24) is in the bottom right corner. + +**Figure 7-1 – Architecture for LVTS using PV and AR** + +The entire system includes the publisher module, the management module, the video delivery module, the data storage module and the audience module. These five parts cooperate with each other to realize the panoramic scenes publishing of tourist attractions, PV stream transmission, AR function management and PV rendering. + +## 7.2 Functional modules + +### 7.2.1 Publisher module + +The publisher module consists of three functional units which are respectively the PV acquisition unit, stream publishing unit and user authentication unit. + +#### 7.2.1.1 Panoramic video acquisition unit + +The panoramic video acquisition unit is responsible for accessing the panoramic camera devices which can capture the PV stream in the scenic area. The device can be fixed or hand-held. After acquiring the live PV stream, this unit will transmit the stream to the stream publishing unit. + +#### 7.2.1.2 Stream publishing unit + +The stream publishing unit is responsible for publishing the live video stream to the video delivery module. After receiving the live stream from the panoramic video acquisition unit, this unit will + +execute the up-streaming preparation function to process this stream in a suitable format in order to send the stream to the video delivery module successfully. + +#### **7.2.1.3 User authentication unit** + +The user authentication unit authenticates the account's identity to verify whether the account has the access rights to the system or not. Only verified publisher accounts can publish the live streams of scenic spots through the system. + +### **7.2.2 Management module** + +The management module consists of five functional units which are respectively the media stream management unit, media resource management unit, AR editing unit, account management unit and system management unit. + +#### **7.2.2.1 Media stream management unit** + +The media stream management unit is responsible for managing the live PV stream. The media stream management unit can: + +- Manage the live PV channel; +- Handle the interaction with the video delivery module; +- Allocate the address of the live PV stream. + +#### **7.2.2.2 Media resource management unit** + +The media resource management unit is responsible for saving, deleting and modifying the different kinds of stored media data used in the system. + +#### **7.2.2.3 AR editing unit** + +The AR editing unit is responsible for editing the AR interactive elements in the PV, and the functions of the AR editing unit include creation, deletion and modification of the AR interactive elements. + +#### **7.2.2.4 Account management unit** + +The account management unit is responsible for creating, deleting, modifying and viewing the user account information of the system. The account management unit can also perform user authentication. + +#### **7.2.2.5 System management unit** + +The system management unit is responsible for monitoring the system running status, warning and handling of anomalies. + +### **7.2.3 Video delivery module** + +The video delivery module consists of two functional units which are respectively the video distribution unit and video transcoding unit. + +#### **7.2.3.1 Video distribution unit** + +The video distribution unit is responsible for delivering the video content to the different viewing terminals through the network. The video distribution unit can determine the video distribution mechanism according to the different transmission conditions. + +#### **7.2.3.2 Video transcoding unit** + +The video transcoding unit is responsible for converting the video from the publisher module into a suitable format for playing in the audience module. + +### **7.2.4 Data storage module** + +The data storage module consists of three functional units which are respectively the video storage unit, AR element storage unit and configuration file storage unit. + +#### **7.2.4.1 Video storage unit** + +The video storage unit is responsible for storing the PV used in the system. + +#### **7.2.4.2 AR element storage unit** + +The AR element storage unit is responsible for storing the AR elements used in the system. The AR elements can be pictures, videos, audios and three-dimensional models. These elements will be used in the AR interactive functions. + +#### **7.2.4.3 Configuration file storage unit** + +The configuration file storage unit is responsible for storing the configuration files of the PV and AR elements. The configuration information in the configuration files can be the information of basic video information and AR effects superimposed on the PV, automatic navigation data and other interactive function information. + +### **7.2.5 Audience module** + +The audience module consists of four functional units which are respectively panoramic video display unit, AR element rendering unit, service control unit and user authentication unit. + +#### **7.2.5.1 Panoramic video display unit** + +The panoramic video display unit is responsible for requesting, decoding and displaying the PV for the users, and it allows users to watch the different fields of view of the online PV. + +#### **7.2.5.2 AR element rendering unit** + +The AR element rendering unit is responsible for requesting and rendering the AR interactive elements according to the previously defined configuration information. The audience can watch the various AR elements superimposed on the PV. + +#### **7.2.5.3 Service control unit** + +The service control unit is responsible for realizing the interactive functions with the audiences. The audience can click the interactive elements rendered on the screen to trigger the corresponding functions such as field of view (FOV) switching of PV, AR hotspot display and automatic navigation. The service control unit can communicate with the other system modules to realize the media service control functions. Specifically, the service control unit can: + +- Handle the media service requests of the audiences; +- Request and receive the media resources from the video delivery module and data storage module; +- Control and execute the interactive functions with the audiences; +- Receive and execute the service control commands sent from the management module. + +#### **7.2.5.4 User authentication unit** + +The user authentication unit authenticates the user's identity to verify whether the user has access rights to the system or not. Only verified audiences can use the live virtual tour functions provided by the system. + +## **7.3 Reference points** + +Figure 7-1 illustrates the reference points related to architecture for LVTS using PV and AR. + +### **7.3.1 Reference point C1** + +The reference point C1 is between the publisher module and the management module. This reference point is used by the publisher module to complete the publisher authentication and request the up-streaming address for the PV. + +### **7.3.2 Reference point C2** + +The reference point C2 is between the publisher module and the video delivery module. This reference point is used by the publisher module to send the live PV stream to the video delivery module. + +### **7.3.3 Reference point C3** + +The reference point C3 is between the management module and the video delivery module. This reference point is used by the management module to send the configuration requests for the video delivery and obtain the video delivery information. + +### **7.3.4 Reference point C4** + +The reference point C4 is between the management module and the data storage module. This reference point is used by the management module to send the storage requests for the media data and configuration files used in the system and obtain the data storage information. + +### **7.3.5 Reference point C5** + +The reference point C5 is between the video delivery module and the data storage module. This reference point is used for the transmission of the panoramic and non-panoramic video. The live PV stream can be sent from the video delivery module to the data storage module for storing the PV data. In addition, the panoramic or non-panoramic videos stored in the data storage module can be sent to the video delivery module for distributing the video to users. + +### **7.3.6 Reference point C6** + +The reference point C6 is between the video delivery module and the audience module. This reference point is used by the video delivery module to send the video streams to the audience module. + +### **7.3.7 Reference point C7** + +The reference point C7 is between the management module and the audience module. This reference point is used by the audience module to complete audience authentication and transmit the media service control commands. + +### **7.3.8 Reference point C8** + +The reference point C8 is between the data storage module and the audience module. This reference point is used by the data storage module to send the AR elements and media configuration files to the audience module. + +# Appendix I + +## Use cases for LVTS using PV and AR + +(This appendix does not form an integral part of this Recommendation.) + +## I.1 Live tour guide through PV + +In this use case, the tour guide captures the PV of the scenic spot in real time through the handheld PV capturing device. The users can follow the location and perspective of the tour guide online through mobile phone, personal computer and other terminal devices to visit the scenic spots in real time. The user can use the terminal device to switch the panoramic viewing angle. By rotating the panoramic viewing angle left and right with the mouse and keyboard, the scenic spots in the corresponding FOV are displayed in the current viewing window. Automatic view switching is also supported. The users can interact and chat with guides in real time to learn about local customs, and they can meet local people and experience the places they want to visit without leaving their homes. + +## I.2 Roaming of multiple PV scenes + +In this use case, multiple panoramic cameras are deployed at the different locations of the scenic spot to collect the real-time PV of the different scenic spots. After processing, the PV is transmitted to the user's viewing terminal. The system supports the roaming of PV scenes, and thus the users can switch the PV scenes according to their personal preferences by clicking the scene label. + +## I.3 Live tour guide through PV with AR effects + +During PV rendering, the LVTS applies AR technology to overlay the automatic and manual explanations of scenic spots (including text, pictures, videos and web pages) on the picture of the corresponding scenic spots in the form of graphic and text hot labels, to enhance the interactive ability of the system. The system can also add particle effects to simulate rainy and snowy scenes. By integrating the various interactive AR elements on the PV frames, users can not only better understand the characteristics and customs of the local scenic spots, but also improve their sense of participation and experience. + +# Bibliography + +- [b-ITU-T J.301] Recommendation ITU-T J.301 (2014), *Requirements for augmented reality smart television systems.* +- [b-ITU-T Y.2012] Recommendation ITU-T Y.2012 (2010), *Functional requirements and architecture of next generation networks.* + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.741-200509-I_PDF-E/0b87abe67b21a93777287649c33e755d_img.jpg b/marked/F/T-REC-F.741-200509-I_PDF-E/0b87abe67b21a93777287649c33e755d_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..84a0ed461a23c44a6bce3145f9f8942779d96ae1 --- /dev/null +++ b/marked/F/T-REC-F.741-200509-I_PDF-E/0b87abe67b21a93777287649c33e755d_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bb2f07f6780ada1ad9ae8595e20124ab4e0a41513e1b1fe87424a1702884ecbd +size 50869 diff --git a/marked/F/T-REC-F.741-200509-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg b/marked/F/T-REC-F.741-200509-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..36cf51f8ba07e44aa7a4b009f3ba08eb2b570b1c --- /dev/null +++ b/marked/F/T-REC-F.741-200509-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:11780faeee75ba39ab4ed9de67a9e67abec0d9ba3068ebec661dea70c16e29d3 +size 4338 diff --git a/marked/F/T-REC-F.741-200509-I_PDF-E/raw.md b/marked/F/T-REC-F.741-200509-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..623ef3f48e34ec6a03edde784e84381a6b0a3a51 --- /dev/null +++ b/marked/F/T-REC-F.741-200509-I_PDF-E/raw.md @@ -0,0 +1,456 @@ + + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.741** + +(09/2005) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Audiovisual services + +--- + +**Service description and requirements for +audiovisual on-demand services** + +ITU-T Recommendation F.741 + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +## **TELEGRAPH SERVICE** + +| | | +|-----------------------------------------------------------------|-------------| +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | + +## **MOBILE SERVICE** + +| | | +|--------------------------------------------------------|-------------| +| Mobile services and multdestination satellite services | F.110–F.159 | +|--------------------------------------------------------|-------------| + +## **TELEMATIC SERVICES** + +| | | +|-------------------------------------------|-------------| +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | + +## **MESSAGE HANDLING SERVICES** + +F.400–F.499 + +## **DIRECTORY SERVICES** + +F.500–F.549 + +## **DOCUMENT COMMUNICATION** + +| | | +|--------------------------------------|-------------| +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | + +## **DATA TRANSMISSION SERVICES** + +F.600–F.699 + +## **AUDIOVISUAL SERVICES** + +**F.700–F.799** + +## **ISDN SERVICES** + +F.800–F.849 + +## **UNIVERSAL PERSONAL TELECOMMUNICATION** + +F.850–F.899 + +### **HUMAN FACTORS** + +F.900–F.999 + +*For further details, please refer to the list of ITU-T Recommendations.* + +## **ITU-T Recommendation F.741** + +# **Service description and requirements for audiovisual on-demand services** + +## **Summary** + +Audiovisual on-demand services are one class of the generic multimedia services identified in ITU-T Rec. F.700. Following the methodology proposed in ITU-T Rec. F.701, this Recommendation describes the functional model, application scenarios, a variety of requirements, and interworking and interoperation of the audiovisual on-demand services. + +###### **Source** + +ITU-T Recommendation F.741 was approved on 13 September 2005 by ITU-T Study Group 16 (2005-2008) under the ITU-T Recommendation A.8 procedure. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications. The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure e.g. interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementors are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database. + +© ITU 2005 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## CONTENTS + +###### Page + +| | | | +|------|-------------------------------------------------------------------------------------------------|---| +| 1 | Scope ..... | 1 | +| 2 | References..... | 1 | +| 2.1 | Normative references..... | 1 | +| 2.2 | Informative references..... | 1 | +| 3 | Definitions ..... | 1 | +| 4 | Abbreviations..... | 2 | +| 5 | Conventions ..... | 2 | +| 6 | Prose description..... | 3 | +| 7 | Functional model of audiovisual on-demand services..... | 3 | +| 7.1 | Functional model ..... | 3 | +| 7.2 | Participants ..... | 4 | +| 8 | Description of audiovisual on-demand service application scenarios..... | 4 | +| 8.1 | Point-to-point unicasting applications ..... | 4 | +| 8.2 | Point-to-multipoint multicasting applications ..... | 5 | +| 8.3 | Audiovisual download and value-added services..... | 6 | +| 9 | Requirements for audiovisual on-demand services ..... | 6 | +| 9.1 | Requirements for services ..... | 6 | +| 9.2 | Requirements for management..... | 7 | +| 9.3 | Requirements for navigation ..... | 7 | +| 9.4 | Requirements for security authentication..... | 7 | +| 9.5 | Requirements for accounting..... | 7 | +| 9.6 | Requirements for QoS ..... | 7 | +| 9.7 | Requirements for DRM ..... | 8 | +| 10 | Related applications..... | 8 | +| 11 | Interworking and interoperation ..... | 8 | +| 11.1 | Interworking and interoperation between different audiovisual on-demand service providers ..... | 8 | +| 11.2 | Interworking between audiovisual on-demand and audiovisual conferencing systems..... | 9 | + + + +# Service description and requirements for audiovisual on-demand services + +# 1 Scope + +This Recommendation describes the audiovisual on-demand services and defines the functional model and application scenarios. Based on these descriptions, it gives the requirements for services, managements, etc. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions, which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +## 2.1 Normative references + +None. + +## 2.2 Informative references + +- ITU-T Recommendation F.700 (2000), *Framework Recommendation for multimedia services*. +- ITU-T Recommendation F.701 (2000), *Guideline Recommendation for identifying multimedia service requirements*. +- ITU-T Recommendation F.740 (1993), *Audiovisual interactive services*. +- ITU-T H-series Recommendations – Supplement 1 (1999), *Application profile – Sign language and lip-reading real-time conversation using low bit-rate video communication*. + +# 3 Definitions + +This Recommendation defines the following terms: + +**3.1 authentication and authorization:** Authentication and authorization refers to validating the identities of users who want to use the services and authorizing the legal users to prevent unauthorized access to the services. + +**3.2 audiovisual multicasting:** When users request the same content at the same time, the audiovisual on-demand services platform lists them in the same multicasting group. The platform sends the requested content media streams to that multicasting group, the users within the same group then may receive the same requested contents. + +**3.3 audiovisual on-demand services:** A kind of network-delivered service that offers the functionality of the home VCR (as play only) without having a need to get a copy of the chosen material. + +**3.4 audiovisual on-demand services platform:** The application environment which supplies audiovisual on-demand services to users. + +**3.5 audiovisual on-demand presented:** One value-added service supplied by the audiovisual on-demand services platform. The person who consumes the content is not the person who requests it. The person who requests the content (as a gift to the viewer) pays for it instead of the person who consumes it. + +**3.6 download services:** Users store the chosen contents in their terminals by downloading them from service platforms in order to play them at any time they like. + +**3.7 electronic program guide (EPG):** Electronic program guide is the assistance tool which helps users to look up the contents they want. It is a kind of metadata of contents which usually consists of program name, program provider, language used by program, etc. Service provider can determine the information the EPG delivers. + +**3.8 near VoD:** Near VoD is a special multicast application that improves the availability of video information, without requiring a dedicated point-to-point connection to each viewer. + +**3.9 information resources retrieval:** It is the retrieval for contents by finding related contents based on the keywords such as, content titles, authors, schedules or actors names. + +**3.10 uniform service portal:** A portal which provides all the users with audiovisual on-demand services information such as authentication and contents retrieval information. + +**3.11 VCR functions:** The functionalities of video cassette recorder, such as select/cancel, start, stop, pause (with or without freeze frame), fast forward, reverse, scan forward or reverse (both with image), and setting and resetting memory marks. + +# 4 Abbreviations + +This Recommendation uses the following abbreviations: + +| | | +|------|------------------------------| +| CP | Content Provider | +| DRM | Digital Right Management | +| EPG | Electronic Program Guide | +| IPTV | Internet Protocol Television | +| PDA | Personal Digital Assistant | +| QoS | Quality of Service | +| SP | Service Provider | +| STB | Set Top Box | +| VCR | Video Cassette Recorder | +| VoD | Video on Demand | + +# 5 Conventions + +In this Recommendation the following conventions are used: + +- "shall" indicates a mandatory requirement. +- "should" indicates a suggested but optional course of action. +- "may" indicates an optional course of action rather than a recommendation that something take place. + +# 6 Prose description + +Audiovisual on-demand services involve asymmetric interactive multimedia communications between users and machines (audiovisual resources) which may provide users with audiovisual information from all around the world, at any time, through telecommunication networks. Users can find the information needed with the assistance of a navigation subsystem. The audiovisual on-demand services platform sends contents information with some grade of guaranteed QoS after the users have chosen the service mode. Users can then use VCR-like functionalities to view the received information. + +The contents provided by the audiovisual on-demand services platform may be textual, aural, graphical or imagery, or video information. Media resources may be stored at any locations on the networks. Audiovisual on-demand services platform integrates media resources, performs DRM and provides navigation service and content delivery through telecommunication networks. + +Users may use audiovisual on-demand services at fixed or mobile locations, such as at homes, offices, or on moving trains. Users should be able to access broadband network environment with streaming media communication capabilities to get audiovisual on-demand service applications using PC, STB plus TV set, mobile phone, PDA or other equipment. + +# 7 Functional model of audiovisual on-demand services + +## 7.1 Functional model + +Audiovisual on-demand services may provide information contents stored at different locations of networks to users by managing and integrating information resources. Audiovisual on-demand services may provide content navigation and retrieval service, provide local caching for remote information resources, and have authentication and accounting functions and security mechanisms. Audiovisual on-demand services may provide IPTV, VoD, Near VoD and download services to users. The functional model of audiovisual on-demand services system is shown in Figure 1. + +![Functional model of audiovisual on-demand services system diagram](0b87abe67b21a93777287649c33e755d_img.jpg) + +The diagram illustrates the functional model of an audiovisual on-demand services system. At the top, three rectangular boxes represent 'Content provider 1', 'Content provider 2', and 'Content provider n'. These are connected by solid lines to a central oval labeled 'Multi-audiovisual on-demand platform'. This platform is enclosed within a larger rectangular box labeled 'Service provider'. Inside this box, to the left of the platform, is a rectangular box labeled 'Control and management', which is connected to the platform by a dashed line. Below the 'Service provider' box, an arrow points down from the 'Control and management' box to the word 'Accounting'. At the bottom center, an oval labeled 'User equipment' is connected to the 'Multi-audiovisual on-demand platform' by a solid line. A dotted line connects the 'Control and management' box to the 'User equipment' oval. A legend at the bottom left defines the line types: a solid line for 'Multimedia stream', a dashed line for 'Control stream', and a dotted line for 'User information (Retrieval and authentication)'. The label 'F.741\_F01' is located in the bottom right corner of the diagram area. + +Functional model of audiovisual on-demand services system diagram + +**Figure 1/F.741 – Functional model of audiovisual on-demand services system** + +An audiovisual on-demand services system is made of content providers, service providers (control and management system and multi-audiovisual on-demand service providers) and end users. + +The control and management system provides the following operational support functions for audiovisual on-demand services: + +- Operational support: it refers to user management, authentication and authorization, accounting and charging management functions; +- Service support: it refers to navigation, ordering contents, resources management, DRM and multicast control functions; +- Network support: it refers to functions such as device management, optimization policy, system maintenance and network security. + +Multi-audiovisual on-demand services provide three modes of services: point-to-point unicasting services, point-to-multipoint multicasting services and download services. Using any one of these three modes of services, users can retrieve the desired contents and get related services. Because the techniques used for these three services modes are different, they may have different features: + +- Point-to-point unicasting services: It takes client/server point-to-point configuration, such as VoD services. Each user has a unicasting media streaming. This mode of services supports VCR-like functionalities such as play, pause, replay, fast forward and fast reverse. +- Point-to-multipoint multicasting services: It takes point-to-multipoint configuration. The services platform multicasts media streams to a multicasting group based on users' requests. Users who are in the multicasting group receive the same media streams. This mode of services does not support VCR-like functionalities, but users can exit from the multicasting group during the use of the services. +- Download services: A kind of service which provides contents to users by downloading and storing. It may use distributed storage methods and multipoint download techniques to avoid a server bottleneck resulting from multiple pieces of content downloading at the same time. +- Service providers may provide all kinds of value-added services, such as audiovisual on-demand as a gift, gaming or pre-ordered contents. + +## 7.2 Participants + +Participants of audiovisual on-demand services have the following three main roles as shown in Figure 1: + +- Content Providers (CP): CP refers to the owner of information resources. They may be issuers of music, movies, live programs or common information resources (such as museums and public libraries). Generally speaking, content providers provide contents lists and information contents through networks or other paths to service providers, for users' retrieval and request at any time. +- Service Providers (SP): SP provides information navigation and retrieval windows and requested contents information to users, and takes charge of security authentication and accounting of the whole system, network resources management and maintains management functions. +- End users: End users are audiovisual on-demand services requesters and content information consumers, who pay for the services they get. + +# 8 Description of audiovisual on-demand service application scenarios + +## 8.1 Point-to-point unicasting applications + +Users can request contents through an electronic program guide or contents retrieval, according to their needs, anytime and anywhere. They can use VCR-like functionalities, such as fast-forward, fast-reverse and pause, to control the receipt of the contents. The information resources are general + +stored contents. The users' locations may be fixed or mobile. The following shows a typical case of a point-to-point unicasting application. + +User A is a tour guide who needs to learn something about the historical, natural and cultural knowledge of certain countries around the world. Most of the time, he obtains such knowledge with the assistance of audiovisual on-demand services and by watching the related information from audiovisual resources. He is able to view such topics as news of current affairs, scenes of natural beauty, nature and films or movies about the culture. The following week, he will be taking a group of people to visit Europe. The application scenario is as follows: + +Step 1: User A turns on his PC, makes a connection to an SP through a telecommunication network at his office. User A wants to request one short program introducing the landscape of his destination. He begins by selecting the name of the program, the language he prefers and indicating whether or not he would like to have the text synchronized with the video. + +Step 2: After he sends the request command to the SP, the SP prompts him to authenticate. As soon as he passes the authentication he can receive the film with high QoS. + +Step 3: While watching the film, he receives a telephone call. So he presses the pause button and stops the transmission whilst he answers the phone. After a few minutes, he resumes watching. But he has forgotten the last part of the content before the phone call. He decides to reverse the content to review the portion he forgot, and resumes watching from where he paused. + +Step 4: After he finishes watching the requested content, User A requests an old film. Because he does not exit from the SP, he may request the new film without again having to authenticate himself. The SP assigns a media server nearest to him to provide media streaming to him. + +Step 5: User A needs to go outside before he finishes watching the whole film. He sends a request to the SP and asks to roam to his multimedia mobile phone in order to continue to watch the rest of the film. As soon as the film ends, User A exits from the SP. + +Step 6: User A returns to his office, he accesses the original SP again. He finds another interesting program on another SP through the navigation system. He requests that program. The second SP prompts him to authenticate. After he passes the authentication, he gets the authorization to request the program with the accounting function performed between the two SPs. + +Step 7: While he is waiting for the requested film, he receives some advertisements from the SP before the film. Then he receives the content and watches it. User A exits from the system after he finishes watching the film. + +## **8.2 Point-to-multipoint multicast applications** + +IPTV and Near VoD take a point-to-multipoint multicast configuration. For IPTV applications, contents are multicast or broadcast in fixed-time and fixed-channel (multicast IP address). While users use IPTV applications, they can view that content by joining that channel at broadcasting time. For Near VoD applications, service providers decide to open channels according to users' requests. The information resources of multicast contents may be real-time contents or stored contents with authorization. The following describes a multicast application scenario: + +Step 1: User A often watches the programs broadcast on one particular IPTV channel in the evening. One evening he decides to watch some programs from audiovisual on-demand services providers. He makes a connection to an SP. After he passes authentication, he chooses the "Today's News" channel and begins to watch this program on his terminal. Because User A is learning English, and because "Today's News" is in Chinese, he chooses to have the text in English with the video shown on the screen. + +Step 2: User A sits on his sofa watching the news just the same way as he watches the news in front of a TV set. + +Step 3: User A has a habit of watching the programs introductions, and hopping between the channels. He finds there will be a live football match on another channel within a few minutes. He switches to that channel. Because the football match will only begin in a few minutes, he switches back to the news channel again. + +Step 4: User A wants to watch two programs at the same time, so he sends a request to the SP to watch the second channel with the football match and divides the screen into two parts. He may watch two programs on one screen at the same time. However, if he wants to watch one of them clearly, he clicks the window of that program and turns it into the whole screen. If he is interested in more programs at the same time, he can request the information from several channels provided by the audiovisual on-demand services, and watch several programs at the same time. + +Step 5: The next day, one of user A's colleagues tells him that there is a videoconference on intellectual property rights. User A is interested in it, so he requests to listen to the live videoconference through an SP which connects to a videoconference services platform. + +Step 6: The videoconference ends, User A exits and disconnects from the service provider. + +## **8.3 Audiovisual download and value-added services** + +To protect the digital copyright of information resources, they should be encrypted. Audiovisual on-demand value-added services include audiovisual pre-ordering, audiovisual on-demand as a gift service, etc. The initiators and consumers of services may be the same persons or different persons. The scenario is as follows: + +Step 1: User A is on vacation. He turns on his PC and finds an e-mail notice from an audiovisual on-demand services provider saying that "Your friend "XX" has ordered a film "XXXX" for you, the password is XXXXXXX. Hope you enjoy it. If you want to get it now, please click on its name." User A wants to continue to read other e-mails, so he does not click immediately. + +Step 2: After User A exits from his e-mail system, he clicks on the film's name shown on a moving bar and enters the password to watch the film after he finishes reading the notice with the film. + +Step 3: The film his friend ordered for him is one of a series of films. He is very interested in it, so he searches for the next one of the series while he is watching the first one. He chooses the next film on the list of films and selects to receive it in download mode. The SP prompts him to authenticate. After user A passes the authentication, he gets the security key to the film. User A downloads that film while he is watching the first film. + +Step 4: User A orders a new film from EPG and hopes to watch it in the evening. He opens a pre-order window and enters the content name, play time, etc. + +Step 5: He wishes to send this new film to his friend as a present. So he chooses audiovisual on-demand presentation services to order the new film for his friend. He directly enters the film name together with his friend's e-mail address and pays for the gift content, since he has already passed the authentication. User A exits from the SP. + +# **9 Requirements for audiovisual on-demand services** + +## **9.1 Requirements for services** + +To realize the applications scenarios described in clause 8, their services requirements include the following: + +- 1) To provide flexible requesting modes, including requesting on-demand, multicasting and download. +- 2) To provide EPG and program introductions, allowing customized services of program contents ordering and retrieval. + +- 3) To provide the roaming of the media streaming between different terminals of one user. +- 4) To have control capabilities to achieve VCR-like functionalities. +- 5) SP may provide value-added services e.g., content as a gift. +- 6) To provide insertion of advertisements. +- 7) To provide the interoperation and internetworking with other services. + +## **9.2 Requirements for management** + +To ensure QoS and protect the copyrights of digital contents, the following requirements for management should be met: + +- 1) Audiovisual on-demand service providers can order contents from contents providers online or offline. +- 2) There should be a reasonable mechanism to optimize contents caching and load balancing. +- 3) There should be methods to manage terminals. +- 4) There should be some methods for managing network security. + +## **9.3 Requirements for navigation** + +An audiovisual on-demand services platform should provide a uniform services portal and an EPG and provide searching functions based on key words entered. An EPG should include the program name and audio and text synchronized with the video in different languages. + +## **9.4 Requirements for security authentication** + +An SP should provide authentication services to users. Only users who receive authorization can use audiovisual on-demand services in order to ensure effective accounting and to prevent illegal users from accessing the audiovisual on-demand services. An SP also should be able to prevent the audiovisual on-demand services from illegal malicious attack. + +The audiovisual on-demand services may support the mobility of users terminals. A user can get authorization as long as he/she enters his/her personal registration number at any location on any terminal. + +## **9.5 Requirements for accounting** + +Accounting for users includes two parts. One part is for content providers who provide contents to users, and the other part is for service providers who provide services to users. + +An SP should have reasonable distribution policies that enable charging users to ensure the content providers' profit, and use scientific accounting policies to ensure that users are fairly charged for the services and content they use. + +Accounting and charge settlement is essential for the interoperation between an SP and other service and content providers. To support the efficient operation of sharing audiovisual on-demand services, SPs should provide reasonable accounting services between different service providers and should be able to offer different accounting and pricing policies to users. + +## **9.6 Requirements for QoS** + +The QoS of audiovisual on-demand services is mainly determined by the quality of media devices, system storage policies, edge pre-caching and optimization, media compression techniques, etc. Cooperation between the network environments and services environments affects the QoS too. The control and management capability for multicasting services is also an important factor to QoS. The requirements for services are: + +- Clear and fluent picture effect. + +- The same viewing method as that of TV. +- Ease of operation and ease and speediness of the channel switchover. + +The main parameters of QoS are quality of audiovisual media, response time, speed of program switchover, response capability of interactive operation, controlling multicasting period, etc. + +## **9.7 Requirements for DRM** + +To ensure the content providers profit, an SP should employ necessary DRM mechanisms for the resources of content providers and save the authorization information of agreements between participants, distribute security keys and send the necessary information for using resources each time. + +The resources of content providers may be filtered according to the service provider's policy. + +# **10 Related applications** + +Audiovisual on-demand services can provide the following related applications: + +- Entertainment applications: live broadcasting, movies and TV film series on-demand, Karaoke on-demand and interactive games. +- Monitoring-type applications: Special on-demand contents, regarding e.g., homes, schools and day-care environments. +- Advertisement-type applications: related to general audiovisual advertisement services applications. +- Audiovisual-type applications: museums on net and exhibition of auction goods on net; +- Text-type applications: digital libraries, digital painting and calligraphy, etc. +- Educational-type applications: distance learning, telemedicine, etc. +- Electronic commerce: goods exhibitions at electronic shops. + +# **11 Interworking and interoperation** + +## **11.1 Interworking and interoperation between different audiovisual on-demand service providers** + +Many service providers can provide audiovisual on-demand services on the same open telecommunication network with their own access, navigation, authentication and accounting systems. All the service providers can be accessed by all the users equally. Interworking and interoperation between different audiovisual on-demand service providers can use the following methods: + +- Users of one audiovisual on-demand service provider can get the services from another audiovisual on-demand service provider. An accounting service should be provided based on the agreements between the two SPs. +- One SP may provide navigation services to users, while another provides the contents. + +## **11.2 Interworking between audiovisual on-demand and audiovisual conferencing systems** + +For both audiovisual on-demand services and videoconference services handling audiovisual information, most of the time users do not store the information during the videoconference session and cannot watch it after the conference. If the audiovisual on-demand services platform stores the information of the videoconference during the videoconference session, users can request either to join the videoconference during the conference or retrieve it after the videoconference. Because each of these services has its own special characteristics, the necessary conversion functions are as follows: + +- Transforming from one encoding format to another. +- Transforming from one transmission protocol to another. +- If audiovisual conference records are to be used as resources for VoD or Near VoD services, the audiovisual on-demand service platform should provide the necessary equipment to store the audiovisual conference information and transform the information formats to suit VoD and Near VoD services requirements. + + + + + +# SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|---------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | General tariff principles | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects and next-generation networks | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.742-202503-I_PDF-E/0b87abe67b21a93777287649c33e755d_img.jpg b/marked/F/T-REC-F.742-202503-I_PDF-E/0b87abe67b21a93777287649c33e755d_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f7f9493bffc9c33496ad4d5faf635a57ab6358ac --- /dev/null +++ b/marked/F/T-REC-F.742-202503-I_PDF-E/0b87abe67b21a93777287649c33e755d_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:322af49e53828cfc7b08805f236d03c1e2eb04f7673abdf304b332f51963c678 +size 15856 diff --git a/marked/F/T-REC-F.742-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.742-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..abda52cbb1047d6649f89e40c898a1c33bed7267 --- /dev/null +++ b/marked/F/T-REC-F.742-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6f3b7e07231c0f3e116c99c493bdda98ecd52313b6735faa6cea23815cf3162b +size 7190 diff --git a/marked/F/T-REC-F.742.1-202212-I_PDF-E/33a8f3f01dfa8bce75d23017855a13c5_img.jpg b/marked/F/T-REC-F.742.1-202212-I_PDF-E/33a8f3f01dfa8bce75d23017855a13c5_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7761b057c2af2c0eabde33c1e064af1413f62167 --- /dev/null +++ b/marked/F/T-REC-F.742.1-202212-I_PDF-E/33a8f3f01dfa8bce75d23017855a13c5_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6be54e298da846f50f7be7b4fad088f489d557d9c842c0373ba74cc752ac1613 +size 98887 diff --git a/marked/F/T-REC-F.742.1-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.742.1-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d4d8969f805414689ab79cfd08c333d454d482b3 --- /dev/null +++ b/marked/F/T-REC-F.742.1-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:636c3839edff955a5b6208530b05fa1c983ab153d90f00216517a7a99478b988 +size 7074 diff --git a/marked/F/T-REC-F.742.1-202212-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg b/marked/F/T-REC-F.742.1-202212-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..863607152a0634529898c62088fa920773826fc7 --- /dev/null +++ b/marked/F/T-REC-F.742.1-202212-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2189f2404d662593a1b0db17328c47e1f9ff9e54d7400e4e340dc215c69fb9ac +size 42903 diff --git a/marked/F/T-REC-F.742.1-202212-I_PDF-E/c78c2eefd86269d1740ab85a916f24f2_img.jpg b/marked/F/T-REC-F.742.1-202212-I_PDF-E/c78c2eefd86269d1740ab85a916f24f2_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..52535151139e8ac31d963a43eec2e0c5d6a5cc9d --- /dev/null +++ b/marked/F/T-REC-F.742.1-202212-I_PDF-E/c78c2eefd86269d1740ab85a916f24f2_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:43d347f945b8a56ecb6a4ae7f776a20dd5cb83768832e8650e738d8e52799e41 +size 74765 diff --git a/marked/F/T-REC-F.742.2-202401-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg b/marked/F/T-REC-F.742.2-202401-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d4d8969f805414689ab79cfd08c333d454d482b3 --- /dev/null +++ b/marked/F/T-REC-F.742.2-202401-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:636c3839edff955a5b6208530b05fa1c983ab153d90f00216517a7a99478b988 +size 7074 diff --git a/marked/F/T-REC-F.742.2-202401-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.742.2-202401-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f0bdc8d0cabea7838748870cff557024dc8e76f4 --- /dev/null +++ b/marked/F/T-REC-F.742.2-202401-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6d15f473197cddd04c9f247075f583ef8562ddfcce9764ec5cd1289a384252cc +size 101710 diff --git a/marked/F/T-REC-F.743.1-201504-I_PDF-E/10d19f166f9b7f5961f09f8041896943_img.jpg b/marked/F/T-REC-F.743.1-201504-I_PDF-E/10d19f166f9b7f5961f09f8041896943_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..0318e9c4413a848e0415fc664cd12d6ab8a2c5d3 --- /dev/null +++ b/marked/F/T-REC-F.743.1-201504-I_PDF-E/10d19f166f9b7f5961f09f8041896943_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:44308fa8021492401572dd08c2bdb3ee000fde1266ee0e449af53d14e53acaf6 +size 31544 diff --git a/marked/F/T-REC-F.743.1-201504-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.743.1-201504-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9002ee68e2b9ebf63d27d39ba8c73bdafb0e6be6 --- /dev/null +++ b/marked/F/T-REC-F.743.1-201504-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8df701536c25ab8246d2868de3c20c8a4e25168bb52c5eb9078278ab1b5b4fd8 +size 6859 diff --git a/marked/F/T-REC-F.743.1-201504-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg b/marked/F/T-REC-F.743.1-201504-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3e06a45cc5248621abb69409ceee03e335851257 --- /dev/null +++ b/marked/F/T-REC-F.743.1-201504-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e77758de74b6dd451dbb6818f3c3a94340c823a65e1b700e70dbea1377519e4f +size 57803 diff --git a/marked/F/T-REC-F.743.1-201504-I_PDF-E/349ca0a6a9c2e2651a4deeeaf8be6da1_img.jpg b/marked/F/T-REC-F.743.1-201504-I_PDF-E/349ca0a6a9c2e2651a4deeeaf8be6da1_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ec7ab13ddb2a89c313ef6b8fc00f97d76c34fca3 --- /dev/null +++ b/marked/F/T-REC-F.743.1-201504-I_PDF-E/349ca0a6a9c2e2651a4deeeaf8be6da1_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:045a5caf75720648b8d3d5db6ae9e6ad7dc60032350a7ab462ef96f24e6d8b0a +size 35473 diff --git a/marked/F/T-REC-F.743.1-201504-I_PDF-E/9999a06627f39a35b608912ed95b09ac_img.jpg b/marked/F/T-REC-F.743.1-201504-I_PDF-E/9999a06627f39a35b608912ed95b09ac_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9626a15d63ba0ade27a0b02e1e47e12562e66003 --- /dev/null +++ b/marked/F/T-REC-F.743.1-201504-I_PDF-E/9999a06627f39a35b608912ed95b09ac_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:49438b18ca78183c79c5fced13c145115f0f450334213f00431ba51584c1b935 +size 33301 diff --git a/marked/F/T-REC-F.743.1-201504-I_PDF-E/9c6461e1e94afae4dec455e69a2ce152_img.jpg b/marked/F/T-REC-F.743.1-201504-I_PDF-E/9c6461e1e94afae4dec455e69a2ce152_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..703635c789c09b44efee39d905618756566e74b0 --- /dev/null +++ b/marked/F/T-REC-F.743.1-201504-I_PDF-E/9c6461e1e94afae4dec455e69a2ce152_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6f25b7cc4fa4940295483ca4c29651a0359ad923f72c7e16f4fb24faf1e18159 +size 71636 diff --git a/marked/F/T-REC-F.743.1-201504-I_PDF-E/a8a65eb4968947846b288d693535f03a_img.jpg b/marked/F/T-REC-F.743.1-201504-I_PDF-E/a8a65eb4968947846b288d693535f03a_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..e3961fb1c89f403745e790a8d15799f02efe8a01 --- /dev/null +++ b/marked/F/T-REC-F.743.1-201504-I_PDF-E/a8a65eb4968947846b288d693535f03a_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0c498096bf85c6d21df886f42ffe2aa55b17d3a96a5b4c15400de0499f496f01 +size 24209 diff --git a/marked/F/T-REC-F.743.1-201504-I_PDF-E/ac852a162572ca8a8c8478c49b571af5_img.jpg b/marked/F/T-REC-F.743.1-201504-I_PDF-E/ac852a162572ca8a8c8478c49b571af5_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..e7401c3b67997e4072a2c62b7c9ba98ad32249a3 --- /dev/null +++ b/marked/F/T-REC-F.743.1-201504-I_PDF-E/ac852a162572ca8a8c8478c49b571af5_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2fefe879140ff4c4ff8ada3ae3a6192a0549b47ffcc211ab40e0a62027d61428 +size 79921 diff --git a/marked/F/T-REC-F.743.1-201504-I_PDF-E/b7c99649d8de39a814e0bad989d50b12_img.jpg b/marked/F/T-REC-F.743.1-201504-I_PDF-E/b7c99649d8de39a814e0bad989d50b12_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..fee6cebe31a5348996c69e2d7574c8a475198f01 --- /dev/null +++ b/marked/F/T-REC-F.743.1-201504-I_PDF-E/b7c99649d8de39a814e0bad989d50b12_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:29235d64a948a67cbf480123890f9047c10a6a074c3b03ec000c19a57aef0fb8 +size 29846 diff --git a/marked/F/T-REC-F.743.1-201504-I_PDF-E/b8d3d85b1575ec6abdcf1bd9ce438701_img.jpg b/marked/F/T-REC-F.743.1-201504-I_PDF-E/b8d3d85b1575ec6abdcf1bd9ce438701_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..e2e23b460f632c91c426e9bdf1580b8c35bdb5eb --- /dev/null +++ b/marked/F/T-REC-F.743.1-201504-I_PDF-E/b8d3d85b1575ec6abdcf1bd9ce438701_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d47877fd0ec19acef92cb9bec6318c5e5069d4ffd1b969f39639e2126da88a52 +size 80111 diff --git a/marked/F/T-REC-F.743.1-201504-I_PDF-E/bb6d33498937738ff5dac8d91c9ebaad_img.jpg b/marked/F/T-REC-F.743.1-201504-I_PDF-E/bb6d33498937738ff5dac8d91c9ebaad_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..46a10f1c1b4b9b11f209afe7bbf8b4ca4f9840e0 --- /dev/null +++ b/marked/F/T-REC-F.743.1-201504-I_PDF-E/bb6d33498937738ff5dac8d91c9ebaad_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e54bbbd94b61936583343b1f6356186ef798152cd2d8f2ead8a6e11dce4ebcca +size 28257 diff --git a/marked/F/T-REC-F.743.1-201504-I_PDF-E/d29cfbf30a471dc06a78be27f86bd1cf_img.jpg b/marked/F/T-REC-F.743.1-201504-I_PDF-E/d29cfbf30a471dc06a78be27f86bd1cf_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f54d271e665f57535f104503126411edd0e28274 --- /dev/null +++ b/marked/F/T-REC-F.743.1-201504-I_PDF-E/d29cfbf30a471dc06a78be27f86bd1cf_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c913a20346316e44c48ece2ec7a113d2a8592b3e7626edb3b29c5c9e97299d5b +size 56135 diff --git a/marked/F/T-REC-F.743.1-201504-I_PDF-E/d336e7ffee4f537d0805ca840ec28582_img.jpg b/marked/F/T-REC-F.743.1-201504-I_PDF-E/d336e7ffee4f537d0805ca840ec28582_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ce6952ed1496c8f62fca62a96d8b3b0dc3aae1e3 --- /dev/null +++ b/marked/F/T-REC-F.743.1-201504-I_PDF-E/d336e7ffee4f537d0805ca840ec28582_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ce26ca3dd62d3b3ec9a95381d3a8ffd3ec71d593a1a5ca2a2822666471f293cc +size 81385 diff --git a/marked/F/T-REC-F.743.1-201504-I_PDF-E/d5a837fa4f4675e5ee596003cf55985c_img.jpg b/marked/F/T-REC-F.743.1-201504-I_PDF-E/d5a837fa4f4675e5ee596003cf55985c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1b102dee2b496efd07f63e3145be1ecd35d41427 --- /dev/null +++ b/marked/F/T-REC-F.743.1-201504-I_PDF-E/d5a837fa4f4675e5ee596003cf55985c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dcf67f8a30ddeaebba3d2cdcf9448ceabdddc87c7562c60393cd3b83576bf6aa +size 26949 diff --git a/marked/F/T-REC-F.743.1-201504-I_PDF-E/dc22d132f81aafd8d8e2ebae77bdbedd_img.jpg b/marked/F/T-REC-F.743.1-201504-I_PDF-E/dc22d132f81aafd8d8e2ebae77bdbedd_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..62bb23e906ce390e0270b90a78c3c56ad0d3dea0 --- /dev/null +++ b/marked/F/T-REC-F.743.1-201504-I_PDF-E/dc22d132f81aafd8d8e2ebae77bdbedd_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ab3d54653f4a5d2688da6895a3445a6c33654b0e4de377c979878ffdd86730b4 +size 31133 diff --git a/marked/F/T-REC-F.743.1-201504-I_PDF-E/e3b8510f6a2194e250205ab7bc38076d_img.jpg b/marked/F/T-REC-F.743.1-201504-I_PDF-E/e3b8510f6a2194e250205ab7bc38076d_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..0a3b6db43dcef8cdf21f9b6cc0d335034972440f --- /dev/null +++ b/marked/F/T-REC-F.743.1-201504-I_PDF-E/e3b8510f6a2194e250205ab7bc38076d_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d625d2e2d4196d75bb73c4003382404f5a0598373a47b7598b84ed97bf8c30fb +size 24007 diff --git a/marked/F/T-REC-F.743.1-201504-I_PDF-E/e7cb11f042fc58088dff4b6d9306845e_img.jpg b/marked/F/T-REC-F.743.1-201504-I_PDF-E/e7cb11f042fc58088dff4b6d9306845e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..427b90f123519667a17202e7fda32413891fa946 --- /dev/null +++ b/marked/F/T-REC-F.743.1-201504-I_PDF-E/e7cb11f042fc58088dff4b6d9306845e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:817c66a2d99a93c30611f592dc342c38325fdbcacc2cc48c05c4f0d8db678670 +size 20021 diff --git a/marked/F/T-REC-F.743.1-201504-I_PDF-E/eafe8d3fa89d770007168b5d2e09e87a_img.jpg b/marked/F/T-REC-F.743.1-201504-I_PDF-E/eafe8d3fa89d770007168b5d2e09e87a_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..daecce60c39458b9dfe43a11ae67d897998254d9 --- /dev/null +++ b/marked/F/T-REC-F.743.1-201504-I_PDF-E/eafe8d3fa89d770007168b5d2e09e87a_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:05c83c1c9cd6b562d4f252a62ba68ee4f89ca69d4ca459fbfd1bfbe19f783396 +size 43642 diff --git a/marked/F/T-REC-F.743.1-201504-I_PDF-E/fbfaa723129f8b8e3be651c6fe7cfab7_img.jpg b/marked/F/T-REC-F.743.1-201504-I_PDF-E/fbfaa723129f8b8e3be651c6fe7cfab7_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..4bc187ec3f9ff45ac893e9a988751b14044efd7c --- /dev/null +++ b/marked/F/T-REC-F.743.1-201504-I_PDF-E/fbfaa723129f8b8e3be651c6fe7cfab7_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d429b25a94b45336efebb7e96375f8ca19e11e6e5a8219603d80b14c7be90560 +size 65200 diff --git a/marked/F/T-REC-F.743.1-201504-I_PDF-E/raw.md b/marked/F/T-REC-F.743.1-201504-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..794000e6ad614cb8ef566a6800494ed5138ebd93 --- /dev/null +++ b/marked/F/T-REC-F.743.1-201504-I_PDF-E/raw.md @@ -0,0 +1,638 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.743.1** + +(04/2015) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Audiovisual services + +--- + +**Requirements for intelligent visual surveillance** + +Recommendation ITU-T F.743.1 + +ITU-T + +![ITU logo](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) features a globe with a red lightning bolt striking it, symbolizing global connectivity and telecommunications. The letters "ITU" are prominently displayed in blue and red. + +ITU logo + +International +Telecommunication +Union + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| AUDIOVISUAL SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +## Recommendation ITU-T F.743.1 + +## Requirements for intelligent visual surveillance + +## Summary + +Recommendation ITU-T F.743.1 defines requirements for intelligent visual surveillance (IVS). The requirements are based on identifying specific objects, behaviours or attributes in video signals. The IVS system transforms the video signals into structured data, which can be transmitted or archived so that the video surveillance system can act accordingly. + +This Recommendation defines the scenarios, the reference architecture and the requirements for IVS. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.743.1 | 2015-04-29 | 16 | 11.1002/1000/12450 | + +## Keywords + +Intelligent visual surveillance, IVS requirement, IVS scenarios. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2015 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations..... | 2 | +| 5 Scenarios for intelligent visual surveillance ..... | 3 | +| 5.1 Target recognition ..... | 3 | +| 5.2 Event recognition and intelligent alarm ..... | 4 | +| 5.3 Data analysis..... | 7 | +| 5.4 Intelligent video diagnosis..... | 10 | +| 5.5 Intelligent video retrieval ..... | 11 | +| 5.6 Intelligent video synopsis ..... | 12 | +| 6 Reference architecture ..... | 12 | +| 7 Requirements for an IVS ..... | 14 | +| 7.1 User requirements..... | 14 | +| 7.2 Service requirements ..... | 16 | +| Bibliography..... | 17 | + + + +## Recommendation ITU-T F.743.1 + +## Requirements for intelligent visual surveillance + +## 1 Scope + +This Recommendation defines requirements for intelligent visual surveillance (IVS), which are based on identifying specific objects, behaviours or attribute in video signals. + +The visual surveillance service (see [b-ITU-T F.743]) is a telecommunication service focusing on audiovisual application technology that is used to capture remote multimedia information (such as audio signals, video signals, images and various alarm signals) and to present it to end users in friendly manner, based on managed broadband networks with ensured quality, security and reliability. + +IVS systems can automatically identify specific objects, behaviours or attributes in video signals. They extract data from the video signals, which are then transmitted or archived so that the visual surveillance system can act accordingly. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T H.626] Recommendation ITU-T H.626 (2011), *Architectural requirements for visual surveillance*. + +## 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following term defined elsewhere: + +**3.1.1 visual surveillance** [ITU-T H.626]: A telecommunication service focusing on video (but including audio) application technology, which is used to remotely capture multimedia (such as audio, video, image, alarm signal, etc.) and present them to the end user in a user-friendly manner, based on a managed broadband network with ensured quality, security and reliability. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 boundary detection:** see *perimeter prevention*. + +**3.2.2 directed tripwire intrusion:** Alarm output that is based on a polygonal region defined by the user. An alarm can be triggered when an object enters or leaves the polygon. + +**3.2.3 event recognition:** Method that recognizes and analyses the events in the scene, and describes them using natural language. + +**3.2.4 facial recognition:** Detection of the human face in video and extraction of its characteristics. The extracted information and/or image are compared with other images or a database of individuals' images, generating comparison results. Facial recognition can be used in airports, ports, railway stations, coach stations and other public places. + +**3.2.5 intelligent video diagnosis:** Automatic recognition of video images failures, such as snowflakes, scrolling or fuzzy video, partial colour, picture freeze, gain imbalance, pan/tilt/zoom (PTZ) control problems and video signal loss. + +**3.2.6 intelligent video retrieval:** Identification of useful or required information from videos. The search criteria can include target category, colour and velocity. + +**3.2.7 intelligent video synopsis:** A simple summary of video content that can be generated in an automatic or semi-automatic way. First, the target motion is analysed and moving targets are extracted. Then, the different targets are spliced to the same background scenes, and combined in some way. + +**3.2.8 perimeter prevention; boundary detection:** An intelligent surveillance scenario where target objects, such as persons or transportation, can be identified in key areas. Through a virtual cordon or virtual warning region, it can also identify the direction of a target. + +**3.2.9 region detection:** Alarm output that is based on the polygonal regions defined by the user and that is triggered when an object enters or leaves the region. + +**3.2.10 region stranded:** Alarm output that is based on the polygonal regions defined by the user and that is triggered when the retention time of an object exceeds a prescribed threshold. + +**3.2.11 target recognition:** A method that separates an image region from the image background and uses some image feature value to classify the target types. + +**3.2.12 traffic/pedestrian flow analysis:** Motion analysis and feature classification of the video of moving target(s) within the specified single or multiple video monitoring area. The pedestrian flow data can be detected accurately. + +**3.2.13 tripwire intrusion:** The setting off of an alarm when an object crosses a line drawn within the surveillance area. + +## 4 Abbreviations + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-----|---------------------------------| +| APP | Application | +| CIV | Client Intelligent Video | +| CMS | Centre Management System | +| CMU | Centre Management Unit | +| CU | Client Unit | +| ICU | Intelligent Customer Unit | +| IPU | Intelligent Premises Unit | +| IVM | Intelligent Video Management | +| IVS | Intelligent Visual Surveillance | +| IVU | Intelligent Video Unit | +| MDU | Media Distribution Unit | +| MSU | Media Storage Unit | +| NTP | Network Time Protocol | +| PIV | Premises Intelligent Video | +| PTZ | Pan/Tilt/Zoom | + +| | | +|-----|-----------------------| +| PU | Premises Unit | +| SCU | Service Control Unit | +| SMS | Short Message Service | + +## 5 Scenarios for intelligent visual surveillance + +This clause describes various scenarios for intelligent visual surveillance (IVS). + +## 5.1 Target recognition + +Target recognition separates the image region from the image background, using some image feature value to achieve the classification of the target type; the characteristics of target classification include space and time attributes. The *space attribute* comprises target contour, target size and target texture. The *time attribute* includes the change of target size and velocity. Accordingly, all targets are classified as follows: persons, vehicles or miscellaneous objects. + +#### 5.1.1 Person identification + +The person identification system identifies human characteristics, which are different from those of vehicles and miscellaneous objects. + +#### 5.1.2 Vehicle (car, ship, aircraft) recognition + +Vehicle recognition can identify the attributes of a vehicle, such as the type and colour. The type can be recognized using rules about the size of the vehicle. A vehicle smaller than a certain size might be a car and something larger might be a truck. In addition, the colour of the vehicle can be used for recognition, but it might be influenced by the ambient light. See Figure 5-1. + +![A photograph of a black car on a city street, with a yellow outline highlighting its shape for target recognition. In the background, there is a red and white bus and other vehicles.](b7c99649d8de39a814e0bad989d50b12_img.jpg) + +A photograph of a black car on a city street, with a yellow outline highlighting its shape for target recognition. In the background, there is a red and white bus and other vehicles. + +Figure 5-1 – Target recognition (vehicle) + +#### 5.1.3 Miscellaneous objects + +As shown in Figures 5-2 and 5-3, miscellaneous objects include moving objects that are not persons or vehicles; e.g., animals, smoke, fire and pollutants. Specifically, smoke and fire recognition can be used in forests, tunnels, oil, petrol and gas stations. As shown in Figure 5-3, this system can distinguish between smoke and other objects that are moving in the video. + +![A small, white, fluffy object, possibly a bird or a piece of clothing, lying on a paved surface. It is highlighted by a yellow irregular outline, indicating it has been recognized as a target object.](e7cb11f042fc58088dff4b6d9306845e_img.jpg) + +A small, white, fluffy object, possibly a bird or a piece of clothing, lying on a paved surface. It is highlighted by a yellow irregular outline, indicating it has been recognized as a target object. + +Figure 5-2 – Target recognition (miscellaneous object) + +![Two side-by-side images. The left image shows a red dumpster with a large plume of white smoke rising from it, outlined in red. The word 'Smoke' is written in red at the bottom. The right image shows a fire inside a building, with a person standing nearby. The word 'FIRE' is written in red at the bottom.](349ca0a6a9c2e2651a4deeeaf8be6da1_img.jpg) + +Two side-by-side images. The left image shows a red dumpster with a large plume of white smoke rising from it, outlined in red. The word 'Smoke' is written in red at the bottom. The right image shows a fire inside a building, with a person standing nearby. The word 'FIRE' is written in red at the bottom. + +Figure 5-3 – Target recognition (miscellaneous object) + +### 5.2 Event recognition and intelligent alarm + +#### 5.2.1 Perimeter prevention + +Perimeter prevention (also known as *boundary detection*) identifies target objects (such as persons or vehicles) in key areas defined by a virtual cordon or virtual warning region. Perimeter prevention can also identify the direction of the target. + +- 1) Region detection: Alarm output must be based on the polygonal regions defined by the user. When an object enters or leaves a region, an alarm can be triggered. See Figure 5-4. + +![A surveillance image showing a paved area with a large green rectangular region outlined. Inside this region, there are two people. One person is walking towards the left, and another is standing. The region is marked with a blue 'X' and a green line, indicating it is a monitored area. The image also shows a bicycle rack in the bottom left corner.](dc22d132f81aafd8d8e2ebae77bdbedd_img.jpg) + +A surveillance image showing a paved area with a large green rectangular region outlined. Inside this region, there are two people. One person is walking towards the left, and another is standing. The region is marked with a blue 'X' and a green line, indicating it is a monitored area. The image also shows a bicycle rack in the bottom left corner. + +Figure 5-4 – Region detection + +- 2) Region stranded: Alarm output must be based on the polygonal regions defined by the user. When the retention time of an object exceeds a prescribed threshold, an alarm can be triggered. See Figure 5-5. + +![Figure 5-5 – Region stranded](d336e7ffee4f537d0805ca840ec28582_img.jpg) + +Two side-by-side surveillance images of a parking lot. In the left image, a silver car is parked, and a red rectangular region is drawn around its windshield. In the right image, a person in an orange shirt is standing next to the car, and the red region has expanded to include the person. Red wavy lines indicate the movement of the object within the region. + +Figure 5-5 – Region stranded + +**Figure 5-5 – Region stranded** + +- 3) Tripwire intrusion: A line that is drawn within the surveillance area. When an object crosses that line, an alarm is triggered. See Figure 5-6. + +![Figure 5-6 – Tripwire intrusion](d29cfbf30a471dc06a78be27f86bd1cf_img.jpg) + +A surveillance image showing a person in a white shirt bending over a fence. A red line, labeled "Tripwire", is drawn across the image. The person is positioned such that they appear to be crossing or interacting with the line. A grey rectangular box with the word "Tripwire" in red text is overlaid on the image. + +Figure 5-6 – Tripwire intrusion + +**Figure 5-6 – Tripwire intrusion** + +- 4) Directed tripwire intrusion: A line is drawn within the surveillance area. When an object travelling in a specified direction crosses the line, an alarm can be triggered. A virtual cordon is drawn within the video image. When a target crosses the line in the specified direction, an alarm can be triggered. As shown in Figure 5-7, the direction of the virtual cordon can be unidirectional or bidirectional. + +![Surveillance image showing a person crossing a virtual tripwire line (red line) in a grassy area near a body of water. A green arrow indicates the direction of movement, and a purple box highlights the person.](d5a837fa4f4675e5ee596003cf55985c_img.jpg) + +Surveillance image showing a person crossing a virtual tripwire line (red line) in a grassy area near a body of water. A green arrow indicates the direction of movement, and a purple box highlights the person. + +(a) Intruder detection + +![Surveillance image showing a person moving in the wrong direction across a virtual tripwire line (yellow line) in a dark, enclosed space. A red box highlights the person, and a yellow arrow indicates the direction of movement.](e3b8510f6a2194e250205ab7bc38076d_img.jpg) + +Surveillance image showing a person moving in the wrong direction across a virtual tripwire line (yellow line) in a dark, enclosed space. A red box highlights the person, and a yellow arrow indicates the direction of movement. + +(b) Wrong direction detection + +**Figure 5-7 – Directed tripwire intrusion** + +#### 5.2.2 Region security alarm and linkage + +- Step 1: Bob is a security clerk. His job is to find suspicious persons and abnormal behaviour by watching real-time video from residential areas of interest using a surveillance system. +- Step 2: Bob uses an IVS system, where an electronic fence is defined based on a virtual line or a polygon in the video images and that is linked to an alarm. When objects cross the electronic fence and their direction of movement is consistent with the setting, the system generates sound and light alarms to alert related personnel, who then deal with the situation. +- Step 3: Leonard lives in this residential area. He enters the underground garage from the exit. Since Bob had set an electronic fence, the IVS system identifies Leonard's abnormal behaviour and triggers an alarm. Bob receives the alarm and warns Leonard through a radio device to correct his behaviour. In addition, the public security on patrol in the area receives at the same time a short message service (SMS) sent by the alarm linkage system to indicate that something abnormal happened in the garage. + +Similarly, this IVS system can detect other abnormal behaviours, such as climbing over a fence or entering a dangerous zone; see Figure 5-8. The user can also adjust the degree of sensitivity of the system in order to reduce false alarms, such as those due to interference by birds and changes in illumination. + +![3D simulation of a garage entrance with a red box highlighting a vehicle entering through the exit lane, which is a security violation.](a8a65eb4968947846b288d693535f03a_img.jpg) + +3D simulation of a garage entrance with a red box highlighting a vehicle entering through the exit lane, which is a security violation. + +Enter the garage using the exit + +![Surveillance image of a person climbing over a fence. A red line indicates the fence line, and a red arrow points to the person climbing over it.](10d19f166f9b7f5961f09f8041896943_img.jpg) + +Surveillance image of a person climbing over a fence. A red line indicates the fence line, and a red arrow points to the person climbing over it. + +Climb over the fence + +F.743.1(15)\_F5-8 + +**Figure 5-8 – Examples of regional security alarm and linkage** + +#### 5.2.3 Intelligent traffic violation monitoring + +- Step 1: Sheldon is a traffic police officer, and his job is to monitor traffic violations using surveillance video. The traffic monitoring system can capture vehicle violation images as evidence. + +- Step 2: Raj drives under a monitoring camera that is part of the intelligent traffic violation-monitoring system. His illegal driving behaviour in a wrong lane is captured by the monitoring camera. See Figure 5-9. +- Step3: The system recognizes Raj's vehicle licence plate and automatically sends a picture to Raj by e-mail or SMS, to warn him that the traffic violation has been detected and that he will receive a penalty. + +This system can also detect automatically illegal parking, speeding and other violations. + +![A black SUV driving on a multi-lane road, with a red rectangle highlighting the vehicle, indicating a traffic violation detection.](bb6d33498937738ff5dac8d91c9ebaad_img.jpg) + +A black SUV is shown from a rear perspective, driving away from the camera on a multi-lane road. A red rectangular box is drawn around the vehicle, highlighting it. In the background, other cars are visible on the road, and a green overhead road sign indicates directions for '14' and '52'. + +A black SUV driving on a multi-lane road, with a red rectangle highlighting the vehicle, indicating a traffic violation detection. + +**Figure 5-9 – Traffic violation** + +### **5.3 Data analysis** + +#### **5.3.1 Traffic/pedestrian flow analysis** + +By using intelligent tracking and recognition algorithms, a traffic/pedestrian flow analysis system can accurately acquire pedestrian flow data based on the video of a moving target, motion analysis and feature classification within a specified single or multiple video monitoring areas. This is illustrated in Figure 5-10. + +![A pedestrian flow analysis scene showing two people walking past a building entrance, with blue vertical lines indicating the flow path.](9999a06627f39a35b608912ed95b09ac_img.jpg) + +A high-angle view of two people walking past a building entrance. The person on the left is wearing a dark shirt and light pants, while the person on the right is wearing a white shirt and light pants. Two vertical blue lines are overlaid on the image, one on the left and one on the right, indicating the flow path of the pedestrians. + +A pedestrian flow analysis scene showing two people walking past a building entrance, with blue vertical lines indicating the flow path. + +**Figure 5-10 – Pedestrian flow analysis** + +Real-time monitoring of pedestrian flow can be used to identify patterns that enable planning of efficient pedestrian flow across different surveillance regions during different times of the day. An intelligent analysis service can help managers to more quickly understand pedestrian flow, to make better planning, scheduling and layout decisions; and to help customers get more business + +opportunities. An intelligent pedestrian flow analysis service is useful in a variety of scenarios, which include statistics and analysis of pedestrian flows in shopping malls, bus stations, parking areas, places of entertainment, stadiums, theatres and other public places. + +##### **Statistics and analysis of pedestrian flow in shopping malls** + +Detection and statistics of pedestrian flow can provide an objective, scientific basis for companies to develop and adjust tactics that will help them achieve their commercial goals. These tactics can include sales to businesses and individuals, promotions, pricing decisions, and options for stocking products, store layouts, service modes and content. + +##### **Statistics and analysis of pedestrian flow in bus stations** + +Through automatic, intelligent, and accurate acquisition of in/out pedestrian flow, it is possible to record and manage vehicle operation information in each site. Transportation owners and managers should use traffic statistics as the basis of vehicle planning, vehicle scheduling and operations management. + +##### **Statistics and analysis of pedestrian flow in parking areas** + +Pedestrian flow statistics help security departments to determine peak times, to identify areas of pedestrian flow, and to help managers study pedestrian flow distribution. + +##### **Statistics and analysis of pedestrian flow in places of entertainment, stadiums and theatres** + +With accurately characterized pedestrian flow data, managers can deploy staff effectively and plan future activities accurately through time and personnel assignment optimization. Based on variations in pedestrian flow, it is possible to make adequate arrangements on the busiest day or the busiest week of every month, or at a busy entrance. + +Pedestrian flow statistics are generally based on information from cameras at passageways. Most cameras are installed indoors, so the position of staff can be accessed and illumination can be controlled. + +Traffic statistics are generally based on information from outdoors, such as wide multi-lane roads. It is therefore difficult to control the orientation of vehicles or the light and weather conditions. Intelligent traffic analysis techniques are currently more developed in pedestrian flow statistics; however, but the effectiveness of traffic statistics can be improved. + +#### **5.3.2 Facial recognition** + +Facial recognition detects facial images in video and extracts their characteristics. The extracted information is then compared with other images or a database of individuals' images, and comparison results are generated. This application can be used in airports, ports, railway stations, coach stations and other public places. See Figure 5-11. + +![Figure 5-11 – Facial recognition. The diagram illustrates two paths for facial recognition. The top path shows an input 'Human image' of a man walking, followed by a 'Face recognition' block, resulting in a zoomed-in face image labeled 'Result of face recognition'. This is followed by a 'Face location' block, resulting in the same face image with white crosshairs on facial features, labeled 'Result of face location'. The bottom path shows the 'Human image' being processed by a 'Facial feature extraction' block, which outputs a vector of features labeled '(x1, x2, ..., xn)' and 'Facial feature'. This vector is then processed by a 'Comparison' block which also receives input from a 'Face database' (represented by a yellow cylinder). The output of the comparison is a zoomed-in face image labeled 'Final result'.](33ed1f9b27c7c21c797aa928b0f06851_img.jpg) + +Figure 5-11 – Facial recognition. The diagram illustrates two paths for facial recognition. The top path shows an input 'Human image' of a man walking, followed by a 'Face recognition' block, resulting in a zoomed-in face image labeled 'Result of face recognition'. This is followed by a 'Face location' block, resulting in the same face image with white crosshairs on facial features, labeled 'Result of face location'. The bottom path shows the 'Human image' being processed by a 'Facial feature extraction' block, which outputs a vector of features labeled '(x1, x2, ..., xn)' and 'Facial feature'. This vector is then processed by a 'Comparison' block which also receives input from a 'Face database' (represented by a yellow cylinder). The output of the comparison is a zoomed-in face image labeled 'Final result'. + +F.743.1(15)\_F5-11 + +**Figure 5-11 – Facial recognition** + +The process is as follows: + +- Step 1: Bill works in the human resources department and work attendance management is one of his job responsibilities. +- Step 2: Two high-definition cameras are installed to capture video at the entrance/exit every day. The facial recognition system can automatically recognize faces. +- Step 3: The system compares those facial images with the company face database, and labels them with the time of passage. Thus Bill can acquire daily reports and easily manage staff attendance. + +#### 5.3.3 Vehicle licence plate recognition + +- Step 1: Peter is a traffic police officer who works in a vehicle information management department. With a recognition system, Peter can get licence plate snapshots at an intersection. +- Step 2: First, the system locates the position of the licence plate; see Figure 5-12. Then, it identifies the plate colour, characters or numbers, and vehicle types using intelligent image analysis algorithms and stores the information in the database. + +Peter can use this database for fake plate vehicles management, vehicle tracking, etc. + +![A collage of various vehicle license plates from China and Hong Kong, demonstrating license plate recognition. The plates include: 冀A 18290, 津港 10762, 皖K 10728, 粤C A3158, 粤C B6924, MH-66-99, 粤C 21298, 沪B L7428, 津 A 72468, 沪D R2251, 沪D R2251 China, 沪D 4808, and 沪D 4808 Hong Kong.](b8d3d85b1575ec6abdcf1bd9ce438701_img.jpg) + +A collage of various vehicle license plates from China and Hong Kong, demonstrating license plate recognition. The plates include: 冀A 18290, 津港 10762, 皖K 10728, 粤C A3158, 粤C B6924, MH-66-99, 粤C 21298, 沪B L7428, 津 A 72468, 沪D R2251, 沪D R2251 China, 沪D 4808, and 沪D 4808 Hong Kong. + +**Figure 5-12 – Vehicle licence plate recognition** + +### **5.4 Intelligent video diagnosis** + +This scenario is illustrated in Figure 5-13: + +- Step 1: Howard is a video surveillance maintenance staff member. He needs to ensure that 100 surveillance cameras in an office building are working normally. +- Step 2: The intelligent video system can diagnose whether cameras within the building are in normal working condition. This system helps Howard find any defective camera, and prompts him with the location and type during a routine inspection. +- Step 3: Howard sends an equipment maintenance worker to repair the defective camera. + +Similarly, Howard can get alarms for picture quality abnormalities, such as snow, blur and colour cast. + +![Figure 5-13 – Video diagnosis. A 2x2 grid of video frames showing different quality issues: Definition exception (top-left, blurry), Brightness exception (top-right, overexposed), Video Interference (bottom-left, noisy), and Video Roll (bottom-right, tilted).](ac852a162572ca8a8c8478c49b571af5_img.jpg) + +The figure displays four video frames arranged in a 2x2 grid, each illustrating a different video quality issue. The top-left frame shows a blurry scene labeled 'Definition exception'. The top-right frame shows an overexposed scene labeled 'Brightness exception'. The bottom-left frame shows a noisy scene labeled 'Video Interference'. The bottom-right frame shows a tilted scene labeled 'Video Roll'. + +Figure 5-13 – Video diagnosis. A 2x2 grid of video frames showing different quality issues: Definition exception (top-left, blurry), Brightness exception (top-right, overexposed), Video Interference (bottom-left, noisy), and Video Roll (bottom-right, tilted). + +**Figure 5-13 – Video diagnosis** + +### 5.5 Intelligent video retrieval + +This scenario is illustrated in Figure 5-14: + +- Step 1: Steven is a police officer working in a criminal investigation department. When a crime occurs, Steven and his colleagues collect signals from all the surveillance videos located in that area at that time. +- Step 2: Steven uses an intelligent video retrieval system to process the videos. The system extracts moving object information, such as size, colour and motion direction, and establishes a target index database. This database transforms the non-structured video into a structured index description. +- Step 3: Using scattered target feature clues, the suspicious objects in the video can be quickly located. For example, Steven searches for a red car in several surveillance videos. With the help of an intelligent video retrieval system, Steven can quickly locate all the red cars appearing in the video. The system assists Steven in searching important information in the mass of video data. + +![Figure 5-14: Intelligent video retrieval. The image shows two side-by-side screenshots of a video surveillance system interface. The left screenshot, titled 'Preprocessing', shows a single video frame of a street intersection with a red line drawn across it. The right screenshot, titled 'Search', shows a grid of multiple video frames from the same intersection, with various vehicles highlighted by red bounding boxes, indicating detected objects.](fbfaa723129f8b8e3be651c6fe7cfab7_img.jpg) + +Figure 5-14: Intelligent video retrieval. The image shows two side-by-side screenshots of a video surveillance system interface. The left screenshot, titled 'Preprocessing', shows a single video frame of a street intersection with a red line drawn across it. The right screenshot, titled 'Search', shows a grid of multiple video frames from the same intersection, with various vehicles highlighted by red bounding boxes, indicating detected objects. + +Figure 5-14 – Intelligent video retrieval + +### 5.6 Intelligent video synopsis + +This scenario is illustrated in Figure 5-15: + +- Step 1: Amy is a company staff member. In order to oversee the elderly and children, she has installed a camera in the house. She wants to remotely see key information of what happened at home. +- Step 2: First, Amy imports the daytime videos into the video synopsis system. After target recognition is finished, targets and related information are indexed. +- Step 3: As Amy's requirement, different targets are stitched to the same background scene and combined together. After that, the duration of the video is compressed and Amy can view a 5 min compressed video instead of 1 h. + +![Figure 5-15: Intelligent video synopsis. The image shows a screenshot of a video synopsis system interface. On the left, there are two video thumbnails with progress bars. The main window displays a compressed video feed showing a collage of different scenes from a house, including people sitting at a table and a person in a red jacket. The interface includes various control buttons and settings for video playback and target recognition.](eafe8d3fa89d770007168b5d2e09e87a_img.jpg) + +Figure 5-15: Intelligent video synopsis. The image shows a screenshot of a video synopsis system interface. On the left, there are two video thumbnails with progress bars. The main window displays a compressed video feed showing a collage of different scenes from a house, including people sitting at a table and a person in a red jacket. The interface includes various control buttons and settings for video playback and target recognition. + +Figure 5-15 – Intelligent video synopsis + +## 6 Reference architecture + +Figure 6-1 shows the overall function architecture for IVS, based on that defined in [ITU-T H.626]. On the right side, the functional architecture for ITU-T F.743.1 focuses on a visual surveillance system with intelligent units, including an intelligent video management (IVM) server, intelligent video unit (IVU) and application (APP), shown on the left side of Figure 16. + +NOTE – See also [b-ITU-T H.627] for signalling and protocols for visual surveillance, and [b-ITU-T H.626.1] for architectural requirements for mobile visual surveillance. + +![Figure 6-1: Functional components of an IVS. The diagram shows a hierarchical architecture. At the top is the 'IT system'. Below it, the 'APP' (Application) and 'CMU' (Central Management Unit) are connected via dashed blue arrows (Management). The 'APP' also connects to 'IVM' (Intelligent Video Management) via solid red arrows (Control). The 'IVM' connects to 'IVU' (Intelligent Video Unit) via solid red arrows (Control). The 'IVU' connects to 'IPU' (Intelligent Premises Unit) via solid black arrows (Media). The 'CMU' connects to 'SCU' (Storage Control Unit) via solid red arrows (Control). The 'SCU' connects to 'MDU' (Media Distribution Unit) via solid red arrows (Control). The 'MDU' connects to 'MSU' (Media Storage Unit) via solid red arrows (Control). The 'MSU' connects to 'ICU' (Intelligent Customer Unit) via solid red arrows (Control). The 'ICU' connects to 'IPU' via solid black arrows (Media). The 'IPU' contains a 'PIV' (Premises Intelligent Video) module. The 'ICU' contains a 'CIV' (Client Intelligent Video) module. Various data flows are indicated by arrows labeled with variables: Pm, Dm, Sc, Sm, Dm', Cm, Ps, Sd, Sd', Cs, Mp, Pd, Cu, Mc, Pc, Pc', Cd', Mc'.](9c6461e1e94afae4dec455e69a2ce152_img.jpg) + +Figure 6-1: Functional components of an IVS. The diagram shows a hierarchical architecture. At the top is the 'IT system'. Below it, the 'APP' (Application) and 'CMU' (Central Management Unit) are connected via dashed blue arrows (Management). The 'APP' also connects to 'IVM' (Intelligent Video Management) via solid red arrows (Control). The 'IVM' connects to 'IVU' (Intelligent Video Unit) via solid red arrows (Control). The 'IVU' connects to 'IPU' (Intelligent Premises Unit) via solid black arrows (Media). The 'CMU' connects to 'SCU' (Storage Control Unit) via solid red arrows (Control). The 'SCU' connects to 'MDU' (Media Distribution Unit) via solid red arrows (Control). The 'MDU' connects to 'MSU' (Media Storage Unit) via solid red arrows (Control). The 'MSU' connects to 'ICU' (Intelligent Customer Unit) via solid red arrows (Control). The 'ICU' connects to 'IPU' via solid black arrows (Media). The 'IPU' contains a 'PIV' (Premises Intelligent Video) module. The 'ICU' contains a 'CIV' (Client Intelligent Video) module. Various data flows are indicated by arrows labeled with variables: Pm, Dm, Sc, Sm, Dm', Cm, Ps, Sd, Sd', Cs, Mp, Pd, Cu, Mc, Pc, Pc', Cd', Mc'. + +**Figure 6-1 – Functional components of an IVS** + +The architecture also contains an intelligent customer unit (ICU) and an intelligent premises unit (IPU) shown on the right side of Figure 6-1. + +Descriptions of IVS components follow. + +**IVU (intelligent video unit):** automatically identifies specific objects, and outputs recognition results to IVM. Recognized information includes that from event triggers and data acquisition. An IVU can include more than one intelligent analysis algorithm that can be added or deleted according to requirements. + +**IVM (intelligent video management):** supports users in configuring intelligent application strategies and dynamically scheduling video sources according to intelligent application strategies. IVM dynamically schedules the IVU and accepts registration, deletion and capability reports from it. IVM can be invoked by an intelligent application from the CMU (see below). Then IVM sends interface information to the CMU. + +**APPs (applications):** third-party data application analysis systems, which in general are not standard units. An APP can accept data from the platform, and then configure data analysis algorithms and strategies. It can analyse data. An APP can send alarm information to the platform if needed. In addition, an APP can be built to meet a specific user requirement. Taking into consideration that alarm processing is a main function with a heavy load; an APP should be stripped from the CMU, which makes the function of platform units clear and allows the platform to be scalable. + +**ICU (intelligent customer unit):** in order to achieve comprehensive intelligent video recognition, this unit adds a client intelligent video (CIV) module to the control unit (CU). A CIV recognizes required information from input videos, and outputs recognition results. In addition, a CIV can search designated records from video files. The identifiable information includes event trigger information and data acquisition information. + +**IPU (intelligent premises unit):** this unit adds a premises intelligent video (PIV) module to a premises unit (PU). A PIV is an intelligent recognition module of the PU, which can be achieved and integrated in PU devices. A PIV can recognize required information from input videos, and output recognition results. The recognized information includes that from event triggers and data acquisition. + +SCU (service control unit): is used to access service control and signal call control between the PU and the CU. + +MDU (media distribution unit): its main functions include media reception, media processing, media routing, media transmission, media forwarding and media replication. + +MSU (media storage unit): its main functions include media storage, media serving, media indexing and media downloading. + +CMU (centre management unit): is used for centralized system management, service operation management, etc. + +# **7 Requirements for an IVS** + +### **7.1 User requirements** + +#### **7.1.1 User level requirements** + +USR-001: IVS system is recommended to classify users into several levels. Users of different levels have different operational rights; high-level users have more operation privileges than low-level ones. Low-level users can do basic operations that meet the basic requirements described in clause 7.1.2, and high-level users can do advanced operations that meet the advanced requirements described in clause 7.1.3. + +#### **7.1.2 Basic requirements** + +##### **7.1.2.1 Common category** + +USR-002: The IVS system is required to have the ability to obtain the standard video streams transmitted or distributed by the platform, to analyse images, and to send the result back to the original system. + +USR-003: The IVS system is required to have the ability to directly access the platform using embedded devices with intelligent analysis functions. The device analyses the image and sends the result back for the system to complete the recognition process, as well as the real-time media stream. + +USR-004: The IVS system is required to have the ability to determine a certain area for picture snapshots through the CU or other controls, should support the drawing of trip lines, and optionally specify coordinates for designation. + +USR-005: The IVS system is required to have the ability to designate the whole or a part of the area for intelligent detection. + +USR-006: The IVS system is required to have the ability to filter redundant information from the image, including jitter and weather factors. + +##### **7.1.2.2 Perimeter alarm category** + +USR-007: The IVS system is required to have the ability to draw one or more trip lines on a video image. + +USR-008: The IVS system is required to have the ability to support in/out two-way intrusion detection in both enclosed and unenclosed regions, and to support two-way intrusion detection for every warning line. The system generates an alarm when the target object moves out of the object detection region. + +USR-009: The IVU/PIV is required to have the ability to determine automatically the features of the target object based on the type detected. The IVU/PIV should support at least person and vehicle feature detection, but it is not required to report the extracted feature information. Other information for the object detection function is optional; users can choose it according to the maturity of the algorithm. + +USR-010: The IVS system is required to have the ability to set the detection period; the minimum unit is the minute. Alarms can be enabled at certain alarm periods or fixed time intervals within a day or week. + +USR-011: The alarm result is required to include alarm information and one or more image(s). + +##### **7.1.2.3 Passenger flow category** + +USR-012: The IVS system is required to have the ability to set the counting region on a snapshot image through the CU or other controls. + +USR-013: The passenger flow function is required to have the ability to be set either one-way or two-way for direction detection and passenger flow statistics. + +USR-014: The IVS system is required to have the ability to set the density threshold, which triggers an alarm when the density of passengers exceeds a threshold. + +##### **7.1.2.4 Target recognition category** + +USR-015: The IVS system is required to have the ability to classify objects into persons, vehicles and other objects, based on target shape, aspect ratio and texture feature characteristics. + +USR-016: The IVS system is required to have the ability to recognize the colour, velocity, direction of movement and other features of the target. + +##### **7.1.2.5 Video synopsis category** + +USR-017: The IVS system is required to have the ability to select a target of interest in the concentrated video; clicking on the original target will play the corresponding video, both the original video and the concentrated form, with the option of displaying them simultaneously. + +USR-018: The IVS system is required to have the ability to output a shortened video including all targets in this video. + +##### **7.1.2.6 Video retrieval category** + +USR-019: The IVS system is required to have the ability to choose the target category and/or the target colour as keywords for searching. + +USR-020: The IVS system is required to have the ability to search targets in the video, by drawing a circle on the sample picture, inside which is the target sample. + +USR-021: The IVS system is required to have the ability to display search results in the form of key frames and video segments, sorted according to the matched similarities or in the order of appearance. + +#### **7.1.3 Advanced requirements** + +USR-022: The IVS system is required to have the ability to upload alarm video segments, if requested. + +USR-023: The IVS system is required to have the ability to develop new algorithms, such as detection of abandoned items (to detect dangerous or suspicious items in a certain area), item preservation (to guard against certain items being moved) and licence plate recognition. + +USR-024: The IVS system is required to have the ability to set the concentration density in real time while playing the concentrated video. + +USR-025: The IVS system is required to have the ability to expand the search criteria, such as the switch value data from the front-end sensor. + +USR-026: The IVS system is required to have the ability to track moving targets within the image. + +## **7.2 Service requirements** + +#### **7.2.1 Basic requirements** + +SRV-001: IVU/PIV is required to provide at least two types of control of algorithm parameter configuration: one for an administrator and another for ordinary users. IVU/PIV also requires the administrator to finish the initial scene studying, advanced parameter setting and other professional operations, while the business is opened via controls, in order to improve alarm accuracy. + +SRV-002: It is required that users have the ability to set alarm rules via controls, when the background scene remains the same, but the warning area changes. + +SRV-003: IVM is required to have the ability to provide the network time protocol (NTP) clock synchronization service to the IVM managed by it. + +SRV-004: The alarm analysing strategies (including the type of algorithm and detection period) require setting by the CU and are stored on the platform. The platform is responsible for starting or stopping the intelligence analysis. + +SRV-005: CMS (centre management system) is required to generate a temporary key valid for a certain period. It can be in form of a temporary username and password, when the intelligent analysing rule requested by the CU is transmitted by the CMU. + +SRV-006: IVS system is required to keep a user log, including user log-in time, log-off time and operating options (video synopsis). + +SRV-007: IVS system is required to keep a system log, including system start-up, system freeze and system failure. It should support log inquiring in accordance with both the content and the timing, as well as exporting logs and querying results. + +SRV-008: IVS system is required to have the ability to manage user authority by setting the user role. It can add, delete and modify users; additionally, the system can customize user roles, set authority for each type of role, and classify users into the appropriate type. + +#### **7.2.2 Advanced requirements** + +SRV-009: The IVS system is required to have the ability to keep record of all synopsis strategies set by users. It should support strategy inquiring and exporting. + +## Bibliography + +- [b-ITU-T H.626.1] Recommendation ITU-T H.626.1 (2013), *Architecture for mobile visual surveillance.* +- [b-ITU-T H.627] Recommendation ITU-T H.627 (2012), *Signalling and protocols for visual surveillance.* +- [b-ITU-T F.743] Recommendation ITU-T F.743 (2009), *Requirements and service description for visual surveillance.* + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|----------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | General tariff principles | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Terminals and subjective and objective assessment methods | +| Series Q | Switching and signalling | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects and next-generation networks | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.743.11-202008-I_PDF-E/raw.md b/marked/F/T-REC-F.743.11-202008-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..1aa06b51a4ff51417f57856534a0abe2470124ed --- /dev/null +++ b/marked/F/T-REC-F.743.11-202008-I_PDF-E/raw.md @@ -0,0 +1,433 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.743.11** + +(08/2020) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Requirements for video surveillance with mobile +premises units** + +Recommendation ITU-T F.743.11 + +**ITU-T** + +![ITU logo: A globe with a red lightning bolt and the text 'ITU International Telecommunication Union'.](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a stylized globe with a red lightning bolt striking across it. To the right of the globe, the text "ITU" is written in a bold, blue font, and below it, "International Telecommunication Union" is written in a smaller, blue font. + +ITU logo: A globe with a red lightning bolt and the text 'ITU International Telecommunication Union'. + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +### **TELEGRAPH SERVICE** + +| | | +|-----------------------------------------------------------------|-------------| +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | + +## **MOBILE SERVICE** + +| | | +|--------------------------------------------------------|-------------| +| Mobile services and multdestination satellite services | F.110–F.159 | +|--------------------------------------------------------|-------------| + +## **TELEMATIC SERVICES** + +| | | +|-------------------------------------------|-------------| +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | + +## **MESSAGE HANDLING SERVICES** + +F.400–F.499 + +## **DIRECTORY SERVICES** + +F.500–F.549 + +## **DOCUMENT COMMUNICATION** + +| | | +|--------------------------------------|-------------| +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | + +## **DATA TRANSMISSION SERVICES** + +F.600–F.699 + +## **MULTIMEDIA SERVICES** + +**F.700–F.799** + +## **ISDN SERVICES** + +F.800–F.849 + +## **UNIVERSAL PERSONAL TELECOMMUNICATION** + +F.850–F.899 + +## **ACCESSIBILITY AND HUMAN FACTORS** + +F.900–F.999 + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.743.11 + +## Requirements for video surveillance with mobile premises units + +## Summary + +Recommendation ITU-T F.743.11 defines requirements for video surveillance with mobile premises units. Mobile premises units can effectively enhance the flexibility of surveillance perspectives, expand surveillance scenarios and the application scope of video surveillance technology. Meanwhile, since the current wireless communication technology can support the mobile wireless transmission of video or image data, the application of wireless communication technology to mobile premises units can greatly improve the flexibility of video surveillance and construct a comprehensive video surveillance system. This Recommendation describes the application scenarios and the requirements for video surveillance with mobile premises units. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.743.11 | 2020-08-13 | 16 | 11.1002/1000/14324 | + +## Keywords + +Handheld device, in-vehicle device, mobile premises units, unmanned aerial vehicle, video surveillance. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2020 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Scenarios..... | 2 | +| 6.1 UAV as mobile premises unit..... | 2 | +| 6.2 Wearable device as mobile premises unit ..... | 3 | +| 6.3 In-vehicle information system as mobile premises unit..... | 4 | +| 6.4 Mobile robot as mobile premises unit ..... | 4 | +| 6.5 Spacecraft detection system as mobile premises unit..... | 5 | +| 7 Requirements for video surveillance with mobile premises units ..... | 5 | +| 7.1 User requirements..... | 5 | +| 7.2 Service requirements ..... | 6 | +| 7.3 Security requirements ..... | 6 | +| 7.4 Management requirements ..... | 7 | +| 7.5 Network and control requirements ..... | 8 | +| 7.6 Quality of service requirements..... | 8 | +| 7.7 Reliability requirements ..... | 8 | +| Bibliography..... | 10 | + + + +# Recommendation ITU-T F.743.11 + +## Requirements for video surveillance with mobile premises units + +## 1 Scope + +This Recommendation describes the requirements for video surveillance with mobile premises units. In particular, the scope of this Recommendation includes the application scenarios and requirements for video surveillance with mobile premises units. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 premises unit** [b-ITU-T F.743V2]: A device located at the remote part of a video surveillance system and used to capture multimedia information (such as audio, video, image, alarm signal, etc.) from a surveilled object. + +**3.1.2 video surveillance system** [b-ITU-T H.626V2]: A telecommunication service focusing on video (including audio and image) application technology, which is used to remotely capture multimedia (such as audio, video, image, alarm signals, etc.) and present them to the end user in a user-friendly manner, based on a managed broadband network with ensured quality, security and reliability. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 mobile premises unit (MPU)**: A mobile device located at the remote part of a video surveillance system and used to capture multimedia information (such as audio, video, image, alarm signal, etc.) from a surveilled object. It has wireless communication capability and can transmit the captured multimedia information through wireless channels. In addition, it can move autonomously or by manual control. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-----|-----------------------------| +| AR | Augmented Reality | +| CPE | Customer Premises Equipment | +| GPS | Global Positioning System | +| IP | Internet Protocol | +| MEC | Mobile Edge Computing | +| MPU | Mobile Premises Unit | +| PC | Personal Computer | +| PU | Premises Unit | +| UAV | Unmanned Aerial Vehicle | +| VS | Video Surveillance | +| VSS | Video Surveillance System | + +## 5 Conventions + +In this Recommendation: + +- The keywords "**is required to**" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The keywords "**is recommended**" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. + +## 6 Scenarios + +This clause describes typical service scenario examples illustrating video surveillance with mobile premises units and deriving its service requirements. + +### 6.1 UAV as mobile premises unit + +With the development of flight control technology, the applications of UAV have gradually expanded from military fields to civil and industrial fields such as inspection of power infrastructure, border patrol, road testing, water monitoring, port management, accident investigation, fire command and disaster relief. UAVs have many advantages in dealing with specific monitoring tasks, such as rapid deployment and aerial monitoring. The UAVs need to be equipped with a ground station, which is responsible for controlling the UAVs and forwarding video data collected by the UAVs to the video surveillance platform. UAVs with autonomous or controlled flight capability can provide a full range of video shooting perspectives from the air to the ground and transmit the monitoring video to the ground station through the wireless air interface. The ground station can process the video data and transmit it to the video surveillance platform through a wired network or mobile communication networks such as 4G or 5G. + +#### Case 1: UAVs in security industry + +When emergencies such as a security incident, traffic accidents or natural disasters occur, the police department can use UAVs to conduct the uninterrupted monitoring of the target accident area from the aerial perspective, and help police officers to fully control the situation. + +#### **Case 2: UAVs in petrol chemical industry** + +At present, the application of UAV based video surveillance in the petroleum and petrochemical industry mainly includes the monitoring of oil pipelines, topographic exploration along pipelines and emergency inspection of pipeline accidents. The remote mobile video sensing solution provided by the UAV enables the people in the monitoring centre to observe directly the situation and status of the site, greatly improving the work efficiency and saving labour costs. + +#### **Case 3: UAVs in big data analysis** + +The UAV equipped with a camera, which is connected to a 5G mobile terminal or CPE of a patrol vehicle, can capture target images and send them to MEC. Relevant big data statistical analysis is performed on the MEC side and then the results are pushed to the screen or command centre. For example, UAVs are deployed in large sports venues to capture real-time vehicle images, and then send them to the MEC terminal through a 5G network to analyse the traffic flow in real time, further driving big data analysis results to large screens or command and alarm centre. + +### **6.2 Wearable device as mobile premises unit** + +The wearable camera is a portable device that is directly worn on the body or integrated into the clothes or accessories of users, which can be widely used in many fields such as business, entertainment, daily life, national defence, scientific research, industry, agriculture and medical treatment. In video surveillance systems, the wearable sensing device with video capturing, transmission, storage and other computing functions can be used as a mobile premises unit, which can observe the surrounding environment or specific targets with a patrol at any time, anywhere. While the user moves freely, the collected video data is transmitted to the remote monitoring platform through the mobile communication technology to realize interaction with the remote users. + +#### **Case 1: Head-mounted network camera** + +The head-mounted video sensing device is a smart terminal device with data collection and processing capabilities which can be worn. The collected real-time monitoring video can be transmitted to the monitoring platform through mobile communication technologies (such as 4G/5G and/or WLAN). Based on its portability, it is widely used in monitoring and inspection. For example, a head-mounted video surveillance device for the elderly or children can effectively prevent accidental loss. The head-mounted sensing device records the real-time walking route and the action track of the ward and transmits it to the remote monitoring centre through the mobile network. The guardian can check the movement track of the ward at any time and anywhere through the monitoring platform to know whether the current state of the ward is safe or not. + +#### **Case 2: Handheld network camera** + +The handheld video sensing device, a terminal device with data storage, computing and communication capabilities, can collect, process and transmit the monitoring video or image. The handheld devices support different mobile communication technologies (such as 4G/5G and/or WiFi) for convenient and fast interaction with the central video surveillance platform. In addition, handheld video sensing devices can be conveniently carried anywhere, and are mainly used in many industries such as mobile enforcement and the power industry. For example, in police and law enforcement applications, police officers go to the accident scene with the handheld video sensing devices, and use these mobile handheld devices to capture and record the on-site visual information. Meanwhile, the captured visual data can be transmitted to the remote video surveillance centre in real time through the wireless communication module equipped in the handheld sensing device. Furthermore, the handheld video sensing devices usually have a voice communication function which can be used for communication between the on-site police officers and the commander at the surveillance centre. + +#### **Case 3: AR glasses with network cameras** + +AR glasses equipped with a camera are connected to a 5G mobile terminal through WIFI or the CPE of a patrol vehicle, and they send the captured images (such as a human face) to MEC. Relevant big data statistical analysis is performed on the MEC side and then the results are pushed to the AR glasses. For example, the AR glasses wearer can take a picture of a vehicle passing in front, then send the images to the MEC terminal through a 5G network to conduct vehicle recognition, further pushing the target vehicle information to AR glasses in real time. + +### **6.3 In-vehicle information system as mobile premises unit** + +The in-vehicle video surveillance device can collect and process monitoring video or images in real time according to the internal and external conditions of the vehicle, and transmit them to a monitoring centre through mobile communication networks. The in-vehicle surveillance information system plays a significant role in multiple fields, such as vehicle anti-theft, anti-robbery, driving route monitoring and rapid response of accidents, which aim to solve the problem of dynamic management of existing vehicles. + +#### **Case 1: In-vehicle surveillance system for public transportation** + +The public vehicle on-board surveillance information system can provide favourable assistance for facilitating the driver's work, enhancing the driving safety factor and reducing the occurrence of accidents. It also provides support for strengthening the service specifications of crew members, reducing public security incidents and crimes in public vehicles, public safety forensics, anti-terrorism and so on. Bus on-board video surveillance terminals can obtain the video and images of the bus in real time, and transmit it to the remote monitoring centre platform through the mobile communication network. The management personnel through the remote monitoring of public transportation can provide timely dispatch and command. The wide use of the vehicle video network monitoring system has realized the safety supervision of the whole process of bus operation. It not only saves on management costs for the bus company, it also improves the efficiency and execution of the management. + +#### **Case 2: In-vehicle surveillance system for law enforcement** + +With the advancement of scientific and technological law enforcement, the in-vehicle information system loaded on law enforcement vehicles has played a significant role in the daily enforcement of police officers. For example, smart patrol cars and fire engines, which can instantly capture and record the road conditions and illegal vehicle videos then transmit live videos for accidents in real time to the remote monitoring centre via the mobile communication network, so that the platform supervisor can view the security situation alongside the patrol car at any time. Based on its high mobility and flexibility, significant real-time performance and fast response, the in-vehicle video surveillance system for public security plays an important role in the command and dispatch of law enforcement personnel, traffic management and social security. + +### **6.4 Mobile robot as mobile premises unit** + +With the comprehensive development of information technology, intelligent mobile robots have been applied to human production and life on a large scale. The mobile robot with autonomous navigation and positioning capability equipped with a sensing device can be deployed in industrial production and human living environments such as hydropower stations, production workshops and airport halls by modelling the scene in advance. The mobile robot monitoring system enables the monitoring of large areas based on a single recording unit. At the same time, the mobile shooting platform can achieve multi-angle and multiposition target monitoring. The mobile robot is used as a carrier for video capture, and the video content can be analysed to obtain corresponding environmental information. The real-time surveillance video or image acquired by the mobile robot in different + +environments is transmitted to the monitoring centre through the mobile communication network, so that the monitoring personnel can view it at any time. + +#### **Case 1: Robot in industry** + +Due to the complex terrain or environmental hazard of most industrial scenes, it is not suitable for workers to conduct field surveys and operate on the spot. At this time, the industrial mobile robot will show its great strength. For example, for underground mine monitoring with high risks, the industrial robots equipped with video surveillance systems can easily realize the detection of downhole conditions and video surveillance without the direct use of people, and they send the video to the monitoring centre on the ground through appropriate wireless communication technology. This remote video surveillance scheme enables personnel in the monitoring centre to observe the underground coal-mine conditions directly, thereby greatly improving industrial production efficiency. + +#### **Case 2: Robot in service** + +At present, the visual monitoring technology based on service robots is mainly applied to house cruise monitoring and automatic alarms. The house security inspection robot can freely move and remotely transmit real-time visual data collected from wireless sensor nodes arranged in different locations in the environment by mobile patrol, which is handed over to the data centre for analysis. Users can watch real-time surveillance videos through smart devices such as PCs and mobile phones. If an abnormality is found, the real-time alarm information is sent to facilitate the user knowing of the situation in the first instance. Moreover, the users can query the saved sensor data at the monitoring terminal and control the state of the mobile robot. + +### **6.5 Spacecraft detection system as mobile premises unit** + +With the development of space exploration technology, more and more space satellites and airships equipped with detection systems perform planetary exploration missions in space; this has meant that the mysterious veil of the universe is gradually being unveiled. The high-definition precision camera mounted on the space probe can capture real visual data of the surface of the star, observe the overall geomorphology, geological processes and geological records (such as mineralogy, texture, structure and stratum) of the detector, and assess the current atmosphere and astronomic environments to analyse the associated processes of surface-atmosphere interaction with each other, which significantly supports the navigation, communication, sampling, extraction and storage of spacecraft detectors and other detection activities. The space exploration system needs to be configured with the corresponding ground station. The image information collected by the spacecraft detector is sent to the ground station through the space communication system to process the telemetry data and transmit it to the command and control centre through the terrestrial communication network. + +## **7 Requirements for video surveillance with mobile premises units** + +### **7.1 User requirements** + +#### **7.1.1 User level requirements** + +- USR-1: A video surveillance system (VSS) with MPU is recommended to give users different operation rights, which allow the high-level users to have more operation privileges than the low-level ones. The low-level users can do the basic operations which meet the following basic requirements (see clause 7.1.2), and the high-level users can do the advanced operations which meet the following advanced requirements (see clause 7.1.3). For example, the low-level users can only view the movement track of the MPU, while the high-level users can control the MPU as well as view the movement track. + +#### **7.1.2 Basic requirements** + +- USR-2: A VSS with MPU is recommended to support the presentation of the real-time video captured by the MPU for end users. +- USR-3: A VSS with MPU is recommended to support the playback of the historical video recorded in the MPU for end users. +- USR-4: A VSS with MPU is recommended to support end users to check the remaining power of the MPU at any time and determine whether to turn on other power modes. +- USR-5: A VSS with MPU is recommended to support end users to view the historical movement trajectory and real-time positioning information of the MPU. + +#### **7.1.3 Advanced requirements** + +- USR-6: A VSS with MPU is recommended to support end users to control the view angle of the MPU remotely to capture the comprehensive video information. +- USR-7: A VSS with MPU is recommended to support end users to manually control the motion trajectory and moving speed of the MPU. +- USR-8: A VSS with MPU is recommended to support the voice intercom for end users. +- USR-9: A VSS with MPU is recommended to support sending alert messages to end users when an alert event occurs. + +### **7.2 Service requirements** + +#### **7.2.1 Mobile video capturing service requirements** + +- SER-1: The MPU is recommended to support the 360-degree rotation of the heading, presenting a comprehensive video surveillance picture. + +#### **7.2.2 Video storage service requirements** + +- SER-2: The MPU is recommended to support the local storage of video streams to save video data directly in the MPU. +- SER-3: The MPU is recommended to support sending the video streams to the remote monitoring centre or the cloud platform for data storage. +- SER-4: The MPU is recommended to support the video data management capabilities, including browsing, querying, playback and deleting of locally stored video files. +- SER-5: The MPU is recommended to support setting configuration information for local video storage, including setting the time period for recording video, setting the time for video retention, etc. + +#### **7.2.3 Other value-added service requirements** + +- SER-6: The MPU is recommended to support recording, cluster intercom, video intercom and group messages. +- SER-7: The MPU is recommended to provide a wide range of expansion interfaces to support peripheral access such as printers, smart display terminals and fire storage boxes. +- SER-8: The MPU is recommended to support the automatic turn on of the dynamic fill light in low-light environments, to assist visual positioning and ensure travel safety, as well as capturing clear video surveillance pictures. + +### **7.3 Security requirements** + +- SEC-1: A VSS with MPU is required to provide the mechanisms for authentication and authorization. And, it is required to only permit the authorized users to access the system and use the related video applications. + +- SEC-2: A VSS with MPU is required to provide the hierarchical authority control and hierarchical password settings to ensure the system security. +- SEC-3: A VSS with MPU is required to ensure the security of accessing and controlling MPUs. Each MPU has access attributes. Only authorized users can access MPUs. The types of access include real-time surveillance, replaying the recorded video, power monitoring, remote controlling, trajectory planning, and so on. An end user can access an MPU according to his/her access privilege of the MPU. +- SEC-4: A VSS with MPU is required to ensure the security of information such as recorded video, digital maps, location information, etc. +- SEC-5: A VSS with MPU is recommended to ensure the security of media stream delivery. It is required to provide security mechanisms, such as video encryption for the media stream and signalling encryption, to ensure the integrity and secrecy of the media data. +- SEC-6: A VSS with MPU is recommended to provide the private protocol protection and ensure the security of encrypted storage of data in local and cloud space. +- SEC-7: A VSS with MPU is recommended to provide the mechanisms for troubleshooting and data backup. + +### **7.4 Management requirements** + +#### **7.4.1 Terminal management requirements** + +- MAN-1: A VSS with MPU is required to manage the registration/deregistration, authentication, connection and troubleshooting of MPUs. +- MAN-2: A VSS with MPU is recommended to support MPU management, such as capability parameters and property configuration, status control and monitoring (i.e., MPU activation and deactivation). + +#### **7.4.2 Location and mobility management requirements** + +- MAN-3: A VSS with MPU is recommended to support the position acquisition capability for MPUs. +- MAN-4: A VSS with MPU is recommended to support independent positioning means such as visual odometer to realize autonomous stable movement without a GPS environment. +- MAN-5: A VSS with MPU is recommended to support the offline map caching tool to ensure that map resources can still be obtained under no network conditions. +- MAN-6: A VSS with MPU is recommended to provide real-time positioning information, and encapsulate the positioning information into the video stream simultaneously when recording. +- MAN-7: A VSS with MPU is recommended to provide the remote manual control mechanism of movement trajectory for MPUs (such as UAVs, robots). The operator can control the direction and speed of MPUs in real time. +- MAN-8: A VSS with MPU is recommended to provide the remote automatic control mechanism of movement trajectory for MPUs (such as UAVs, robots). The moving routes of MPUs can be intelligently planned in real time. +- MAN-9: A VSS with MPU is recommended to support intelligent obstacle avoidance and guarantee the route of travel. + +#### **7.4.3 Power management requirements** + +- MAN-10: The MPU powered by battery is recommended to support the multilevel power threshold control and low-power alarm. + +- MAN-11: The MPU is recommended to support the multiple energy supply modes, including battery power, solar power and so on. When the battery is low, the alternative charging mode can be turned on. +- MAN-12: The MPU is required to monitor real-time power and support the remote video surveillance centre to query the status of the power. + +### **7.5 Network and control requirements** + +- NET-1: The MPU is required to support wireless data transmission technologies, such as WLAN, 4G/5G, and digital microwave transmission, and the media data and control signals are required to be transmitted via two separate wireless channels. +- NET-2: The MPU is required to access the IP-based video surveillance (VS) network directly through wireless communication modes or through a gateway. +- NET-3: The MPU is recommended to support dynamic wireless transmission mode adjustment. For example, WLAN transmission mode is changed to 5G network transmission mode. +- NET-4: The MPU is recommended to support adaptive bit rate adjustment according to the current wireless network bandwidth resources. + +### **7.6 Quality of service requirements** + +- QUA-1: A VSS with MPU is recommended to support the adaptive wireless channel detection technology to ensure the stability of over-the-horizon data transmission. +- QUA-2: A VSS with MPU is recommended to support the automatic selection of the best wireless channel so that it can switch channels and adjust the video bit rate dynamically according to network conditions to ensure the stability of the video stream transmission. +- QUA-3: A VSS with MPU is recommended to support long-distance control and multichannel data transmission to obtain powerful anti-interference ability and ensure the stability of data transmission. + +### **7.7 Reliability requirements** + +- REL-1: A VSS with MPU is recommended to provide the core sensor redundancy and sensor fault detection, which can automatically switch to redundant sensors if an abnormality occurs. +- REL-2: A VSS with MPU is recommended to provide omnidirectional sensing and intelligent obstacle detection. The MPU is equipped with visual or infrared sensors in different directions for auxiliary positioning and obstacle perception to avoid the movement malfunctions in a narrow and complex environment. +- REL-3: A VSS with MPU is recommended to provide strong environmental adaptability and anti-interference ability. +- REL-4: The MPU is recommended to support the power-off protection function, which guarantees the integrity of important data in case of sudden power failure. +- REL-5: The MPU is recommended to provide stable anti-shake performance with the characteristics of waterproof, dustproof and anti-fall, which is suitable for all-weather field work. + +## Bibliography + +- [b-ITU-T F.743V2] Recommendation ITU-T F.743V2 (2019), *Requirements and service description for video surveillance.* +- [b-ITU-T H.626V2] Recommendation ITU-T H.626V2 (2019), *Architectural requirements for video surveillance system.* + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.743.12-202106-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.743.12-202106-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7fc0ce00090f66b87404c12debf099b47a838ec0 --- /dev/null +++ b/marked/F/T-REC-F.743.12-202106-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0efd3ea366c9c151bbcfa01a99d20b4bfbdb579443e1a9dd88122c3e286dae51 +size 6108 diff --git a/marked/F/T-REC-F.743.13-202203-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.743.13-202203-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7fc0ce00090f66b87404c12debf099b47a838ec0 --- /dev/null +++ b/marked/F/T-REC-F.743.13-202203-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0efd3ea366c9c151bbcfa01a99d20b4bfbdb579443e1a9dd88122c3e286dae51 +size 6108 diff --git a/marked/F/T-REC-F.743.13-202203-I_PDF-E/367926125450c2bc3f4bdca9d59a62ba_img.jpg b/marked/F/T-REC-F.743.13-202203-I_PDF-E/367926125450c2bc3f4bdca9d59a62ba_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..8cdab1208875e42feba26fa3ac812d56ba6329ce --- /dev/null +++ b/marked/F/T-REC-F.743.13-202203-I_PDF-E/367926125450c2bc3f4bdca9d59a62ba_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4d65bdb71d9a091a93f646ac1ab687c715a57d84ae496807634c97052be8f024 +size 69297 diff --git a/marked/F/T-REC-F.743.13-202203-I_PDF-E/cfda9df1319e04207eb28bcefd1dab7b_img.jpg b/marked/F/T-REC-F.743.13-202203-I_PDF-E/cfda9df1319e04207eb28bcefd1dab7b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1b1fc1a3910568aa85542d2a5d900b7e4c6b1d66 --- /dev/null +++ b/marked/F/T-REC-F.743.13-202203-I_PDF-E/cfda9df1319e04207eb28bcefd1dab7b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6d432fe0ead82c6f37c1c5c200fef0421d5d631ace506db13c928f2d5bfe88a7 +size 23000 diff --git a/marked/F/T-REC-F.743.13-202203-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.743.13-202203-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c78ed55532386a352d7be7bef7b7c13dcaf46f4b --- /dev/null +++ b/marked/F/T-REC-F.743.13-202203-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bae9ab2a5c3b00e9e530bc2dec007dd70c7127ff4271c25f44157dc52e606aab +size 75869 diff --git a/marked/F/T-REC-F.743.14-202203-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.743.14-202203-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1e32f9b8f1141bed243c16e02a6cd1411cbb87f1 --- /dev/null +++ b/marked/F/T-REC-F.743.14-202203-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c395c4bbde83b1e232652a43f51538278c116f51f710c37e20a073f7333726c8 +size 6134 diff --git a/marked/F/T-REC-F.743.14-202203-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.743.14-202203-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ead0adafbe50a059b5497c362b9a37b4fc69f096 --- /dev/null +++ b/marked/F/T-REC-F.743.14-202203-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:164940f14147268db8d9be6fb08b8449b12767045210aa48ba3c6bbc49c844db +size 133560 diff --git a/marked/F/T-REC-F.743.15-202203-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.743.15-202203-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1e32f9b8f1141bed243c16e02a6cd1411cbb87f1 --- /dev/null +++ b/marked/F/T-REC-F.743.15-202203-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c395c4bbde83b1e232652a43f51538278c116f51f710c37e20a073f7333726c8 +size 6134 diff --git a/marked/F/T-REC-F.743.15-202203-I_PDF-E/5cab96b2d23174c25919840ecd50aa48_img.jpg b/marked/F/T-REC-F.743.15-202203-I_PDF-E/5cab96b2d23174c25919840ecd50aa48_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d0f168a31acf9488d6ccf5f4f3d177b4e86dfb11 --- /dev/null +++ b/marked/F/T-REC-F.743.15-202203-I_PDF-E/5cab96b2d23174c25919840ecd50aa48_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a314455411d59f35f5265e6e2b21a9b4ead8afa87e00e8f8c4c037e9b491fe00 +size 67134 diff --git a/marked/F/T-REC-F.743.15-202203-I_PDF-E/a26e142d3df5bef41a84a9dd099d7825_img.jpg b/marked/F/T-REC-F.743.15-202203-I_PDF-E/a26e142d3df5bef41a84a9dd099d7825_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..514bc00d46ba783a86b7ebe5f3bceae3203e620c --- /dev/null +++ b/marked/F/T-REC-F.743.15-202203-I_PDF-E/a26e142d3df5bef41a84a9dd099d7825_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:eaa4cd81df42058bcebfc7d6300ccc867147f31781e4365525ff6981efbfa49f +size 21151 diff --git a/marked/F/T-REC-F.743.15-202203-I_PDF-E/af6be343f0c0a8f155f965dcf337b8af_img.jpg b/marked/F/T-REC-F.743.15-202203-I_PDF-E/af6be343f0c0a8f155f965dcf337b8af_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7d2b37452d05f5a412fc1d2b72dfea58a2a61ed9 --- /dev/null +++ b/marked/F/T-REC-F.743.15-202203-I_PDF-E/af6be343f0c0a8f155f965dcf337b8af_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8300e2fc17d79a6cdb1737427b0c546f0017c2de90c45acdbfb1a1356f5ff9e8 +size 79589 diff --git a/marked/F/T-REC-F.743.15-202203-I_PDF-E/bd0b93e7a46ede276d0a3b79ac487bd9_img.jpg b/marked/F/T-REC-F.743.15-202203-I_PDF-E/bd0b93e7a46ede276d0a3b79ac487bd9_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..65d72a446925b2623c4bafd0c84ceb7e4f6ecea8 --- /dev/null +++ b/marked/F/T-REC-F.743.15-202203-I_PDF-E/bd0b93e7a46ede276d0a3b79ac487bd9_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f1a1df46313dcb646c90b5d8825aa448ae286e40605e5558112ab9fb968ab0a3 +size 62652 diff --git a/marked/F/T-REC-F.743.15-202203-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.743.15-202203-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..eb9be5273ea2cf6667c0ccab0d24cafe1187c500 --- /dev/null +++ b/marked/F/T-REC-F.743.15-202203-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:811dd99cece90261e6f1b67f6e2115db01241889b389dc48afd39ea8ba07d601 +size 129842 diff --git a/marked/F/T-REC-F.743.15-202203-I_PDF-E/raw.md b/marked/F/T-REC-F.743.15-202203-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..88a9d777b66e5665d7544dc2eda667716836995c --- /dev/null +++ b/marked/F/T-REC-F.743.15-202203-I_PDF-E/raw.md @@ -0,0 +1,471 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.743.15** + +(03/2022) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Requirements for multi-operator core network +enabled multimedia services** + +Recommendation ITU-T F.743.15 + +ITU-T + +![ITU logo](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with a white grid pattern, overlaid by a stylized 'ITU' text in blue. + +ITU logo + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.743.15 + +## Requirements for multi-operator core network enabled multimedia services + +## Summary + +Recommendation ITU-T F.743.15 specifies the requirements for multi-operator core network (MOCN) enabled multimedia services (MOCN-MS). This Recommendation not only defines a high-level functional framework of MOCN enabled multimedia services, but also specifies the sharing capability information unit (SCIU) function requirements and MOCN-MS system requirements according to this framework. In addition, several scenarios for MOCN enabled multimedia services are provided in this Recommendation. + +MOCN enabled multimedia service defined in this Recommendation can improve the conventional multimedia service quality by taking advantage of the network sharing capability of the co-construction network, without bringing any additional impact on the base station. The MOCN enabled multimedia service providers (SP) will benefit from the requirements and reference framework defined in this Recommendation. This way the extension and optimization of the existing multimedia service can be estimated efficiently. In addition, it is possible for the multimedia service provider to discover and develop new features while the MOCN-MS is employed. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.743.15 | 2022-03-16 | 16 | 11.1002/1000/14956 | + +## Keywords + +Co-construction network, MOCN, multimedia services (MS), multi-operator core network. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2022 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|--------------------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Overview ..... | 2 | +| 7 Requirements for multi-operator core network enabled multimedia services..... | 4 | +| 7.1 SCIU functional requirements ..... | 4 | +| 7.2 MOCN-MS system requirements ..... | 5 | +| 8 High level MOCN-MS functional framework..... | 6 | +| Appendix I – Scenarios ..... | 8 | +| I.1 User moving through MOCNs and unshared networks ..... | 8 | +| I.2 MS platform associated with the Operator dedicated SCIU deployed in a MOCN ..... | 9 | +| I.3 User moving across different MOCNs ..... | 10 | +| Bibliography..... | 11 | + + + +# Recommendation ITU-T F.743.15 + +## Requirements for multi-operator core network enabled multimedia services + +## 1 Scope + +This Recommendation specifies the requirements of multi-operator core network enabled multimedia services. + +The scope of this Recommendation includes: + +- High-level functional framework of MOCN enabled multimedia services; +- SCIU function requirements for MOCN enabled multimedia services; +- System requirements for MOCN enabled multimedia services; +- MOCN enabled multimedia services scenarios. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following term defined elsewhere: + +**3.1.1 multi-operator core network (MOCN)** [b-3GPP TS 23.501]: MOCN is one of the co-construction network architectures. It is a radio access network (RAN) sharing-based solution that allows two or more core networks to use a shared RAN. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 operations and maintenance centre – RAN (OMC-R)**: In mobile networks, an operations and maintenance centre – RAN is the central location to operate and maintain the radio access network (RAN). OMC-R can provide wireless network data to applications through northbound interfaces. + +**3.2.2 sharing capability information (SCInfo)**: The information about the capabilities of the shared network. + +**3.2.3 sharing capability information unit (SCIU)**: A platform used to collect the network information provided by all shared network elements, and generate the SCInfo for the multimedia service platforms and/or edge multimedia service platforms to optimize the multimedia service in the co-construction and sharing networks. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-------|-----------------------------------------| +| 5GC | 5G Core | +| AR | Augmented Reality | +| MNO | Mobile Network Operator | +| MOCN | Multi-Operator Core Network | +| MS | Multimedia Service | +| NSA | Non-Standalone | +| OMC-R | Operations and Maintenance Centre – RAN | +| PLMN | Public Land Mobile Network | +| QoS | Quality of Service | +| RAN | Radio Access Network | +| SA | Standalone | +| SCIU | Sharing Capability Information Unit | +| SLA | Service Level Agreement | +| SP | Service Provider | +| UAV | Unmanned Aerial Vehicle | +| UE | User Equipment | +| UPF | User Plane Function | +| VR | Virtual Reality | +| WLAN | Wireless Local Area Network | + +## 5 Conventions + +In this Recommendation: + +- The keywords "is required to" indicate a requirement that must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "is recommended" indicate a requirement that is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. +- The keywords "can optionally" and "may" indicate an optional requirement that is permissible, without implying any sense of being recommended. These terms are not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +## 6 Overview + +MOCN is one of the co-construction network architectures proposed by other SDOs (e.g., MOCN in [b-3GPP TS 23.251]). A co-construction network is deployed by multiple operators in collaboration to jointly make use of network resources, which not only improves network capabilities but also reduces infrastructure investment for each operator. Figure 6-1 shows the overview of the multimedia services enabled by MOCN. + +![Figure 6-1: Overview of the multi-operator core network enabled multimedia services. The diagram shows Operator A 5GC connected to a Co-construction network with three RAN areas (Operator A, B, and C). Each area has a base station broadcasting multiple PLMN IDs. Below the RAN areas are moving units (drones) and fixed wireless units (cameras/sensors). These units are connected to a Sharing capability information unit (SCIU), which is part of a Multimedia service platform. The SCIU is also connected to an Edge multimedia service platform. The entire system is labeled 'MOCN enabled multimedia services'.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +The diagram illustrates the architecture of multi-operator core network (MOCN) enabled multimedia services. At the top, 'Operator A 5GC' is shown as a yellow cloud. It connects via solid blue lines to a 'Co-construction network' box. Inside this box are three base stations, each with a list of PLMN IDs: + +- Left base station: #1 PLMN-A, #2 PLMN-B, #3 PLMN-C +- Middle base station: #1 PLMN-B, #2 PLMN-C, #3 PLMN-A +- Right base station: #1 PLMN-C, #2 PLMN-A, #3 PLMN-B + +Below the base stations are three colored hexagonal areas representing RAN areas: 'Operator A construction area' (blue), 'Operator B construction area' (red), and 'Operator C construction area' (green). Below these areas, there are 'Moving unit' (drone) and 'Fixed wireless unit' (camera/sensor) icons. Arrows show data flow from these units through a 'Sharing capability information unit' (red square) to a 'Multimedia service platform' (containing a monitor and phone icon) and an 'Edge multimedia service platform' (blue cloud). Dashed yellow lines represent signaling or data paths between the SCIU and the platforms. The bottom section is labeled 'MOCN enabled multimedia services'. A reference code 'F.743.15(22)' is in the bottom right corner. + +Figure 6-1: Overview of the multi-operator core network enabled multimedia services. The diagram shows Operator A 5GC connected to a Co-construction network with three RAN areas (Operator A, B, and C). Each area has a base station broadcasting multiple PLMN IDs. Below the RAN areas are moving units (drones) and fixed wireless units (cameras/sensors). These units are connected to a Sharing capability information unit (SCIU), which is part of a Multimedia service platform. The SCIU is also connected to an Edge multimedia service platform. The entire system is labeled 'MOCN enabled multimedia services'. + +**Figure 6-1 – The overview of the multi-operator core network enabled multimedia services** + +Multi-operator core network enabled multimedia services, deployed in the co-construction network, are an extension of existing telecommunication services to handle at least two types of media in a synchronized way, where a co-construction network can take full advantage of the network resources of multiple operators involved in the construction and extend related service capabilities. The overview of such multimedia services is shown in Figure 6-1. The co-construction network contains several RAN areas, in which the base stations of the areas can broadcast public land mobile network (PLMN) IDs of all operators for network sharing. + +In the example shown in Figure 6-1, a moving unit, which is a client of multimedia service is required to support multiple wireless data transmission technologies, and the media data and control signalling are required to transmit via two separate radio channels. With the deployment of MOCN, the SCInfo is generated by the SCIU and transferred to the multimedia service platform and/or the edge multimedia service platform (which is closer to the end-user), based on the request. The SCIU allows the multimedia service platform to acquire the shared network capabilities at different levels. Besides, the map of co-construction areas needs to be provided to reduce the switch latency of transmitting modes and handover delay in shared border areas. The distribution information of these sharable edge network elements also needs to be included in the SCInfo for the multimedia service platform, especially in some low-latency service application scenarios. + +This comprehensive information may bring about a big chance in helping conventional multimedia service providers (SPs) to provide a more optimized service capability, such as more accurate bandwidth reservation, customized service for the specific user and dynamic quality of service (QoS) adjusting. + +## **7 Requirements for multi-operator core network enabled multimedia services** + +### **7.1 SCIU functional requirements** + +#### **7.1.1 SCIU general requirements** + +FUN-001: The SCIU component is required to provide a southbound interface to OMC-R to acquire all kinds of RAN sharing information in the MOCN. + +FUN-002: The SCIU component is required to provide a northbound interface to the multimedia service platform to provide SCInfo for a multimedia service platform. + +FUN-003: The SCIU component is recommended to support management of the collected network information, such as creating, deleting and updating. + +FUN-004: The SCIU is recommended to be able to process massive requests from the multimedia service platforms with reasonable and acceptable response time to fulfil specific performance requirements. These requirements may include response time, efficiency, information format, information type and precision. + +FUN-005: The SCIU is recommended to provide a failure recovery mechanism to maintain services if certain functions become unavailable. + +FUN-006: The SCIU proxy can optionally be used for exposing the SCIU information to the upper-layer service. It can be implanted as an open network capability function, such as IMT-2020 (5G) NEF (network exposure function) or a dedicated interface. + +#### **7.1.2 Sharing information collection** + +FUN-007: The SCIU component is required to support the collection of all shared network elements and all shared network information provided by co-construction operators on demand. + +FUN-008: The SCIU component is recommended to acquire the types of shared radio access technologies in each operator's co-construction area, such as IMT-Advanced (4G), IMT-2020 (5G) and wireless local area network (WLAN). + +FUN-009: The SCIU component is recommended to acquire the available radio frequency, radio bandwidth and radio resource priority of each shared RAN area. + +FUN-010: The SCIU component is recommended to acquire the distribution information of available shared edge network elements, including network locations and adjacency relationships of the shared edge nodes. + +FUN-011: The SCIU component is recommended to acquire the map of co-construction network deployment, especially including the map of borders between different shared networks. + +FUN-012: The SCIU component is recommended to acquire the statistical network load information of the shared networks. + +FUN-013: The SCIU component is recommended to support the collection of the co-construction network deployment information and some sharing policies enforced by operators. + +#### **7.1.3 Sharing information process** + +FUN-014: The SCIU component is required to process all kinds of shared network information to generate SCInfo. + +FUN-015: The SCIU component is recommended to be capable of categorizing the network sharing capability at different levels in terms of the features such as capacity, coverage and delay. These sharing capability levels are defined as available network resource priorities in different operators' co-construction areas. + +#### **7.1.4 SCInfo storage/cache** + +FUN-016: The SCIU component is recommended to be capable of storing the network information from all shared network elements and co-construction operators on demand. + +FUN-017: The SCIU component is recommended to encrypt SCInfo to protect the essential network operator information, and sensitive information collected from the shared networks. Appropriate methods should be deployed to prevent attacks targeted at the SCIU. + +FUN-018: The SCIU component is recommended to duplicate SCInfo in the cloud to avoid data loss or modification. + +#### **7.1.5 SCInfo requirements** + +FUN-019: The network SCInfo is recommended to support the indication of the types of shared radio access technologies in each operator's co-construction area, such as IMT-Advanced (4G), IMT-2020 (5G) and WLAN. In addition, the information on shared IMT-2020 (5G) radio access technologies are recommended to refer to the non-standalone (NSA) mode or standalone (SA) mode. + +FUN-020: The network SCInfo is recommended to contain the available radio frequency, radio bandwidth and radio resource priority of each shared RAN area. + +FUN-021: The network SCInfo is recommended to contain the distribution information of available shared edge network elements, including network locations and adjacency relationships of the shared edge nodes. + +FUN-022: The network SCInfo is recommended to contain the map of co-construction network deployment, especially including the map of borders between different shared networks. + +FUN-023: The network SCInfo is recommended to contain the statistical network load information of the shared networks. + +FUN-024: The network SCInfo can optionally be provided in the formats specified by multimedia service platforms or edge multimedia service platforms. + +### **7.2 MOCN-MS system requirements** + +#### **7.2.1 General requirements** + +FUN-025: A MOCN-MS system is recommended to be capable of providing a unified multimedia service to an operator's subscriber that are connected either to an operator independent network or a multi-operator core network. + +FUN-026: A MOCN-MS system is recommended to provide continuous service for the end-users when commanding handover between operator independent network and multi-operator core network or between different multi-operator core networks. + +FUN-027: A MOCN-MS system is recommended to be able to identify the different operator's subscribers who are connected to the multi-operator core network. + +#### **7.2.2 Network management function** + +FUN-028: Network management function is recommended to be able to retrieve the SCInfo from SCIU via an open interface or an SCIU proxy. + +FUN-029: Network management function is recommended to be able to retrieve the customized SCInfo based on the specific operator's identification. + +FUN-030: Network management function is recommended to be able to transfer the retrieved SCInfo to the multimedia application function via a pre-defined internal interface, based on the SCInfo query. + +FUN-031: The network SCInfo is recommended to be cached in the local entities of the network management function to avoid frequent querying. + +#### 7.2.3 Multimedia application function + +FUN-032: Multimedia application function is recommended to be able to retrieve the SCInfo from the network management function. + +FUN-033: Multimedia application function is optional to be able to retrieve the SCInfo from SCIU via an open interface or an SCIU proxy. + +FUN-034: Multimedia application function is recommended to be able to provide or adjust the service capability at different levels related to its corresponded SLA, based on the SCInfo. + +FUN-035: If the multimedia application function is implemented as the virtualized network application, the service capability is optional to be adjusted by changing the virtual hardware/software resource, based on the SCInfo. + +FUN-036: Multimedia application function is recommended to have the capability of scheduling the service request to the most appropriate service node according to the SCInfo. + +FUN-037: Multimedia application function is optionally able to create or change traffic routing policy, which is applied on the routing device, according to the SCInfo, to offload its service traffic into the edge network. + +FUN-038: Multimedia application function is recommended to have the capability of migrating services between different multimedia service platforms to ensure service continuity when user equipments (UEs) switch to MOCN areas. + +## 8 High level MOCN-MS functional framework + +According to the overview of the MOCN enabled multimedia service shown in Figure 6-1, a MOCN-MS system is composed of four components, which include the end-user, MOCN, SCIU and MOCN-MS platform. The high-level MOCN-MS functional framework is illustrated in Figure 8-1. + +![Figure 8-1: The high-level MOCN-MS functional framework diagram. The diagram shows four stacked layers. The top layer is the 'MOCN-MS platform' containing 'Network management function', 'Multimedia application function', and a dashed box 'SCIU proxy'. The second layer is 'SCIU' containing 'Sharing information process', 'Sharing information collection', and 'SCInfo storage/cache'. The third layer is 'MOCN' containing 'OMC-R', 'Core network', and 'Shared RAN'. The bottom layer is 'End-user'. Arrows indicate 'SCInfo' (dashed) and 'Multimedia content' (solid) flow between layers. A legend at the bottom defines the arrow types.](af6be343f0c0a8f155f965dcf337b8af_img.jpg) + +The diagram illustrates the high-level MOCN-MS functional framework. It consists of four main layers stacked vertically: + +- MOCN-MS platform** (top layer): Contains the **Network management function** and **Multimedia application function**. A dashed box labeled **SCIU proxy** is positioned between them. This layer has bidirectional dashed arrows (SCInfo) with the SCIU layer below and a solid arrow (Multimedia content) pointing down to the MOCN layer. +- SCIU** (second layer): Contains **Sharing information process**, **Sharing information collection**, and **SCInfo storage/cache**. It has bidirectional dashed arrows (SCInfo) with the MOCN-MS platform above and the MOCN layer below. +- MOCN** (third layer): Contains **OMC-R**, **Core network**, and **Shared RAN**. It has a solid arrow (Multimedia content) pointing down to the End-user layer. +- End-user** (bottom layer): The final recipient of multimedia content. + +Legend: + +- Dashed double-headed arrow: SCInfo +- Solid double-headed arrow: Multimedia content + +F.743.15(22) + +Figure 8-1: The high-level MOCN-MS functional framework diagram. The diagram shows four stacked layers. The top layer is the 'MOCN-MS platform' containing 'Network management function', 'Multimedia application function', and a dashed box 'SCIU proxy'. The second layer is 'SCIU' containing 'Sharing information process', 'Sharing information collection', and 'SCInfo storage/cache'. The third layer is 'MOCN' containing 'OMC-R', 'Core network', and 'Shared RAN'. The bottom layer is 'End-user'. Arrows indicate 'SCInfo' (dashed) and 'Multimedia content' (solid) flow between layers. A legend at the bottom defines the arrow types. + +Figure 8-1 – The high-level MOCN-MS functional framework + +Basically, the MOCN-MS platform contains a multimedia application function, network management function and probably an SCIU proxy. The multimedia application function is responsible for delivering various types of multimedia services and content over the co-construction network, such as audio / video service and web service. According to the current deployment cases, a MOCN area may have multiple operator-dedicated multimedia service platforms severed for the different operator's subscribers and they are isolated from each other. The network management function, especially in this Recommendation, is responsible for the MOCN status monitoring and MOCN information retrieving. The multimedia application function can directly obtain the SCInfo from SCIU or interact with the network management function to obtain the SCInfo if the network capacity is not intended to be exposed to the upper-layer applications. The SCIU proxy is able to transfer the SCInfo from the SCIU to the MOCN-MS platform. According to the SCInfo, the multimedia application function can select the most appropriate service mode or optimize its service quality, for example, to limit the maximum connection and bandwidth based on the local MOCN status. + +The SCIU component is responsible for collecting and processing the shared network capability information from the MOCN and providing accurate and reliable SCInfo for the multimedia service platform. SCIU contains shared information collection functions, shared information process and SCInfo storage/cache functions. The shared information collection function is responsible for collecting the underlying network information related to network sharing in MOCN. The shared information process function is responsible for the processing of acquired sharing network information and generating SCInfo. SCInfo storage/cache function is responsible for storing or caching SCInfo in SCIU, avoiding frequent data acquisition and processing from OMC-R. + +It is noted that, in practice, there are possible multiple SCIU functional entities existing by associating with the different MOCNs, which belong to the different operators or regions. + +MOCN is a co-construction network and is responsible for providing a shared access network for the end-users which can be the subscribers belonging to the different operators. The MOCN comprises a shared RAN, associated with multiple independent core networks and OMC-Rs. The shared networks can be deployed by different operators in co-construction areas. Compared with the operator dedicated network, the MOCN can provide different levels of resources and service capabilities based on the agreements between operators. + +The end-user function represents the multiple types of devices, e.g., personal computer, intelligent mobile device, augmented reality (AR) / virtual reality (VR) device and unmanned aerial vehicle (UAVs) with a camera, which is authorized to access MOCN and obtain multimedia services. + +## Appendix I + +### Scenarios + +(This appendix does not form an integral part of this Recommendation.) + +This clause describes various scenarios for multi-operator core network enabled multimedia services. + +### I.1 User moving through MOCNs and unshared networks + +![Diagram showing a user moving between a MOCN and an unshared network, connected to a multimedia service platform via an SCIU.](a26e142d3df5bef41a84a9dd099d7825_img.jpg) + +The diagram illustrates a user (represented by a smartphone icon) connected to two network types: a MOCN (Multi-Operator Core Network) and an Unshared network. Both networks are connected to a central Multimedia service platform. An SCIU (Shared Core Interface Unit) is shown above the MOCN, connected to it and also connected to the Multimedia service platform. The SCIU is represented by a circle with four arrows pointing outwards. The Multimedia service platform is represented by a rectangle with a circle inside. The text 'F.743.15(22)' is written below the Multimedia service platform. + +Diagram showing a user moving between a MOCN and an unshared network, connected to a multimedia service platform via an SCIU. + +Figure I.1 – User moving through MOCNs and unshared networks + +MOCN is a RAN sharing solution that allows two or more operators' core networks to use the same RAN. MOCN can aggregate multiple operators' resources to achieve better network performance. Users in the unshared network and the MOCN usually have different SLAs (service level agreements). Figure I.1 shows an example of a user moving through MOCNs and unshared networks. When the users move from an unshared network to the MOCN, the multimedia platform needs to acquire SCInfo, which contains the shared network capability information with the corresponding level of service. + +| Level | Capacity | Delay | Coverage | ... | +|-------|----------|-------|----------|-----| +| 1 | | | √ | | +| 2 | √ | | | | +| 3 | | √ | | | +| 4 | | | | √ | +| 5 | | | | | +| ... | | | | | + +F.743.15(22) + +Figure I.2 – Example table of sharing capability information + +In this scenario, SCIU is deployed to obtain the shared network information from the RAN side and generates SCInfo for the multimedia server platform. Figure I.2 shows an example of SCInfo. SCInfo contains the levels of resource allocation that a shared network can provide such as capacity, delay, coverage, etc. The multimedia service platform acquires the SCInfo and then adjusts the QoS (quality of service), such as video resolution and round-trip time and can be adaptively adjusted in advance to ensure the stability of the video stream transmission. + +### I.2 MS platform associated with the Operator dedicated SCIU deployed in a MOCN + +![Diagram of Operator dedicated SCIU deployed in a MOCN showing two operators (A and B) sharing a RAN but having dedicated UPF, 5GC, OMC-R, and SCIU components.](bd0b93e7a46ede276d0a3b79ac487bd9_img.jpg) + +The diagram illustrates a Multi-Operator Core Network (MOCN) scenario. On the left, two mobile users, 'Operator A user' and 'Operator B user', are shown. Each user connects to a 'Base station' located within a 'Shared RAN' (indicated by a dashed box). From each base station, there are two types of connections: a red line representing 'Data flow' to a User Plane Function (UPF) and a black line representing 'Control flow' to a 5G Core (5GC). The UPF for Operator A is labeled 'UPF A' and for Operator B is 'UPF B'. Each UPF is connected to a 'Multimedia service platform'. Each 5GC is labeled '5GC-A' and '5GC-B' and is connected to an Operations Management and Control - Radio (OMC-R) unit. Each OMC-R unit is connected to a Shared Cell Information Unit (SCIU). A legend at the bottom indicates that red lines represent 'Data flow' and black lines represent 'Control flow'. The reference 'F.743.15(22)' is noted at the bottom right. + +Diagram of Operator dedicated SCIU deployed in a MOCN showing two operators (A and B) sharing a RAN but having dedicated UPF, 5GC, OMC-R, and SCIU components. + +**Figure I.3 – Operator dedicated SCIU deployed in a MOCN** + +In this scenario, the multimedia service platforms from different operators are independent from each other and the media data is transmitted over the corresponding operators' own user plane function (UPF). MNO (mobile network operator) can deploy the SCIU near the OMC-R (e.g., on the same server), so that the SCIU can efficiently acquire the shared network information from the OMC-R. This information will be processed and generated as a piece of accurate and reliable SCInfo for the multimedia service platform. SCInfo can be stored/cached in the SCIU and periodically updated. + +Figure I.3 illustrates an example of an operator dedicated SCIU deployment for multimedia service in a MOCN. In the Figure, Operator-A user and Operator-B user are able to connect to the same MOCN but they might be attached to the different UPFs which are allocated by their registered 5G Core (5GCs). An Operator-A dedicated multimedia service platform can only provide a certain level of service to the Operator-A user by retrieving the shared network capability described from SCIU belonging to Operator A. + +### I.3 User moving across different MOCNs + +![Diagram illustrating user moving across different MOCNs (MOCN1 and MOCN2) within a Shared RAN. The diagram shows two mobile network cores (MOCN1 and MOCN2) connected to a shared RAN. A user (Operator A user) moves from MOCN1 to MOCN2 via handover. Each MOCN is connected to a UPF (UPF A and UPF B), which is connected to a Multimedia service platform (Multimedia service platform A and Multimedia service platform B). Each MOCN also has a 5GC (5GC-A and 5GC-B), OMC-R, and SCIU. Data flows are shown as red lines, and control flows as black lines.](5cab96b2d23174c25919840ecd50aa48_img.jpg) + +The diagram illustrates a network architecture for user mobility across different Mobile Network Cores (MOCNs). A central 'Shared RAN' (indicated by a dashed box) contains two radio units, 'MOCN1' and 'MOCN2'. An 'Operator A user' is shown performing a 'Handover' from MOCN1 to MOCN2. Each MOCN is connected to its own User Plane Function (UPF): 'UPF A' for MOCN1 and 'UPF B' for MOCN2. These UPFs are connected to respective 'Multimedia service platform A' and 'Multimedia service platform B'. Each MOCN also includes a 5G Core (5GC-A and 5GC-B), an Operation and Management Controller (OMC-R), and a Service Capability Interface (SCIU). Red lines represent 'Data flow', and black lines represent 'Control flow'. The reference 'F.743.15(22)' is noted at the bottom right. + +Diagram illustrating user moving across different MOCNs (MOCN1 and MOCN2) within a Shared RAN. The diagram shows two mobile network cores (MOCN1 and MOCN2) connected to a shared RAN. A user (Operator A user) moves from MOCN1 to MOCN2 via handover. Each MOCN is connected to a UPF (UPF A and UPF B), which is connected to a Multimedia service platform (Multimedia service platform A and Multimedia service platform B). Each MOCN also has a 5GC (5GC-A and 5GC-B), OMC-R, and SCIU. Data flows are shown as red lines, and control flows as black lines. + +**Figure I.4 – User moving across the different MOCNs** + +This scenario is very similar to the case described in Appendix I.2. In this scenario, MOCN1 and MOCN2 belong to the same Operator-A but they are built in the different geographic co-construction areas. The multimedia service platforms, which may be provided by the same SP, can be distributed, deployed and associated with the MOCN area. + +Figure I.4 shows an example of a user moving across the different MOCNs. When the Operator-A user handover from MOCN1 to MOCN2, the current multimedia service status can be migrated from multimedia platform A to multimedia platform B in order to ensure the service continuity. But multimedia platform B should determine if the service level needs to be changed or not, according to the current SCInfo retrieved from SCIU in MOCN2. + +## Bibliography + +- [b-3GPP TS 23.251] 3GPP TS 23.251 v16.0.0 (2020), *Network Sharing; Architecture and functional description (Release 16)*. + +- [b-3GPP TS 23.501] 3GPP TS 23.501 (2020), *System architecture for the 5G System (5GS)*. + + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.743.16-202203-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.743.16-202203-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1e32f9b8f1141bed243c16e02a6cd1411cbb87f1 --- /dev/null +++ b/marked/F/T-REC-F.743.16-202203-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c395c4bbde83b1e232652a43f51538278c116f51f710c37e20a073f7333726c8 +size 6134 diff --git a/marked/F/T-REC-F.743.16-202203-I_PDF-E/acfc53eca625d62b38aa2563efa95c3e_img.jpg b/marked/F/T-REC-F.743.16-202203-I_PDF-E/acfc53eca625d62b38aa2563efa95c3e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..528adfcecffae8ac1fa9032018c8832508f051fd --- /dev/null +++ b/marked/F/T-REC-F.743.16-202203-I_PDF-E/acfc53eca625d62b38aa2563efa95c3e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bb999895e12c126a902f219d0afb40709b405aac226bacaa5c1dacdd47eb5c6a +size 54867 diff --git a/marked/F/T-REC-F.743.16-202203-I_PDF-E/raw.md b/marked/F/T-REC-F.743.16-202203-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..454377f72dfdd13302aef138fffc1d4e4b25f082 --- /dev/null +++ b/marked/F/T-REC-F.743.16-202203-I_PDF-E/raw.md @@ -0,0 +1,377 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.743.16** + +(03/2022) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Requirements for communication resource +management in intelligent visual surveillance +systems** + +Recommendation ITU-T F.743.16 + +ITU-T + +![ITU logo](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with a white grid pattern, overlaid by a stylized 'ITU' text in blue. + +ITU logo + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +### **TELEGRAPH SERVICE** + +| | | +|-----------------------------------------------------------------|-------------| +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | + +### **MOBILE SERVICE** + +| | | +|--------------------------------------------------------|-------------| +| Mobile services and multdestination satellite services | F.110–F.159 | +|--------------------------------------------------------|-------------| + +## **TELEMATIC SERVICES** + +| | | +|-------------------------------------------|-------------| +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | + +## **MESSAGE HANDLING SERVICES** + +F.400–F.499 + +## **DIRECTORY SERVICES** + +F.500–F.549 + +## **DOCUMENT COMMUNICATION** + +| | | +|--------------------------------------|-------------| +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | + +## **DATA TRANSMISSION SERVICES** + +F.600–F.699 + +## **MULTIMEDIA SERVICES** + +**F.700–F.799** + +## **ISDN SERVICES** + +F.800–F.849 + +## **UNIVERSAL PERSONAL TELECOMMUNICATION** + +F.850–F.899 + +## **ACCESSIBILITY AND HUMAN FACTORS** + +F.900–F.999 + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.743.16 + +# Requirements for communication resource management in intelligent visual surveillance systems + +## Summary + +Recommendation ITU-T F.743.16 defines the architecture and specifies the requirements for communication resource management in intelligent visual surveillance (IVS) systems, including the requirements of communication resource monitoring, resource provisioning and resource scheduling. Communication resource management in the IVS system aims to make rational use of communication resources, so that it can complete the tasks which are generated by the IVS system efficiently. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.743.16 | 2022-03-16 | 16 | 11.1002/1000/14957 | + +## Keywords + +Communication resource management, intelligent visual surveillance. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2022 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-------------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Architecture of the communication resource management in an IVS system..... | 3 | +| 7 Requirements of communication resources monitoring in an IVS system..... | 4 | +| 7.1 Requirements of path congestion status monitoring ..... | 4 | +| 7.2 Requirements of path bandwidth occupancy monitoring..... | 4 | +| 7.3 Requirements of path traffic monitoring ..... | 4 | +| 8 Requirements of communication resource provisioning in an IVS system..... | 4 | +| 8.1 Requirements of communication resource discovery..... | 4 | +| 8.2 Requirements of communication path selection..... | 5 | +| 8.3 Requirements of bandwidth reservation ..... | 5 | +| 9 Requirements of communication resource scheduling in an IVS system..... | 5 | +| 9.1 Requirements of bandwidth adaptation in the original path..... | 5 | +| 9.2 Requirements of path reselection ..... | 5 | + + + +# Recommendation ITU-T F.743.16 + +# Requirements for communication resource management in intelligent visual surveillance systems + +## 1 Scope + +This Recommendation specifies the requirements for communication resource management in IVS systems, including communication resource monitoring, communication resource provisioning and communication resource scheduling. Computing resource management and storage resource management in the IVS systems are out of the scope of this Recommendation. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T F.743.1] Recommendation ITU-T F.743.1 (2015), *Requirements for intelligent visual surveillance*. + +[ITU-T H.626 (V2)] Recommendation ITU-T H.626 (2019), *Architecture requirements for video surveillance system*. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 customer unit** [ITU-T H.626 (V2)]: A device located at the customer part of a visual surveillance system and used to present multimedia information (such as audio, video, image, alarm signal, etc.) to the end user. + +**3.1.2 intelligent visual surveillance system** [ITU-T F.743.1]: A system that can automatically identify specific objects, behaviours, or attributes in video signals. They extract data from the video signals, which are then transmitted or archived so that the visual surveillance system can act accordingly. + +**3.1.3 video surveillance system** [ITU-T H.626 (V2)]: A telecommunication service focusing on video (but including audio) application technology, which is used to remotely capture multimedia (such as audio, video, image, alarm signal, etc.) and present them to the end user in a user-friendly manner, based on a managed broadband network with ensured quality, security and reliability. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 communication resource information table**: A table that is used to store all the available communication resources of the IVS system. + +**3.2.2 communication resource management:** The monitoring, provisioning and scheduling of communication resources, which can complete the tasks of the IVS system, such as intrusion detection and portrait recognition. + +**3.2.3 communication resource monitoring:** A module that is used to monitor path status and collect the resources usages by measuring tools in the IVS system. + +**3.2.4 communication resource provisioning:** A module that provides the demanded communication resources for the new tasks to be performed according to tasks' requirements, such as QoS. + +**3.2.5 communication resource provisioning agent:** An agent which performs communication resource discovery and selects suitable communication resources based on the tasks' requirements, such as QoS. + +**3.2.6 communication resource scheduling:** A module that is used for dynamically scheduling communication resources. + +**3.2.7 traffic classification:** An automatic process that classifies data flow into multiple categories based on various parameters (such as port numbers or protocols). + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|-------------------------------------------| +| CRIT | Communication Resource Information Table | +| CRM | Communication Resource Management | +| CRMT | Communication Resource Monitoring | +| CRP | Communication Resource Provisioning | +| CRPA | Communication Resource Provisioning Agent | +| CRS | Communication Resource Scheduling | +| CU | Customer Unit | +| IVS | Intelligent Visual Surveillance | + +## 5 Conventions + +In this Recommendation: + +- The keywords "is required to" indicate a requirement that must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "is recommended" indicate a requirement that is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. +- The keywords "can optionally" and "may" indicate an optional requirement that is permissible, without implying any sense of being recommended. These terms are not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +## 6 Architecture of the communication resource management in an IVS system + +![Figure 6-1: The architecture of communication resource management in an IVS system. The diagram shows the interaction between the Communication Resource Management (CRM) block, the Communication Control System (CCS), the Management System (MS), the Processing Unit (PU), and the Customer Unit (CU). The CRM block contains CRMT, CRP, and CRS modules. The CCS block contains CMU and SCU modules. The MS block contains MDU and MSU modules. The PU and CU are external units. Arrows indicate data (Media) and control (Control) flows between these components. The legend indicates that black arrows represent Media and red arrows represent Control. The reference F.743.16(22) is shown in the bottom right corner of the diagram area.](acfc53eca625d62b38aa2563efa95c3e_img.jpg) + +Figure 6-1: The architecture of communication resource management in an IVS system. The diagram shows the interaction between the Communication Resource Management (CRM) block, the Communication Control System (CCS), the Management System (MS), the Processing Unit (PU), and the Customer Unit (CU). The CRM block contains CRMT, CRP, and CRS modules. The CCS block contains CMU and SCU modules. The MS block contains MDU and MSU modules. The PU and CU are external units. Arrows indicate data (Media) and control (Control) flows between these components. The legend indicates that black arrows represent Media and red arrows represent Control. The reference F.743.16(22) is shown in the bottom right corner of the diagram area. + +**Figure 6-1 – The architecture of communication resource management in an IVS system** + +Figure 6-1 shows the architecture of communication resource management (CRM) in the IVS system, based on the functional architecture of the video surveillance system in [ITU-T H.626 (V2)]. This framework adds three new modules, including the communication resource monitoring (CRMT) module, communication resource provisioning (CRP) module and communication resource scheduling (CRS) module. + +The CRMT module is the basis of efficient data transmission. The CRMT module works on all the communication paths, and is in charge of monitoring the changes in the communication path state in real time, including the congestion status, bandwidth occupancy and the traffic classification transmitted in the paths between the IVS system and the devices. The CRMT module obtains the monitoring data from the MS module of the IVS system. So that the CRMT module can quantify the usage of the communication resources in real time, and guide the provisioning and scheduling of communication resources. + +The CRP module offers three functions, including communication resource discovery, communication path selection and communication resource providing. The CRP module updates the communication resources status of the IVS system and reserves the communication resources for new tasks according to the current communication resources status and the requirements of the tasks. The CRP module consists of communication resource information table (CRIT) and communication resource provisioning agent (CRPA). CRIT is a table that stores all the available communication resources, such as the amount of usable bandwidth on a certain path. Customer unit (CU) can interact with CRPA and submit the QoS requirements of the workload. Based on the requirements (such as QoS) and the communication resource information delivered by the CRIT, CRPA checks the available communication resources of the IVS system. It provisions the demanded communication resources for the new tasks to be performed only if the desired communication resources are available in the IVS system. + +After provisioning the communication resources effectively, the data will be transmitted in the selected path. During transmission, the communication resources may not be able to meet the task requirements, for example, a congested path is detected by the communication resource monitoring module, or the problem of long end-to-end delay is reported by the external devices. Then the CRS module will schedule the appropriate communication resources for the task in time, so that the tasks can utilize the communication resources effectively. The CRS module comprises two functions: bandwidth adaptation and path reselection. + +## **7 Requirements of communication resources monitoring in an IVS system** + +### **7.1 Requirements of path congestion status monitoring** + +PCSM-01: CRMT is required to monitor the number of bits and packets of all the paths of the IVS system in real time. + +PCSM-02: CRMT is required to obtain the delay of each task of the IVS system according to the monitored information. + +PCSM-03: CRMT is required to analyse which flow stresses certain routers or paths according to the monitored information of all the paths of the IVS system. + +PCSM-04: CRMT is required to determine which path is congested and the congestion state of the path according to the obtained transmission delay and traffic. + +PCSM-05: CRMT is recommended to send an alert message to notify the communication resource provisioning module and the communication resource scheduling module when a congested path is detected. + +### **7.2 Requirements of path bandwidth occupancy monitoring** + +PBOM-01: It is required to monitor the bandwidth usage of all paths of the IVS system so that the IVS system can compute path bandwidth occupancy of all paths. + +PBOM-02: Monitoring path bandwidth occupancy is required to use the tools which can measure the bandwidth usage of the path. + +### **7.3 Requirements of path traffic monitoring** + +PTM-01: It is required to get the relevant parameters of the traffic of the paths in the IVS system, such as port number, QoS, etc. + +PTM-02: It is required to determine the traffic type according to the relevant parameters of the traffic of the paths in the IVS system. For example, the IVS system can divide traffic into delay-sensitive traffic and non-delay-sensitive traffic according to the QoS requirements of the traffic, so that the IVS system can obtain the dominant traffic type and analyse the traffic situation. + +PTM-03: Different types of traffic have different requirements, so it is required to mark the identified traffic type. + +PTM-04: The IVS system is required to deploy a communication resource allocation policy for the different traffic according to the traffic types. For example, the IVS system can allocate different communication resources for different traffics. + +## **8 Requirements of communication resource provisioning in an IVS system** + +### **8.1 Requirements of communication resource discovery** + +CRD-01: It is required to obtain information about the available communication resources of the IVS system by using the communication resource monitoring. + +CRD-02: It is required to store all the available communication resources of the IVS system in a table. For example, the types (such as video, audio, etc.) and bandwidth occupancy of each traffic are recorded in the table. + +CRD-03: It is required to update the communication resource information table regularly because the communication resources of the IVS system are dynamic. + +### **8.2 Requirements of communication path selection** + +CPS-01: It is required to select the communication path for the new tasks of the IVS system. + +CPS-02: It is required to submit the related requirements (such as monitoring queries, alarms, etc.) of the tasks to CRPA. + +CPS-03: The IVS system may serve multiple tasks simultaneously, so CRPA will receive tasks submitted by multiple users at the same time. + +CPS-04: According to the relevant requirements of different tasks and the real-time status of the communication resources in the IVS system, it is required to provide appropriate communication resources for different tasks. + +CPS-05: In the IVS system, there are multiple hops during the transmission process of a task, it is required to select the communication path of the next hop for the task according to the current path status and the task's requirements, to achieve the QoS of different tasks. + +CPS-06: It is required to record the selected path for different tasks. + +### **8.3 Requirements of bandwidth reservation** + +BR-01: In order to ensure that the data can be successfully transmitted in the path, adequate bandwidth is required to be reserved for the tasks to be processed in the IVS system. + +BR-02: The communication path of the data transmission may contain multiple hops, it is required to reserve bandwidth for the task in each hop of the communication path. + +BR-03: Two factors are required to be considered when determining the amount of bandwidth to be reserved for a task, including the bandwidth requirements of the task and the total bandwidth requirements of all tasks within the hop. For example, bandwidth is reserved for each task in proportion according to the bandwidth requirements of each task and the total bandwidth requirements of all data transmitted in the path. + +## **9 Requirements of communication resource scheduling in an IVS system** + +### **9.1 Requirements of bandwidth adaptation in the original path** + +BA-01: Bandwidth adaptation in the original path as an efficient way is required to utilize communication resources efficiently in the IVS system. + +BA-02: During data transmission, the reserved bandwidth in the communication resource provisioning of some hops in the path may not be able to meet the task requirements. When this problem occurs, bandwidth is required to be adjusted and the free bandwidth in the hop are required to be considered. + +BA-03: If there is idle bandwidth reserved for other tasks, that is, part of the reserved bandwidth is not used, and the idle bandwidth is recommended to be temporarily adjusted to the task with insufficient reserved bandwidth. + +BA-04: Bandwidth adaptation in the original path is required to reallocate bandwidth for the tasks to meet the requirements of transmission. + +### **9.2 Requirements of path reselection** + +PR-01: It is required to reselect the path if the data transmitted in the original path cannot meet its QoS requirements, and more bandwidth cannot be scheduled through bandwidth adaptation in the original path. + +PR-02: Path reselection is required to reselect some hops of the path or the reselection of the whole path. + +PR-03: When reselecting a hop of a path, it is required to consider whether there is an alternative hop between two points. The hop can be replaced if there is an alternative hop and the bandwidth of the alternative hop can meet the transmission requirements of the task. + +PR-04: The whole path is required to be reselected when the single-hop replacement cannot be carried out. + +PR-05: A new path with sufficient bandwidth is required to be selected to meet the requirements. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.743.17-202203-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.743.17-202203-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1e32f9b8f1141bed243c16e02a6cd1411cbb87f1 --- /dev/null +++ b/marked/F/T-REC-F.743.17-202203-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c395c4bbde83b1e232652a43f51538278c116f51f710c37e20a073f7333726c8 +size 6134 diff --git a/marked/F/T-REC-F.743.17-202203-I_PDF-E/e1a0d046fbe7f28f5e93a47091851747_img.jpg b/marked/F/T-REC-F.743.17-202203-I_PDF-E/e1a0d046fbe7f28f5e93a47091851747_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..10c48604c876af3fc4df6babab19cccb411f1229 --- /dev/null +++ b/marked/F/T-REC-F.743.17-202203-I_PDF-E/e1a0d046fbe7f28f5e93a47091851747_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:42e869a8aea61d98761703bd978e7688fef3904beccc3c1d2c523b0eb28df3c5 +size 50970 diff --git a/marked/F/T-REC-F.743.17-202203-I_PDF-E/raw.md b/marked/F/T-REC-F.743.17-202203-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..a1fbbdab9356c50c2e0b726c46692cabb820e155 --- /dev/null +++ b/marked/F/T-REC-F.743.17-202203-I_PDF-E/raw.md @@ -0,0 +1,424 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.743.17** + +(03/2022) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +# --- **Requirements for cloud gaming systems** + +Recommendation ITU-T F.743.17 + +ITU-T + +![ITU logo](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with a white grid pattern, and the letters 'ITU' in white, bold, sans-serif font, superimposed on the globe. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +## Recommendation ITU-T F.743.17 + +# Requirements for cloud gaming systems + +## Summary + +Recommendation ITU-T F.743.17 describes the requirements for a cloud gaming system including user requirements, service requirements, performance requirements, management requirements, security requirements, network requirements and terminal requirements for a cloud gaming system. It also describes the typical scenarios including cloud game distribution scenario, cloud game running scenario, cloud game cross-platform scenario and cloud game live streaming scenario for cloud gaming system in Appendix I. + +This Recommendation is intended to provide a reference for cloud gaming service providers, platform providers, and developers when they build and operate a cloud gaming system. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.743.17 | 2022-03-16 | 16 | 11.1002/1000/14958 | + +### Keywords + +Application scenarios, cloud gaming system, requirements. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2022 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|-----|-------------------------------------------------------------------------------|---| +| 1 | Scope ..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 1 | +| 4 | Abbreviations and acronyms ..... | 1 | +| 5 | Conventions ..... | 2 | +| 6 | Overview ..... | 2 | +| 7 | User requirements ..... | 3 | +| 7.1 | CGU requirements..... | 3 | +| 7.2 | CGSP requirements ..... | 3 | +| 7.3 | CGD requirements..... | 3 | +| 8 | Service requirements ..... | 3 | +| 9 | Performance requirements ..... | 4 | +| 10 | Management requirements..... | 4 | +| 11 | Security requirements ..... | 4 | +| 12 | Network requirements ..... | 5 | +| 13 | Terminal requirements..... | 5 | +| | Appendix I – Typical application scenarios of cloud gaming system (CGS) ..... | 6 | +| I.1 | Game as a Service (GaaS) ..... | 6 | +| I.2 | Click-and-play scenario..... | 6 | +| I.3 | Cloud game cross-platform scenario ..... | 6 | +| I.4 | Live cloud game streaming scenario ..... | 7 | +| I.5 | Other scenarios ..... | 7 | +| | Bibliography..... | 9 | + + + +# Requirements for cloud gaming systems + +# 1 Scope + +This Recommendation describes the requirements for cloud gaming system. + +The scope of this Recommendation includes: + +- 1) User requirements for the cloud gaming system. +- 2) Service requirements for the cloud gaming system. +- 3) Performance requirements for the cloud gaming system. +- 4) Management requirements for the cloud gaming system. +- 5) Security requirements for the cloud gaming system. +- 6) Network requirements for the cloud gaming system. +- 7) Terminal requirements for the cloud gaming system. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following term defined elsewhere: + +**3.1.1 cloud computing** [b-ITU-T Y.3500]: Paradigm for enabling network access to a scalable and elastic pool of shareable physical or virtual resources with self-service provisioning and administration on-demand. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following term: + +**3.2.1 cloud gaming**: A way of playing the game on a cloud server. In the "cloud gaming" mode, game storage, computing, rendering, and so on are completed in the cloud, players can experience high-quality games through any terminal, anytime anywhere. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +CGD Cloud Gaming Developer + +CGPP Cloud Gaming Platform Provider + +CGS Cloud Gaming System + +CGSP Cloud Gaming Service provider + +CGU Cloud Gaming User + +# 5 Conventions + +In this Recommendation: + +- The keywords "is recommended" indicate a requirement that is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. + +# 6 Overview + +Cloud game is a game mode based on cloud computing. Under the operation mode of cloud games, all games are running on the cloud side, and the game images are compressed and transmitted to users through the network after rendering. On the client side, the user's game device does not need any high-end processors or graphics cards, simply a basic video decompression capability is needed. In the cloud gaming system (CGS), there are four roles: cloud gaming developer (CGD), cloud gaming platform provider (CGPP), cloud gaming service provider (CGSP) and cloud gaming user (CGU). CGD completes the development of the game content and technical implementation on the game development platform, CGPP builds the cloud gaming platform to provide the cloud game service and management; CGSP distributes and operates the cloud games in the form of cloud game platforms or channels, including the development of client / terminal devices for the business side and the optimization of cloud game capabilities for the terminals. CGU uses cloud game apps to play games. The relationship between these roles is shown in Figure 6-1. + +![Diagram illustrating the roles in the cloud gaming system (CGS). The diagram shows four main components: Cloud gaming user (top), Cloud gaming service provider and Cloud gaming developer (middle), Cloud gaming platform provider (bottom), and Cloud computing platform and network (bottom-most, dashed box). Arrows indicate the flow of data and services: Infrastructure from the bottom to the platform provider; Basic ability from the platform provider to the service provider; Game content from the developer to the platform provider; and Game distribution from the service provider to the user. A bracket on the left groups the service provider, developer, and platform provider under the label CGS.](e1a0d046fbe7f28f5e93a47091851747_img.jpg) + +The diagram illustrates the roles in the cloud gaming system (CGS). It consists of four main components arranged vertically: + +- Cloud gaming user** (top): Represented by an icon of three people. +- Cloud gaming service provider** and **Cloud gaming developer** (middle): The service provider is represented by an icon of a person, and the developer by a gear icon. These two are grouped together by a bracket labeled **CGS**. +- Cloud gaming platform provider** (bottom): Represented by a cloud icon with a game controller. +- Cloud computing platform and network** (bottom-most, dashed box): Represented by a cloud icon. + +Arrows indicate the flow of data and services: + +- Infrastructure**: An arrow points from the bottom-most box to the platform provider. +- Basic ability**: An arrow points from the platform provider to the service provider. +- Game content**: An arrow points from the developer to the platform provider. +- Game distribution**: An arrow points from the service provider to the user. + +F.743.17(22) + +Diagram illustrating the roles in the cloud gaming system (CGS). The diagram shows four main components: Cloud gaming user (top), Cloud gaming service provider and Cloud gaming developer (middle), Cloud gaming platform provider (bottom), and Cloud computing platform and network (bottom-most, dashed box). Arrows indicate the flow of data and services: Infrastructure from the bottom to the platform provider; Basic ability from the platform provider to the service provider; Game content from the developer to the platform provider; and Game distribution from the service provider to the user. A bracket on the left groups the service provider, developer, and platform provider under the label CGS. + +**Figure 6-1 – Roles in cloud gaming system (CGS)** + +This Recommendation describes the requirements for the CGS. For cloud gaming system requirements, it describes typical requirements including user requirements, service requirements, performance requirements, management requirements, security requirements, network requirements, and terminal requirements. For application scenarios, it describes typical scenarios including cloud game distribution scenario, cloud game running scenario, cloud game cross-platform scenario and cloud game live streaming scenario in Appendix I. + +# **7 User requirements** + +## **7.1 CGU requirements** + +- [USR-01] The CGS is recommended to support user account registration, account cancellation, and personal information modification. All these features need to be presented to the CGU in a simple, intuitive UI. +- [USR-02] The CGS is recommended to support user login and logout. Multiple login forms support is recommended, including account password, fingerprint identification, face recognition and voiceprint recognition, etc. +- [USR-03] The CGS is recommended to support user points, achievements, rankings, forums and comments, etc., to fully mobilize consumer engagement and enhance the sociality of the game. +- [USR-04] The CGS is recommended to support user account exchange. CGUs can host their own account to the CGS, and others can easily borrow or lease these accounts. + +## **7.2 CGSP requirements** + +- [USR-05] The CGS is recommended to support business account registration, account cancellation and corporate information modification. +- [USR-06] The CGS is recommended to support business account login and logout. +- [USR-07] The CGS is recommended to support game titles management, including game release, update, revocation, etc. +- [USR-08] The CGS is recommended to support funds management, including top up service, resource leasing and game operations. + +## **7.3 CGD requirements** + +- [USR-09] The CGS is recommended to support the channel access function, which defines the API connection of various functions such as account, payment and advertisement used in different channels to provide a unified environment for the development and operation of the cloud games and reduce the developer's access cost. +- [USR-10] The CGS is recommended to support platform hardware capability, to make hardware capabilities of client devices for use on cloud devices, such as cameras, microphones, Bluetooth, accelerometers, compasses, gyroscopes, motors, and other capabilities add more playability to the game. +- [USR-11] The CGS is recommended to support user input interaction, for transferring input operations from cloud devices to client devices so that local input methods can be used more smoothly and cleanly. + +# **8 Service requirements** + +- [SER-01] The CGS is recommended to support game distribution. The CGS can assign a game instance to the CGU in response to the request, transmitting the game video stream to the designated terminal, and receiving game operation instructions for the user. +- [SER-02] The CGS is recommended to support the game resource supply function. Based on the flexible resource supply capability of the cloud computing system, the CGS can automatically allocate operations, storage, network resources, etc. to game instances according to the requirements. +- [SER-03] The CGS is recommended to support game rendering. Cloud game instances can perform operations and graphics rendering in the cloud and generate real-time game video streams. + +- [SER-04] The CGS is recommended to support cross-platform adaptation. In CGS, game parameters such as resolution, display scale, image effects, HDR, etc. can be configured to adapt to the different platforms and terminals. +- [SER-05] The CGS is recommended to support preloading. The cloud game system can pre-instantiate popular games according to game popularity statistics and predictions, reducing user waiting time and improving the quality of the game experience. +- [SER-06] The CGS is recommended to support the game's live broadcasting function. Combined with the game live broadcast, the cloud game video stream can have the functions of barrage interaction, anchor interaction, game relay, etc., increasing the sociability and playability of the cloud game. + +# **9 Performance requirements** + +- [PFM-01] The CGS is recommended to support short-term waiting to achieve click-and-play, reducing game launching time. +- [PFM-02] The CGS is recommended to support low latency to achieve a fast response of user operations on the game screen, and the end-to-end time delays are better to no higher than 70 ms. +- [PFM-03] The CGS is recommended to support the adaptive resolution. Users can manually or automatically select the game resolution suitable due to the different network environments. The resolution is set to be smooth, standard definition, high definition and ultra-definition. +- [PFM-04] The CGS is recommended to support a minimum of 30 frames of game screens. The ultra- definition image quality needs to support a minimum of 60 frames of game screens to ensure the smoothness of the game. +- [PFM-05] The CGS is recommended to support variable bitrate transmission to dynamically adjust the bitrate based on the clarity and the quality of the game. + +# **10 Management requirements** + +- [MAN-01] The CGS is recommended to support resource managements, such as resource usage dashboard, statistics and resource elastic configuration. +- [MAN-02] The CGS is recommended to support system management, for example, providing a unified system management UI to implement the functions of the cloud gaming system. +- [MAN-03] The CGS is recommended to support the operation and maintenance management, including recording the operating status of the system and maintaining the stable operation of the system. + +# **11 Security requirements** + +- [SEC-01] The CGS is recommended to support permission security management. The CGS provides users with an authentication mechanism to ensure user rights and prevent unauthorized users from accessing the CGS. +- [SEC-02] The CGS is recommended to support system security management. The CGS provides the system with the ability to prevent viruses and attacks. +- [SEC-03] The CGS is recommended to support content security management, including content information security, content distribution security and content audit and other capabilities. + +- [SEC-04] The CGS is recommended to support personal information protection management, including in the entire life cycle of information collection, transmission, storage, sharing and other protection capabilities. +- [SEC-05] The CGS is recommended to support network security management, including access control, slice security, edge computing security and other capabilities. +- [SEC-06] The CGS is recommended to support business and application security management, including identity authentication, data transmission, business security, application security and other capabilities. + +# **12 Network requirements** + +- [NET-01] The CGS is recommended to support wide network coverage, indoor, outdoor, and hot spots coverage should be realized. In some extreme scenarios, there will be a high demand for outdoor high-speed coverage capability of the network. +- [NET-02] The CGS is recommended to support network slicing, different types of games differentiation ability and SLA web service demand. Network slice of network operator needs to satisfy the cloud game differentiation deterministic network capacity, and fulfill the need to give business operators the provision to provide flexible quick open network business ability. This can be flexible according to the need to deploy and manage, thereby meeting the needs of the cloud game business scenario. +- [NET-03] The CGS is recommended to support MEC, with distributed sharing mode of computing node edges, which brings lower interaction delay and higher image response speed, thus further improving user experience, including edge rendering capability, business operation environment, graphics API interface, resource management interface and other capabilities. + +# **13 Terminal requirements** + +- [TER-01] The CGS is recommended to support various game terminals, including mobile terminals, such as intelligent terminal products; immersive terminals, such as VR / AR devices and products; console terminals, such as game controllers; large-screen terminals, such as PC, TV and other large-screen products. +- [TER-02] The CGS is recommended to support key technical capabilities which have consistent decoding capability, screen touch delay, operation mode and other common hardware performance on terminal devices, providing open access capability for all kinds of cloud game terminals. + +# Appendix I + +## Typical application scenarios of cloud gaming system (CGS) + +(This appendix does not form an integral part of this Recommendation.) + +This appendix describes the typical application scenarios of cloud gaming systems (CGS). + +### I.1 Game as a Service (GaaS) + +The CGS provides GaaS to implement the servitization of games for CGSP and CGU. CGSP can publish cloud games as a service swiftly by uploading game packages, configuring archive paths, resolutions and other simple operations. CGU can play a cloud game via the Internet as a service. + +Step 1: The CGSP logs in to the CGS, uploads the game package body and fills in the game description metadata information, including game type, game introduction, game icon, game screenshots and videos, etc. + +Step 2: The CGS analyses the uploaded game package body and allocates resources to the game. + +Step 3: The CGS automatically or manually configures the game save path, resolution, resource requirements, etc. + +Step 4: After CGS review and manufacturer test, the cloud game is successfully released. + +### I.2 Click-and-play scenario + +The CGS provides CGU with click-and-play, free-to-install gaming services. The computing and rendering of the game are completed in the cloud. The CGU can receive the game video stream and send control commands to the cloud after installing and running the cloud game system client on the terminal devices, such as mobile phones, tablets, TVs, game consoles, computers, etc. + +Step 1: CGU logs in to the cloud game client and the client sends an authentication request to the CGS. + +Step 2: After the CGS authenticates the user, the client sends a game list request to the CGS and the CGS sends the corresponding game list to the client. + +Step 3: The CGU selects the game and the client sends a request for a game instance to CGS. + +Step 4: The CGS receives the application request for the game instance, establishes the instance in the cloud, calls the corresponding game and archived information, generates the video stream through coding and sends the address of the video stream to the client. + +Step 5: The client receives the game video stream, and the CGU starts the game and sends control commands to the instance through the controller. + +### I.3 Cloud game cross-platform scenario + +Cloud games provide cross-platform gaming services to CGUs. Users can play games when at home, during work, or when having fun. CGU can easily play games anytime anywhere with all the information, including user information, VIP rights, and archive data, saved in the cloud and shared across all terminals. + +#### Case 1: One terminal, all games + +CGU can play all types of games such as console games and PC games, on a certain terminal, for example, a mobile phone, through the proprietary or virtual peripheral controller. Consistent gaming experience is guaranteed. + +Step 1: CGU logs in to the cloud game system client on one terminal, select games on other terminals, and initiate game requests. + +Step 2: The CGS receives the game request, generates the game instance, adapts the parameters according to the difference in terminals and returns the address of the game instance. + +Step 3: The client receives the game instance address, starts the game, initializes the corresponding terminal controller and sends a game video stream request to the instance address. + +Step 4: The instance receives the video stream request reads the game archive and sends the game stream to the client. + +Step 5: The client receives the game video stream, and the CGU starts the game and sends control commands to the instance through the controller. + +#### **Case 2: One game, all terminals** + +GC can share the same game archive through the cloud game system and use different terminals to play the same game on different occasions. For example, on the way to work, CGU plays the game using a mobile phone. At home, the user can continue the game through a computer or a console for a better gaming experience. + +Step 1: CGU uses a terminal to log in to the CGS client and selects a game archive to initiate a game request. + +Step 2: After completing steps 2-5 in Case 1, the CGU plays the game. + +Step 3: The CGU ends the game and initiates a game termination request. + +Step 4: The CGS receives the game termination request, saves the game and ends the video stream transmission. + +Step 5: CGU uses other terminals to log in to the CGS client and selects the game archive to initiate the game requests. + +Step 6: After completing steps 2-5 in Case 1, the CGU continues with the game. + +### **I.4 Live cloud game streaming scenario** + +As a video streaming, the cloud game itself can be integrated with existing live streaming platforms. Combined with the interactive nature of the cloud game video streaming and the socialization of the live video streaming, a fusion scene based on various functions such as barrage interaction, anchor interaction and game relay is formed. + +Steps: [TBD] + +### **I.5 Other scenarios** + +Two important features of cloud gaming are: 1) game screen is transmitted through streaming media; 2) game running and operation instruction input are separated. Using these two characteristics, the following scenarios can be derived: + +#### **I.5.1 Interactive live broadcast scenario** + +Similar to the traditional live game, the streamer plays games and other users watch it. On this basis, we can take advantage of the separation of game running and operation instruction input in cloud gaming to support the transfer of game control from the streamer to one audience. This can provide stronger interactivity. + +#### **I.5.2 Screen clone scenario** + +The game screen generated by a cloud gaming instance can be transmitted to different user terminals at the same time. Multiplayer games can realize one screen for one person. + +#### **I.5.3 Instruction splitting scenario** + +Game operation instruction originally issued by one terminal can change to the instruction issued by multiple terminals, and the instruction synthesis is carried out at the server. The non-multiplayer game can support the multiplayer in this scenario. + +# Bibliography + +[b-ITU-T Y.3500] Recommendation ITU-T Y.3500 (2014) | ISO/IEC 17788:2014, *Information technology – Cloud computing – Overview and vocabulary*. + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.743.18-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.743.18-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ac295b1b0c7f14b1c5a33ef6682b1b017c812477 --- /dev/null +++ b/marked/F/T-REC-F.743.18-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37aed4fc114a39843438175326e0b7e1d09a9d6075723d82fea32040fd560ee4 +size 7192 diff --git a/marked/F/T-REC-F.743.18-202212-I_PDF-E/raw.md b/marked/F/T-REC-F.743.18-202212-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..5cf7b34f0e90e505ab17c94b5c2906df78b259ed --- /dev/null +++ b/marked/F/T-REC-F.743.18-202212-I_PDF-E/raw.md @@ -0,0 +1,483 @@ + + +# Recommendation **ITU-T F.743.18 (12/2022)** + +SERIES F: Non-telephone telecommunication services +Multimedia services + +## --- **Requirements for IMT-2020 ultra-high definition surveillance cameras** + +![ITU logo](390120de4fe440c42fea8154fcaad334_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +### **TELEGRAPH SERVICE** + +| | | +|-----------------------------------------------------------------|-------------| +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | + +### **MOBILE SERVICE** + +| | | +|--------------------------------------------------------|-------------| +| Mobile services and multdestination satellite services | F.110–F.159 | +|--------------------------------------------------------|-------------| + +### **TELEMATIC SERVICES** + +| | | +|-------------------------------------------|-------------| +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | + +### **MESSAGE HANDLING SERVICES** + +F.400–F.499 + +### **DIRECTORY SERVICES** + +F.500–F.549 + +### **DOCUMENT COMMUNICATION** + +| | | +|--------------------------------------|-------------| +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | + +### **DATA TRANSMISSION SERVICES** + +F.600–F.699 + +### **MULTIMEDIA SERVICES** + +**F.700–F.799** + +### **ISDN SERVICES** + +F.800–F.849 + +### **UNIVERSAL PERSONAL TELECOMMUNICATION** + +F.850–F.899 + +### **ACCESSIBILITY AND HUMAN FACTORS** + +F.900–F.999 + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.743.18 + +# Requirements for IMT-2020 ultra-high definition surveillance camera + +## Summary + +Recommendation ITU-T F.743.18 specifies functional, performance and security requirements of IMT-2020 ultra-high definition (UHD) surveillance cameras, in order to ensure reliable UHD video transmission in IMT-2020. This Recommendation also specifies services for classification of IMT-2020 UHD surveillance, service level agreement ranking of IMT-2020 UHD surveillance and network requirements for IMT-2020 UHD video surveillance, which are very relevant to IMT-2020 surveillance scenarios, to meet the actual user's UHD video capture and transmission requirements. Recommendation ITU-T F.743.18 also describes typical use cases. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.743.18 | 2022-12-14 | 16 | 11.1002/1000/15185 | + +## Keywords + +IMT-2020, surveillance, UHD. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2023 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|---------------------------------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation ..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Overview of IMT-2020 UHD video business ..... | 2 | +| 6.1 Classification of IMT-2020 UHD surveillance business ..... | 2 | +| 6.2 SLA rank of IMT-2020 UHD surveillance business..... | 3 | +| 6.3 Network construction requirements for IMT-2020 UHD video surveillance business ..... | 3 | +| 7 Use cases of IMT-2020 UHD surveillance cameras ..... | 4 | +| 7.1 Remote control of gantry cranes with an IMT-2020 UHD surveillance camera in a smart port..... | 4 | +| 7.2 Smart environmental protection ..... | 4 | +| 7.3 Smart inspection for power system ..... | 5 | +| 7.4 Smart travel..... | 5 | +| 7.5 HD broadcast ..... | 5 | +| 7.6 Security for temporary event ..... | 6 | +| 7.7 Smart farm ..... | 6 | +| 7.8 Smart fishery..... | 6 | +| 8 Requirements for IMT-2020 UHD surveillance camera ..... | 6 | +| 8.1 Functional requirements..... | 6 | +| 8.2 Performance requirements ..... | 8 | +| 8.3 Security requirements..... | 8 | + + + +# Requirements for IMT-2020 ultra-high definition surveillance camera + +# 1 Scope + +This Recommendation specifies the functional, performance and security requirements of IMT-2020 ultra-high definition (UHD) cameras. The requirements include those for service level agreement (SLA) monitoring, video coder-decoder (codec), application layer transport protocols and video transmission performance. This Recommendation also specifies the classification of IMT-2020 UHD surveillance services, SLA ranking of IMT-2020 UHD surveillance business and network requirements for IMT-2020 UHD video surveillance business, which are very relevant to IMT-2020 surveillance scenarios. + +This Recommendation includes: + +- classification and SLA ranking of IMT-2020 UHD surveillance business; +- network requirements for IMT-2020 UHD video surveillance business; +- use cases of IMT-2020 UHD surveillance cameras; +- functional requirements for IMT-2020 UHD surveillance cameras; +- performance requirements for IMT-2020 UHD surveillance cameras; +- security requirements for IMT-2020 UHD surveillance cameras. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +- [ITU-T H.222.0] Recommendation ITU-T H.222.0 (2021), *Information technology – Generic coding of moving pictures and associated audio information: systems*. +- [ITU-T H.264] Recommendation ITU-T H.264V14 (2021), *Advanced video coding for generic audiovisual services*. +- [ITU-T H.265] Recommendation ITU-T H.265V8 (2021), *High efficiency video coding*. +- [ITU-T H.626] Recommendation ITU-T H.626V2 (2019), *Architecture requirements for video surveillance system*. +- [ITU-T H.627] Recommendation ITU-T H.627V2 (2020), *Signalling and protocols for a video surveillance system*. +- [ITU-T T.81] Recommendation ITU-T T.81 (1992), *Information technology – Digital compression and coding of continuous-tone still images – requirements and guidelines*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following term defined elsewhere: + +**3.1.1 premises unit** [ITU-T H.626]: A device located at the remote part of a video surveillance system and used to capture multimedia information (such as audio, video, image, alarm signal, etc.) from a surveilled object. + +## **3.2 Terms defined in this Recommendation** + +None. + +# **4 Abbreviations and acronyms** + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|----------|----------------------------------------------| +| 3G | third Generation | +| 4G | fourth Generation | +| codec | coder-decoder | +| E2E | End to End | +| HD | High Definition | +| IMT-2020 | International Mobile Telecommunications-2020 | +| IP | Internet Protocol | +| O&M | Operation and Maintenance | +| PU | Premises Unit | +| SLA | Service Level Agreement | +| TCP | Transmission Control Protocol | +| UHD | Ultra-High Definition | +| WAN | Wide-Area Network | + +# **5 Conventions** + +In this Recommendation: + +- The phrase "**is required to**" indicates a requirement that must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The phrase "**is recommended**" indicates a requirement that is recommended but which is not absolutely required. Thus, this requirement need not to be present to claim conformance. +- The phrases "**can optionally**" and "**may**" indicate an optional requirement that is permissible, without implying any sense of being recommended. These terms are not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator or service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +# **6 Overview of IMT-2020 UHD video business** + +This clause specifies the classification, SLA rank and network performance requirements for IMT-2020 UHD surveillance business, which are the prerequisites for defining the functional and performance requirements for IMT-2020 UHD camera. + +## **6.1 Classification of IMT-2020 UHD surveillance business** + +IMT-2020 UHD surveillance business can be divided into the following video business categories: + +- a) type A: Real-time control level, the receiving buffer does not exceed 150 ms; + +- b) type B: Interactive real-time broadcast level, the receiving buffer is not more than 500 ms; +- c) type C: Security monitoring level, the receiving buffer is not greater than 1 000 ms. + +## 6.2 SLA rank of IMT-2020 UHD surveillance business + +This clause specifies the SLA grade for IMT-2020 UHD surveillance business, Table 6-1 shows the different SLA grades that suit type A, type B and type C surveillance business. + +**Table 6-1 – SLA rank of IMT-2020 UHD surveillance business** + +| SLA grade | Frame receiving rate (%) | Frame loss rate | Average frame loss time (s) | Acceptability | +|-----------|--------------------------|-----------------|-----------------------------|----------------------------------| +| 1 | 99.9 | 1/1 000 | 40 | Unacceptable | +| 2 | 99.99 | 1/10 000 | 400 | | +| 3 | 99.999 | 1/100 000 | 4 000 | Acceptable for type B | +| 4 | 99.999 9 | 1/1 000 000 | 40 000 | Acceptable for type A and type C | +| 5 | 99.999 99 | 1/10 000 000 | 400 000 | Acceptable for type A, B and C | +| 6 | 99.999 999 | 1/100 000 000 | 4 000 000 | Acceptable for type A, B and C | + +NOTE – Video frame is the objective metric for IMT-2020 surveillance business, this table is based on a rate of 25 frame/s. + +## 6.3 Network construction requirements for IMT-2020 UHD video surveillance business + +In the IMT-2020 surveillance scenario, the UHD video is transmitted in the wireless situation, due to the different coverage and capacity of the user's location, the IMT-2020 network provides different carrying capacity. The classifications and requirements of the network for specific video businesses are shown in Table 6-2. + +**Table 6-2 – Classification and requirement for the IMT-2020 UHD video business** + +| No. | Network | Description | Network performance | | | Supportable business type | +|-----|---------|----------------------------------------------------------------------------------------------------------------------------------------------------|--------------------------------------------------|----------------|-------------------------------|---------------------------| +| | | | Uplink/downlink bandwidth (single user) [Mbit/s] | E2E delay (ms) | Network package loss rate (%) | | +| 1 | I | Supports carriage of high bit rate and low latency video business, e.g., near the point of a wide-area network (WAN) or IMT-2020 dedicated network | 70/500 | ≤ 30 | ≤ 1 | A, B and C | + +**Table 6-2 – Classification and requirement for the IMT-2020 UHD video business** + +| No. | Network | Description | Network performance | | | Supportable business type | +|-----|---------|---------------------------------------------------------------------------------------------------------|--------------------------------------------------|----------------|-------------------------------|---------------------------| +| | | | Uplink/downlink bandwidth (single user) [Mbit/s] | E2E delay (ms) | Network package loss rate (%) | | +| 2 | II | Supports carriage of medium bit rate and delay insensitive video business, e.g., the mid-point of a WAN | 40/300 | $\leq 50$ | $\leq 3$ | B (SLA $\geq 4$ ) and C | +| 3 | III | Supports carriage of low bit rate and delay-insensitive video services, e.g., remote locations on a WAN | 15/100 | $\leq 80$ | $\leq 10$ | B (SLA $\geq 3$ ) and C | + +NOTE 1 – Mobile handover can exist for all the above network scenarios, which is manifested in that the bit rate is damaged within 100 ms, the time delay becomes larger, the packet loss rate increases, and the bandwidth decreases. +NOTE 2 – Uplink refers to the uplink channel in which message data is carried from the cellular communication terminal to the server, while downlink refers to the channel in which data is carried from the server to the cellular communication terminal. + +# 7 Use cases of IMT-2020 UHD surveillance cameras + +## 7.1 Remote control of gantry cranes with an IMT-2020 UHD surveillance camera in a smart port + +Traditional ports mainly deploy optical fibre and industrial wireless fidelity for key services and communication systems, covering gantry cranes, container trucks and video surveillance systems. However, these methods often cause high construction and operation and maintenance (O&M) costs, inflexible deployment, and low stability and reliability. + +IMT-2020 UHD surveillance cameras installed on each gantry crane can solve these issues by sending surveillance video feedback to the central control room over the network. IMT-2020 wireless networks can also enable multiple gantry cranes to be operated together and solve the problem of optical fibre winding, effectively reducing device purchase costs (such as those for cable) and costs of cable replacement. The ultra-high bandwidth ensures that the uplink rate meets the requirements of concurrent high definition (HD) video services, and the ultra-low latency ensures the precision of remote control effectively. + +## 7.2 Smart environmental protection + +Traditional environmental detection faces problems with high labour resources cost, fast response, and lack of sampling points. In addition, detection personnel may miss certain alarms, or alarms may fail to be reported. + +To combat this, an IMT-2020 multi-dimensional environmental surveillance system installed with 4K/8K HD IMT-2020 UHD surveillance cameras and environmental detection sensors at key points, such as drainage ports, river courses and forests, can efficiently detect the environmental issue and solve these traditional problems. Using IMT-2020 edge computing, the system can: analyse and process real-time video data; automatically identify and warn against unauthorized emissions and abnormal objects; summarize and generate visualized analysis results; and coordinate with + +environmental protection management departments at all levels to implement all-round, efficient, and multi-dimensional environmental surveillance. + +## 7.3 Smart inspection for power system + +### Case 1: Power distribution room inspection + +A power distribution room contributes to the last phase for power supply, with clear correlations between its management level and the power supply capability and quality. Currently, most power distribution rooms are unattended and lack intelligent surveillance and analysis methods. Because these distribution rooms are usually located in remote regions, personnel cannot conduct onsite predictive maintenance and periodic manual inspection costs are high. In the event of a fault, a large number of sites will be affected, and the fault recovery is not immediate and time-consuming. + +To resolve this issue, an IMT-2020 UHD surveillance camera with power distribution room inspection applications can be easily deployed to replace manual inspection. An IMT-2020 UHD surveillance camera can realize real-time video feedback to the analysis central platform to identify abnormalities. Alternatively, this IMT-2020 UHD surveillance camera can execute an intelligent inspection algorithm by itself and immediately identify any abnormal devices or issues. This intelligent modernization solution can reduce manual inspection time by half, significantly decreasing the frequency of emergency handling, and improve power supply quality. + +### Case 2: Power line inspection + +Traditional power line inspection is required to successfully detect abnormal situations, such as a bird's nest or honeycomb in the tower and power pole, absence of a suspension clamp pin, a broken end of a shockproof hammer and insulator self-explosion. The pictures and videos are required to be in UHD for an analysis centre platform to detect a fault. It is vital to report these issues to the control centre as soon as possible to prevent power accidents. Usually, workers need to climb on to the tower and check them one by one, which is time-consuming and dangerous. Even with third generation (3G) and fourth generation (4G) networks, high-resolution images and videos cannot be transmitted to the terminal in real time due to the limitation of bandwidth of the transmission channel. It cannot meet the requirement for immediate feedback. + +IMT-2020 network and IMT-2020 UHD surveillance cameras can be deployed on a drone, which can be used for ultra-long range remote control as long as there is IMT-2020 network coverage in the region. Drones in the connected region may communicate with each other. The inspector can control the drone to take UHD pictures or videos of the electric power line and the electricity tower, and then transmit them to the analysis centre platform. The analysis centre platform with computation resources can determine abnormal situations and alarms. Therefore, the worker can fix the problem immediately without economic losses occurring. + +## 7.4 Smart travel + +Nowadays, more and more shuttle buses, tour buses, and dangerous goods vehicles are required to deploy real-time surveillance cameras to report fatigued drivers and help prevent social public safety incidents. Currently, vehicle-mounted video surveillance cameras mainly use 3G or 4G networks for real-time data transmission. Due to limited uplink bandwidth, limited coverage, and high latency, surveillance video feeds and images are of poor quality. All of these issues can be solved by IMT-2020 UHD surveillance cameras; with HD image quality, more intelligent algorithms can be executed accurately to prevent fatigued driving, and identify the driver or abnormal driving behaviours, etc. + +## 7.5 HD broadcast + +IMT-2020 UHD cameras can be used for webcasting of remote regions (such as Mount Everest), providing users with a UHD video experience at convenient times or places. Due to the harsh + +environmental and climatic conditions of such areas, traditional camera deployment cost is high, while the deployment of IMT-2020 UHD surveillance cameras will be more flexible, using high bandwidth and low latency features to achieve real-time and efficient live broadcast of beautiful scenery. + +## **7.6 Security for temporary event** + +Sometimes, a large-scale exhibition or concert is temporarily held in a certain area where the existing monitoring system is incomplete, and lacks coverage and intelligent functions. To achieve full monitoring coverage and intelligent surveillance in the short term, the traditional wired monitoring deployment method is expensive and needs a long construction time. In contrast, IMT-2020 network coverage with IMT-2020 UHD surveillance cameras deployed is more simple and efficient, flexible to achieve multi-directional, no dead-angle surveillance deployment and can deliver intelligent monitoring transformation in a short time. IMT-2020 UHD surveillance cameras can also be dismantled afterwards, without causing waste of resource investment. + +## **7.7 Smart farm** + +With the development of urbanization, animal cultivation is becoming more and more specialized. Usually, farmers have a large number of animals such as cows and sheep in a wide region. An IMT-2020 UHD surveillance camera can be used in intelligent functions such as animal quantity detection and location tracking, and action recognition. + +Traditionally, a farmer needs to check the number of animals when they depart and enter within a fence. With the deployment of an IMT-2020 UHD surveillance camera and IMT-2020 network, the animal count data can be easily transmitted to the cloud platform. By installing an IMT-2020 UHD surveillance camera mounted on an electronic fence and grassland, the animals can be identified in real time with high accuracy. 3G and 4G networks could not achieve the requirement for low latency and high accuracy. In addition, if the animals cross a prescribed area, there may be a security issue. With an IMT-2020 UHD surveillance camera, farmers can take measures to protect animals if they are stolen. Furthermore, animal growth monitoring, animal environment and grass degradation detection can be implemented with the deployment of an IMT-2020 UHD surveillance camera. + +## **7.8 Smart fishery** + +With the deployment of an IMT-2020 UHD camera, IMT-2020 network and underwater camera system, fishery management can be more efficient in breeding and save labour. It can improve underwater management by preventing cultured fish from escaping and pest diseases infecting them. + +For example, a marine fish farming enterprise can use IMT-2020 UHD surveillance cameras to assist in feeding the fish, detect aquatic product growth and achieve panoramic detection due to the low latency and high bandwidth of the IMT-2020 network. Fishery managers can monitor aquatic growth at home or office. Furthermore, the images and videos captured by an IMT-2020 UHD surveillance camera are uploaded to an intelligent fishery management cloud platform, the algorithm working on the platform can detect pests and diseases, e.g., those affecting carp, and analyse them to prevent massive fish deaths. + +# **8 Requirements for IMT-2020 UHD surveillance camera** + +## **8.1 Functional requirements** + +### **8.1.1 Video codec requirements** + +FUN-001: The IMT-2020 UHD surveillance camera is required to support video codecs such as those in [ITU-T H.264], [ITU-T H.265] or [ITU-T H.222.0], and the common image file format such as that in [ITU-T T.81]. + +FUN-002: The IMT-2020 UHD surveillance camera is recommended to support intelligent video codecs such as region of interest. + +### **8.1.2 SLA monitoring requirements** + +FUN-003: The IMT-2020 UHD surveillance camera is required to support the SLA monitoring function that supports the receiver in obtaining business SLA data and network real-time performance data, including frame loss rate, packet success rate, network packet loss rate and network end-to-end (E2E) delay. + +FUN-004: The IMT-2020 UHD surveillance camera is recommended to support sending SLA data to other analysis platforms for O&M, statistics, and analyses. + +### **8.1.3 Transmission protocol requirements** + +FUN-005: The IMT-2020 UHD surveillance camera is recommended to support transmission protocols with anti-packet loss, retransmission mechanism, data encryption, forward error correction, congestion control capability and network traffic shaping control. + +NOTE – Traffic shaping is a bandwidth management technique used on computer networks to delay some or all datagrams to prevent datagram burst, thus ensuring that datagrams are sent out at a relatively uniform rate. + +FUN-006: The IMT-2020 UHD surveillance camera in network I (see Table 6-2) is recommended to support a real-time streaming protocol over a transmission control protocol (TCP) or a real-time transport protocol over TCP. + +FUN-007: The IMT-2020 UHD surveillance camera in network II or III (see Table 6-2) is recommended to support user datagram protocol. + +### **8.1.4 Bandwidth adaptive requirements** + +FUN-008: The IMT-2020 UHD surveillance camera is recommended to support adaptively reducing video bit rates based on the reduction of networks (such as the reduction of uplink bandwidth, the increase of E2E delay and the packet loss rate). + +### **8.1.5 Key frame timing adjustment requirements** + +FUN-009: The IMT-2020 UHD surveillance camera is recommended to support dynamically adjusting the key frame sequence, in order to reduce the peak video bit rate and maintain the smooth transmission of the video bit rate. + +### **8.1.6 UHD video and image capture function requirements** + +FUN-010: The IMT-2020 UHD surveillance camera is recommended to support the environment adaptive function to ensure high-quality videos and images are captured in any environmental situation such as low illumination, dust, fog, harsh backlighting or large change in illumination. + +FUN-011: The IMT-2020 UHD surveillance camera is recommended to support structured information extraction of interested targets (including persons, vehicles, objects and events). + +### **8.1.7 Remote maintenance and upgrade requirements** + +FUN-012: The IMT-2020 UHD surveillance camera is required to support remote maintenance operations including camera status query and initial setting restoration. + +FUN-013: The IMT-2020 UHD surveillance camera is required to support system upgrade remotely. If an abnormal situation such as power failure or disconnection occurs, the camera is required to be able to restore the state before the upgrade. + +### **8.1.8 Video and audio access and control requirements** + +FUN-014: The IMT-2020 UHD surveillance camera is required to comply with the functional requirements of a premises unit (PU), which are specified in [ITU-T H.626]. + +FUN-015: The IMT-2020 UHD surveillance camera is required to comply with the signalling and protocols specified in [ITU-T H.627] related to a PU (specified in [ITU-T H.626]). + +## **8.2 Performance requirements** + +### **8.2.1 Basic performance requirements** + +PER-001: The highest resolution of IMT-2020 UHD surveillance camera is required not to be lower than $3\,840 \times 2\,160$ pixels. + +PER-002: The minimum illumination of IMT-2020 UHD surveillance camera is required not to be lower than 0.01 lx (in colour mode) and 0.001 lx (in monochrome mode). + +NOTE – A colourful camera may work in both colour and monochrome mode; only black and white photographs are generated in monochrome mode. + +PER-003: The maximum frame rate of an IMT-2020 UHD surveillance camera is required not to be lower than 25 frames/s. + +PER-004: The IMT-2020 UHD surveillance camera is required to support the standard bit rate of 10 Mbit/s, maximum bit rate not lower than 16 Mbit/s. + +PER-005: The IMT-2020 UHD surveillance camera is required to support not less than three videos streams. + +PER-006: The maximum out-of-synchronization time between audio and video is required not to be larger than 1 s. + +### **8.2.2 Video transmission performance requirements** + +PER-007: The video transmission performance of an IMT-2020 UHD surveillance camera is required to meet SLA rank not lower than grade 3 (specified in Table 6-1). + +PER-008: The IMT-2020 UHD surveillance camera is required to support the peak to average power ratio of a single stream not larger than 6. + +## **8.3 Security requirements** + +### **8.3.1 Authentication requirements** + +SEC-001: The IMT-2020 UHD surveillance camera's authentication control capability is required to meet all the following requirements: + +- a) the IMT-2020 UHD surveillance camera is required to support user authorization and authentication; +- b) the administrator ID is required not to be easily used fraudulently, and the password is required to be complex and changed regularly; +- c) the login failure handling function is required to be enabled, including session ending, number of illegal logins limited and automatic logout; +- d) authentication control is required to be performed for important operations of the administrator, such as restoring configuration and modifying the administrator's information, when the IMT-2020 UHD surveillance camera is remotely managed. + +### **8.3.2 Access control requirements** + +SEC-002: The IMT-2020 UHD surveillance camera's access control capability is required to meet all the following requirements: + +- a) the Internet protocol (IP) address or media access control address range of the access device is required to be restricted according to the security policy or management needs; +- b) the number of users with remote access authority is required to be limited; + +- c) user rights are required to be separated, and the rights of administrator users and ordinary users are required to be separated. + +### **8.3.3 Data integration and encryption requirements** + +SEC-003: The IMT-2020 UHD surveillance camera is required to support cryptographic technology or other mechanisms to ensure the integrity of data in the communication process according to business needs, e.g., by supporting encrypted transmission. + +SEC-004: The IMT-2020 UHD surveillance camera is required to support important data encrypted, including the session process, entire message body and important authentication data during the transmission process. + +### **8.3.4 Hacker intrusion detection requirements** + +SEC-005: The IMT-2020 UHD surveillance camera is required to be able to detect and deal with hacker intrusions, including but not limited to, sensitive system files tampering, abnormal super accounts, camera hijacking and process privilege escalation. + +### **8.3.5 Personally identifiable information protection requirement** + +SEC-006: The IMT-2020 UHD surveillance camera is required to support video and image privacy mask transmission according to the user configuration. + +NOTE – Privacy mask is a feature used in IP cameras to protect personally identifiable information by concealing parts of the image from viewing a masked area, usually according to the user's configuration. + +### **8.3.6 Operation log query requirement** + +SEC-007: The IMT-2020 UHD surveillance camera is required to support the operation of log queries and uploading to the log server or monitoring management platform. + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.743.19-202212-I_PDF-E/0f2a1e4a7b12fe5b8749882ecd636f5c_img.jpg b/marked/F/T-REC-F.743.19-202212-I_PDF-E/0f2a1e4a7b12fe5b8749882ecd636f5c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7cab34bc76ddd6b3a3e884fc0c01fb8a6e1b05b4 --- /dev/null +++ b/marked/F/T-REC-F.743.19-202212-I_PDF-E/0f2a1e4a7b12fe5b8749882ecd636f5c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7f55c1d3b20004445cc66f2d4f4063801ad9405dd70d5f009fc0acc8bd061577 +size 38416 diff --git a/marked/F/T-REC-F.743.19-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.743.19-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ac295b1b0c7f14b1c5a33ef6682b1b017c812477 --- /dev/null +++ b/marked/F/T-REC-F.743.19-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37aed4fc114a39843438175326e0b7e1d09a9d6075723d82fea32040fd560ee4 +size 7192 diff --git a/marked/F/T-REC-F.743.19-202212-I_PDF-E/4d2624b1b2871a23382b647cd71e1c1c_img.jpg b/marked/F/T-REC-F.743.19-202212-I_PDF-E/4d2624b1b2871a23382b647cd71e1c1c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..05e6aaf3a262dcb82769f642bb32f58670cbc402 --- /dev/null +++ b/marked/F/T-REC-F.743.19-202212-I_PDF-E/4d2624b1b2871a23382b647cd71e1c1c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3872c5fe33d477c4a2a1651e33e2325addb7c28394bbfeade39e0c3fe3f1adc5 +size 58011 diff --git a/marked/F/T-REC-F.743.19-202212-I_PDF-E/a83ba9e3e2c1e21dd69953a7b09e45b4_img.jpg b/marked/F/T-REC-F.743.19-202212-I_PDF-E/a83ba9e3e2c1e21dd69953a7b09e45b4_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..6e9c29dc736b0dc3e0ca1fa02dcbb6b1cb2dd089 --- /dev/null +++ b/marked/F/T-REC-F.743.19-202212-I_PDF-E/a83ba9e3e2c1e21dd69953a7b09e45b4_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:83462b6a69fb6671fb31da69853c7c10ada63daed05051dc68a2498402ec5311 +size 45296 diff --git a/marked/F/T-REC-F.743.19-202212-I_PDF-E/c092f712a80ce3310c5e29d0fa0e454a_img.jpg b/marked/F/T-REC-F.743.19-202212-I_PDF-E/c092f712a80ce3310c5e29d0fa0e454a_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..4bd2eb01398544575dd39ae33fa3d38ad140a6c6 --- /dev/null +++ b/marked/F/T-REC-F.743.19-202212-I_PDF-E/c092f712a80ce3310c5e29d0fa0e454a_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:401c57650ef3e2336f78c77976749a585dd81cd66c7b23b02c13806e80a655f3 +size 24849 diff --git a/marked/F/T-REC-F.743.19-202212-I_PDF-E/d5a837fa4f4675e5ee596003cf55985c_img.jpg b/marked/F/T-REC-F.743.19-202212-I_PDF-E/d5a837fa4f4675e5ee596003cf55985c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..6582f628e752fc2ff15a6075e24563df8c1f0e07 --- /dev/null +++ b/marked/F/T-REC-F.743.19-202212-I_PDF-E/d5a837fa4f4675e5ee596003cf55985c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4bf8933b09a69375e0cf30ce46873aa217fd22a2982b0dc45715a27123664ded +size 57290 diff --git a/marked/F/T-REC-F.743.19-202212-I_PDF-E/raw.md b/marked/F/T-REC-F.743.19-202212-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..e35538f18ff20358cfa0cdafc2afe112e20b3487 --- /dev/null +++ b/marked/F/T-REC-F.743.19-202212-I_PDF-E/raw.md @@ -0,0 +1,765 @@ + + +# Recommendation **ITU-T F.743.19 (12/2022)** + +SERIES F: Non-telephone telecommunication services +Multimedia services + +--- + +## **Requirements for intelligent surveillance cameras in intelligent video surveillance systems** + +![ITU logo](390120de4fe440c42fea8154fcaad334_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +### **TELEGRAPH SERVICE** + +| | | +|-----------------------------------------------------------------|-------------| +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessages service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | + +### **MOBILE SERVICE** + +| | | +|--------------------------------------------------------|-------------| +| Mobile services and multdestination satellite services | F.110–F.159 | +|--------------------------------------------------------|-------------| + +## **TELEMATIC SERVICES** + +| | | +|-------------------------------------------|-------------| +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | + +### **MESSAGE HANDLING SERVICES** + +F.400–F.499 + +### **DIRECTORY SERVICES** + +F.500–F.549 + +### **DOCUMENT COMMUNICATION** + +| | | +|--------------------------------------|-------------| +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | + +### **DATA TRANSMISSION SERVICES** + +F.600–F.699 + +### **MULTIMEDIA SERVICES** + +**F.700–F.799** + +## **ISDN SERVICES** + +F.800–F.849 + +## **UNIVERSAL PERSONAL TELECOMMUNICATION** + +F.850–F.899 + +## **ACCESSIBILITY AND HUMAN FACTORS** + +F.900–F.999 + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.743.19 + +# Requirements for intelligent surveillance cameras in intelligent video surveillance systems + +## Summary + +Recommendation ITU-T F.743.19 specifies the classification of, and scenarios for intelligent analysis functions, in addition to intelligent analysis function and grading requirements for intelligent surveillance cameras (ISCs). The related intelligent analysis functions include video diagnosis, tampering detection, video enhancement, target detection and feature extraction, and object behaviour identification. The basic functions of a camera (see the premises unit specified in Recommendation ITU-T H.626), such as multimedia capturing, multimedia encoding, output alarm signal and parsing pan/tilt/zoom command, lie outside the scope of this Recommendation. + +This Recommendation specifies the relevant intelligent analysis function and grading requirements for ISCs used in intelligent video surveillance. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.743.19 | 2022-12-14 | 16 | 11.1002/1000/15186 | + +## Keywords + +Grading, intelligent surveillance camera, requirements. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2023 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|------|-------------------------------------------------------------|----| +| 1 | Scope ..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 2 | +| 3.1 | Terms defined elsewhere ..... | 2 | +| 3.2 | Terms defined in this Recommendation..... | 2 | +| 4 | Abbreviations and acronyms ..... | 3 | +| 5 | Conventions ..... | 4 | +| 6 | Overview of intelligent surveillance cameras in IVS ..... | 4 | +| 6.1 | Classification of intelligent surveillance cameras ..... | 4 | +| 6.2 | Intelligent analysis classification of ISC ..... | 4 | +| 6.3 | Intelligent level classification of ISC ..... | 5 | +| 7 | Typical scenarios for intelligent surveillance cameras..... | 5 | +| 7.1 | Intelligent perception and adaptability scenarios ..... | 5 | +| 7.2 | Vehicle analysis-related scenarios..... | 7 | +| 7.3 | Human feature-related scenarios ..... | 8 | +| 7.4 | Behaviour analysis scenarios..... | 9 | +| 8 | Functional architecture of ISC..... | 10 | +| 9 | Requirements for consumer-level ISC..... | 10 | +| 9.1 | Basic requirements ..... | 10 | +| 9.2 | Intelligent analysis requirements ..... | 11 | +| 10 | Requirements for public-level ISC ..... | 12 | +| 10.1 | Basic requirements ..... | 12 | +| 10.2 | Intelligent analysis requirements..... | 13 | +| 10.3 | Advanced requirements ..... | 16 | + + + +## Requirements for intelligent surveillance cameras in intelligent video surveillance systems + +## 1 Scope + +This Recommendation specifies the classification of, and scenarios for intelligent analysis function, in addition to the intelligent analysis function and grading requirements for intelligent surveillance cameras (ISCs). The basic functions of a camera (see the premises unit (PU) specified in [ITU-T H.626]) such as multimedia capturing, multimedia encoding, output alarm signal and parsing pan/tilt/zoom (PTZ) command, lie outside the scope of this Recommendation. + +The scope of this Recommendation includes: + +- overview of ISCs for intelligent video surveillance (IVS); +- typical scenarios for ISCs; +- intelligent analysis function and grading requirements for ISCs. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +- [ITU-T F.743.1] Recommendation ITU-T F.743.1 (2015), *Requirements for intelligent visual surveillance*. +- [ITU-T G.711] Recommendation ITU-T G.711 (1988), *Pulse code modulation (PCM) of voice frequencies*. +- [ITU-T H.222.0] Recommendation ITU-T H.222.0V8 (2021), *Information technology – Generic coding of moving pictures and associated audio information: systems*. +- [ITU-T H.264] Recommendation ITU-T H.264V14 (2021), *Advanced video coding for generic audiovisual services*. +- [ITU-T H.265] Recommendation ITU-T H.265V8 (2021), *High efficiency video coding*. +- [ITU-T H.626] Recommendation ITU-T H.626V2 (2019), *Architecture requirements for video surveillance system*. +- [ITU-T H.626.5] Recommendation ITU-T H.626.5V2 (2022), *Architecture for intelligent video surveillance systems*. +- [ITU-T H.627] Recommendation ITU-T H.627V2 (2020), *Signalling and protocols for a video surveillance system*. +- [ITU-T H.627.2] Recommendation ITU-T H.627.2 (2022), *Requirements and protocols for home surveillance systems*. + +- [ITU-T T.81] Recommendation ITU-T T.81 (1992), *Information technology – Digital compression and coding of continuous-tone still images – Requirements and guidelines*. +- [ITU-T T.800] Recommendation ITU-T T.800 (2019), *Information technology – JPEG 2000 image coding system: Core coding system*. +- [ISO/IEC 14496-3] ISO/IEC 14496-3:2019, *Information technology – Coding of audio-visual objects – Part 3: Audio*. + +## 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 facial recognition** [ITU-T F.743.1]: Detection of the human face in video and extraction of its characteristics. The extracted information and/or image are compared with other images or a database of individuals' images, generating comparison results. Facial recognition can be used in airports, ports, railway stations, coach stations and other public places. + +**3.1.2 intelligent video diagnosis** [ITU-T F.743.1]: Automatic recognition of video images failures, such as snowflakes, scrolling or fuzzy video, partial colour, picture freeze, gain imbalance, pan/tilt/zoom (PTZ) control problems and video signal loss. + +**3.1.3 premises intelligent video (PIV)** [ITU-T H.626.5]: An intelligent identification module in the premises unit (PU). It identifies required information from input video and outputs the result. + +**3.1.4 region detection** [ITU-T F.743.1]: Alarm output that is based on the polygonal regions defined by the user and that is triggered when an object enters or leaves the region. + +**3.1.5 region stranded** [ITU-T F.743.1]: Alarm output that is based on the polygonal regions defined by the user and that is triggered when the retention time of an object exceeds a prescribed threshold. + +**3.1.6 traffic/pedestrian flow analysis** [ITU-T F.743.1]: Motion analysis and feature classification of the video of moving target(s) within the specified single or multiple video monitoring area. The pedestrian flow data can be detected accurately. + +**3.1.7 tripwire intrusion** [ITU-T F.743.1]: The setting off of an alarm when an object crosses a line drawn within the surveillance area. + +**3.1.8 video surveillance system** [ITU-T H.626]: A telecommunication service focusing on video (but including audio and image) application technology, which is used to remotely capture multimedia (such as audio, video, image, alarm signal, etc.), and present them to the end user in a user-friendly manner, based on a managed broadband network with ensured quality, security and reliability. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 detection rate:** The ratio of the number of correct targets or correct events detected to the number of targets or events that should be detected. The detection rate, $r_d$ , is calculated by + +$$r_d = \frac{n_{t+}}{n_{f-} + n_{t+}} \times 100\%$$ + +where: + +$n_{t+}$ is the number of true positive results + +$n_{\text{f-}}$ is the number of false negative results. + +**3.2.2 humanoid detection:** Activation of an alarm when a human body passes through the monitoring area of a camera. + +**3.2.3 intelligent video management (IVM):** A unit that supports strategies configuration of intelligent application by users and video sources schedule dynamically. The IVM accepts registration, deletion, capability report from IVU and schedule IVU dynamically. It can also store, manage, and schedule those capabilities dynamically. IVM can be invoked by an intelligent application from the centre management unit (CMU). + +NOTE – Paraphrased from clause 6 of [ITU-T F.743.1]. + +**3.2.4 intelligent visual surveillance (IVS):** A system that can automatically identify specific objects, behaviours or attributes in visual signals. IVS extracts data from visual signals, which are then transmitted or archived so that the visual surveillance system can act accordingly. + +NOTE – Paraphrased from clause 1 of [ITU-T F.743.1]. + +**3.2.5 intelligent premises unit (IPU):** A unit that adds a premises intelligent video (PIV) module to a premises unit (PU). A PIV is an intelligent recognition module of the PU, which can be achieved and integrated in PU devices. A PIV can recognize required information from input videos, and output recognition results. The recognized information includes that from event triggers and data acquisition. + +NOTE – Paraphrased from clause 6 of [ITU-T F.743.1]. + +**3.2.6 object enhancement:** An image processing function realizes target object enhancement from the whole video image. + +**3.2.7 recognition accuracy:** The ratio of the number of correctly identified targets or events to the total number of targets or events that should be identified correctly. The recognition accuracy, $a_{\text{r}}$ , is calculated by + +$$a_{\text{r}} = \frac{n_{\text{t+}}}{n_{\text{ic}}} \times 100\%$$ + +where: + +$n_{\text{t+}}$ is the number of true positive results + +$n_{\text{ic}}$ is the number of results that should be identified correctly. + +**3.2.8 tampering detection:** The discovery of a process of malicious alteration of video content, so as to conceal an object, an event or change the meaning conveyed by the imagery in the video. + +**3.2.9 intelligent surveillance camera (ISC):** A kind of intelligent premises unit that can not only capture video, audio, and images but also process the captured images or video and execute particular analysis algorithms, recognize required information and output analysis results including alarms, video structure data and recognition results. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-----|---------------------------------| +| IPU | Intelligent Premises Unit | +| ISC | Intelligent Surveillance Camera | +| IVM | Intelligent Video Management | +| IVS | Intelligent Video Surveillance | +| PIV | Premises Intelligent Video | +| PTZ | Pan/Tilt/Zoom | + +## 5 Conventions + +In this Recommendation: + +- The phrase "**is required to**" indicates a requirement that must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The phrase "**is recommended**" indicates a requirement that is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. +- The phrases "**can optionally**" and "**may**" indicate an optional requirement that is permissible, without implying any sense of being recommended. These terms are not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator or service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +## 6 Overview of intelligent surveillance cameras in IVS + +An ISC is defined in clause 3. + +This clause specifies the typical intelligent analysis function classification, intelligent-level classification and each intelligent analysis category-dedicated code. + +## 6.1 Classification of intelligent surveillance cameras + +This Recommendation divides ISCs into two categories, whose descriptions follow. + +- Consumer-level ISC: An ISC dedicated to home surveillance and private use. The consumer-level ISC is usually used in remote surveillance scenarios in homes, private stores, orchards, courtyards, fish ponds, etc.; +- Public-level ISC: An ISC dedicated to traditional public security scenarios (including public transportation surveillance and public security enhancement), and enterprise smart surveillance scenarios (including in communities, supermarkets, construction sites, kitchens and power infrastructure). Public security ISCs are commonly deployed by governments to monitor public areas to prevent security incidents, while enterprise-used ISCs are commonly deployed by enterprises to improve the public security of their community, and the intelligent operation and management of their workflow. + +## 6.2 Intelligent analysis classification of ISC + +The typical intelligent analysis functions for ISCs are divided into several categories that are shown in Figure 6-1. Usually intelligent video diagnosis, tampering detection and video enhancement are basic analysis functions for an ISC to achieve HD video and image in any environment, while the use of intelligent features such as object behaviour analysis, target detection and attribute extraction depends on actual scenarios. For simplification, this Recommendation classifies these intelligent analysis functions as follows. + +- Intelligent perception and adaptability capability: This function contains the capability for intelligent video diagnosis, video image enhancement and tampering detection. The intelligent video diagnosis analysis includes that of brightness anomalies, as well as detection of blurring, fogging and colour casts. Video and image enhancement employs process technology to improve quality, e.g., of definition, resolution and contrast, as well as enhancement of the target object in the whole image. Tampering analysis indicates some changes in images that are triggered by some external factors, such as camera coverage by a tree felled by storm winds, disconnection from power supply, theft and physical coverage; + +- Vehicle analysis category: This function refers to the capability to automatically detect, identify and analyse traffic target information and traffic flow status of vehicles, including detection of motor vehicles and their features, plus their recognition (e.g., of licence plates, brand, type and colour), traffic statistics, and non-motor vehicle detection and recognition; +- Face or body analysis category: This function refers to the capability to automatically detect and recognize target information, e.g., faces and bodies, including their detection and attribute recognition. +- Behaviour and event analysis category: This function refers to abnormal behaviour, e.g., physical violence, unexpected gatherings and falling, as well as behavioural events occurring in the scene, including monitored region intrusion, when a specific item moves or disappears, a baby cries or an elderly person experiences a fall; + +| Intelligent perception and adaptability capability | | | Behaviour and event analysis | +|----------------------------------------------------|--------------------------------------|-----------------------------|--------------------------------| +| Brightness anomaly diagnosis | Scene change detection | Low-light image enhancement | Loitering detection | +| Noise diagnosis and restoration | Blurring detection | Defogging enhancement | Legacy detection | +| Fogging detection and restoration | Camera covered detection | De-blurring enhancement | Vehicle/person flow statistics | +| Colour cast detection and restoration | Video occlusion detection | Object enhancement | Abnormal gather detection | +| Others | | | Tripwire detection | +| Target detection and attribute extraction | | | Intrusion detection | +| Motorized vehicle detection | Vehicle individual feature detection | Face feature recognition | Tumbling detection | +| Vehicle licence plate recognition | Vehicle basic feature detection | Body detection | Region detection | +| Non-motorized vehicle detection | Body feature recognition | Face detection | Abnormal behaviour detection | +| Others | | | Others | + +F.743.19(22) + +**Figure 6-1 – Intelligent analysis function classification of ISC** + +## 6.3 Intelligent level classification of ISC + +The intelligence level of the camera is graded and assessed according to the performance of the intelligent analysis function. This Recommendation specifies two levels for ISC: + +- general level: level I, which can meet general commercial use of the market; +- enhanced level: level II, which can meet advanced commercial use of the market. + +# 7 Typical scenarios for intelligent surveillance cameras + +This clause defines typical intelligence scenarios and analysis functions for ISC. + +## 7.1 Intelligent perception and adaptability scenarios + +#### 7.1.1 Brightness anomaly diagnosis and restoration + +Brightness anomaly diagnosis is a basic function for intelligent transportation monitoring; especially at night, vehicle headlights are overexposed in captured images, which makes recognition analysis of vehicle licence plates difficult, due to excessive brightness close to them. This use case is referred to as strong light suppression, which is illustrated in Figure 7-1. + +![Figure 7-1: Strong light suppression. A side-by-side comparison of two night-time street scenes. The 'Before' image on the left shows extreme lens flare and overexposure from bright headlights and streetlights, making details difficult to discern. The 'After' image on the right shows the same scene with significantly reduced glare and improved visibility of the road markings and vehicles.](d5a837fa4f4675e5ee596003cf55985c_img.jpg) + +Figure 7-1: Strong light suppression. A side-by-side comparison of two night-time street scenes. The 'Before' image on the left shows extreme lens flare and overexposure from bright headlights and streetlights, making details difficult to discern. The 'After' image on the right shows the same scene with significantly reduced glare and improved visibility of the road markings and vehicles. + +**Figure 7-1 – Strong light suppression** + +A camera with a strong light suppression function can effectively improve the concealment of dark details by the halo around a strong light source, and can make the outline and part of the details of the backlit objects visible without any increase in signal noise. + +Another typical scenario related to brightness anomaly diagnosis and restoration is the backlit scene; when the rear of a target object is exposed to sunlight or any light source, there will be parts of the object affected by low brightness, which is illustrated in Figure 7-2. + +![Figure 7-2: Backlit enhancement. A side-by-side comparison of two indoor scenes. The 'Before' image on the left shows a dark, underexposed area in the upper part of the frame, circled in red, where details are lost. The 'After' image on the right shows the same scene with enhanced brightness and clarity in the previously dark area, revealing objects like a plant and a fan.](4d2624b1b2871a23382b647cd71e1c1c_img.jpg) + +Figure 7-2: Backlit enhancement. A side-by-side comparison of two indoor scenes. The 'Before' image on the left shows a dark, underexposed area in the upper part of the frame, circled in red, where details are lost. The 'After' image on the right shows the same scene with enhanced brightness and clarity in the previously dark area, revealing objects like a plant and a fan. + +**Figure 7-2 – Backlit enhancement** + +Backlit enhancement can process and enhance the brightness and clarity of images of scenes with dark subjects on which it is difficult to distinguish details, and the resolution does not decrease while the signal noise is not increased. + +#### **7.1.2 Blurring detection and restoration** + +Licence plate and vehicle features are recognized at highway turnpikes and at urban road vehicle checkpoints by taking high-definition vehicle images and analysing them. High vehicle speed affects and blurs image quality, which will cause licence plate recognition to fail. In this situation, blur detection and restoration is needed. + +#### **7.1.3 Fogging detection and restoration** + +When a camera is deployed in a foggy outside environment, the initial video and image will be as shown in Figure 7-3-a, in which the image is low resolution and low definition. After processing by the fogging detection and restoration function (also called the intelligent defogging function) the resolution and definition are improved; in addition, there is no significant increase in signal noise and colour casting. The processed image is shown in Figure 7-3-b. + +![Figure 7-3: Intelligent defogging function. The image is split into two panels, 'a' (Before) and 'b' (After). Panel 'a' shows a hazy, low-contrast scene with a building and trees. Panel 'b' shows the same scene after defogging, with much clearer details and higher contrast. The labels 'Before' and 'After' are in the top right of each panel, and 'a' and 'b' are centered below each panel.](0f2a1e4a7b12fe5b8749882ecd636f5c_img.jpg) + +Figure 7-3: Intelligent defogging function. The image is split into two panels, 'a' (Before) and 'b' (After). Panel 'a' shows a hazy, low-contrast scene with a building and trees. Panel 'b' shows the same scene after defogging, with much clearer details and higher contrast. The labels 'Before' and 'After' are in the top right of each panel, and 'a' and 'b' are centered below each panel. + +**Figure 7-3 – Intelligent defogging function** + +#### 7.1.4 Object enhancement + +Useful information in the image of an object can be enhanced and its visual effect improved for a given application (shown in Figure 7-4). For example, new coding processing can be performed on moving vehicles and human face regions, so that the regional targets that the surveillance camera focuses on are more detailed and clearer, and the overall coding efficiency is not reduced. + +![Figure 7-4: Object enhancement. The image shows a street scene with a red bus. A legend on the left indicates 'Back ground' in blue and 'Target ground' in red. The bus is highlighted with a red rectangle, and the road surface is highlighted with a red overlay. The label 'F.743.19(22)' is in the bottom right corner.](c092f712a80ce3310c5e29d0fa0e454a_img.jpg) + +Figure 7-4: Object enhancement. The image shows a street scene with a red bus. A legend on the left indicates 'Back ground' in blue and 'Target ground' in red. The bus is highlighted with a red rectangle, and the road surface is highlighted with a red overlay. The label 'F.743.19(22)' is in the bottom right corner. + +**Figure 7-4 – Object enhancement** + +#### 7.1.5 Colour cast detection and correction + +Sometimes external factors, e.g., power abnormalities, low temperature and special colour temperature of the environment, will induce colour cast in a video or camera image. A dedicated correction function can detect the colour cast of an image and automatically adjust several colour-related configurations to make the colour of an output image closer to that in reality. + +#### 7.1.6 Detection of a camera obstruction + +Obstruction mainly affects the lens. If a camera lens is obstructed, an alarm is generated to alert the user to perform related operations. Detection of a camera obstruction is used to alert the user if a camera is: + +- vandalized with spray paint covering the lens to prevent recording; +- physically covered by a fallen tree or other objects. + +#### 7.1.7 Global scene change detection + +Global scene change detection is used when a monitoring scene changes, e.g., when someone hits the camera. If the global scene changes, an alarm is generated. This use case can be applied to outdoor scenes that have a number of moving objects, e.g., roads and car parks. Global scene change detection could be used to alert the user if the camera is physically moved or hit by something. + +### 7.2 Vehicle analysis-related scenarios + +This clause specifies scenarios related to vehicle detection and recognition. + +#### **7.2.1 Vehicle detection and recognition** + +This scenario is illustrated in clause 5.1.2 of [ITU-T F.743.1]. + +#### **7.2.2 Vehicle licence plate recognition** + +Vehicle licence plate recognition is specified in clause 5.3.3 of [ITU-T F.743.1]. + +#### **7.2.3 Vehicle basic feature recognition** + +Basic vehicle features, including brand, type, mode, colour, can be detected for recognition. + +#### **7.2.4 Vehicle individual feature recognition** + +Individual features related to a specific vehicle (e.g., detecting whether there are pendants, tissue box or vehicle inspection signs), and any person within it (including a driver or co-pilot not wearing a seat belt or making phone calls) can be detected for recognition. + +#### **7.2.5 Vehicle flow statistics** + +Vehicle flow statistics count traffic flow according to lanes and cycles, including the number of vehicles, vehicle types, and average speeds. + +#### **7.2.6 Non-motor vehicle detection** + +Non-motor vehicles, like bicycles, electric vehicles and tricycles, can be recognized and analysed in non-motor vehicle detection. This kind of non-motor vehicle detection is used to assist traffic analysis and statistics at traffic intersections, as well as to identify electric bicycles and then analyse whether cyclists are wearing helmets. + +### **7.3 Human feature-related scenarios** + +This clause describes specific scenarios for analysis of the face and body from video or image including their detection, feature recognition and recognition. + +#### **7.3.1 Face detection** + +Face detection captures and tracks human faces in areas monitored, and surrounds a target with a highlight rectangle in the image. The scenarios and face recognition are introduced in clause 5.3.2 of [ITU-T F.743.1]. + +#### **7.3.2 Mask-wearing recognition** + +Recognition of facial features, e.g., those indicating age, gender or clothing worn, is widely used in transportation security. Especially during a pandemic or in hospital, people may be required to wear masks. It is useful for a camera to integrate a mask detection algorithm, so that mask wearing as an attribute of the human face is integrated with face detection. + +#### **7.3.3 Body detection** + +Body detection is used to distinguish the human body from other target objects, like vehicles or animals, and is widely used to detect perimeter intrusion. It is necessary to exclude certain features, so as to distinguish a human from a dog, for example. + +#### **7.3.4 Body feature recognition** + +Body feature recognition involves analysis of the attributes of the clothing and accessories of personnel detected in the scene monitored. Body feature detection may be used in the following scenarios: + +- detection of helmets for people who ride a non-motor vehicle; +- detection of helmets and uniforms worn by workers during security checks on a smart construction site. + +### **7.4 Behaviour analysis scenarios** + +#### **7.4.1 Tripwire detection** + +Tripwire detection is used to detect whether a target crosses a line. If so, an alarm is generated. Tripwire detection may be used in the following intrusion prevention scenarios: + +- at key locations; +- in restricted areas. + +#### **7.4.2 Region detection** + +For an introduction to region detection, see clause 5.2.1 of [ITU-T F.743.1]. + +#### **7.4.3 Region stranded detection** + +For an introduction to region stranded detection, see clause 5.2.1 of [ITU-T F.743.1]. + +#### **7.4.4 Traffic/pedestrian flow analysis** + +For an introduction to traffic or pedestrian flow analysis, see clause 5.3.1 of [ITU-T F.743.1]. + +#### **7.4.5 Abnormal behaviour detection** + +Abnormal behaviours (e.g., unplanned gatherings of people, running, falls to the ground, climbing and travel in the wrong direction), can be detected. For example, gathering detection determines the density of people in a current scene by counting their images in specified area. When the density of people reaches a threshold, an alarm is generated. Gathering detection can be used to prevent threats to safety in public areas, e.g., bus stations, railway stations or scenic spots, where the maximum flow of people is limited or to monitor the density of people to quickly prevent accidents, including outbreaks of violence, due to overcrowding. + +#### **7.4.6 Abnormal sound detection** + +Abnormal sounds, including babies crying and sudden high-decibel noise, can be detected and tracked. This function is typically used in a home surveillance scenario. When a camera detects abnormal sound, an alarm with a captured image of sound source can be generated and reported to the client of the user. Users can be alerted about what happened in their homes and respond in a timely fashion. + +#### **7.4.7 Detection of falls** + +A fall is a kind of change in posture. Changes in human posture, including falling, slipping and fainting, can be detected. This function is typically used in a home surveillance scenario to look after the elderly or young children. When a camera detects people falling, an alarm with a captured image of the spot is generated and reported to the user. Users can be alerted about what happened and respond in a timely fashion. + +#### **7.4.8 Humanoid motion detection** + +Humanoid motion detection will only raise an alarm when human movement is detected. This type of intelligent analysis is used in warehouses, private houses, banks, etc. Usually, after such detection, the camera can capture an image of the human body and send it to the client. In the meantime, the camera broadcasts an alarm and turns on a double source light-emitting diode source to deter intruders. + +#### **7.4.9 High-altitude throwing detection** + +Objects thrown from the top of buildings can be detected and tracked, and a real-time alarm sent to the client. These objects include but are not limited to rubbish bags, paper balls, cigarette butts, mineral water bottles, knives, porcelain, pots and bowls, as well as bricks. At the same time, high- + +altitude throwing detection can effectively exclude environmental factors that affect the results, such as trees, weather change, birds and vibration. + +## 8 Functional architecture of ISC + +The functional architecture of an ISC, which is specified in clause 6.2.1 of [ITU-T H.626.5], is shown in Figure 8-1. PIV module is the intelligent analysis function module used to implement specific intelligent analysis, it: + +- identifies required information from captured videos and image and outputs analysis results; +- reports the analysis results such as triggered events, alarms and acquired data to the service platform. + +For detailed functions of other modules, please see [ITU-T H.626.5]. + +![Functional architecture of an ISC diagram](a83ba9e3e2c1e21dd69953a7b09e45b4_img.jpg) + +The diagram illustrates the functional architecture of an Intelligent Surveillance Camera (ISC). It is organized into three vertical columns within a rectangular frame. The left column contains six stacked rectangular boxes representing individual functions: 'Audio/video capture function', 'Audio play function', 'PTZ control function', 'Alarm input function', 'Alarm output function', and 'Media storage function'. The middle column contains three stacked rectangular boxes: 'Audio/video encoding function', 'Audio decoding function', and a larger box at the bottom labeled 'PIV module' in red text. The right column consists of a single large rectangular box labeled 'Control and management function'. All boxes are connected by lines to a common vertical bus on the right side of the frame. + +Functional architecture of an ISC diagram + +F.743.19(22) + +**Figure 8-1 – Functional components of an ISC** + +# 9 Requirements for consumer-level ISC + +### 9.1 Basic requirements + +### 9.1.1 Common requirements + +BR-001: The highest resolution is required to be not lower than $1\,920 \times 1\,080$ pixels. + +BR-002: A consumer-level ISC is required to support chip computing power not lower than 0.5 T operations/s. + +BR-003: A consumer-level ISC is required to support video codec, e.g., as specified in [ITU-T H.264], [ITU-T H.265] or MPEG-4 [ITU-T H.222.0], and support the image format, e.g., as specified in [ITU-T T.81] or [ITU-T T.800]. + +BR-004: A consumer-level ISC is required to support at least two video streams. + +BR-005: A consumer-level ISC is required to support a bi-directional voice intercom function. + +BR-006: A consumer-level ISC is recommended to support an infrared night vision function, to ensure the user can obtain the clear video at night. + +BR-007: A consumer-level ISC is required to support multiple network access, including wired network connection and wireless connection such as Wi-Fi, 4G and 5G. + +BR-008: A consumer-level ISC is required to support both local micro non-volatile flash memory card storage and cloud storage. + +BR-009: A consumer-level ISC is required to support two- or three-dimensional digital noise reduction. + +### **9.1.2 Signalling and protocols requirements** + +PR-001: A consumer-level ISC is required to support the signalling and protocols specified in [ITU-T H.627.2]. + +### **9.1.3 Security requirements** + +SR-001: The authentication and authorization control capability of a consumer-level ISC is required to meet all the following requirements: + +- a) it is required to support user authorization and authentication; +- b) the administrator ID is required to not be easily used fraudulently, and the password is required to be complex and changed regularly; +- c) the login failure handling function is required to be enabled, including session ending, number of illegal logins limited and automatic logout; +- d) authentication control is required to be performed for important operations of administrator such as restoring configuration and modifying administrator's information when the ISC is remotely managed; +- e) the Internet protocol address or media access control address range of the access device is required to be restricted according to the security policy or management needs; +- f) the number of users with remote access authority is required to be limited; +- g) user rights are required to be separated, and the rights of administrator users and ordinary users are required to be separated. + +SR-002: A consumer-level ISC is required to support cryptographic technology or other mechanisms to ensure the integrity of data in the communication process according to business needs, e.g., supporting encrypted transmission. + +SR-003: A consumer-level ISC is required to support important data encrypted, including session process, entire message body and important authentication data during the transmission process. + +SR-004: A consumer-level ISC is recommended to support video and image privacy mask transmission according to the user configuration. + +SR-005: A consumer-level ISC is required to be able to detect and deal with hacker intrusions, including but not limited to tampering with sensitive system files, abnormal super accounts, camera hijacking and process privilege escalation. + +SR-006: A consumer-level ISC is required to support querying of operation logs and uploading to the log server or monitoring management platform. + +## **9.2 Intelligent analysis requirements** + +#### **9.2.1 Humanoid detection** + +IAR-001: A consumer-level ISC is recommended to support the detection and capture of a humanoid target larger than $80 \times 80$ pixels in the monitoring image; the humanoid detection response time is required to be no greater than 50 ms indoors and no greater than 150 ms outdoors, and the humanoid accuracy is required to be no less than 95%. + +### **9.2.2 Face detection** + +IAR-002: A consumer-level ISC is recommended to support the detection of a face target larger than $120 \times 120$ pixels in the monitoring image, detection of multiple faces at the same time, and outputs the number, size (width, height, and pixels), location and other information about the face. + +IAR-003: A consumer-level ISC is recommended to support the capture of no less than 10 effective face targets at the same time. A level-I ISC is required to support a detection rate of no less than 90%. A level-II ISC is required to support a detection rate of no less than 95%. + +### **9.2.3 Motor vehicle detection** + +IAR-004: A consumer-level ISC is recommended to support the detection and capture of motor vehicle targets larger than $120 \times 120$ pixels in the monitoring image; the vehicle detection response time should be no greater than 100 ms. A level-I consumer-level ISC is required to support a detection accuracy rate of no less than 95%. A level-II consumer-level ISC is required to support a detection accuracy rate of no less than 99%. + +#### **9.2.4 Motor vehicle licence plate recognition** + +IAR-005: A consumer-level ISC is recommended to support the recognition of the number, colour and type of motor vehicle licence plates with a width greater than $100 \times 30$ pixels in the monitoring area. The accuracy rate of motor vehicle licence plate recognition should be no less than 95% during the day and no less than 90% at night. + +### **9.2.5 Motor vehicle basic feature recognition** + +IAR-006: A consumer-level ISC is recommended to support the detection and capture of motor vehicle targets larger than $100 \times 80$ pixels in the monitoring image; the vehicle detection response time should be no greater than 200 ms. A level-I consumer-level ISC is required to support a detection accuracy rate of no less than 90%. A level-II consumer-level ISC is required to support a detection accuracy rate of no less than 95%. + +### **9.2.6 Non-motor vehicle detection** + +IAR-007: A consumer-level ISC is recommended to support the detection and recognition of non-motor vehicle targets (including electric bicycles, bicycles, scooters, trolleys), larger than $30 \times 40$ pixels in the monitoring image, the detection accuracy rate should not be less than 90%, and the vehicle detection response time should be no greater than 100 ms. + +#### **9.2.7 High altitude throwing detection** + +IAR-008: A consumer-level ISC is recommended to support the detection and capture of high-altitude throwing targets larger than $5 \times 5$ pixels in the surveillance image. A level-I consumer-level ISC is required to support a detection accuracy rate of no less than 90%. A level-II consumer-level ISC is required to support a detection accuracy rate of no less than 95%. + +## **10 Requirements for public-level ISC** + +### **10.1 Basic requirements** + +### **10.1.1 Common requirements** + +BR-001: A public-level ISC is required to support a basic PU feature (see clause 6 of [ITU-T F.743.1]) including video or audio acquisition, image snapshot, recording or playback, event and alarm, and PTZ control. + +BR-002: A public-level ISC is required to support a chip computing power not lower than 1 T operations/s. + +BR-003: A public-level ISC is required to support video codec, e.g., as specified in [ITU-T H.264], [ITU-T H.265] or [ITU-T H.222.0], and support an image format, e.g., as specified in [ITU-T T.81] or [ITU-T T.800]. + +BR-004: A public-level ISC is required to support audio codec, e.g., pulse code modulation [ITU-T G.711] or advanced audio coding [ISO/IEC 14496-3]. + +BR-005: A public-level ISC is required to support at least two video streams. + +### **10.1.2 Signalling and protocols requirements** + +PR-001: A public-level ISC is required to support the signalling and protocols specified in [ITU-T H.627], in order to support video and audio transmission, access and control from IVS. + +### **10.1.3 Security requirements** + +A public-level ISC is required to support the security requirements specified in clause 9.1.3. + +### **10.2 Intelligent analysis requirements** + +#### **10.2.1 Intelligent perception and adaptability capability requirements** + +##### **10.2.1.1 Brightness anomaly diagnosis and restoration** + +IAR-001: A public-level ISC is recommended to support brightness anomaly diagnosis and a restoration function including adjustment when the brightness of an image is too high or too low, enhancement of the whole or part of the brightness of a video or image and strong light suppression. + +IAR-002: A public-level ISC is recommended to support automatic adjustment of the high-dynamic range parameter according to the brightness change in the environment and support the backlight compensation function to ensure the object monitored is clear and recognizable. + +##### **10.2.1.2 Blurring detection and restoration** + +IAR-003: A public-level ISC is recommended to support the detection and processing of image blur caused by target movement or an unfocused lens. + +IAR-004: A public-level ISC is recommended to have an adaptable de-blurring function, and to improve the resolution and definition of images. + +##### **10.2.1.3 Fogging detection and restoration** + +IAR-005: A public-level ISC is recommended to have an adaptive defogging function that can process video images with low clarity and visibility in foggy weather, and to improve the resolution and definition of video images; obvious colour cast should not be caused after defogging. + +##### **10.2.1.4 Object enhancement** + +IAR-006: A public-level ISC is recommended to enhance the target-related region of an image by improving the resolution and definition of the related region. + +IAR-007: A public-level ISC is recommended to support the ability to locally enhance regions related to moving target objects (e.g., human faces and motor vehicle licence plates). + +##### **10.2.1.5 Colour cast detection and correction** + +IAR-008: A public-level ISC is recommended to support processing the colour cast of an image that is caused by the wrong distribution of colour in a specific range, and making the colour of the output image closer to that of reality. + +##### **10.2.1.6 Obstructed camera detection** + +IAR-009: A public-level ISC is recommended to support the detection of obstructed cameras; a level-I public-level ISC is required to support detection of 100% of obstructed cameras; a level-II public-level ISC is required to support detection of more than 50% of obstructed cameras. + +##### **10.2.1.7 Scene change detection** + +IAR-010: A public-level ISC is recommended to support detection of scene changes, e.g., due to the camera being moved or rotated. A level-I public-level ISC is required to support detection of more + +than 100% of scenes changed. A level-II public-level ISC is required to support detection of more than 50% of scenes changed. + +#### **10.2.2 Vehicle analysis capability requirement** + +##### **10.2.2.1 Motor vehicle detection** + +IAR-011: A public-level ISC is recommended to support detection of motor vehicles, and to support the detection and capture of multiple motor vehicle targets at the same time. A level-I public-level ISC is required to support detection and capture of motor vehicle targets larger than $120 \times 120$ pixels in the monitoring image and achieve a detection rate $\geq 95\%$ – the detection accuracy rate should not be less than 99%. A level-II public-level ISC is required, in addition to meeting level-I specifications, to achieve a detection rate no less than 95% in the vehicle forward and reverse scenes; furthermore, the accuracy rate should be no less than 99%. + +##### **10.2.2.2 Motor vehicle licence plate recognition** + +IAR-012: A public-level ISC is recommended to support the recognition of the number, colour and type of motor vehicle licence plates with a width greater than $100 \times 30$ pixels in the monitoring area. In addition, the following recognition accuracy requirements for level-I and level-II ISC should be met: + +- a) level-I: the accuracy rate of motor vehicle licence plate recognition should be no less than 95% during the day and no less than 90% at night; +- b) level-II: on the basis of conforming to level-I specifications, it is required to have the ability to identify whether a licence plate is altered or covered. + +##### **10.2.2.3 Motor vehicle basic feature recognition** + +IAR-013: A public-level ISC is recommended to support the recognition of basic characteristics of motor vehicles and meet the following recognition accuracy requirements for level-I and level-II ISC: + +- a) level I: an ISC is required to support the analysis and output of attribute information such as type and colour of motor vehicle, the average recognition accuracy of motor vehicle basic characteristics should not be less than 90%; +- b) level II: on the basis of level I, the ISC is still required to support the recognition of the brand of motor vehicle, and the average recognition accuracy of the basic characteristics of the motor vehicle should not be less than 90%. + +##### **10.2.2.4 Motor vehicle individual feature recognition** + +IAR-014: A public-level ISC is recommended to support the identification of individual characteristics of motor vehicles, such as whether the main or co-driver is wearing a seat belt, or holding a phone, etc., the average recognition accuracy of individual characteristics of motor vehicles is not less than 80%. + +##### **10.2.2.5 Motor vehicle flow statistics** + +IAR-015: A public-level ISC is recommended to support the counting of motor vehicle traffic targets larger than $120 \times 120$ pixels and the compilation of statistics of motor vehicle flow according to lane and time conditions, including the number of motor vehicles – the accuracy rate of motor vehicle flow statistics is not less than 90%. + +##### **10.2.2.6 Non-motor vehicle detection** + +IAR-016: A public-level ISC is recommended to support the identification of non-motor vehicles targets with a width greater than 80 pixels in the video image – the detection rate of non-motor vehicles should not be less than 90%, and the accuracy rate should not be less than 90%. + +#### **10.2.3 Person analysis capability requirement** + +##### **10.2.3.1 Face detection** + +IAR-017: A public-level ISC is recommended to support the detection of face targets and their positions and sizes in scenes monitored, detection of multiple faces at the same time, and outputs the number, size (width, height, and pixels), location and other information about the face. A level-I public-level ISC is required to support the capture of no less than 10 effective face targets at the same time, the detection rate of face detection should be no less than 95%, and the accuracy rate of face detection should be no less than 99%. A level-II public-level ISC is required to support the capture of no less than 50 at the same time. For a valid face target, the rate of face detection should be no less than 95%, and the accuracy rate of face detection should be no less than 99%. + +##### **10.2.3.2 Body detection** + +IAR-018: A public-level ISC is recommended to support the detection of the portrait, as well as its position and size in the monitoring scene, supporting the detection of multiple human targets at the same time, and outputting information such as the number of human targets and the size of each target (width and height pixels). A level-I public-level ISC is required to support the capture of no less than 10 effective portrait targets at the same time. A level-II public-level ISC is required to support the capture of no less than 50 effective portrait targets at the same time. + +##### **10.2.3.3 Body feature recognition** + +IAR-019: A public-level ISC is recommended to support the analysis of and judgement on the attributes of people's clothing, accessories and objects carried that are detected in the monitoring scene, as well as the identification of attributes such as hairstyle, clothing style and colour, body shape, objects carried and irection of travel. + +#### **10.2.4 Behaviour analysis capability requirement** + +##### **10.2.4.1 Tripwire detection** + +IAR-020: A public-level ISC is recommended to support the detection of tripwires of targets larger than $32 \times 32$ pixels in the area of interest of the video screen and output event information, and meet the detection rate of not be less than 90% when the false detection rate is not greater than 5%. + +##### **10.2.4.2 Region detection** + +IAR-021: A public-level ISC is recommended to support the detection of the intrusion of targets larger than $32 \times 32$ pixels in the area of interest of the video screen and output event information; when the false detection rate is not greater than 5%, the detection rate is required to not be less than 90%. + +##### **10.2.4.3 Region stranded detection** + +IAR-022: A public-level ISC is recommended to support the detection of the intrusion of targets larger than $32 \times 32$ pixels into the area of interest of the video screen that stay longer than the specified time, and output event information; when the false detection rate is not greater than 5%, the detection rate is required to be not less than 90%. + +##### **10.2.4.4 Traffic or pedestrian flow analysis** + +IAR-023: A public-level ISC is recommended to support the detection of the number of people, whose shoulder width is greater than 32 pixels in the target statistical area, alarm and output event information if the threshold value is exceeded; the accuracy of crowd density detection is not less than 90%. + +##### **10.2.4.5 Abnormal behaviour detection** + +IAR-024: A public-level ISC is recommended to support the detection of one or more abnormal behaviours (including running, falling to the ground, climbing, travelling in the wrong direction), of personnel targets larger than $64 \times 128$ pixels in the area of interest of the video screen and output event information. + +### **10.3 Advanced requirements** + +#### **10.3.1 Intelligent algorithm sustainability requirements** + +ADR-001: A public-level ISC is recommended to support on-demand deployment and online upgrade of applications without service interruption. + +ADR-002: A public-level ISC is recommended to support the life cycle management of applications, including installation or uninstallation, activation or deactivation, upgrade and licence management. + +ADR-003: A public-level ISC is recommended to support licence verification and algorithm model file encryption of an application to protect its intellectual property. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.743.2-201607-I_PDF-E/raw.md b/marked/F/T-REC-F.743.2-201607-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..e88d5d2f9ea617d2122a93bc593a83932bbbc184 --- /dev/null +++ b/marked/F/T-REC-F.743.2-201607-I_PDF-E/raw.md @@ -0,0 +1,505 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.743.2** + +(07/2016) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +# --- **Requirements for cloud storage in visual surveillance** + +Recommendation ITU-T F.743.2 + +**ITU-T** + +![ITU logo](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) features a globe with a red lightning bolt striking across it. To the right of the globe, the text "International Telecommunication Union" is written in blue, with "ITU" in a larger, bold font above it. + +ITU logo + +International +Telecommunication +Union + +## ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +### **TELEGRAPH SERVICE** + +| | | +|-----------------------------------------------------------------|-------------| +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | + +### **MOBILE SERVICE** + +| | | +|--------------------------------------------------------|-------------| +| Mobile services and multdestination satellite services | F.110–F.159 | +|--------------------------------------------------------|-------------| + +## **TELEMATIC SERVICES** + +| | | +|-------------------------------------------|-------------| +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | + +### **MESSAGE HANDLING SERVICES** + +| | | +|--|-------------| +| | F.400–F.499 | +|--|-------------| + +### **DIRECTORY SERVICES** + +| | | +|--|-------------| +| | F.500–F.549 | +|--|-------------| + +## **DOCUMENT COMMUNICATION** + +| | | +|--------------------------------------|-------------| +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | + +### **DATA TRANSMISSION SERVICES** + +| | | +|--|-------------| +| | F.600–F.699 | +|--|-------------| + +### **MULTIMEDIA SERVICES** + +| | | +|--|--------------------| +| | F.700–F.799 | +|--|--------------------| + +### **ISDN SERVICES** + +| | | +|--|-------------| +| | F.800–F.849 | +|--|-------------| + +## **UNIVERSAL PERSONAL TELECOMMUNICATION** + +| | | +|--|-------------| +| | F.850–F.899 | +|--|-------------| + +## **HUMAN FACTORS** + +| | | +|--|-------------| +| | F.900–F.999 | +|--|-------------| + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.743.2 + +# Requirements for cloud storage in visual surveillance + +## Summary + +Recommendation ITU-T F.743.2 defines the cloud storage service requirements in visual surveillance. Cloud storage enables service users to have ubiquitous, convenient, on-demand network access to a shared pool of configurable storage resources, which can be rapidly provisioned and released with minimal management effort or service-provider interaction. Cloud storage can realize flexible and reliable data storage for large-scale visual surveillance, and its components are modularized and allocated dynamically based on real usage. This Recommendation provides the application scenarios and requirements for cloud storage in visual surveillance. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.743.2 | 2016-07-14 | 16 | 11.1002/1000/12895 | + +## Keywords + +Cloud storage, picture storage, video file uploading, video metadata management, video stream storage, visual surveillance. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2016 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|---------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Overview..... | 2 | +| 7 Scenarios..... | 4 | +| 7.1 Video stream storage ..... | 4 | +| 7.2 Video file uploading..... | 4 | +| 7.3 Video metadata management ..... | 5 | +| 7.4 Picture storage ..... | 5 | +| 8 Requirements for cloud storage in visual surveillance ..... | 6 | +| 8.1 User requirements..... | 6 | +| 8.2 Service requirements ..... | 6 | +| 8.3 Security requirements ..... | 7 | +| 8.4 Management requirements ..... | 8 | +| 8.5 Scalability requirements ..... | 8 | +| 8.6 Reliability requirements ..... | 9 | +| 8.7 Performance requirements..... | 9 | +| Bibliography..... | 10 | + + + +# Recommendation ITU-T F.743.2 + +# Requirements for cloud storage in visual surveillance + +# 1 Scope + +This Recommendation describes the brief functional model, application scenarios and requirements for cloud storage in visual surveillance (VS) systems, based on the requirements and architectures defined by [ITU-T F.743], [ITU-T H.626] and [ITU-T H.626.1]. + +A visual surveillance service is a telecommunication service focusing on video (and audio) application technology, which is used to remotely capture multimedia (e.g., audio, video, image, various alarm signals), and present this to end users in a friendly manner (including accessibility aspects), based on a broadband network with ensured quality, security and reliability. Cloud storage is a data storage model for enabling service users to have ubiquitous, convenient and on-demand network access to a shared pool of configurable storage resources, which can be rapidly provisioned and released with minimal management effort or service-provider interaction. In cloud storage systems, the physical and virtual resources can be dynamically assigned and reassigned according to user demand. Cloud storage can realize scalable, flexible and reliable data storage for large-scale visual surveillance. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +- [ITU-T F.743] Recommendation ITU-T F.743 (2009), *Requirements and service description for visual surveillance*. +- [ITU-T H.626] Recommendation ITU-T H.626 (2011), *Architectural requirements for visual surveillance*. +- [ITU-T H.626.1] Recommendation ITU-T H.626.1 (2013), *Architecture for mobile visual surveillance*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +- 3.1.1 application** [b-ITU-T Y.101]: A structured set of capabilities, which provide value-added functionality supported by one or more services. +- 3.1.2 customer** [b-ITU-T M.60]: An entity which receives services offered by a service provider based on a contractual relationship. It may include the role of a network user. +- 3.1.3 customer unit** [ITU-T F.743]: A device located at the customer part of a visual surveillance system and used to present multimedia information (such as audio, video, image, alarm signal, etc.) to the end user. + +**3.1.4 premises unit** [ITU-T F.743]: A device located at the remote part of a visual surveillance system and used to capture multimedia information (such as audio, video, image, alarm signal, etc.) from a surveilled object. + +**3.1.5 service** [b-ITU-T Y.101]: A structure set of capabilities intended to support applications. + +**3.1.6 visual surveillance** [ITU-T F.743]: A telecommunication service focusing on video (but including audio) application technology, which is used to remotely capture multimedia (such as audio, video, image, alarm signals, etc.) and present them to the end user in a friendly manner, based on a managed broadband network with quality, security and reliability ensured. + +## **3.2 Terms defined in this Recommendation** + +This Recommendation defines the following term: + +**3.2.1 cloud storage**: A data storage model for enabling service users to have ubiquitous, convenient and on-demand network access to a shared pool of configurable storage resources, which can be rapidly provisioned and released with minimal management effort or service-provider interaction. In cloud storage systems, the physical and virtual resources can be dynamically assigned and reassigned according to user demand. + +# **4 Abbreviations and acronyms** + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|-----------------------------------| +| AS | Application System | +| CU | Customer Unit | +| DVR | Digital Video Recorder | +| IPU | Intelligent Premises Unit | +| MCU | Mobile Customer Unit | +| MSU | Media Storage Unit | +| NVR | Network Video Recorder | +| PU | Premises Unit | +| VS | Visual Surveillance | +| VSCS | Visual Surveillance Cloud Storage | + +# **5 Conventions** + +In this Recommendation: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this recommendation is to be claimed. +- The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus this requirement need not be present to claim conformance. + +# **6 Overview** + +Currently, large-scale visual surveillance system deployment has contributed to the explosive growth of surveillance video data. Moreover, the massive amount of surveillance video data contains a lot of valuable information which can be mined to provide more intelligent services. Attention is needed to the efficient storage of video data and fast access to information of interest to users. + +When the surveillance industry changes to all IP network usage, the network video recorder (NVR), as a main component for surveillance video storage, will gradually replace the existing digital video recorder (DVR). However, an NVR which is installed on single physical server may reach the maximum storage or throughput limitation due to predefined resource limitations. In addition, the current NVR-based storage model does not efficiently support the analysis and mining of the massive amount of video data. + +![Figure 1 – Cloud storage model for visual surveillance. The diagram illustrates a hierarchical system. At the top, a 'VS platform' (Visual Surveillance platform) is connected to a user interface box containing 'CU' (Customer Unit), 'MCU' (Media Control Unit), and 'Third party AS' (Application System). Below this, a 'Cloud storage system' is represented by a cloud containing several server icons. At the bottom, a yellow box contains various surveillance devices: 'IPU' (IP Camera), 'PU' (Premises Unit), and 'MSU' (Media Storage Unit). Arrows indicate bidirectional data flow between the VS platform and the cloud, between the user interface and the cloud, and between the cloud and the surveillance devices. The MSUs are shown connected to multiple IPUs and PUs.](acfc53eca625d62b38aa2563efa95c3e_img.jpg) + +Figure 1 – Cloud storage model for visual surveillance. The diagram illustrates a hierarchical system. At the top, a 'VS platform' (Visual Surveillance platform) is connected to a user interface box containing 'CU' (Customer Unit), 'MCU' (Media Control Unit), and 'Third party AS' (Application System). Below this, a 'Cloud storage system' is represented by a cloud containing several server icons. At the bottom, a yellow box contains various surveillance devices: 'IPU' (IP Camera), 'PU' (Premises Unit), and 'MSU' (Media Storage Unit). Arrows indicate bidirectional data flow between the VS platform and the cloud, between the user interface and the cloud, and between the cloud and the surveillance devices. The MSUs are shown connected to multiple IPUs and PUs. + +F.743.2(16)\_F01 + +**Figure 1 – Cloud storage model for visual surveillance** + +Cloud storage is a new model for enabling service users to have ubiquitous, convenient and on-demand network access to a shared pool of configurable storage resources, which can be rapidly provisioned and released with minimal management effort or service-provider interaction. From a telecommunication perspective, users do not buy resources but rather purchase cloud services provided by cloud environments. This can improve the system flexibility and reduce the expenditure of users' systems. + +Cloud storage is used for aggregating and managing a particular data set. Figure 1 shows a cloud storage model for a visual surveillance system. A cloud storage system is a scalable, flexible and reliable video recording system that can optimize the storage resources for massive video sources. In a cloud storage system, each component is modularized and allocated dynamically based on actual usage, and data protection mechanisms are supported to improve system reliability. A cloud storage system enables convenient and fast access to a wealth of data across distributed and heterogeneous data sources in the cloud. Furthermore, this data storage system can be easily integrated with cloud computing frameworks which support efficient intelligent video analysis tasks. For the premises unit (PU) devices, video data can be uploaded to a cloud storage system by online/offline modes using streaming or other network data transfer protocols. The data sources can be IP cameras, mobile devices or other media storage units (MSUs), such as DVRs with networking capability, NVRs, and vehicle DVRs installed on cars. The visual surveillance (VS) platform can call the cloud storage service through standard interfaces provided by the cloud storage system. All types of authorized customer units (CUs) or authorized third-party application systems (ASs) can access the data stored in the cloud storage system. + +# 7 Scenarios + +This clause describes typical service scenario examples illustrating the cloud storage in visual surveillance and deriving its service requirements. + +## 7.1 Video stream storage + +Surveillance cameras capture video data continuously. User Bob has subscribed to the visual surveillance cloud storage (VSCS) service from the VS service provider, and wants to store streaming video data in the cloud directly. + +Step 1: Bob logs into the VS system via his PC. He clicks a cloud storage menu function and goes to a video stream storage plan form. A camera list appears on the screen, which displays detailed information for each camera that Bob can access. Bob chooses the cameras whose captured video data needs to be stored in the cloud, and then sets the time interval during which the data should be recorded. The VS management system receives Bob's submitted video stream cloud storage plan, and forwards the plan to the cloud storage system. + +Step 2: According to the video stream storage plan, the cloud storage system gets the video stream directly from the monitoring cameras, and then writes the video data to the cloud storage resource pool built on the storage device clusters. + +Step 3: Bob can browse the camera's video data recorded in the cloud; he can also choose the camera name and the video starting time, and replay the video of interest stored in the cloud. + +## 7.2 Video file uploading + +### Case 1: Video file uploading from NVR + +An NVR is usually used as the local storage devices for continuously recording surveillance video from network cameras. NVR has a maximum storage capacity or throughput limitation due to predefined resource limitations. For example, an NVR can receive video streams from 16 cameras simultaneously, and can record the video data from those cameras for one month. In addition, the data reliability is difficult to guarantee when the NVR fails. + +The public security department of a city has deployed a VS system in the city. Tom is the system administrator and is responsible for operating the visual surveillance system to carry out public security tasks. Currently, the surveillance video is recorded on the local NVRs. However, Tom's supervisor lets him retain the video data from some cameras for at least six months. + +Step 1: Tom logs into the VS system. He chooses the surveillance cameras, sets the time interval to 12 months and sets the video file uploading period to 12 hours. Tom submits the video file uploading plan to the visual surveillance management system. This plan is then forwarded to the cloud storage system. + +Step 2: According to the video file uploading plan, the cloud storage system retrieves the video files from the corresponding NVRs every 12 hours, and then writes this video data to the cloud storage resource pool built on the storage device clusters. The storage space assigned to this plan is large enough to store the video data for 12 months. + +Step 3: Tom can download the video files from the cloud; he can also replay the videos stored in the cloud on demand. + +### Case 2: Video file uploading from mobile devices + +Mobile visual surveillance devices are widely used today. The storage capacity of a mobile device is limited, and the user wants to access the video captured by a mobile device using any terminals, e.g., PC, anywhere. The user can then upload the captured video files from the mobile device to the cloud. + +Step 1: Bob logs into the VS system through a mobile device. He often uses the camera of the mobile device to capture videos; these video files are stored in the mobile device. + +Step 2: Since the storage space of the mobile device is limited, Bob applies for video cloud storage space in order to save more surveillance video data. He chooses the video files from his mobile device, and then uploads the files to the cloud through wireless channels. + +Step 3: Bob logs into the visual surveillance system using a PC, and can download the video files (which were uploaded from his mobile device) from the cloud. He can also replay the videos, stored in the cloud, on demand. + +## **7.3 Video metadata management** + +The VS system can provide various intelligent services by using video/image analysis technologies. The intelligent traffic service is an example. For realizing these intelligent services, some video/image analysis algorithms are embedded in the traffic surveillance cameras, and these cameras can directly output video content metadata, such as: vehicle plate number, vehicle type, speed of vehicle, colour of vehicle. In addition, video analysis servers can be deployed in the data centre, and the surveillance video stream or the recorded video files can be processed to obtain the video content metadata on these servers. + +Step 1: Model the surveillance video metadata, and design and implement the video metadata management system in cloud. + +Step 2: The surveillance video metadata is obtained from the intelligent video/image analysis devices, and is sent to the cloud. + +Step 3: An authorized user can browse and retrieve the surveillance video metadata stored in the cloud. Likewise, authorized application systems can access the video metadata in the cloud, and provide the various intelligent services based on the video metadata. + +## **7.4 Picture storage** + +### **Case 1: Picture upload from surveillance camera** + +High-definition cameras are widely deployed on streets to capture high-resolution pictures. These pictures may contain useful information and can be processed further. For example, vehicle plate numbers can be extracted accurately from a picture by using a vehicle plate recognition algorithm. + +Step 1: Network cameras continuously capture pictures, and send these pictures directly to the cloud. + +Step 2: The cloud storage system receives the pictures, and writes the pictures to the cloud storage resource pool built on the storage device clusters. + +Step 3: An authorized user or application system can access the pictures stored in the cloud. + +### **Case 2: Picture uploading from mobile device** + +John is a traffic police officer, and one of his routine tasks is to find illegally parked vehicles and record the infraction. + +Step 1: John logs into the surveillance system through a mobile device. He captures the picture of an illegally parked vehicle using the camera of the mobile device. + +Step 2: The captured picture is then sent to the cloud through wireless channels. + +Step 3: An authorized user or application system can access the pictures stored in the cloud. + +# **8 Requirements for cloud storage in visual surveillance** + +## **8.1 User requirements** + +There are two types of cloud storage service users : the service consumer and the service provider. + +### **8.1.1 Cloud storage service consumer requirements** + +- USR-001: A VSCS system is required to support registration and de-registration of the end user through the interface provided by the system, and the end user can view and modify personal information. +- USR-002: A VSCS system is required to support end-user login and logout from the system conveniently. The user name and password are required when an end user logs into the system. +- USR-003: A VSCS system is recommended to support end-user view of the user access logs or other system logs. +- USR-004: A VSCS system is required to support user's data uploading from a client device, including video files, pictures and video metadata. +- USR-005: A VSCS system is required to support the user's view and downloading of stored data via a client device, including video files, pictures and video metadata. +- USR-006: A VSCS system is recommended to support the video on demand and picture presentation for end users. +- USR-007: A VSCS system is recommended to support information retrieval for end users. +- USR-008: A VSCS system is recommended to support flexible storage space application for end users. +- USR-009: A VSCS system is required to support the view of a user's storage space status, including the storage space occupancy rate, the remaining storage space. + +### **8.1.2 Cloud storage service provider requirements** + +- USR-010: A VSCS system is required to support the provider's login and logout from the system conveniently. The provider's name and password are required when logging into the system. +- USR-011: A VSCS system is required to support the provider's view of the cloud storage system operating status from the beginning to then-present time, including the system storage space occupancy rate, the system's remaining storage space, the storage space occupancy rate of individual users, the remaining storage space of individual users, etc. +- USR-012: A VSCS system is required to support flexible storage space assignment for individual users. + +## **8.2 Service requirements** + +### **8.2.1 Video storage service requirements** + +- SER-001: A VSCS system is required to support directly writing video streams, from multiple PUs, through the network. +- SER-002: A VSCS system is required to support video file uploading from multiple local video storage devices through the network, including NVRs, mobile video capturing devices, etc. +- SER-003: A VSCS system is required to support video file deletion. +- SER-004: A VSCS system is required to support video file browsing and searching; the searching conditions can be the source of video files, the capture time of video files, etc. + +- SER-005: A VSCS system is required to support video file downloading through the network. +- SER-006: A VSCS system is required to support automatic video file overwriting when a user's storage space is full; the overwriting principle is that the oldest data is replaced first. +- SER-007: A VSCS system is recommended to support video playback for end users according to the source of the video file and the capture time of the video file. Video playback operations include fast forward, slow forward, pause, and stop. + +### **8.2.2 Picture storage service requirements** + +- SER-008: A VSCS system is required to support direct writing of pictures, from multiple PUs, through the network. +- SER-009: A VSCS system is required to support picture uploading from multiple local video storage devices through the network, including PCs, mobile picture capturing devices, etc. +- SER-010: A VSCS system is required to support picture deletion. +- SER-011: A VSCS system is required to support picture browsing and searching; the searching conditions can be the source of the pictures, the capture time of the pictures, etc. +- SER-012: A VSCS system is required to support picture downloading through the network. + +### **8.2.3 Video metadata storage service requirements** + +- SER-013: A VSCS system is required to support video metadata writing through the network. +- SER-014: A VSCS system is required to support video metadata deletion. +- SER-015: A VSCS system is required to support video metadata browsing and searching; the searching condition can be the source of video metadata, the creation time of video metadata, video metadata content, etc. + +## **8.3 Security requirements** + +### **8.3.1 Authentication security requirements** + +- SEC-001: A VSCS system is required to provide the mechanisms for authentication and authorization, and it is required to only permit authorized users to access the system and use system services. A VSCS system is required to forbid an unauthorized user to handle any resources of the system. + +### **8.3.2 Access security requirements** + +- SEC-002: A VSCS system is required to operate in environments where network address translation (NAT) and/or firewall devices are present. It is recommended to utilize specified firewalls, gatekeepers and other network devices to ensure security for access to some special cloud storage services. + +### **8.3.3 Content security requirements** + +- SEC-003: A VSCS system is required to ensure the security of stored data such as video, picture, video metadata, etc. It is recommended to provide the mechanisms to protect the copyrights of the stored data, and to protect the stored data from being corrupted. +- SEC-004: A VSCS system is required to provide data protection mechanisms such as data backup, coding, etc. It is required to be able to recover the destroyed data. +- SEC-005: A VSCS system is required to protect user privacy. + +### **8.3.4 System security requirements** + +- SEC-006: A VSCS system is required to have the capability of resisting various attacks. + +- SEC-007: A VSCS system is required to provide the mechanisms for troubleshooting. It is required that a structural single-node problem be avoided (i.e., a problem at a single node should not cause failure of the entire system). + +## **8.4 Management requirements** + +### **8.4.1 Storage management requirements** + +- MAN-001: A VSCS system is required to support storage space management. The storage space can be increased and decreased flexibly. + +### **8.4.2 Equipment management requirements** + +- MAN-002: A VSCS system is required to provide the unified management of the storage equipment. + +### **8.4.3 Service management requirements** + +- MAN-003: A VSCS system is required to provide the various storage service subscription means for users, and to provide the capabilities of querying, viewing and modifying their subscription information. +- MAN-004: A VSCS system is recommended to provide the capability of accounting, charging and billing for the service operation. +- MAN-005: A VSCS system is recommended to provide various alternative accounting modes, and to support flexible combinations of payment modes, billing modes, billing cycles, preferential pricing, etc. + +### **8.4.4 Data management requirements** + +- MAN-006: A VSCS system is required to support data management in the cloud, including adding, deleting, browsing, indexing, searching. + +### **8.4.5 System management requirements** + +- MAN-007: A VSCS system is required to provide a unified system management interface which can be called conveniently. +- MAN-008: A VSCS system is required to provide a visual interface for users. + +### **8.4.6 Operation management requirements** + +- MAN-009: A VSCS system is required to monitor, record and display the running status of the system. + +## **8.5 Scalability requirements** + +- SCA-001: A VSCS system is required to provide storage resource scalability. When the storage equipment is increased or decreased in the system, the storage capacity is increased or decreased accordingly, and the system service is uninterrupted. +- SCA-002: A VSCS system is required to provide storage space scalability for users. The storage space of individual users can be increased or decreased according to each user's demands. +- SCA-003: A VSCS system is required to provide user scalability. The number of supported users can be increased or decreased dynamically. + +## **8.6 Reliability requirements** + +- REL-001: A VSCS system is required to ensure service reliability. When storage equipment fails or the storage system network operation is abnormal, the system service can be used normally. +- REL-002: A VSCS system is required to ensure data reliability. When data is destroyed, the system can recover it based on the data protection mechanism, e.g., data backup, data coding. + +## **8.7 Performance requirements** + +- PER-001: A VSCS system is required to support concurrent user operations. The system can serve a large number of users simultaneously, while ensuring the service quality. + +# Bibliography + +- [b-ITU-T M.60] Recommendation ITU-T M.60 (1993), *Maintenance terminology and definitions.* +- [b-ITU-T Y.101] Recommendation ITU-T Y.101 (2000), *Global Information Infrastructure terminology: Terms and definitions.* +- [ITU-T FG TR] ITU-T FG Technical Report: Part 1 (2012), *Introduction to the cloud ecosystem: definitions, taxonomies, use cases and high-level requirements.* + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | General tariff principles | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Terminals and subjective and objective assessment methods | +| Series Q | Switching and signalling | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects and next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.743.20-202008-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.743.20-202008-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9002ee68e2b9ebf63d27d39ba8c73bdafb0e6be6 --- /dev/null +++ b/marked/F/T-REC-F.743.20-202008-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8df701536c25ab8246d2868de3c20c8a4e25168bb52c5eb9078278ab1b5b4fd8 +size 6859 diff --git a/marked/F/T-REC-F.743.20-202008-I_PDF-E/367926125450c2bc3f4bdca9d59a62ba_img.jpg b/marked/F/T-REC-F.743.20-202008-I_PDF-E/367926125450c2bc3f4bdca9d59a62ba_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d3f2bdd2777932b396426ccdf75d81dcbea09bb6 --- /dev/null +++ b/marked/F/T-REC-F.743.20-202008-I_PDF-E/367926125450c2bc3f4bdca9d59a62ba_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d9c2be715012a330d6c33026eb375c2c95fb95380b3b33e40ffcac43b07c3ab9 +size 26826 diff --git a/marked/F/T-REC-F.743.20-202008-I_PDF-E/562f471e8153729557e6a4ee6343c32c_img.jpg b/marked/F/T-REC-F.743.20-202008-I_PDF-E/562f471e8153729557e6a4ee6343c32c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..e7ef0157ee93e5961d1eb5671c835f04a0cca019 --- /dev/null +++ b/marked/F/T-REC-F.743.20-202008-I_PDF-E/562f471e8153729557e6a4ee6343c32c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6f16f432c70c5d50c39c59053fb9edd131e549d2f102bed349d665c833aef395 +size 66787 diff --git a/marked/F/T-REC-F.743.21-202008-I_PDF-E/1c94fd3cebf58af136144f14160d128e_img.jpg b/marked/F/T-REC-F.743.21-202008-I_PDF-E/1c94fd3cebf58af136144f14160d128e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a548a0d815ca4088be2b2a268844ea3fac085919 --- /dev/null +++ b/marked/F/T-REC-F.743.21-202008-I_PDF-E/1c94fd3cebf58af136144f14160d128e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3aca2c88c3fde99849890f8427545eee8f703a9fac792efa75c75d10c40f3b52 +size 34609 diff --git a/marked/F/T-REC-F.743.21-202008-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.743.21-202008-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9002ee68e2b9ebf63d27d39ba8c73bdafb0e6be6 --- /dev/null +++ b/marked/F/T-REC-F.743.21-202008-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8df701536c25ab8246d2868de3c20c8a4e25168bb52c5eb9078278ab1b5b4fd8 +size 6859 diff --git a/marked/F/T-REC-F.743.21-202008-I_PDF-E/raw.md b/marked/F/T-REC-F.743.21-202008-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..3cfd8f6a773b482a362aedee3f37e53f82cd47d3 --- /dev/null +++ b/marked/F/T-REC-F.743.21-202008-I_PDF-E/raw.md @@ -0,0 +1,468 @@ + + +International Telecommunication Union + +# **ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +# **F.743.21** + +(08/2020) + +### SERIES F: NON-TELEPHONE TELECOMMUNICATION SERVICES + +Multimedia services + +## **Framework for data asset management** + +Recommendation ITU-T F.743.21 + +**ITU-T** + +![ITU logo](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) features a blue globe with a red lightning bolt striking across it. The letters 'ITU' are written in blue inside the globe. To the right of the logo, the text 'International Telecommunication Union' is written in blue. + +International +Telecommunication +Union + +ITU logo + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +### **TELEGRAPH SERVICE** + +| | | +|-----------------------------------------------------------------|-------------| +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | + +## **MOBILE SERVICE** + +| | | +|--------------------------------------------------------|-------------| +| Mobile services and multdestination satellite services | F.110–F.159 | +|--------------------------------------------------------|-------------| + +## **TELEMATIC SERVICES** + +| | | +|-------------------------------------------|-------------| +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | + +## **MESSAGE HANDLING SERVICES** + +F.400–F.499 + +## **DIRECTORY SERVICES** + +F.500–F.549 + +## **DOCUMENT COMMUNICATION** + +| | | +|--------------------------------------|-------------| +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | + +## **DATA TRANSMISSION SERVICES** + +F.600–F.699 + +## **MULTIMEDIA SERVICES** + +**F.700–F.799** + +## **ISDN SERVICES** + +F.800–F.849 + +## **UNIVERSAL PERSONAL TELECOMMUNICATION** + +F.850–F.899 + +## **ACCESSIBILITY AND HUMAN FACTORS** + +F.900–F.999 + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.743.21 + +# Framework for data asset management + +## Summary + +Recommendation ITU-T F.743.21 defines a data asset management framework with its corresponding objects, activities and supports. Objects of data asset management are data assets, which include master data, metadata, and other data assets. Activities include data standards management, data model management, data quality management, data security management, data valuation management, and data sharing management. To ensure the proper level of management, the corresponding people in charge, rules and regulations, and technology tools are needed. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.743.21 | 2020-08-13 | 16 | 11.1002/1000/14326 | + +## Keywords + +Big data, data asset management. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2020 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-----------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Overview ..... | 2 | +| 7 Management objects ..... | 3 | +| 7.1 Metadata management..... | 3 | +| 7.2 Master data management..... | 3 | +| 8 Management activity ..... | 4 | +| 8.1 Data standards management ..... | 4 | +| 8.2 Data model management ..... | 4 | +| 8.3 Data quality management ..... | 5 | +| 8.4 Data security management ..... | 6 | +| 8.5 Data valuation management ..... | 6 | +| 8.6 Data sharing management ..... | 6 | +| 9 Management support ..... | 7 | +| 9.1 People ..... | 7 | +| 9.2 Rules and regulations ..... | 7 | +| 9.3 Technical tools..... | 7 | + + + +# Recommendation ITU-T F.743.21 + +# Framework for data asset management + +# 1 Scope + +This Recommendation defines a data asset management framework with its corresponding objects, activities and supports, which can be used as a reference for a government and enterprises to apply data asset management, as well as a reference framework for data management platform and product design. The scope of this Recommendation includes: + +- Management object, +- Management activities, +- Management support. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T Y.3519] Recommendation ITU-T Y.3519 (2018), *Cloud computing – Functional architecture of big data as a service*. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following term defined elsewhere: + +**3.1.1 metadata:** [ITU-T Y.3519]: Data about data or data elements, possibly including their data descriptions, and data about data ownership, access paths, access rights and data volatility. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 data standards:** Normative constraints to guarantee consistency and accuracy of data usage and exchange. + +**3.2.2 data quality:** The integrity, normalization, consistency, accuracy, uniqueness, and relevance of data, which is the premise and guarantee for data mining. + +**3.2.3 data asset management:** A set of business functions for planning, controlling and providing data and information assets, including developing, executing and monitoring data related plans, policies, programs, projects, processes, methods and procedures, so as to control, protect, deliver and improve the value of data assets. Data asset management requires a full integration of business, technology, and management to ensure that the value of data assets is maintained and added. + +**3.2.4 data management platform:** Software products that provide some functions for data management, including data source management, metadata management, data standard management, master data management, data quality management, and data flow management. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +DBMS Database Management System + +## 5 Conventions + +In this Recommendation, the following conventions are used: + +- The keywords "is required" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement needs not be present to claim conformance. +- The keywords "can optionally" indicate an optional requirement which is permissible, without implying any sense of being recommended. This term is not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +## 6 Overview + +The framework of data asset managements is shown in Figure 1, and includes three main aspects: management objects, management activities and management support. + +**Object:** Objects of data asset management are data assets, which include master data, metadata, and other data assets. The management of metadata and master data is the key activities of data asset management. + +**Activity:** Activities of data asset management include data standards management, data model management, data quality management, data security management, data valuation management and data sharing management. + +**Support:** In order to ensure the proper level of management, the corresponding people in charge, rules and regulations, and technology tools are needed. + +![Figure 1 – Overview of data asset management. The diagram shows three concentric circles. The innermost circle is labeled 'Data asset' and contains 'Metadata' and 'Master data'. The middle circle is labeled 'Data quality' and contains 'Data model', 'Data standards', 'Data security', 'Data valuation', and 'Data sharing'. The outermost circle is labeled 'People' and contains 'Rules and regulation' and 'Technology tools'.](1c94fd3cebf58af136144f14160d128e_img.jpg) + +The diagram illustrates the framework of data asset management as three concentric circles. The innermost circle is labeled "Data asset" and contains the terms "Metadata" and "Master data". The middle circle is labeled "Data quality" and contains the terms "Data model", "Data standards", "Data security", "Data valuation", and "Data sharing". The outermost circle is labeled "People" and contains the terms "Rules and regulation" and "Technology tools". The label "F.743.21(20)\_F01" is located at the bottom right of the diagram. + +Figure 1 – Overview of data asset management. The diagram shows three concentric circles. The innermost circle is labeled 'Data asset' and contains 'Metadata' and 'Master data'. The middle circle is labeled 'Data quality' and contains 'Data model', 'Data standards', 'Data security', 'Data valuation', and 'Data sharing'. The outermost circle is labeled 'People' and contains 'Rules and regulation' and 'Technology tools'. + +**Figure 1 – Overview of data asset management** + +## 7 Management objects + +Objects of data asset management are data assets. Master data and metadata can be abstracted from data assets. + +Master data is the data representing the core business object and the transaction business of an enterprise. These specific data will be shared and applied across various business processes, departments and systems. + +Metadata is the data about data, which describes the data itself (e.g., database, data models), the concepts the data represents (e.g., application system, business process) and the relationships between them. + +## 7.1 Metadata management + +Metadata management is planning, implementation, and control of activities to enable access to high quality integrated metadata. Metadata management ensures that people understand data content and can use data consistently based on documenting and managing organizational knowledge of data-related terminology, and on collecting and integrating metadata from diverse sources. + +Metadata can be categorized into three types, business, technical and operational metadata, based on how metadata is used. + +The main activities of metadata management are recommended to include: + +- Establishing and maintain metadata standards, +- Building metadata management tools, +- Creating, collecting, and integrating metadata, +- Managing metadata repository, +- Distributing and delivering metadata, +- Metadata analysis (e.g., data lineage and impact analysis, data mapping). + +#### 7.1.1 Technical metadata + +Technical metadata describes information about the technical details of data, data storage systems, and processes that move data within and between systems, including database object properties, access permissions, file format schema definitions, etc. + +#### 7.1.2 Business metadata + +Business metadata describes the concepts, relationships, rules, and other details of data governance, including the non-technical names and definitions of concepts, subject areas, entities, attributes, and calculations. + +#### 7.1.3 Operational metadata + +Operational metadata describes details of the processing and accessing data, including history of data extracts, schedule anomalies, data archiving and retention, etc. + +## 7.2 Master data management + +Master data management is a set of rules, applications, and technologies to ensure the uniformity, accuracy, semantics, consistency, and accountability of the enterprise's officially shared master data assets. + +The main activities of master data management are recommended to include: + +- Understanding and defining the requirements of master data management, +- Identifying and assessing master data sources, +- Maintaining data integration architecture, + +- Defining and maintaining data matching rules, +- Establishing process approval mechanism for master data collection and creation, +- Easy to modify, monitor and update the main data across the integrated systems. + +# **8 Management activity** + +### **8.1 Data standards management** + +Data standards refer to the normative constraints that guarantee the consistency and accuracy of data use and exchange. Data standards management is planning and publishing a set of standards regarding data items, datasets and data models to make sure that data is complete, valid, consistent, standard, and easy to be shared and managed. + +Data standards can usually be categorized into basic data standards and indicator data standards. Basic data standards generally include metadata standards, master data standards, data dimension (measurement) standards, logical data model standards, physical data model standards, and common code standards. Indicator data standards generally concern basic indicators and calculation indicators. Calculation indicator is a combination of two or more basic indicators. + +The main activities of data standards management are recommended to include: + +- Understanding data standardization requirements, +- Constructing data standards system and specification, +- Planning the implementation roadmap and scheme of data standardization, +- Establishing data standards management methods and process requirements, +- Building data standards management tools, +- Evaluating the development of data standardization. + +### **8.2 Data model management** + +Data model management is the process of discovering analysing, and scoping data requirements, and then representing and communicating these data requirements in a precise form called the data model. This process is iterative and may include a conceptual, logical, and physical model. + +The main activities of data model management are recommended to include: + +- Identifying the information and business requirements, +- Building data model management process standards, +- Selecting or creating data model management tools, +- Defining entities, attributes, and business-oriented key relationships, +- Building diagrams to represent definitions, +- Building a set of conceptual, logical, and physical models, +- Reviewing and improving the data models (scope, content, and structure), +- Maintaining the data models. + +#### **8.2.1 Conceptual model** + +Conceptual model is oriented by objective-world, and it mainly describes each entity and the relationships between entities, which captures the high-level data and is independent from database management system (DBMS). + +#### **8.2.2 Logical model** + +Logical model is a business-oriented model based on business processes and business scenarios, and it can be used to guide implementation in different DBMS systems. The logical model includes network data model, hierarchical data model, etc. + +#### **8.2.3 Physical model** + +The physical model describes the organizational structure of data in a storage media. The design of the physical model is based on the logical model, and is adapted to function considering a set of specific DBMS, operating system and hardware. It also should be designed with the relevant capabilities and requirements of system performance. + +### **8.3 Data quality management** + +High data quality guarantees that the data application process is carried out correctly. Data quality management means planning, implementation, and control of activities that apply quality management techniques to data in order to excavate the data value and to meet the needs of data consumers. Producing high quality data is a necessary premise for developing data products and services, and this process requires cross-functional commitment and coordination, including building quality assessment system, managing data through its lifecycle by setting standards, etc. + +A data quality dimension is a measurable feature of the characteristic of data, which can be used as a basis for an initial data quality assessment. Dimensions of data quality include: + +- Accuracy: The degree to which data correctly describes the real object that is being described. +- Completeness: The proportion of data stored against the potential for 100%. +- Uniqueness: The characteristic of an entity instance whereby it will not be recorded more than once based on how it is identified. +- Timeliness: The degree to which data represents reality at the required point in time. +- Validity: Data is valid if it conforms to the syntax (format, type, range) of its definition. +- Consistency: The absence of difference, when comparing two or more representations of a thing against a definition. + +The main activities of data quality management are recommended to include: + +- Realizing the importance of data quality improvement, +- Defining data quality requirements, +- Analysing and assessing data quality, +- Defining measurements in terms of data quality dimensions, +- Identifying critical data and business rules, +- Testing and validating data quality requirements, +- Assessing data quality service levels, +- Constant measuring and monitoring of data quality, +- Diagnosing data quality issues, +- Analysing root cause of poor data quality, +- Defining a data quality improvement strategy, +- Cleaning and correcting data quality flaws, +- Designing and operating data quality management tools. + +### **8.4 Data security management** + +Data security includes the privacy, confidentiality regulations, contractual agreements, and business requirements of data assets. It involves not just preventing inappropriate access, but also enabling appropriate access to data. + +Data security management refers to the planning, development, and execution of security policies and procedures to provide proper authentication, authorization, access, and auditing of data assets. + +The main activities of data security management are recommended to include: + +- Identifying data security and regulatory requirements, +- Defining data confidentiality levels, +- Defining data security strategy and standards, +- Defining security roles and corresponding assignment hierarchy, +- Defining data security control scope and procedures, +- Monitoring user identity authentication and access behaviour, +- Deploying data security control systems or tools, +- Managing security policy compliance, +- Auditing data security and compliance activities. + +### **8.5 Data valuation management** + +Data valuation management is a measurement of the intrinsic value of data, including the cost and application value of data. The cost of data includes collection and storage cost (labour costs, IT equipment and other direct and indirect costs) and operation and maintenance costs (business operation fees, technical operation fees, etc.). Data valuation mainly involves classification of data assets, assessment of frequency of usage, identification of the objects and assessment of the effects of use. + +The main activities of data valuation management are recommended to include: + +- Determining the level of enterprise data integration, +- Describing the application scenarios of data, +- Calculating the cost of enterprise data assets, +- Calculating the income of data in different application scenarios, +- Calculating the overall value of enterprise data assets. + +### **8.6 Data sharing management** + +Data sharing management refers to a series of activities of sharing and exchanging data to realize the external value of data. In this process, data owner excavates the hidden information by building data model and sharing or releasing an application product as a combination of these massive information in a form of compliance security. + +The main activities of data sharing management are recommended to include: + +- Defining data asset operation and circulation of monitoring indicators, +- Designing data asset operation and circulation of management plan, +- Formulating data resource operation and circulation of management methods and implementation process requirements, +- Monitoring data asset operation indicators, +- Supervising the implementation of compliance management requirements such as data circulation, + +- Analysing operational circulation indicators, evaluating operational performance, and improving it. + +## **9 Management support** + +People, rules and regulations, and technology tools are three main factors to ensure that data the management process is launched in the correct order. + +## **9.1 People** + +People are the core of data management process. The main activities related to people are to: + +- Establish and maintain enterprise strategy of data management and form data management culture, +- Establish organizational structure with clear rights, duties, and smooth communication, +- Form regular training mechanism for personnel who is engaged in data management. + +## **9.2 Rules and regulations** + +Rules and regulations are requirements of data management process that should be guaranteed. The main activities include to: + +- Establish proper rules and regulations, such as management method, management process, technical specifications, +- Establish audit mechanism and apply it through the life cycle of data management. + +## **9.3 Technical tools** + +Technical tools are fundamentals of data management process. The main activities include to: + +- Select and build technical tools and platforms to meet the needs of enterprises, +- Build technical and operational team to maintain the data management platform. + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.743.22-202212-I_PDF-E/raw.md b/marked/F/T-REC-F.743.22-202212-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..101450135231f1ec393d550b3658e1d34670908a --- /dev/null +++ b/marked/F/T-REC-F.743.22-202212-I_PDF-E/raw.md @@ -0,0 +1,681 @@ + + +# Recommendation **ITU-T F.743.22 (12/2022)** + +SERIES F: Non-telephone telecommunication services +Multimedia services + +--- + +## **Requirements and architecture of an algorithm-training system for intelligent video surveillance** + +![ITU logo](390120de4fe440c42fea8154fcaad334_img.jpg) + +The logo of the International Telecommunication Union (ITU), featuring the letters 'ITU' in blue inside a circle with a globe-like design. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS NON-TELEPHONE TELECOMMUNICATION SERVICES + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessages service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.743.22 + +# Requirements and architecture of an algorithm-training system for intelligent video surveillance + +## Summary + +Recommendation ITU-T F.743.22 specifies requirements and architecture for algorithm-training systems (ATSs) for intelligent video surveillance (IVS) and provides the algorithm-training workflow within it. The intelligent analysis algorithm in IVS needs to train a large amount of actual scene data to improve the accuracy and the recall of identification. The ATS can collect video and image data from the IVS, complete sample data selection and annotate data, as well as training and deploying algorithms in the IVS, so that the IVS has the reasoning capability of new scenarios and the ATS can continuously iteratively improve on algorithm identification performance. Recommendation ITU-T F.743.22 aims to solve the problems of difficult application of algorithm customization in various industries, long algorithm development and iteration cycle, and effective protection of user privacy and improvement of data security. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.743.22 | 2022-12-14 | 16 | 11.1002/1000/15187 | + +## Keywords + +Algorithm-training system, architecture, intelligent video surveillance, requirement, workflow. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2023 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-------------------------------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 3 | +| 4 Abbreviations and acronyms ..... | 3 | +| 5 Conventions ..... | 4 | +| 6 Overview..... | 4 | +| 6.1 Introduction of the algorithm-training system for intelligent video surveillance..... | 4 | +| 6.2 Algorithm-training workflow in the algorithm-training system..... | 5 | +| 7 Requirements of the algorithm-training system for intelligent video surveillance ..... | 6 | +| 7.1 General requirements..... | 6 | +| 7.2 Requirements of sample data collection..... | 6 | +| 7.3 Requirements of sample data annotation..... | 7 | +| 7.4 Requirements of algorithm training..... | 7 | +| 7.5 Requirements of algorithm deployment ..... | 8 | +| 8 Architecture of the algorithm-training system for intelligent video surveillance..... | 8 | +| 8.1 Functional architecture ..... | 8 | +| 8.2 Reference points ..... | 10 | +| Bibliography..... | 13 | + + + +# Recommendation ITU-T F.743.22 + +# Requirements and architecture of an algorithm-training system for intelligent video surveillance + +# 1 Scope + +This Recommendation specifies requirements and architecture for algorithm-training systems (ATSs) for intelligent video surveillance (IVS) and provides the algorithm-training workflow within it. + +The scope of this Recommendation includes: + +- an overview of ATS for IVS; +- requirements for ATS for IVS; and +- architecture and reference points of ATS for IVS. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +- [ITU-T H.222.0] Recommendation ITU-T H.222.0V8 (2021), *Information technology – Generic coding of moving pictures and associated audio information: Systems*. +- [ITU-T H.264] Recommendation ITU-T H.264V14 (2021), *Advanced video coding for generic audiovisual services*. +- [ITU-T H.265] Recommendation ITU-T H.265V8 (2021), *High efficiency video coding*. +- [ITU-T H.626] Recommendation ITU-T H.626V2 (2019), *Architecture requirements for video surveillance system*. +- [ITU-T H.626.5] Recommendation ITU-T H.626.5V2 (2022), *Architecture for intelligent video surveillance systems*. +- [ITU-T T.81] Recommendation ITU-T T.81 (1992), *Information technology – Digital compression and coding of continuous-tone still images – Requirements and guidelines*. +- [IETF RFC 3550] IETF RFC 3550 (2003), *RTP: A transport protocol for real-time applications*. +- [IETF RFC 4571] IETF RFC 4571 (2006), *Framing real-time transport protocol (RTP) and RTP control protocol (RTCP) packets over connection-oriented transport*. +- [IETF RFC 9110] IETF RFC 9110 (2022), *HTTP semantics*. +- [IETF RFC 9112] IETF RFC 9112 (2022), *HTTP/1.1*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 accuracy** [b-ISO/IEC 2382]: Quantitative measure of the magnitude of error, preferably expressed as a function of the relative error, a high value of this measure corresponding to a small error. + +**3.1.2 algorithm** [b-ISO/IEC 11557]: A set of rules for transforming the logical representation of data. + +**3.1.3 data annotation** [b-ISO/IEC 22989]: Process of attaching a set of descriptive information to data without any change to that data. + +NOTE – The descriptive information can take the form of metadata, labels and anchors. + +**3.1.4 data storage and service (DSS)** [ITU-T H.626.5]: The storage and sharing centre for video and image information within the intelligent video surveillance system. The DSS accepts registration, keepalive reports from the intelligent premises unit (IPU), intelligent video management (IVM) or intelligent application service (IAS) and obtains video and image information from them. It can also provide information network sharing services such as information querying, subscription, guard and alarm, etc. + +**3.1.5 intelligent application service (IAS)** [ITU-T H.626.5]: The intelligent application service (IAS) is located at the application service core of the intelligent video surveillance system. Its main functions include providing application services about the video and image information, such as information query and retrieval, subscription and notification, manual annotation, etc.; providing external operations such as the creation and start-up of intelligent analysis tasks based on intelligent video management (IVM); obtaining the list of devices from the central control server (CCS) or data storage and service (DSS) and sharing them with the IVM so that it can schedule intelligent video units (IVUs) for video stream analysis, etc. + +**3.1.6 intelligent premises unit (IPU)** [ITU-T H.626.5]: The premises subsystem within the intelligent video surveillance system. The premises intelligent video (PIV) module is added to the premises unit (PU) for intelligent video analysis. + +**3.1.7 intelligent video unit (IVU)** [ITU-T H.626.5]: The intelligent video unit (IVU) identifies specific objects automatically, and outputs recognition results to an intelligent video management (IVM) system. The recognition information includes triggered events and acquired data. One or more intelligent analysis algorithms can be loaded or unloaded on an intelligent video unit (IVU) according to different requirements. + +**3.1.8 media distribution unit (MDU)** [ITU-T H.626.5]: The media distribution unit (MDU) is used to transport media from the premises unit (PU) to the customer unit (CU). Its main functions include media receiving, media processing, media routing, media transmission, media forwarding and media replication. + +**3.1.9 media server** [ITU-T H.626]: A device located at the centre part of a video surveillance system. It is used to forward real-time media stream as well as store, retrieve and replay historical media stream. The media server receives the media stream from premises unit or other media server and forwards the media stream to other customer unit or media server. It consists of media distribution unit and media storage unit. + +**3.1.10 model** [b-ISO/IEC TR 29119-11]: Output of an ML algorithm trained with a training dataset that generates predictions using patterns in the input data. + +**3.1.11 recall** [b-ISO/IEC TR 29119-11]: Performance metric used to evaluate a classifier, which measures the proportion of actual positives that were predicted correctly. + +**3.1.12 sample** [b-ISO/IEC 22989]: Atomic data element processed in quantities by a machine learning algorithm or an AI system. + +**3.1.13 training dataset** [b-ISO/IEC TR 24029-1]: Set of samples used to fit a machine learning model. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 algorithm management unit (AMU):** A subsystem unit of the algorithm-training system that provides model warehouse management, model adaptation, model packaging and algorithm deployment. + +**3.2.2 algorithm training:** A technology that learns the experience or the environment represented by input data and upgrades the algorithm. + +**3.2.3 algorithm-training management (ATM):** A subsystem unit of the algorithm-training system (ATS), which provides the management of the sample management unit (SMU), the training management unit (TMU) and the algorithm management unit (AMU). + +**3.2.4 algorithm-training system (ATS):** A system composed of algorithm-training management (ATM), sample management unit (SMU), training management unit (TMU) and algorithm management unit (AMU), which provides algorithm-training services for intelligent video surveillance (IVS). + +**3.2.5 sample management unit (SMU):** A subsystem unit of the algorithm-training system that provides training dataset management, data annotation and annotation information storage. + +**3.2.6 training management unit (TMU):** A subsystem unit of the algorithm-training system that provides algorithm-training task management, algorithm training, and model verification and evaluation. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|----------------------------------| +| AMU | Algorithm Management Unit | +| ATM | Algorithm-Training Management | +| ATS | Algorithm-Training System | +| CCS | Central Control Server | +| CIV | Client Intelligent Video | +| CMU | Central Management Unit | +| DSS | Data Storage and Service | +| HTTP | Hypertext Transfer Protocol | +| IAS | Intelligent Application Service | +| ICU | Intelligent Customer Unit | +| IPU | Intelligent Premises Unit | +| IVM | Intelligent Video Management | +| IVS | Intelligent Video Surveillance | +| IVU | Intelligent Video Unit | +| JPEG | Joint Photographic Experts Group | +| JSON | JavaScript Object Notation | +| MDU | Media Distribution Unit | +| MP4 | Moving Picture Experts Group 4 | + +| | | +|------|--------------------------------------| +| MS | Media Server | +| MSU | Media Storage Unit | +| PIV | Premises Intelligent Video | +| RTCP | Real-time Transport Control Protocol | +| RTP | Real-time Transport Protocol | +| SCU | Service Control Unit | +| SMU | Sample Management Unit | +| TMU | Training Management Unit | + +# 5 Conventions + +In this Recommendation: + +- The phrase "**is required to**" indicates a requirement that must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The phrase "**is recommended**" indicates a requirement that is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. +- The phrase "**can optionally**" indicates an optional requirement that is permissible, without implying any sense of being recommended. This term is not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator or service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance to this Recommendation. + +# 6 Overview + +## 6.1 Introduction of the algorithm-training system for intelligent video surveillance + +The intelligent analysis algorithm in an IVS needs to train a large amount of actual scene data to improve the accuracy and recall of identification. At present, there are many mature and general algorithms with high capabilities and wide applications. + +Some algorithms, which may not be so common and mature, are specifically designed for a certain scenario, and may not be trained on a large amount of data at the time of deployment, or the training dataset may be different from the actual scenario. Users can use the same type of algorithm to identify objects, behaviours or attributes in the acquisition environment. + +For example, one algorithm can identify smoke and fire. When smoke identification algorithms are also applied to identify fires in the same scene, they always need to be trained to improve accuracy. An algorithm is only as good as its training data. To make the algorithm accurate, continuous algorithm training is essential during the operation of the video surveillance system. + +In some cases, uploading training data to external platforms for algorithm training is not allowed due to privacy and data security. For example, when an IVS system is deployed in a restricted area, algorithms are used to analyse whether the equipment is damaged or whether there is oil leaking on the ground. The videos and images taken from the restricted area are kept confidential, and they are not allowed to be uploaded to external platforms for algorithm training. + +An ATS can solve this problem. The video and image collected by the IVS can be annotated, trained in the ATS, and deployed into the IVS locally. The purpose of algorithm training is to learn the law hidden behind the data. For data outside the training set having a similar or adapted law, the trained algorithm can give an appropriate output and it is considered that the algorithm has good generalization performance. + +## **6.2 Algorithm-training workflow in the algorithm-training system** + +Algorithm-training workflow in the ATS can be divided into sample data collection, sample data annotation, algorithm training, model verification and evaluation, and algorithm deployment. + +### **6.2.1 Sample data collection** + +The steps in sample data collection follow: + +- 1) to create an algorithm-training task in the ATS according to the application scenarios, such as object classification and object detection; +- 2) to consult and analyse real-time or historical video and image data in the IVS; +- 3) to upload the optimized video and image data to the ATS after verification, cleaning and enhancement; +- 4) to create corresponding training datasets and classification categories according to the data types such as video or image; +- 5) to import the uploaded data into the corresponding training datasets. + +### **6.2.2 Sample data annotation** + +The steps in sample data annotation follow: + +- 1) to select the training dataset to be annotated and determine the method, such as annotation that is manual or automatic; +- 2) to annotate the data in the training dataset. + +NOTE 1 – Manual annotation is performed by hand on new sample data by creating a single person annotation task or multi-person collaborative annotation task or re-annotate by modifying or deleting labels on the annotated sample data. + +NOTE 2 – Automatic annotation is the process of refining existing annotation labels of video and image recordings using existing models in the system for intelligent annotation. + +### **6.2.3 Algorithm training** + +The steps in algorithm training follow: + +- 1) to create a model warehouse; +- 2) to create a training task, select a training dataset and configure the available computing resources; +- 3) to execute all elements in the training task list (e.g., start, stop, recreate, retrieve); +- 4) to store the trained model in the model warehouse. + +### **6.2.4 Model validation and evaluation** + +The steps in model verification and evaluation follow: + +- 1) to upload the verification sample data to the ATS; +- 2) to use the trained model inference to analyse the multiple verification sample data and comprehensively evaluate the model; +- 3) to visualize the results of the analysis; +- 4) to manually verify the results of the analysis. + +### **6.2.5 Algorithm deployment** + +The steps in algorithm deployment follow: + +- 1) to select the optimal model according to its evaluation results; +- 2) to check the hardware and software information about the device to be deployed, and generate a model that matches the device; + +- 3) to pack the model and runtime-related components into the algorithm and upload it to the device to be deployed, and then start the algorithm upgrade – after the upgrade is complete, the device enables the new algorithm for inference. + +NOTE – "The device" is the one in the IVS. + +# **7 Requirements of the algorithm-training system for intelligent video surveillance** + +In this Recommendation, requirements specific to the ATS for IVS are identified. + +## **7.1 General requirements** + +The ATS: + +- is required to support algorithm scheme management, including but not limited to object classification and object detection; +- can optionally support data isolation, data encryption and data protection. + +## **7.2 Requirements of sample data collection** + +### **7.2.1 General requirements** + +- The sample data is recommended to satisfy homology, diversity and balance. + +### **7.2.2 Requirements of training dataset** + +The training dataset is required to: + +- support addition, deletion, modification and querying of datasets. +- support video and image type datasets; +- support the release of the dataset version; +- meet the minimum number of training samples. + +### **7.2.3 Requirements of sample data collection method** + +The system is required to support: + +- video and image collection from the IVS (the collection method is specified in [ITU-T H.626]); +- single and batch manual uploads of local video and image files. + +### **7.2.4 Requirements of sample data quality** + +The sample data is required to satisfy: + +- absence of imaging quality issues, such as overexposure, excess darkness and abnormal deformation; +- the requirements of the algorithm scheme, including but not limited to image resolution and the proportion of the objects in the image; +- that the object is clearly visible and of appropriate size. + +### **7.2.5 Requirements of sample data format** + +The sample data format is required to support: + +- [ITU-T H.264] and [ITU-T H.265] encoded video data in Moving Picture Experts Group 4 (MP4) [ITU-T H.222.0] format; +- JPG format, Joint Photographic Experts Group (JPEG) (see [ITU-T T.81]) and portable network graphics format image data. + +### **7.2.6 Requirements of sample data processing** + +- The system is required to support the validation, cleaning and enhancement of sample data. + +## **7.3 Requirements of sample data annotation** + +### **7.3.1 General requirements** + +The system: + +- is required to support manual annotation and automatic annotation; +- is recommended to support single-person manual annotation and multi-person collaborative manual annotation; +- is required to support multiple annotation types, including but not limited to image classification and object detection. + +### **7.3.2 Requirements of annotation attribute management** + +The system is required: + +- to support annotation attribute management; +- to provide preset annotation attributes for use; +- to support creation, retrieval, updating and deletion of custom annotation attributes. + +### **7.3.3 Requirements of annotation function** + +The system is required to support: + +- object annotation of videos or images; +- multi-object annotation; +- object annotation for location, size and shape; +- multiple attribute annotations for objects; +- scaling annotations. + +### **7.3.4 Requirements of annotation description** + +The annotation description is required to support: + +- the JavaScript object notation (JSON) [b-Wright] description format; +- metadata definition for annotation types. + +## **7.4 Requirements of algorithm training** + +### **7.4.1 Requirements of computing resource management** + +The system is required to support: + +- matching specifications of computing resource pools based on the training objects; +- heterogeneous computing resource pool management; +- load-balancing management of computing resource pools; +- permission control for computing resource pools; +- status monitoring of computing resource pools, including but not limited to fault alarms, and resource utilization statistics. + +### **7.4.2 Requirements of algorithm-training task management** + +The system is required to support: + +- the classification of algorithm-training tasks; + +- algorithm-training task management, including creation, start, stop, and deletion; +- manual and automatic pause and resume of algorithm-training tasks; +- algorithm task status management, including but not limited to execution progress query. + +### **7.4.3 Requirements of model validation and evaluation** + +- The system is required to support model verification, including but not limited to accuracy, recall and mean average precision. + +## **7.5 Requirements of algorithm deployment** + +### **7.5.1 General requirements** + +The system: + +- is required to support model packaging; +- is required to support algorithm release; +- is recommended to support algorithm orchestration. + +### **7.5.2 Requirements of model management** + +The system is required to support: + +- trained model viewing and support appropriate trained model archiving; +- create, retrieve, update, and delete of the archived models. + +### **7.5.3 Requirements of deployment task management** + +- The system is required to support creation, retrieval, updating and deletion of deployment tasks. + +### **7.5.4 Requirements of deployment status management** + +The system is required to support the retrieval of: + +- the deployment status of devices in the IVS; +- the deployment time and version information of devices in the IVS. + +### **7.5.5 Requirements of model warehouse** + +The model warehouse: + +- is required to support the management, and storage of models; +- is recommended to support model conversion and validation. + +NOTE – Model conversion refers to the system that converts the model to adapt to the different chips. + +# **8 Architecture of the algorithm-training system for intelligent video surveillance** + +## **8.1 Functional architecture** + +Figure 1 shows the functional architecture of the ATS for IVS. + +![Functional architecture of the algorithm-training system for intelligent video surveillance. The diagram shows the interaction between the ATS (Algorithm-Training System) and the IVS (Intelligent Video Surveillance) system. The ATS contains ATM, SMU, TMU, and AMU. The IVS system contains IAS, DSS, IVM, IVU, CCS (CMU, SCU), MS (MDU, MSU), IPU (PIV), and ICU (CIV). Arrows indicate control, media, video and image information, and algorithm flows between these components.](33ed1f9b27c7c21c797aa928b0f06851_img.jpg) + +CIV: client intelligent video + +Functional architecture of the algorithm-training system for intelligent video surveillance. The diagram shows the interaction between the ATS (Algorithm-Training System) and the IVS (Intelligent Video Surveillance) system. The ATS contains ATM, SMU, TMU, and AMU. The IVS system contains IAS, DSS, IVM, IVU, CCS (CMU, SCU), MS (MDU, MSU), IPU (PIV), and ICU (CIV). Arrows indicate control, media, video and image information, and algorithm flows between these components. + +**Figure 1 – Functional architecture of the algorithm-training system for intelligent video surveillance** + +The functional architecture of the ATS for IVS is based on the functional architecture of the IVS system, which is specified in [ITU-T H.626.5]. + +Central management unit (CMU), service control unit (SCU), MDU and media storage unit (MSU) are defined in [ITU-T H.626]. + +The terms IPU, intelligent customer unit (ICU), IVM, IVU, DSS and IAS are defined in [ITU-T H.626.5]. + +The ATS is composed of ATM, SMU, TMU and AMU. + +This Recommendation describes units such as the ATM, SMU, TMU and AMU in clauses 8.1.1 to 8.1.4 and reference points in clause 8.2. + +### 8.1.1 Algorithm-training management + +ATM is a subsystem unit within the ATS. + +The functions of the ATM are to: + +- obtain a media, video and image information resource list; +- provide an algorithm list; +- manage algorithm-training tasks; +- manage annotation tasks; +- manage algorithm deployment tasks. + +### 8.1.2 Sample management unit + +An SMU is a subsystem unit within the ATS. + +The functions of the SMU are to: + +- create training datasets of media, video and image information; +- create, retrieve, update and delete annotation tasks; +- annotate the training dataset and store the annotation information. + +### 8.1.3 Training management unit + +A TMU is a subsystem unit within the ATS. + +The functions of the TMU are to: + +- create, retrieve, update, and delete training tasks; +- verify and archive trained models. + +### 8.1.4 Algorithm management unit + +An AMU is a subsystem unit within the ATS. + +The functions of the AMU are to: + +- create, retrieve, update, and delete the model warehouse; +- manage and store the model; +- package models into algorithms; +- deploy the algorithm. + +## 8.2 Reference points + +Reference points Cm and Ds/Ds' are specified in [ITU-T H.626]. + +Reference points PS, Mp/Mp', Pc, Cs, Mc/Mc', Su, Du, Mu/Mu', Ma/Ma', Ad/Ad', Id/Id', Pd/Pd' and Ac are specified in [ITU-T H.626.5]. + +This Recommendation specifies reference points Ai, Ms, Ta, Tu, Md, Mm, Ts/Ts', Tm, Mv, and Mi. + +### 8.2.1 Reference point Ai: IAS – ATM + +Reference point Ai is between the IAS and the ATM. + +It is used to: + +- retrieve the device list for algorithm upgrade from the ATM to the IAS; +- send algorithm upgrade notifications from the ATM to the IAS. + +### 8.2.2 Reference point Ms: SMU – ATM + +Reference point Ms is between the SMU and the ATM. + +It is used to: + +- register or de-register and keepalive from the SMU to the ATM; +- send a training dataset creation command from the ATM to the SMU; +- send a sample annotation task creation command from the ATM to the SMU; +- send a sample annotation task status from the SMU to the ATM; +- send a resources usage status from the SMU to the ATM. + +It uses the hypertext transfer protocol (HTTP) [IETF RFC 9110] and [IETF RFC 9112] + JSON protocol. + +### 8.2.3 Reference point Ta: TMU – ATM + +Reference point Ta is between the TMU and the ATM. + +It is used to: + +- register or de-register and keepalive from the TMU to the ATM; +- send an algorithm-training task creation command from the ATM to the TMU; +- send an algorithm-training task status from the TMU to the ATM; + +– send an algorithm-training capability and resource usage status from the TMU to the ATM. +It uses the HTTP + JSON protocol. + +### **8.2.4 Reference point Tu: AMU – ATM** + +Reference point Tu is between the AMU and the ATM. + +It is used to: + +- register or de-register and keepalive from the AMU to the ATM; +- send a model warehouse to create and retrieve a command from the ATM to the AMU; +- send an algorithm deployment command from the ATM to the AMU; +- send an algorithm deployment status from the AMU to the ATM. + +It uses the HTTP + JSON protocol. + +### **8.2.5 Reference point Md: DSS – SMU** + +Reference point Md is between the DSS and the SMU. + +It is used to: + +- transmit video and image information from the DSS to the SMU through the HTTP + JSON protocol. + +### **8.2.6 Reference point Mm: MS – SMU** + +Reference point Mm is between the media server (MS) and the SMU. + +It is used to: + +- transmit media from the MS to the SMU through the real-time transport protocol (RTP) [IETF RFC 3550] or real-time transport control protocol (RTCP) [IETF RFC 4571]. + +### **8.2.7 Reference point Ts/Ts': SMU – TMU** + +Reference points Ts/Ts' are between the SMU and the TMU. + +Ts is used to: + +- transmit media from the SMU to the TMU through the RTP or RTCP. + +Ts' is used to: + +- transmit video and image information from the SMU to the TMU through HTTP + JSON protocol. + +### **8.2.8 Reference point Tm: TMU – AMU** + +Reference point Tm is between the TMU and the AMU. + +It is used to: + +- transmit a model from the TMU to the AMU through the HTTP. + +### **8.2.9 Reference point Mv: AMU – IVU** + +Reference point Mv is between the AMU and the IVU. + +It is used to: + +- transmit an algorithm from the AMU to the IVU through HTTP. + +### **8.2.10 Reference point Mi: AMU – IPU** + +Reference point Mi is between the AMU and the IPU. + +It is used to: + +- transmit an algorithm from the AMU to the IPU through HTTP. + +## Bibliography + +- [b-ISO/IEC 2382] ISO/IEC 2382:2015, *Information technology – Vocabulary*. +- [b-ISO/IEC 11557] ISO/IEC 11557:1992, *Information technology – 3,81 mm wide magnetic tape cartridge for information interchange – Helical scan recording – DDS-DC format using 60 m and 90 m length tapes*. +- [b-ISO/IEC 22989] ISO/IEC 22989:2022, *Information technology – Artificial intelligence – Artificial intelligence concepts and terminology*. +- [b-ISO/IEC TR 24029-1] ISO/IEC TR 24029-1:2021, *Artificial Intelligence (AI) – Assessment of the robustness of neural networks – Part 1: Overview*. +- [b-ISO/IEC TR 29119-11] ISO/IEC TR 29119-11:2020, *Software and systems engineering – Software testing – Part 11: Guidelines on the testing of AI-based systems*. +- [b-Wright] Wright, A., Andrews, H., Hutton, B., Dennis, G. (2022). *JSON schema: A media type for describing JSON documents*. Fremont, CA: Internet Engineering Task Force. Available [viewed 2023-03-20] at: + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.743.23-202310-I_PDF-E/raw.md b/marked/F/T-REC-F.743.23-202310-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..c9477f507ac24d0b53d57e129e96ba9117760237 --- /dev/null +++ b/marked/F/T-REC-F.743.23-202310-I_PDF-E/raw.md @@ -0,0 +1,543 @@ + + +# Recommendation **ITU-T F.743.23 (10/2023)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +## --- **Security requirements for video surveillance systems** + +![ITU logo](6ed175c791b5e156d9c98a8dbcc3318c_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.743.23 + +## Security requirements for video surveillance systems + +## Summary + +Recommendation ITU-T F.743.23 defines premises unit (PU) device security classification, functional requirements, typical use case and scenario for video surveillance systems. The Recommendation specifies the functional requirements, including PU access security requirements, transmission security requirements, platform security requirements, application security requirements, network security, and security management centre in video surveillance systems. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.743.23 | 2023-10-29 | 16 | 11.1002/1000/15618 | + +## Keywords + +Security requirements, video surveillance. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2023 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|--------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 PU device security classification ..... | 3 | +| 7 Functional requirements ..... | 3 | +| 7.1 PU access security requirements ..... | 3 | +| 7.2 Transmission security requirements ..... | 5 | +| 7.3 Platform security requirements..... | 6 | +| 7.4 Application security requirements..... | 7 | +| 7.5 Network security requirements..... | 7 | +| 7.6 Security management centre requirements..... | 8 | +| Appendix I – Typical use case and scenario ..... | 10 | +| I.1 PU authentication ..... | 10 | +| I.2 Security hardening of video surveillance systems..... | 10 | +| I.3 Detect and block malicious threats and send alerts ..... | 11 | + + + +# Recommendation ITU-T F.743.23 + +## Security requirements for video surveillance systems + +## 1 Scope + +This Recommendation specifies security requirements for video surveillance systems. + +The scope of this Recommendation includes: + +- Premises unit (PU) device security classification; +- PU access security requirements; +- Transmission security requirements; +- Platform security requirements; +- Application security requirements; +- Network security requirements; +- Security management centre requirements; +- Typical use case and scenario. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T H.626] Recommendation ITU-T H.626 (V2) (2019), *Architectural requirements for video surveillance system*. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 premises unit (PU)** [ITU-T H.626]: A device located at the remote part of a video surveillance system and used to capture multimedia information (such as audio, video, image, alarm signal, etc.) from a surveilled object. + +**3.1.2 video surveillance system** [ITU-T H.626]: A telecommunication service focusing on video (including audio and image) application technology, which is used to remotely capture multimedia (such as audio, video, image, alarm signal, etc.) and present them to the end user in a user-friendly manner, based on a managed broadband network with ensured quality, security and reliability. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 asset management module**: An asset management module deployed on the platform side unifies the management of security information of all access devices on the platform. + +**3.2.2 asset whitelist:** A list of premises unit (PU) devices allowed to access and be accessed by the platform is stored in the asset management module of the platform side, mainly containing device ID, device type, manufacturer, model, firmware version number, etc. + +**3.2.3 device fingerprint library:** A device fingerprint library is deployed in the asset management module on the platform side, which can uniquely identify the information about a device, including media access control address (MAC), device type, manufacturer, model, operating system, etc. + +**3.2.4 device address feature code:** A device address feature code stored in the asset management module of the platform side is formed by the device information such as the Internet protocol (IP) and media access control address (MAC) address of the device, which is used to bind with the device identity document (ID). + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|--------|--------------------------------------| +| DHCP | Dynamic Host Configuration Protocol | +| FTP | File Transfer Protocol | +| HTTP | Hypertext Transfer Protocol | +| HTTPS | Hypertext Transfer Protocol Secure | +| IP | Internet Protocol | +| IPC | Internet Protocol Camera | +| MAC | Media Access Control Address | +| NFS | Network File System | +| NVR | Network Video Recorder | +| POP3 | Post Office Protocol-Version 3 | +| PU | Premises Unit | +| QR | Quick Response | +| REST | Representational State Transfer | +| RTCP | Real-time Transport Control Protocol | +| RTP | Real-time Transport Protocol | +| RTSP | Real-time Streaming Protocol | +| SIP | Session Initiation Protocol | +| SMTP | Simple Mail Transfer Protocol | +| SSH | Secure Shell | +| TELNET | Teletype Network | + +## 5 Conventions + +In this Recommendation: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus this requirement needs not be present to claim conformance. + +- The keywords "can optionally" and "may" indicate an optional requirement which is permissible, without implying any sense of being recommended. These terms are not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +# 6 PU device security classification + +According to the strength of the security protection, the security capability of the premises unit (PU) device is divided into three levels, ranging from low to high: A, B and C. The security classification of the PU device is shown in Table 6-1. + +A-level devices are recommended to perform two-way authentication with the platform based on the digital certificates to ensure authentic identity. + +B-level devices are recommended to perform two-way authentication with the platform based on the digital certificates and sign the video data, which can ensure that authentic identity and video data come from the real device and can verify whether the video content has been tampered with. + +C-level devices are recommended to perform two-way authentication with the platform based on the digital certificates, sign and encrypt the video data, which can ensure that authentic identity and video data come from the real device and can verify whether the video content has been tampered with, and can encrypt and protect the video data. + +**Table 6-1 – PU device security classification** + +| Level | Perform two-way authentication with the platform based on digital certificates | Perform two-way authentication with the platform based on digital certificates and sign the video data | Perform two-way authentication with the platform based on digital certificates, sign and encrypt the video data | +|---------|--------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------------------------------| +| A-level | √ | - | - | +| B-level | √ | √ | - | +| C-level | √ | √ | √ | + +## 7 Functional requirements + +### 7.1 PU access security requirements + +PU contains a large amount of video image data and information. Due to the presence of numerous cameras and multiple network branches, there are significant risks associated with the security access control of cameras and other devices. This situation can easily lead to the abuse of network resources and information leakage. + +Network cameras are vulnerable to illegal attacks by hackers, and there is a risk of PU device hijacking in video surveillance. Hackers use hijacked PU devices (such as cameras) to launch DDoS attacks and can also use fake devices connected to a video cloud platform to attack internal systems. Therefore, it is recommended to enhance the system-level security of PU device for video surveillance through technical means, including access control, vulnerability assessment, trusted surveillance, equipment authentication, data encryption, and attack protection. + +#### 7.1.1 Access control + +Security access control achieves effective identification and security management of various types of PU device. It is recommended to support video protocol analysis and establish an [ITU-T H.626] + +protocol feature library and device fingerprint library. Only authorised devices are allowed access, and authorised devices are only allowed to transmit authorised data to ensure the security and control of the PU device access process. Furthermore, and prevent illegal private connections, fraudulent use, and timely alarm and make devices knowable and trusted in the network. + +The security access control policy is recommended to control access separately according to security requirements and terminal data transmission protocols. + +##### **Device access mechanism based on [ITU-T H.626] protocol analysis** + +For the device access mechanism based on [ITU-T H.626] protocol analysis, it is recommended to support identifying whether an Internet protocol camera (IPC) is accessed using [ITU-T H.626] protocol and extracting the device information of session initiation protocol (SIP) signalling protocol in [ITU-T H.626] protocol, including device ID, device type, manufacturer, model, and firmware version number. Then, it is recommended to identify whether the device information is in the platform-side asset whitelist. If not, it is to alert and block it. + +##### **Access mechanism based on device fingerprints collected by the secure access module** + +The secure access module includes forms such as a secure access gateway or secure plug-in to realise PU device access security and transmission security functions. The secure access gateway is deployed in hardware form behind devices (such as IPC) to realise secure authentication and encrypted transmission of terminals. The secure plug-in is loaded in software form into devices (such as IPC) to achieve secure authentication and encrypted transmission of terminals. + +For security IPC, a secure plug-in is recommended to be integrated in the IPC to generate a unique device fingerprint by collecting the MAC address, device type, manufacturer, model, and operating system of the PU device through the secure plug-in. + +For other IPC(s), a secure access gateway is recommended to be connected behind the other IPC to generate a unique device fingerprint by collecting the MAC address, device type, manufacturer, model, and operating system of the PU device through the secure access gateway. + +It is recommended to upload device fingerprint information to the platform-side asset management module for fingerprint identification of video surveillance device access. If the device fingerprint matches the information in the platform-side device fingerprint library, the PU device is allowed to access the video capability platform, and the device with the wrong fingerprint information should be blocked and alerted in real time. + +##### **Access mechanism based on password system** + +For high security demand scenarios, key information such as a digital ID book is recommended to be written in the IPC to achieve device access using a public key cryptosystem. + +##### **Access mechanism based on device address characteristics** + +For IPC or network video recorder (NVR), it is recommended that access control be performed by collecting device information, such as the Internet protocol (IP) and MAC address of the device, generating a device address feature code, and binding it with the device ID. At the same time, it is stored in the platform-side asset management module to realise device information registration. + +#### **7.1.2 Vulnerability assessment** + +It is recommended to support regular vulnerability security scan and weak password scan for PU device to master the security status of devices, while dynamically sorting out asset information. + +##### **Vulnerability security scan** + +Vulnerability security scan is recommended to perform a regular scanning schedule for on-net devices and generate device vulnerability scan reports, with the scope of the scan specified by device type + +and IP. The device vulnerability scan report includes information such as vulnerability name, risk level, discovery time, vulnerability port, and vulnerability description. It is recommended to support vulnerability security interface queries for on-net terminals and support retrieval by conditions, including terminal name, belonging group, IP address, MAC address, and terminal type. + +##### **Weak password detection** + +Weak password detection is recommended to perform a regular scanning schedule for on-net devices and generate device weak password scan reports, with the scope of the scan specified by device type and IP. The device weak password scan report includes information such as terminal name, belonging group, IP address, MAC address, and terminal type. It is recommended to support updating and configuring weak password tables. + +#### **7.1.3 Trusted surveillance** + +Based on trusted computing technology, the system's electronic fingerprint is established through a comprehensive collection of IPC system bootloader, operating system core files, and important configuration parameters to realise the security protection of the operating system. + +Once a breach of trustworthiness is detected, an alarm message will be sent to the management backend. + +#### **7.1.4 Equipment authentication** + +It is recommended to use a variety of technologies to implement device authentication functions such as password, device fingerprint, password technology, etc., to confirm the authenticity of the PU device identity. + +#### **7.1.5 Data encryption** + +Security IPC is recommended to support data encryption based on password technology through the built-in secure access module. + +Other IPC(s) are recommended to support the implementation of data transmission encryption protection through the secure access gateway. + +#### **7.1.6 Attack protection** + +For security IPC, it is recommended to implement a security protection function through the built-in secure access module to block network attacks and abnormal access behaviour against the PU device. + +For other IPC, it is recommended to implement security protection function through the secure access gateway to block network attacks and abnormal access behaviour against the PU device. + +### **7.2 Transmission security requirements** + +It is recommended to use encryption measures to ensure the confidentiality of data and prevent information attacks during data transmission. + +It is recommended to encrypt key video service data during transmission by deploying an SIP signalling gateway and video security gateway to prevent data eavesdropping. + +#### **7.2.1 Transmission encryption** + +Transmission encryption is applicable to security IPC and other IPCs. + +The security IPC is recommended to embed the secure access module. + +Other IPC(s) are recommended to add the secure access gateway. + +Both the secure access module and the secure access gateway are embedded with encryption modules to enable encryption of the video data collected by the IPC. + +The secure access module and the secure access gateway negotiate with the platform based on the key to establish transmission encryption channels (such as virtual private networks, etc.) to protect the integrity and confidentiality of video streams and signalling during transmission. + +#### **7.2.2 Digital signature** + +It is recommended to embed the secure access module on the security IPC or add the secure access gateway to other IPC(s) to provide two-way identity authentication and video data signature capabilities. + +### **7.3 Platform security requirements** + +Platform security is the foundation of business application security. + +It is recommended to provide access control capability of platform resources and formulate access control policies for tenants' IP addresses, target ports, access protocols and access rights. + +It is recommended to use security hardening techniques to protect the platform environment. + +It is recommended to provide security protection capability for tenants on public networks. + +Platform security includes virtual network security, host-network security, storage security, identity security, and video content filtering. + +#### **7.3.1 Virtual network security** + +It is recommended to control the network traffic of the platform for security access, build an isolated virtual network environment, and timely detect and block malicious threats in the network. + +#### **7.3.2 Host-network security** + +The virtual machine agent security plug-in for host protection is recommended to detect traffic between virtual machines, performance status of virtual machines, and risk status. + +It is recommended to detect and alert on the failure of resource isolation between virtual machines for intrusions on critical nodes, and to detect and alert on propagating malicious code infections. + +#### **7.3.3 Storage security** + +Mechanisms such as multi-copy redundancy, data consistency assurance, and virtualisation isolation are recommended to ensure secure and reliable cloud storage. + +It is recommended to ensure that memory and storage space are fully cleared when reclaimed. + +When a tenant deletes business application data, it is recommended that the platform remove all copies from the cloud storage. + +#### **7.3.4 Identity security** + +When managing devices in the platform remotely, it is recommended to establish a two-way authentication mechanism between the management terminal and the platform. + +When a cloud tenant logs into the platform, it is recommended that a two-way authentication mechanism be established between the tenant and the platform. + +It is recommended to unify the management of privileged accounts such as platform operation and maintenance management to reduce the risk of privileged accounts being abused. + +#### **7.3.5 Video content filtering** + +It is recommended to provide the ability to detect viruses, remote control Trojans, attack software packets, etc. that are entrapped in the network traffic of video transmission. + +It is recommended to transmit video data according to the pre-registered video data format. It is recommended to support real-time analysis and filtering of transmitted video data, blocking and alarming of video data that does not conform to the format. + +### **7.4 Application security requirements** + +The platform carries various cloud applications. It is recommended to strictly control access to the application to prevent illegal theft of video resources. + +Application security requirements include video leak proof and authorisation management. + +#### **7.4.1 Video leak proof** + +It is recommended to support the encryption of sensitive video data during video access. + +It is recommended to implement features such as user authentication access, video data encryption, watermark loading (text watermark, quick response (QR) code watermark, dot matrix watermark, etc.), and screenshot prohibition. + +##### **Automatic encryption** + +When the surveillance video is downloaded locally, it is recommended to automatically encrypt it to ensure that it can only be opened on the downloaded computer. + +##### **Screenshot control** + +It is recommended to block all kinds of screenshot software and screen recording software to avoid taking screenshots and recording screens while playing surveillance video. + +##### **Watermark loading** + +It is recommended to provide a watermark loading function, including an invisible dot matrix watermark, visible text watermark, and visible QR code watermark, which can be loaded with self-defined text description, date, time, source IP, MAC address and username to avoid screen recording and ensure the traceability of surveillance video data. + +#### **7.4.2 Authorisation management** + +Authorisation management includes examining and approving surveillance video data and surveillance video data control. + +#### **Examine and approve surveillance video data** + +When someone wants to access surveillance video, it is recommended to support the management and recording of the surveillance video access process, including request, approval, and authorisation management. + +##### **Surveillance video data control** + +When someone is authorised to access surveillance video, it is recommended to restrain the video visitor by setting a password, limiting the access time and the number of times, and loading a watermark on the surveillance video to ensure the security of the surveillance video. + +### **7.5 Network security requirements** + +Network security recommends communicating through authorised interfaces provided by edge devices. + +It is recommended that only authorised devices can connect to the internal network, and that internal users can only connect to the external network after authorisation. + +Network security requirements include access control and network protection. + +#### 7.5.1 Access control + +It is recommended to monitor various access operations occurring on the network and the data packets transmitted on the network. + +It is recommended to analyse protocol behaviour and network traffic data to discover security risks and perform timely access control and alerts according to security policies. + +It is recommended to identify and control the application-layer protocol to achieve malicious code protection and application attack protection. The specific protocol control requirements are as follows: + +- It is recommended to only allow SIP, real-time transport protocol (RTP), real-time streaming protocol (RTSP), real-time transport control protocol (RTCP) and hypertext transfer protocol secure (HTTPS) protocols access, filter the signalling and parameter keywords of these protocols, and only allow three business services and ports: SIP signalling control, RTP and RTSP media transmission, and Representational state transfer (REST) intelligent analysis data transmission. +- It is recommended to prohibit hypertext transfer protocol (HTTP), simple mail transfer protocol (SMTP), file transfer protocol (FTP), teletype network (TELNET), secure shell (SSH), network file system (NFS), post office protocol-version 3 (POP3), dynamic host configuration protocol (DHCP), and other unnecessary protocols access. + +#### 7.5.2 Network protection + +It is recommended to provide the application layer intrusion prevention function to analyse messages from header to content and match them with pre-defined attack characteristics to discover and block hidden attack messages in the data flow. At the same time, it is recommended to analyse the interaction behaviour of protocols and discover abnormal network protocol messages to detect attack behaviour. When a security threat is detected, it can respond in real time and defend automatically. + +It is recommended to monitor attacks at the application layer of the platform including port scanning, trojan horse attacks, denial of service attacks, buffer overflow attacks, IP fragmentation attacks and network worms. + +When an attack is detected, it is recommended to record the attack source IP, attack type, attack purpose, attack time, and provide an alert in case of a serious intrusion. + +### 7.6 Security management centre requirements + +It is recommended to monitor the status of video surveillance devices in real time in order to detect abnormal devices and provide early warning. + +It is recommended to unify the security management of the platform and provide security situational awareness capability. + +#### 7.6.1 Asset management + +It is recommended to collect PU device information through active scanning, passive listening, manual setting, establish a device fingerprint database (mainly containing an IP address, MAC address, device type, manufacturer, model, hostname, operating system, etc.) and conduct classification statistics to realise unified asset management of devices. + +#### 7.6.2 Key infrastructure + +For high security level scenarios, it is recommended to provide full life-cycle management of encryption and decryption keys (including symmetric keys, asymmetric keys, digital certificates, etc.). + +It is recommended to provide encryption capability for cloud platform and application data. + +#### **7.6.3 Security situational awareness** + +It is recommended to dynamically integrate information on platform assets, attacks, threats and vulnerabilities, and analyse them using big data technology to grasp the overall security status of the platform. + +It is recommended that security situation reports be generated automatically on a regular basis to judge and warn of current network and future threats. The security situation report includes statistics of relevant data, including the number of security vulnerabilities, attack IP, attacked assets, attacked domain names, attack events, attack threat levels, etc. + +## Appendix I + +## Typical use case and scenario + +(This appendix does not form an integral part of this Recommendation.) + +### I.1 PU authentication + +- Step 1: Bob is the administrator of the video surveillance platform at X museum. He logs into the video surveillance platform using his account and password in the X museum's dedicated virtual network environment. He can see all the devices currently connected to the platform in the asset management module of the platform. The asset management module contains an asset whitelist and device details, etc. The X museum video surveillance platform is pre-set to allow access only to IPC with [ITU-T H.626] protocol. +- Step 2: Alice wants to connect the A-level IPC at the new point of the X museum to the video surveillance platform. She requests Bob to add the IPC to the platform asset whitelist and provides Bob with the IPC's device ID, device type, manufacturer, model, firmware version number, etc. +- Step 3: After Bob adds the IPC to the platform asset whitelist, Alice connects the A-level IPC into the platform in the virtual network environment dedicated to the X museum. The video surveillance platform first identifies whether the IPC uses the [ITU-T H.626] protocol to access, and after successful identification extracts the device information in the [ITU-T H.626] standard SIP signalling protocol, including device ID, device type, manufacturer, model number, and firmware version number, and matches this information with the platform side asset whitelist. After successful matching, the IPC is successfully connected to the X museum video surveillance platform. + +Figure I.1 illustrates this scenario. + +![Diagram illustrating the PU authentication scenario. A dashed box labeled 'Museum X' contains 'The video surveillance platform'. Inside the platform box, there is an 'Asset management module' which contains an 'Asset whitelist'. Outside the platform box, 'Bob' (represented by a person icon) is connected to the 'Asset management module' via a dashed double-headed arrow. 'Alice' (represented by a person icon) is connected to 'Bob' via a dashed double-headed arrow. 'Alice' is also connected to an 'IPC' (represented by a circle icon) via a solid arrow. The 'IPC' is connected to 'The video surveillance platform' via a solid arrow.](41a438d7e4adc17c3a4005e7c9500091_img.jpg) + +The diagram shows a 'Museum X' environment enclosed in a dashed box. Inside, 'The video surveillance platform' contains an 'Asset management module' with an 'Asset whitelist'. Outside the platform but within the museum environment, 'Bob' (administrator) and 'Alice' (user) are shown. Bob has a dashed bidirectional connection to the Asset management module. Alice has a dashed bidirectional connection to Bob and a solid arrow pointing to an 'IPC' (Internet Protocol Camera). The IPC has a solid arrow pointing to 'The video surveillance platform'. + +Diagram illustrating the PU authentication scenario. A dashed box labeled 'Museum X' contains 'The video surveillance platform'. Inside the platform box, there is an 'Asset management module' which contains an 'Asset whitelist'. Outside the platform box, 'Bob' (represented by a person icon) is connected to the 'Asset management module' via a dashed double-headed arrow. 'Alice' (represented by a person icon) is connected to 'Bob' via a dashed double-headed arrow. 'Alice' is also connected to an 'IPC' (represented by a circle icon) via a solid arrow. The 'IPC' is connected to 'The video surveillance platform' via a solid arrow. + +F.743.23(23) + +**Figure I.1 – PU authentication** + +### I.2 Security hardening of video surveillance systems + +- Step 1: Company M is a leading company in the industry. Company M is at location G and the factory is at location H. Company M uses the video surveillance system to view the operation of the factory in real time. Since the factory involves M company's commercial secrets, M company proposes to strengthen the security of the video surveillance system. +- Step 2: Company M first reinforces the security of the IPCs in the factory. The security capability of the current IPCs in the factory are all B-level. For security IPC, secure access module is added, and for other IPC(s), a secure access gateway is installed to block network attacks and abnormal access behaviours against the PU devices and to realise the security protection function for the IPCs. Among them, the secure access module and secure access gateway are + +embedded with encryption modules to realise the encryption of the video data collected by the IPCs, making the security capability of the factory's IPCs reach the C-level. The secure access module and secure access gateway negotiate with the platform based on keys to establish transmission encryption channels (such as VPN, etc.) to protect the integrity and confidentiality of IPC video streams and signalling in the transmission process. Besides, company M encrypts the IPC's video data during transmission by deploying an SIP signalling gateway and video security gateway to prevent the data from being eavesdropped. + +Step 3: Company M conducts security access control for the network traffic of the video monitoring platform and builds an isolated virtual network environment. At the same time, company M sets multiple platform administrators for the video surveillance platform and divides the authority, and each administrator is responsible for an area in the factory. All platform administrators have to log into the platform in the virtual network environment and verify their identity with the platform. After successful verification, the administrators can only view and manage the devices and video files within their authority, and company M manages these platform management accounts uniformly to reduce the risk of abuse of privileged accounts. + +Figure I.2 illustrates this scenario. + +![Diagram illustrating the security hardening of video surveillance systems. The diagram shows a 'Factory' environment containing two areas, 'Area 1' and 'Area 2'. Each area has a 'Security IPC' and an 'Other IPC'. These IPCs connect to 'Secure access module' and 'Secure access gateway' units. These units then connect to 'Encrypt' units at the top and bottom. The 'Encrypt' units connect to 'The video surveillance system' located within a 'Company M' environment, which also contains two user icons representing administrators.](8307f6b04df072c9332f9987e034272c_img.jpg) + +The diagram illustrates the security architecture for video surveillance. On the left, a large circle labeled 'Factory' contains two dashed rectangular boxes labeled 'Area 1' and 'Area 2'. Each area contains two ovals: 'Security IPC' and 'Other IPC'. Arrows from these IPCs point to a 'Secure access module' and a 'Secure access gateway' located at the bottom of the factory circle. From these modules, arrows point to 'Encrypt' ovals at the top and bottom. These 'Encrypt' units are connected by a horizontal line to 'The video surveillance system' on the right. The surveillance system is enclosed in a dashed rectangle labeled 'Company M' and has two person icons (administrators) pointing to it. Arrows also connect the 'Secure access module' and 'Secure access gateway' directly to 'The video surveillance system'. + +Diagram illustrating the security hardening of video surveillance systems. The diagram shows a 'Factory' environment containing two areas, 'Area 1' and 'Area 2'. Each area has a 'Security IPC' and an 'Other IPC'. These IPCs connect to 'Secure access module' and 'Secure access gateway' units. These units then connect to 'Encrypt' units at the top and bottom. The 'Encrypt' units connect to 'The video surveillance system' located within a 'Company M' environment, which also contains two user icons representing administrators. + +F.743.23(23) + +**Figure I.2 – Security hardening of video surveillance systems** + +### I.3 Detect and block malicious threats and send alerts + +Step 1: Company E provides a common video surveillance management platform called the Q platform. Company A and company B are tenants of the Q platform. Company A and company B connect each company's IPCs to the Q platform for management. Company A and company B are independent of each other on the Q platform. Company E can only see tenant information on the Q platform but cannot view and manage customer's devices and video files. + +Step 2: John is the platform administrator of company A and Smith is the platform administrator of company B. John and Smith log into the Q platform in their respective companies' isolated virtual network environment. They can only see and manage their own company's devices and video files on the Q platform. + +- Step 3: Q platform provides periodic vulnerability security scan and weak password detection for all devices accessing the platform, specifies the scanning scope by terminal type and IP address, and generates a terminal vulnerability scan report and terminal weak password scan report. The content of the terminal vulnerability scan report includes information such as vulnerability name, risk level, discovery time, vulnerability port, and vulnerability description. The terminal weak password scan report includes the terminal name, belonging group, IP address, MAC address, terminal type, and weak password type. +- Step 4: One day, Q platform performed regular security scans and found that a device in company A had a weak password, and immediately generated an alert message and a weak password scan report and sent it to John. John received the alert message and rectified the device based on the device information in the weak password scan report, instantly changing the default password of the device and securing the device. +- Step 5: One day, Q platform performed a regular security scan and found that company B's device had a remote code executable vulnerability, and immediately generated an alert message and vulnerability scan report and sent it to Smith. After receiving the alert message, Smith immediately analysed the vulnerability and initiated a vulnerability fix to ensure the security of the device. + +Figure I.3 illustrates this scenario. + +![Diagram illustrating the detection and blocking of malicious threats and sending alerts via the Q platform.](dfe556fea00682b09a59427aaf72051c_img.jpg) + +The diagram illustrates a security management scenario. At the top, a dashed box labeled 'Company E' contains the 'Q platform'. Above the platform are two ovals labeled 'Weak password detection' and 'Vulnerability scanning'. Below the platform, a box contains 'Tenant 1', 'Tenant 2', and 'Other tenants ...', each represented by a building icon. Below this, two dashed boxes represent 'Company A' and 'Company B'. 'Company A' contains a person icon labeled 'John' and a circle of five devices, each labeled 'C'. 'Company B' contains a person icon labeled 'Smith' and a circle of five devices, each labeled 'C'. Arrows show the flow of information: from the 'Q platform' to 'John' and 'Smith', and from 'John' and 'Smith' back to the 'Q platform'. Yellow arrows point from the 'Q platform' to the devices in 'Company A' and 'Company B'. + +Diagram illustrating the detection and blocking of malicious threats and sending alerts via the Q platform. + +**Figure I.3 – Detect and block malicious threats and send alerts** + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.743.24-202402-I_PDF-E/raw.md b/marked/F/T-REC-F.743.24-202402-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..2b9be3aaf039bfb9628889cd7fe6fdec6a248a42 --- /dev/null +++ b/marked/F/T-REC-F.743.24-202402-I_PDF-E/raw.md @@ -0,0 +1,686 @@ + + +# Recommendation **ITU-T F.743.24 (02/2024)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +## **Scenarios and requirements for distributed ledger technology (DLT) in video surveillance system interworking** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU), featuring the letters 'ITU' in blue inside a circle with a globe-like pattern. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.743.24 + +# Scenarios and requirements for distributed ledger technology (DLT) in video surveillance system interworking + +## Summary + +Recommendation ITU-T F.743.24 provides an overview of video surveillance system interworking (VSSI) based on distributed ledger technology (DLT), for which it specifies application scenarios and capability requirements to realize interoperability, high reliability and high efficiency of VSSI identity authentication and authorization. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.743.24 | 2024-02-13 | 16 | 11.1002/1000/15846 | + +## Keywords + +DLT, requirements, scenarios, video surveillance. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2024 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|--------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------|----| +| 1 | Scope..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation ..... | 2 | +| 4 | Abbreviations and acronyms ..... | 2 | +| 5 | Convention..... | 2 | +| 6 | Overview of DLT in video surveillance system interworking ..... | 3 | +| 7 | General framework of DLT in video surveillance system interworking ..... | 4 | +| 7.1 | Ecosystem and roles ..... | 4 | +| 7.2 | General framework of DLT in video surveillance system interworking..... | 4 | +| 7.3 | General flow of DLT in video surveillance system interworking..... | 5 | +| 8 | Requirements for DLT in video surveillance system interworking..... | 6 | +| 8.1 | Basic requirements for DLT in video surveillance system interworking..... | 6 | +| 8.2 | Advanced video surveillance system interworking based on DLT ..... | 7 | +| Appendix I – Scenarios and use cases of video surveillance system interworking based on DLT..... | | 9 | +| I.1 | Registration in VSSI..... | 9 | +| I.2 | Authentication and authorization ..... | 10 | +| I.3 | Request the surveillance video data in the cloud..... | 11 | +| I.4 | Video resource upload to the DLT and verification ..... | 12 | +| I.5 | Video data resources sharing..... | 13 | +| Bibliography..... | | 15 | + + + +# Recommendation ITU-T F.743.24 + +## Scenarios and requirements for distributed ledger technology (DLT) in video surveillance system interworking + +# 1 Scope + +This Recommendation specifies application scenarios and requirements for distributed ledger technology (DLT) in video surveillance system interworking (VSSI), based on the requirements and architectures specified in [ITU-T F.743], [ITU-T H.626] and [ITU-T H.626.1], which includes: + +- 1) overview; +- 2) scenarios, including registration, authentication and authorization, request and sharing; +- 3) requirements, including basic and advanced. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +- [ITU-T F.743] Recommendation ITU-T F.743 (2019), *Requirements and service description for video surveillance.* +- [ITU-T F.751.0] Recommendation ITU-T F.751.0 (2020), *Requirements for distributed ledger systems.* +- [ITU-T F.751.2] Recommendation ITU-T F.751.2 (2020), *Reference framework for distributed ledger technologies.* +- [ITU-T H.626] Recommendation ITU-T H.626 (2019), *Architectural requirements for video surveillance system.* +- [ITU-T H.626.1] Recommendation ITU-T H.626.1 (2013), *Architecture for mobile visual surveillance.* + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +- 3.1.1 customer unit** [ITU-T F.743]: A device located at the customer part of a video surveillance system and used to present multimedia information (such as audio, video, image, alarm signal, etc.) to the end user. +- 3.1.2 distributed ledger** [b-ITU-T X.1400]: A type of ledger that is shared, replicated, and synchronized in a distributed and decentralized manner. +- 3.1.3 distributed ledger technology (DLT)** [b-ITU-T X.1400]: Technology that enables the operation and use of distributed ledgers. + +**3.1.4 premises unit** [ITU-T F.743]: A device located at the remote part of a video surveillance system and used to capture multimedia information (such as audio, video, image, alarm signal, etc.) from a surveilled object. + +**3.1.5 video surveillance system** [ITU-T H.626]: A telecommunication service focusing on video (including audio and image) application technology, which is used to remotely capture multimedia (such as audio, video, image, alarm signals, etc.) and present them to the end user in a user-friendly manner, based on a managed broadband network with ensured quality, security and reliability. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following term: + +**3.2.1 video surveillance system interworking platform**: A series of devices and subsystems located at the centred part of the interworking video surveillance systems. The platform is used to integrate the capabilities of different video surveillance systems and enable the resource and data sharing of different video surveillance systems. + +NOTE – Definition based on [b-ITU-T F.743.3]; "video" replaces "visual". + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-------|-----------------------------------------------| +| BVSSI | DLT in Video Surveillance System Interworking | +| CU | Customer Unit | +| DLT | Distributed Ledger Technology | +| ID | Identifier | +| IPU | Intelligent Premises Unit | +| ISP | Internet Service Provider | +| MCU | Mobile Customer Unit | +| MSU | Media Storage Unit | +| PU | Premises Unit | +| SC | Smart Contract | +| SHA | Secure Hash Algorithm | +| VS | Video Surveillance | +| VSS | Video Surveillance System | +| VSSI | Video Surveillance System Interworking | + +# 5 Convention + +In this Recommendation: + +- The phrase "is required" indicates a requirement that must be strictly followed and from which no deviation is permitted, if conformance to this Recommendation is to be claimed. +- The phrase "is recommended" indicates a requirement that is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. +- The phrase "can optionally" indicates an optional requirement that is permissible, without implying any sense of being recommended. This term is not intended to imply that the vendor's implementation must provide the option, and the feature can be optionally enabled + +by the network operator or service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with this Recommendation. + +# 6 Overview of DLT in video surveillance system interworking + +This Recommendation specifies scenarios and requirements for DLT [ITU-T F.751.0] in VSSI. VSSI is a new model for enabling users to have ubiquitous, convenient and on-demand network access to a shared pool of configurable storage resources, which can be rapidly provisioned and released with minimal management effort or service-provider interaction. + +DLT enables operation and use by providing a verifiable, append-only chained data structure of transactions, through allowing the data to be stored and updated in distributed way. DLT is essentially a decentralized architecture that no longer relies on a centralized authority. The transactions are approved by nodes and recorded in time-stamped blocks, where each block is identified by a cryptographic hash and chained to preceding blocks in chronological order. In some ways, DLT can establish an irrefutable database (DB), which uses a consensus mechanism to maintain the integrity, security, tamper-proofing and peep-proofing of the data recorded on the blocks. + +In video surveillance systems (VSSs), three key components work together: the VSS itself; DLT; and cloud storage. DLT plays a crucial role in identifying users, hardware equipment and multimedia data (like videos, audio and images) within the VSS. It accomplishes this through a network of nodes equipped with smart contracts (SCs) [ITU-T F.751.2]. Whenever a new entity (e.g., customer unit (CU), premises unit (PU)) joins the VSS and generates multimedia content, it receives a unique, globally recognizable identifier (ID), as shown in Figure 1. Such functions as ID management, identity authentication, rights management and video data authenticity verification are realized through SCs on the ID chain, which is the DLT system in VSSI. Based on a unified global ID, interoperability of VSSs across systems and operators can be realized. + +![Figure 1: DLT in video surveillance system interworking. The diagram illustrates the architecture of a video surveillance system (VSS) interconnected with a Distributed Ledger Technology (DLT) network. On the left, three VSS units (VSS 1, VSS 2, VSS 3) are shown, each containing components like MSU (Media Storage Unit), PU (Premises Unit), IPU (Intelligent Premises Unit), CU (Customer Unit), and MCU (Media Control Unit). These units are connected to a central DLT layer (represented by a green bar with a cloud icon). Above the DLT layer, VSS 1 and VSS 2 are connected to 'Cloud storage' (represented by a green cloud icon) via dashed arrows. Below the DLT layer, VSS 3 is connected to another 'Cloud storage' (represented by an orange cloud icon) via dashed arrows. The DLT layer is connected to a network of nodes (NodeA, NodeB, NodeC, etc.) arranged in a circular fashion. Each node is associated with a Smart Contract (SC) represented by a hexagon. The nodes are connected to different Internet Service Providers (ISPs): ISP_A (NodeA), ISP_B (NodeB), and ISP_C (NodeC). The diagram shows the flow of data and transactions from the VSS units through the DLT layer to the nodes and their smart contracts, and finally to the cloud storage.](367926125450c2bc3f4bdca9d59a62ba_img.jpg) + +IPU: intelligent premises unit; MSU: media storage unit + +Figure 1: DLT in video surveillance system interworking. The diagram illustrates the architecture of a video surveillance system (VSS) interconnected with a Distributed Ledger Technology (DLT) network. On the left, three VSS units (VSS 1, VSS 2, VSS 3) are shown, each containing components like MSU (Media Storage Unit), PU (Premises Unit), IPU (Intelligent Premises Unit), CU (Customer Unit), and MCU (Media Control Unit). These units are connected to a central DLT layer (represented by a green bar with a cloud icon). Above the DLT layer, VSS 1 and VSS 2 are connected to 'Cloud storage' (represented by a green cloud icon) via dashed arrows. Below the DLT layer, VSS 3 is connected to another 'Cloud storage' (represented by an orange cloud icon) via dashed arrows. The DLT layer is connected to a network of nodes (NodeA, NodeB, NodeC, etc.) arranged in a circular fashion. Each node is associated with a Smart Contract (SC) represented by a hexagon. The nodes are connected to different Internet Service Providers (ISPs): ISP\_A (NodeA), ISP\_B (NodeB), and ISP\_C (NodeC). The diagram shows the flow of data and transactions from the VSS units through the DLT layer to the nodes and their smart contracts, and finally to the cloud storage. + +**Figure 1 – DLT in video surveillance system interworking** + +# 7 General framework of DLT in video surveillance system interworking + +## 7.1 Ecosystem and roles + +DLT in video surveillance system interworking (BVSSI) is the system that supports users of their VSS and devices of their VSS to share the video resource for others as business demands. + +This clause describes a VSSI environment, with roles and sub-roles. It also specifies necessary activities for roles performed based on DLT. + +The BVSSI environment includes the following roles. + +- Video resource provider: Provide a video surveillance (VS) resource, whose main activities include resource identity, resource debugging, making sharing regular and an open-resource application programming interface. +- Video resource user: The end user or platform to obtain the resource or results, or services from a provider. The activities include requesting and obtaining the resource. +- VS platform: A PU or CU terminal and DB of user and provider. The activities include equipment registration, and resource authentication and authorization. + +The BVSSI ecosystem includes the following sub-role. + +- DLT infrastructure provider: DLT infrastructure construction, including consensus algorithm design and SCs. The main activities include node admittance, node classification, anomaly detection and node status monitoring. + +## 7.2 General framework of DLT in video surveillance system interworking + +The general framework of DLT in VSSI involves multiple roles, which work together to promote VSSI based on DLT, as shown in Figure 2. + +In order interactively to share video data resources, a provider needs to classify them into general and privacy. Resources are then shared with a data requester based on DLT infrastructure, In which process, it is required that privacy data be stored in a local DB by the provider to sufficiently protect information from disclosure, a shared certificate into DLT infrastructure to trade be recorded, and general data be stored in the DLT infrastructure for sharing, which could involve verification of user information, including data provider and requester, as well as the user VSS device. + +![Figure 2 – General framework of DLT in video surveillance system interworking. The diagram illustrates the interaction between two Video Surveillance Systems (VSS 1 and VSS 2) and a central DLT infrastructure. VSS 1 (Data requester) and VSS 2 (Data provider) each contain a Multimedia data platform, a PU (Person User), a CU (Computer User), and a local DB. VSS 1 sends 'Data upload/data access' to the DLT infrastructure, which in turn sends 'Data upload/data access' to VSS 2. The DLT infrastructure performs 'Resource authentication and authorization' and facilitates 'General video data resources sharing' and 'Private video data resources sharing' between the two systems. VSS 2 also contains a 'Public - private key pair and device ID [K_PU, K_PU, ID_PU]'.](b10763be31553f31cbb795653b731a43_img.jpg) + +The diagram illustrates the general framework of DLT in video surveillance system interworking. It shows two Video Surveillance Systems (VSS 1 and VSS 2) interacting with a central DLT infrastructure. VSS 1 is labeled as 'Data requester' and VSS 2 as 'Data provider'. Each VSS contains a 'Multimedia data platform', a 'PU' (Person User), a 'CU' (Computer User), and a 'local DB'. VSS 1 sends 'Data upload/data access' to the DLT infrastructure, which in turn sends 'Data upload/data access' to VSS 2. The DLT infrastructure performs 'Resource authentication and authorization' and facilitates 'General video data resources sharing' and 'Private video data resources sharing' between the two systems. VSS 2 also contains a 'Public - private key pair and device ID [K\_PU, K\_PU, ID\_PU]'. The diagram is labeled 'F.743.24(24)' in the bottom right corner. + +Figure 2 – General framework of DLT in video surveillance system interworking. The diagram illustrates the interaction between two Video Surveillance Systems (VSS 1 and VSS 2) and a central DLT infrastructure. VSS 1 (Data requester) and VSS 2 (Data provider) each contain a Multimedia data platform, a PU (Person User), a CU (Computer User), and a local DB. VSS 1 sends 'Data upload/data access' to the DLT infrastructure, which in turn sends 'Data upload/data access' to VSS 2. The DLT infrastructure performs 'Resource authentication and authorization' and facilitates 'General video data resources sharing' and 'Private video data resources sharing' between the two systems. VSS 2 also contains a 'Public - private key pair and device ID [K\_PU, K\_PU, ID\_PU]'. + +Figure 2 – General framework of DLT in video surveillance system interworking + +## **7.3 General flow of DLT in video surveillance system interworking** + +Based on the general framework, the general flow of DLT in VSSI includes five steps: registration in VSSI; authentication and authorization; request for surveillance video data; video resource upload to the DLT and verification; and video data resource sharing. + +### **7.3.1 Registration in VSSI** + +Users register their VSS and associated devices in the DLT to prepare for BVSSI. The registration includes registration of user and equipment, the typical detailed flow is described in clause I.1. + +- User registration: users register their information in the DLT infrastructure, generating user identification, user public key and user private key. Considering the different capabilities, users can be divided into providers and requesters of data. +- Equipment registration: users register their VSS devices, generating their identification and public and private keys. + +### **7.3.2 Authentication and authorization** + +A data provider receives an encryption authorization request message sent by a data requester. The encryption authorization request message consists of information about the device identification, as well as the public and private device keys of the data requester, who has encrypted it using the public key of the data provider. + +The data provider uses its device private key to decrypt the message and get the device identification and device public key of data requester, and then authenticates the information by DLT. + +The typical detailed flow is described in clause I.2. + +### **7.3.3 Request the surveillance video data** + +After authentication and authorization by DLT, the data requester requests the resource; the data provider sends its public and private device keys to the data requester to access VSS. + +The typical detailed flow is described in clause I.3. + +### **7.3.4 Video data resource upload to the DLT and verification** + +The data provider divides video data into general and privacy. Privacy data is stored in a local DB and general data is uploaded to the DLT infrastructure. + +The system generates the video resource ID and resource hash information, uploads them to the DLT infrastructure and verifies the uploaded information. + +The typical detailed flow is described in clause I.4. + +### **7.3.5 Video data resource sharing** + +Sharing video data resources has two different scenarios: general and private. + +- In the general video data resources sharing scenario, the data owner uploads the video to cloud storage and sets the SC for access rules, and the data requester obtains video resources from the cloud storage via DLT. +- The private video data resources sharing scenario, which is suitable for extremely confidential data, is not suitable for uploading video to cloud storage. Due to their high privacy, the data owner stores video data in a local private DB. When there is a sharing demand, the data owner actively sends out a private transaction request for the specified target object. The result of the encryption of the private data includes an encrypted file and hash value; hash value is used to verify whether the encrypted file has been tampered with by the data requester. + +The typical detailed flow is described in clause I.5. + +# **8 Requirements for DLT in video surveillance system interworking** + +## **8.1 Basic requirements for DLT in video surveillance system interworking** + +#### **8.1.1 User identification and authentication requirements** + +Req UIAR-1: It is required to provide user identity management and authentication capabilities within VSSI to ensure mutual access and resource sharing among different users. + +Req UIAR-2: It is required to provide a capability of identity authentication and management, including generating public-private key pairs for users. + +### **8.1.2 System requirements** + +Req SyR-1: It is required to provide a capability to VSS for publishing network resources, including video data resource. + +Req SyR-2: It is required to provide a capability to write or read data from ledger precisely used for resource registration and history traceability. + +Req SyR-3: It is recommended to provide the ability to repair faulty nodes and realize the reconstruction of shared resource services. + +Req SyR-4: It is recommended to provide a node redundancy strategy to ensure DLT node availability. + +Req SyR-5: It is required to provide a capability of node control that can detect bad DLT nodes in time and disable them. + +Req SyR-6: It is recommended to provide a capability of an effective transmission of video data resource to ensure the upload and download of video data resources between VSSIs. + +Req SyR-7: It is required to provide a capability of VS equipment management and record VS equipment information in the VSS, such as hardware serial numbers and device IDs. + +Req SyR-8: It is required to provide a capability of reliable public-private key pair management to achieve the identity authentication necessary for interaction with different VSSs. + +Req SyR-9: It is required to provide a secure device interaction interface through which the platform can ensure the management of different PUs in the VSS and the scheduling of VS resource. + +Req SyR-10: It is required to provide SC security audits and keep audit records. + +Req SyR-11: It is required to provide a capability to manage sharing resources information and VSS information, including resources requirements, resources status and VSS registration status. + +### **8.1.3 Service requirements** + +Req SeR-1: It is required to provide the video data resource catalogue to the video data resource requester. + +Req SeR-2: It is required to provide the ability to manage identity and authentication using SCs to reduce the time cost for users to obtain authentication. The system can satisfy the permission of the user to obtain encrypted video data through the simplest process. + +Req SeR-3: It is required to provide the capability to record video monitoring resources, including video resource uploading and sharing, and record the video resource sharing process. + +Req SeR-4: It is required to provide the capability to manage SCs and maintain the life cycle of SCs for resource sharing permission authorization to share video data resources among different VSSs. + +Req SeR-5: It is required to provide that SC version control be provided to ensure that the version of the SC invoked in the video data resource sharing transaction is explicit. + +Req SeR-6: It is required to provide distinct VS resource sharing conditions to data requesters. + +Req SeR-7: It is required to provide the capability to generate the temporary public-private key pair for VSS user authorization safely, and provide temporary permissions for the interaction between CUs and PUs of different VSSs. The data owner has the right to manage the sharing rights of the video data resources generated by the PU of its own VSS, including the setting of conditions such as usage period, usable range and permission. + +Req SeR-8: It is required to provide the transaction desensitization mechanism, realize the privacy protection of video data resource sharing transaction, reduce the density of transaction information and prevent illegal attackers from using transaction information for cluster analysis. + +Req SeR-9: It is required to provide the capability to ensure the interaction between CUs and PUs between different VSSs, such as being able to implement the data sharing request task of requester to a PU under VSS of the provider according to the temporary public-private key pair of device authorization. + +Req SeR-10: It is required to provide a general transmission regulation for VS resource sharing between different VSSs. + +Req SeR-11: It is required to provide a secure resource monitoring and resource encryption mechanism to ensure the data protection and security of video data resources. + +Req SeR-12: It is required to provide a capability to support the general video data resources sharing, and the shared data needs to be encrypted. + +Req SeR-13: It is required to provide the ability to share private video data that is stored in the local private DB. The transaction request needs to include the target node ID, encrypted video data and its hash, and private transaction parameters. The private transaction parameter identifies the sharing transaction as a private transaction, and specifies the relevant node that can participate in the private video data sharing transaction. + +Req SeR-14: It is required to provide the ability to share private data, and only the relevant nodes in the private transaction parameters can participate in the transaction. + +Req SeR-15: It is required to provide the ability to identify whether the shared data has been tampered with. It can be judged by comparing whether the data hash in the node is consistent with that stored in the DLT. + +## **8.2 Advanced video surveillance system interworking based on DLT** + +Req ABVSSI-1: It is recommended to provide an efficient and flexible consensus mechanism to ensure that incremental data synchronization and ledger updates are consistent. + +Req ABVSSI-2: It is required to provide an oversight strategy for the consensus process to ensure that the consensus history is auditable, regulated and immutable. + +Req ABVSSI-3: It is required to provide the ability to share data across chains to provide video data access to users under different platforms. + +Req ABVSSI-4: It is required to provide the ability for cross-platform consistency to ensure a unified interface for different data uploads. + +Req ABVSSI-5: It is required to provide the ability to provide the availability of data. Data shared by different systems need to be completely available. + +Req ABVSSI-6: It is recommended to provide the ability to provide high-speed transmission and communication in order to support secure sharing and transmission of encrypted video data in BVSSI. + +Req ABVSSI-7: It is required to provide a user information collection notification policy if collection is necessary to ensure that the content and purpose of the collection of personal information are clear to users. + +Req ABVSSI-8: It is required to provide a scalable storage mechanism and trusted node deployment hardware facilities, combine cloud storage and DLT to achieve massive resource data storage and sharing, and ensure the sustainability of ledger synchronization. + +Req ABVSSI-9: It required to provide a video data scheduling mechanism to ensure a more efficient sharing strategy for VS data with different usage frequencies. + +Req ABVSSI-10: It is recommended to provide an optional encryption mechanism to ensure that VS data owners perform data encryption operations according to their own needs. + +Req ABVSSI-11: It is required to provide a capability of pricing principle to realize the transaction of video data assets and value transfer of video data elements. + +## Appendix I + +## Scenarios and use cases of video surveillance system interworking based on DLT + +(This appendix does not form an integral part of this Recommendation.) + +### I.1 Registration in VSSI + +Alice registers her VSS and the devices of her VSS through an Internet service provider (ISP) to the DLT, to prepare for BVSSI. + +NOTE – The ISP is responsible for providing access authentication and network access services for users. Here, users can access a BVSSI network through ISP authentication. + +![Figure I.1 – Registration workflow in BVSSI. The diagram illustrates the interaction between Alice, Alice's VSS (containing VS platform, PU, and CU), ISP, and DLT. The steps are: 1. Alice generates a public-private key pair and user ID [K_A, K'_A, ID_A]. 2. Alice sends real identification [Real identity + K_A + ID_A] to ISP. 3. ISP sends an upload request [ID_A + K_A] to DLT. 4. Alice logs in to the VS platform with user ID. 5. Alice's VSS generates a public-private key pair and device ID [K_PU, K'_PU, ID_PU]. 6. Alice's VSS sends device identification [K_PU + ID_PU + Sign_A] to ISP. 7. ISP sends an upload request [ID_PU + K_PU] to DLT.](9ae17964ddd9b814c7d905b1af2fddf2_img.jpg) + +``` +graph LR + Alice((Alice)) -- "1. Generate public-private key pair and user ID [K_A, K'_A, ID_A]" --> Alice + Alice -- "2. Real identification [Real identity + K_A + ID_A]" --> ISP[ISP] + ISP -- "3. Upload request [ID_A + K_A]" --> DLT[DLT] + Alice -- "4. Log in VS platform with user ID" --> VS[VS platform] + subgraph Alice_VSS [Alice's VSS] + VS[VS platform] + PU[PU] + CU[CU] + end + Alice_VSS -- "5. Generate public-private key pair and device ID [K_PU, K'_PU, ID_PU]" --> Alice_VSS + Alice_VSS -- "6. Device identification [K_PU + ID_PU + Sign_A]" --> ISP + ISP -- "7. Upload request [ID_PU + K_PU]" --> DLT +``` + +Figure I.1 – Registration workflow in BVSSI. The diagram illustrates the interaction between Alice, Alice's VSS (containing VS platform, PU, and CU), ISP, and DLT. The steps are: 1. Alice generates a public-private key pair and user ID [K\_A, K'\_A, ID\_A]. 2. Alice sends real identification [Real identity + K\_A + ID\_A] to ISP. 3. ISP sends an upload request [ID\_A + K\_A] to DLT. 4. Alice logs in to the VS platform with user ID. 5. Alice's VSS generates a public-private key pair and device ID [K\_PU, K'\_PU, ID\_PU]. 6. Alice's VSS sends device identification [K\_PU + ID\_PU + Sign\_A] to ISP. 7. ISP sends an upload request [ID\_PU + K\_PU] to DLT. + +F.743.24(24) + +Figure I.1 – Registration workflow in BVSSI + +#### I.1.1 User registration + +**Step 1:** Alice adopts an asymmetric encryption algorithm to generate a public-private key pair and a user ID. The public key ( $K_{pub}$ of Alice) is a tool used by other users to verify the signature of text sent by Alice. The private key ( $K_{pri}$ of Alice) is a tool for Alice to make a digital signature and is stored by Alice. The user ID ( $ID_{Alice}$ ) is Alice's identity in VSS. + +**Step 2:** Alice sends her real identity (such as identity card), public key and user ID to ISP for real identity identification. + +**Step 3:** ISP authenticates Alice's real identity. If the authentication is successful, Alice's public key and user ID are signed by the ISP and form a record. Then the record is uploaded to DLT as a transaction. + +#### I.1.2 Equipment registration + +**Step 4:** Alice logs in to the VS platform of her VSS with user ID to control her PUs and CUs in intra-system and inter-system interaction. + +**Step 5:** Before joining VSS, PUs and CUs newly purchased by Alice also need to generate public-private key pairs (the private key can be mixed by the hardware serial number and the password set by the user) and device IDs. The private key is stored by Alice. + +**Step 6:** Device IDs and public key record are sent to the ISP after being signed by Alice with her private key. + +**Step 7:** The ISP verifies the signature of the received record. If the verification is successful, the ISP uploads the device IDs and the corresponding public key to DLT. + +### I.2 Authentication and authorization + +Bob owns a VSS consisting of multiple CUs and mobile customer units (MCUs). Alice's and Bob's VSSs have been registered on DLT. Bob wants to access the data of PU1 in Alice's VSS with his CU, so he needs Alice's authorization first. + +![Figure I.2 – Authentication and authorization workflow in BVSSI. The diagram shows three main entities: Bob's VSS (left), Alice's VSS (right), and DLT (bottom center). Bob's VSS contains a VSS platform, a CU (computer icon), and an MCU (smartphone icon). Alice's VSS contains a VSS platform, PU1 (blue flag), and PU2 (orange flag). DLT is represented by three connected cubes. Seven numbered arrows show the message flow: 1. Request for PU's public key [ID_PU1] from Bob's VSS to DLT. 2. Response for PU's public key [ID_PU1 + K_PU1] from DLT to Bob's VSS. 3. Operation request (M1) E_KPU1[ID_CU + K_CU + Sign_KCU + nonce] from Bob's VSS to Alice's VSS. 4. Verification request for CU [ID_CU + K_CU] from Alice's VSS to DLT. 5. Verification response for CU [ID_CU + K_CU + OK] from DLT to Alice's VSS. 6. Authentication request (M2) E_KCU[ID_CU + K'_PU1 + Sign_K'PU1 + (nonce+1)] from Alice's VSS to Bob's VSS. 7. Authentication response (M3) E_KPU1[ID_CU + Sign_KCU + (nonce+2)] from Bob's VSS to Alice's VSS.](0f985b39edc1d52ba3600c438bc8f0a5_img.jpg) + +F.743.24(24) + +Figure I.2 – Authentication and authorization workflow in BVSSI. The diagram shows three main entities: Bob's VSS (left), Alice's VSS (right), and DLT (bottom center). Bob's VSS contains a VSS platform, a CU (computer icon), and an MCU (smartphone icon). Alice's VSS contains a VSS platform, PU1 (blue flag), and PU2 (orange flag). DLT is represented by three connected cubes. Seven numbered arrows show the message flow: 1. Request for PU's public key [ID\_PU1] from Bob's VSS to DLT. 2. Response for PU's public key [ID\_PU1 + K\_PU1] from DLT to Bob's VSS. 3. Operation request (M1) E\_KPU1[ID\_CU + K\_CU + Sign\_KCU + nonce] from Bob's VSS to Alice's VSS. 4. Verification request for CU [ID\_CU + K\_CU] from Alice's VSS to DLT. 5. Verification response for CU [ID\_CU + K\_CU + OK] from DLT to Alice's VSS. 6. Authentication request (M2) E\_KCU[ID\_CU + K'\_PU1 + Sign\_K'PU1 + (nonce+1)] from Alice's VSS to Bob's VSS. 7. Authentication response (M3) E\_KPU1[ID\_CU + Sign\_KCU + (nonce+2)] from Bob's VSS to Alice's VSS. + +**Figure I.2 – Authentication and authorization workflow in BVSSI** + +**Step 1:** First, Bob's VSS searches for the public key of Alice's PU1 ( $K_{pub\text{ of PU1}}$ ) on DLT, according to the ID of PU1 ( $ID_{PU1}$ ). + +**Step 2:** DLT receives a legitimate query request from Bob's VSS. The corresponding public key of PU1 ( $K_{pub\text{ of PU1}}$ ), which is generated by PU1 during the registration, is then queried in DLT ledger according to the received $ID_{PU1}$ . If it exists, the $K_{pub\text{ of PU1}}$ is returned to Bob's VSS. + +**Step 3:** When Bob's VSS receives the public key of Alice's PU1 ( $K_{pub\text{ of PU1}}$ ), it initiates an authorization session with Alice's VSS. Bob's VSS sends a cryptographic authorization request message (M1) to Alice's VSS, including $ID_{CU}$ , the public key of CU ( $K_{pub\text{ of CU}}$ ), a signature encrypted by the private key of CU ( $K_{pri\text{ of CU}}$ ), and a nonce used for maintaining the response sequence of message. Bob's VSS encrypts the first message (M1) with the public key of PU1 ( $K_{pub\text{ of PU1}}$ ) and sends the cipher text to Alice's VSS. + +**Step 4:** Alice's VSS decrypts M1 with $K_{pri\text{ of PU1}}$ after receiving it, and obtains $ID_{CU}$ and $K_{pub\text{ of CU}}$ . Alice's VSS then verifies the authenticity of the message through signature and $K_{pub\text{ of CU}}$ . If the message is legitimate, Alice's VSS sends a verification request to DLT to verify the $ID_{CU}$ and its corresponding $K_{pub\text{ of CU}}$ further. + +**Step 5:** DLT receives a legitimate verification request from Alice's VSS. It queries and verifies the $K_{pub\text{ of CU}}$ in DLT ledger according to the $ID_{CU}$ , and then returns the verification result to Alice's VSS. + +**Step 6:** If the verification of CU's information is successful, Alice's VSS generates a temporary public-private key pair ( $K'_{pub\text{ of PU1}}$ , $K'_{pri\text{ of PU1}}$ ) for this authorization of Bob's CU to access Alice's PU1. + +**NOTE** – The temporary key pair is generated based on the root key pair of the PU1 ( $K_{pub\text{ of PU1}}$ , $K_{pri\text{ of PU1}}$ ) and is only valid in this authorization. Its lifecycle is controlled by Alice's VSS. In addition, the $K_{pri\text{ of PU1}}$ cannot be obtained through this temporary key pair. + +**Step 7:** Alice's VSS continues the authorization session, and sends an identity verification request message (M2) to Bob's VSS, consisting of $ID_{PU1}$ , $K'_{pub\text{ of PU1}}$ , a signature encrypted by $K'_{pri\text{ of PU1}}$ , and $nonce+1$ which is used for maintaining the response sequence of the message. Alice's VSS encrypts the entire message with $K_{pub\text{ of CU}}$ and sends the cipher text to Bob's VSS. + +**Step 8:** Bob's VSS decrypts M2 with $K_{pri}$ of CU after receiving it, and obtains the $K'_{pub}$ of PU1 generated for this session. Bob's VSS then verifies the authenticity of the message through signature and $K_{pub}$ of PU1. If the message is legitimate and authentic, Bob's VSS sends a response message (M3) to Alice's VSS. The response message includes $ID_{CU}$ , a signature encrypted by $K_{pri}$ of CU, and nonce+2 used to maintain the response sequence of messages. M3 is encrypted by $K'_{pub}$ of PU1. Alice's VSS verifies the signature and nonce with $K_{pub}$ of CU after receiving and decrypting M3 using $K'_{pri}$ of PU1. + +If the verification is successful, bidirectional authentication is ended. The CU obtains the authorization of Alice's PU and prepares to establish the connection. + +### I.3 Request the surveillance video data in the cloud + +Bob's VSS requests access authorization to video data stored by Alice's VSS in the cloud, which divided into three steps. + +#### Step 1: Get a trusted identity on the DLT + +Bob gets the public key of Alice's PU through the DLT, and then uses it to encrypt the identity information to initiate a video data request to Alice to access the cloud DB. Alice obtains Bob's CU public key through the block chain, as shown in flows 1 to 3 in Figure I.3. + +#### Step 2: Authentication and authorization code issuing + +Alice verifies Bob's identity by verifying the signature, and then sends Bob a cloud DB access authorization code. Alice's authorization code can include certain information, such as $ID_{CU}$ , $ID_{PU}$ and authorized access range, as shown in flows 4 and 5 in Figure I.3. + +#### Step 3: Authorization code verification + +Bob sends video data request to Alice's cloud DB through authorization code, and the cloud DB verifies the authorization code from Alice. If the authorization code is ok, then the whole request process ends, as shown in flows 6 to 8 in Figure I.3. + +![Figure I.3 – Request video data workflow in BVSSI. The diagram illustrates the interaction between Bob's VSS (containing a VSS platform and a CU), Alice's VSS (containing a VSS platform and two PUs, PU1 and PU2), a DLT (Decentralized Ledger Technology) block chain, and Cloud storage of Alice's VSS. The workflow consists of eight steps: 1. Bob's CU requests PU1's public key from the DLT. 2. The DLT responds with the public key and identity. 3. Bob's VSS sends an access request to Alice's VSS. 4. Alice's VSS performs mutual authentication. 5. Alice's VSS sends an access response with an authorization code. 6. Bob's CU sends an access request for PU1's data to the cloud storage. 7. The cloud storage verifies the authorization code. 8. The cloud storage confirms that the CU is authorized by PU1.](90ee16ccc0ad16aeca48087797d7b07f_img.jpg) + +``` + +graph TD + subgraph Bob_VSS [Bob's VSS] + VSS_Bob[VSS platform] <--> CU_Bob[CU] + end + subgraph Alice_VSS [Alice's VSS] + VSS_Alice[VSS platform] <--> PU1[PU1] + VSS_Alice <--> PU2[PU2] + end + DLT[DLT] + Cloud[Cloud storage of Alice's VSS] + + CU_Bob -- "1-1. Request for PU1's public key [ID_PU1]" --> DLT + DLT -- "1-2. Response for PU1's public key [ID_PU1 + K_PU1]" --> CU_Bob + CU_Bob -- "2. Access request to PU1 +E_{K_PU1}[ID_{CU} + K_{CU} + Sign_{K_{CU}}]" --> Alice_VSS + Alice_VSS -- "4. Mutual authentication between CU and PU1 +(Verify both private key through public key)" --> CU_Bob + Alice_VSS -- "5. Access response to CU +E_{K_{CU}}[ID_{PU1} + Sign_{K_{PU1}} + AutCode]" --> CU_Bob + CU_Bob -- "6. Access request for PU1's data +[ID_{CU} + AutCode]" --> Cloud + Cloud -- "7-1. Request verification for AutCode +[ID_{CU} + AutCode]" --> Alice_VSS + Alice_VSS -- "7-2. Response verification for AutCode +[ID_{PU1} + AutCode + OK]" --> Cloud + Cloud -- "8. Confirm that CU is authorized by PU1" --> CU_Bob + CU_Bob -- "3-1. Request for CU's public key [ID_{CU}]" --> DLT + DLT -- "3-2. Response for CU's public key [ID_{CU} + K_{CU}]" --> CU_Bob + +``` + +Figure I.3 – Request video data workflow in BVSSI. The diagram illustrates the interaction between Bob's VSS (containing a VSS platform and a CU), Alice's VSS (containing a VSS platform and two PUs, PU1 and PU2), a DLT (Decentralized Ledger Technology) block chain, and Cloud storage of Alice's VSS. The workflow consists of eight steps: 1. Bob's CU requests PU1's public key from the DLT. 2. The DLT responds with the public key and identity. 3. Bob's VSS sends an access request to Alice's VSS. 4. Alice's VSS performs mutual authentication. 5. Alice's VSS sends an access response with an authorization code. 6. Bob's CU sends an access request for PU1's data to the cloud storage. 7. The cloud storage verifies the authorization code. 8. The cloud storage confirms that the CU is authorized by PU1. + +F.743.24(24) + +Figure I.3 – Request video data workflow in BVSSI + +### I.4 Video resource upload to the DLT and verification + +Considering the authenticity requirement of the video resources which Bob's VSS downloads from the cloud storage by video resources ID, the following scenario based on the DLT can be considered. + +#### Step 1: Video resource ID and resource hash information generation + +The original video files recorded by Alice's VSS are generated encrypted video files through a symmetric key. Meanwhile, using a secure hash algorithm (SHA) series is used to map the encrypted video file to get a hash. In order to get the video resource ID, the multihash encoding (PU number + hash algorithm category + length of hash (number of bytes) + hash value) should then be generated to obtain the final video resource ID through hash mapping, as shown in flow 1 in Figure I.4. + +#### Step 2: Uploading video resource ID and resource hash information to the DLT + +Alice's VSS sends a video resource ID, encrypted video data, PU ID, resource hash information to cloud storage. The resource ID and resource hash information are then uploaded to the DLT by cloud storage server, appending the index value of the video resource stored in the cloud and these processes are recorded on the DLT, as shown in flows 2 to 4 in Figure I.4. + +#### Step 3: Video information verification + +When Bob's VSS CU downloads the video file (called data request in Figure I.4) from the cloud storage service provider, it searches the video resource by resource ID and index to download the video resource file. The CU then matches the hash value of the resource with that of the DLT to verify whether the video resource has been tampered with (called data response in Figure I.4) and these processes are also recorded on the DLT, as shown in flows 5 to 7 in Figure I.4. + +NOTE – In this scenario, all interactions and transmissions are encrypted. + +![Figure I.4 – Information upload and verification workflow in BVSSI. The diagram shows the interaction between Alice's VSS, Police VSS, DLT, and Cloud storage of Alice's VSS. Alice's VSS (containing VSS platform, PU1, and PU2) generates ID_m and Hash_m (Step 1), uploads data to cloud storage (Step 2), and uploads ID_m and Hash_m to DLT (Step 3). Police VSS (containing VSS platform and CU) sends a data request [ID_m] to cloud storage (Step 5), receives a data response E_{k_{PU1}}[file] (Step 6), and validates records on DLT (Step 7). Cloud storage uploads ID_m and Hash_m to DLT (Step 4).](3fa8bfee86764e3c3a1a6fbbe61bbd52_img.jpg) + +Figure I.4 – Information upload and verification workflow in BVSSI. The diagram shows the interaction between Alice's VSS, Police VSS, DLT, and Cloud storage of Alice's VSS. Alice's VSS (containing VSS platform, PU1, and PU2) generates ID\_m and Hash\_m (Step 1), uploads data to cloud storage (Step 2), and uploads ID\_m and Hash\_m to DLT (Step 3). Police VSS (containing VSS platform and CU) sends a data request [ID\_m] to cloud storage (Step 5), receives a data response E\_{k\_{PU1}}[file] (Step 6), and validates records on DLT (Step 7). Cloud storage uploads ID\_m and Hash\_m to DLT (Step 4). + +F.743.24(24) + +Figure I.4 – Information upload and verification workflow in BVSSI + +### I.5 Video data resources sharing + +#### I.5.1 General video data resources sharing + +See Figure I.5. + +Based on the registration step, authentication and authorization, video data information uploading and video data resources can be shared between Alice's VSS and Bob's VSS. + +In VSSI based on DLT, a video data resource can be shared automatically by SC in a trusted and secure fashion by public-private key pairs, traceable by data structure, etc. + +##### Step 1: Sign the smart contract between Alice and Bob + +During video resource upload to the DLT and verification, Alice (resource owner) has created a user SC, including usage period, usable range and permission. To obtain and use these resources, Bob (resource user) should sign this SC with his $K_{pri}$ . + +##### Step 2: Obtain the video data catalogue + +After signing the SC, Bob obtains the data catalogue of Alice's VSS automatically, which can confirm the video data location, and sends a message to DLT, including $ID_{Bob}$ , resource ID and PU number. + +##### Step 3: Encrypt and decrypt video data + +Alice's VSS encrypts the video data resource by Bob's $K_{pub}$ and sends it by SC. Bob's VSS uses his own $K_{pri}$ to decrypt the encrypted video data by his $K_{pub}$ . + +##### Step 4: Smart contract execution records are recorded on DLT + +In video data resource sharing flow, all of SC execution records (including data-using records) are recorded on DLT to trace. + +![Diagram illustrating the general video data resources sharing process in BVSSI. The diagram shows interactions between Bob's VSS (containing a VSS platform and a CU), Alice's VSS (containing a VSS platform and two PUs, PU1 and PU2), a central Smart contract, and a DLT (represented by three blocks).](cf4ac1058c52bc3ca37737740afb7f2c_img.jpg) + +``` +graph TD + subgraph Bob_VSS [Bob's VSS] + BP[VSS platform] <--> CU[CU] + end + subgraph Alice_VSS [Alice's VSS] + AP[VSS platform] <--> PU1[PU1] + AP <--> PU2[PU2] + end + SC((Smart contract)) + DLT[DLT] + + CU -- "1.1 Create smart contract" --> SC + SC -- "1.2 Sign smart contract" --> AP + AP -- "2.1 Obtain the data catalogue" --> SC + SC -- "2.2 Confirm the video data location (ID_Bob, resource ID and PU number)" --> CU + CU -- "3.1 Encrypts the video data resource [Bob's K_pub]" --> SC + SC -- "3.2 Decrypts the encrypted video data [Bob's K_pri]" --> AP + SC -- "4. Call and recorded" --> DLT +``` + +Diagram illustrating the general video data resources sharing process in BVSSI. The diagram shows interactions between Bob's VSS (containing a VSS platform and a CU), Alice's VSS (containing a VSS platform and two PUs, PU1 and PU2), a central Smart contract, and a DLT (represented by three blocks). + +F.743.24(24) + +Figure I.5 – General video data resources sharing in BVSSI + +#### I.5.2 Private video data resources sharing + +See Figure I.6. + +There is a private DB in each user's VSS. Similar to PU and CU, the private DB also has a public-private key pair and a device ID. The private key is stored by the user, and the public key and device ID are recorded in DLT. + +For a private video resource that is extremely confidential, Alice stores it in her local private DB instead of uploading it to cloud storage. When there is a need to share the private video resource, Alice actively initiates a private transaction to a specific target object such as Bob, which process is divided into five steps. + +**Step 1:** Alice's VSS requests the public key of Bob's private DB ( $K_{DB2}$ ) on DLT, according to the ID of Bob's private database ( $ID_{DB2}$ ), and DLT response for public key of Bob's private database [ $K_{DB2}$ , $ID_{DB2}$ ]. + +**Step 2:** Alice's VSS encrypts the private video resource that needs to be shared by the public key of Bob's private database ( $K_{DB2}$ ). Meanwhile, Alice's VSS uses SHA series mapping on the encrypted video resource to get a hash ( $Hash_v$ ). After that, in order to get the video resource ID, the multihash encoding (the number of Alice's private DB + hash algorithm category + length of hash (number of bytes) + hash value) should be generated to obtain the final video resource ID ( $ID_v$ ) through hash mapping. Alice then uploads the resource ID ( $ID_v$ ) and resource hash ( $Hash_v$ ) information to the DLT. + +**Step 3:** Alice initiates a private transaction request. The request contains the ID of Bob's private DB ( $ID_{DB2}$ ), the ID of the resource ( $ID_v$ ), the encrypted resource file ( $E_{K_{DB2}}[file]$ ), the hash value ( $Hash_v$ ) and a private transaction parameter ( $PrivateParm\{ID_{DB1}, ID_{DB2}\}$ ). The private transaction parameter identifies the transaction as private, and includes the relevant node IDs of the transaction, i.e., the IDs of Alice's ( $ID_{DB1}$ ) and Bob's private databases ( $ID_{DB2}$ ). Only the nodes in the parameters can participate in the transaction. + +**Step 4:** Bob receives the private transaction request and compares whether his ID of private database ( $ID_{DB2}$ ) belongs to the IDs in the private transaction parameters. If so, he participates in the transaction, otherwise not. + +**Step 5:** Bob's VSS uses the private key of Bob's private DB to decrypt the video resource, obtains the resource and writes it into his own private DB. After then, Bob matches the hash value of the resource with that on the DLT to verify whether the video resource has been tampered with. + +![Diagram illustrating the private video data resources sharing process in BVSSI. The diagram shows interactions between Bob's VSS (containing VSS platform, CU, and Private DB2), Alice's VSS (containing VSS platform, Private DB1, PU1, and PU2), and the DLT. The process involves: 1. Alice's VSS requesting Bob's public key from DLT. 2. Alice's VSS encrypting a video resource and uploading ID_v and Hash_v to DLT. 3. Alice's VSS sending a private transaction request to Bob's VSS. 4. Bob's VSS matching ID_DB2 in PrivateParm. 5. Bob's VSS decrypting the resource and verifying it against DLT.](3da1a07cb87051bf616c9876db958cf0_img.jpg) + +The diagram illustrates the flow of data and requests between three main components: Bob's VSS, Alice's VSS, and the DLT. + +- Alice's VSS to DLT:** + - 1-1. Request for public key of Bob's private database [ $ID_{DB2}$ ] + - 1-2. Response for public key of Bob's private database [ $ID_{DB2}$ , $K_{DB2}$ ] + - 2-2. Upload $ID_v$ and $Hash_v$ to DLT +- Alice's VSS to Bob's VSS:** + - 2-1. Encrypt private video resource, generate $ID_v$ , $Hash_v$ + - 3. Private transaction request: [ $ID_{DB2}$ , $ID_v$ , $E_{K_{DB2}}[file]$ , $Hash_v$ , $PrivateParm\{ID_{DB1}, ID_{DB2}\}$ ] +- Bob's VSS to DLT:** + - 5-1. Decrypt $E_{K_{DB2}}[file]$ and write it into private DB2 + - 5-2. Verify whether the private video resource has been tampered +- Internal Bob's VSS:** + - 4. Match $ID_{DB2}$ in $PrivateParm$ + +Bob's VSS contains: VSS platform, CU (Computer Unit), and Private DB2. Alice's VSS contains: VSS platform, Private DB1, PU1 (Processing Unit 1), and PU2 (Processing Unit 2). The DLT contains three blocks representing a chain. + +Diagram illustrating the private video data resources sharing process in BVSSI. The diagram shows interactions between Bob's VSS (containing VSS platform, CU, and Private DB2), Alice's VSS (containing VSS platform, Private DB1, PU1, and PU2), and the DLT. The process involves: 1. Alice's VSS requesting Bob's public key from DLT. 2. Alice's VSS encrypting a video resource and uploading ID\_v and Hash\_v to DLT. 3. Alice's VSS sending a private transaction request to Bob's VSS. 4. Bob's VSS matching ID\_DB2 in PrivateParm. 5. Bob's VSS decrypting the resource and verifying it against DLT. + +**Figure I.6 – Private video data resources sharing in BVSSI** + +# Bibliography + +- [b-ITU-T F.743.3] Recommendation ITU-T F.743.3 (2016), *Requirements for visual surveillance system interworking*. +- [b-ITU-T X.1400] Recommendation ITU-T X.1400 (2020), *Terms and definitions for distributed ledger technology*. + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.743.25-202406-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.743.25-202406-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d8b9892f78b0e2874841f59e03c0c4ce65b75bd1 --- /dev/null +++ b/marked/F/T-REC-F.743.25-202406-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:68790afd0434f832bc5db83cce6775355efa0d89785ad99324ab9ba72055239f +size 7145 diff --git a/marked/F/T-REC-F.743.25-202406-I_PDF-E/b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg b/marked/F/T-REC-F.743.25-202406-I_PDF-E/b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c9805865919c45fb615a1651a632e897f257695f --- /dev/null +++ b/marked/F/T-REC-F.743.25-202406-I_PDF-E/b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:06e20bfe7db02c02820c6f7f27756430b2bf7e9f812b61a09866a38b8cd24125 +size 73308 diff --git a/marked/F/T-REC-F.743.25-202406-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg b/marked/F/T-REC-F.743.25-202406-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..375518441992ff49b9cae18460e15dc0c0cfffe1 --- /dev/null +++ b/marked/F/T-REC-F.743.25-202406-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:42500ed4abfc41827f0162daa687f7f39d1e4a387f3cbdc26c7e8c73daa39786 +size 111216 diff --git a/marked/F/T-REC-F.743.25-202406-I_PDF-E/raw.md b/marked/F/T-REC-F.743.25-202406-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..9ec86e1194fd1ab88b26b940a350109dc88cac0e --- /dev/null +++ b/marked/F/T-REC-F.743.25-202406-I_PDF-E/raw.md @@ -0,0 +1,602 @@ + + +# Recommendation + +# **ITU-T F.743.25 (06/2024)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +## --- **Procedures and requirements for multimedia data asset management** + +![ITU logo](84a1d09fb489061482111515543b60dc_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with white grid lines and the letters 'ITU' in a bold, blue, sans-serif font. The globe is stylized with a white arrow pointing upwards and to the right, symbolizing global communication and progress. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.743.25 + +# Procedures and requirements for multimedia data asset management + +## Summary + +Recommendation ITU-T F.743.21 provides a high-level overview of a data asset management framework with its corresponding objects, activities and supports. A standard multimedia data asset management procedure is needed in order to manage data assets effectively so that the value of multimedia data assets can be maintained and added. The procedures can help different enterprises to identify, recognize, share, exchange and apply multimedia data assets in a unified and standardized way. + +Recommendation ITU-T F.743.25 contains the procedures for the life cycle management of multimedia data assets. The procedures include multimedia data asset catalogue management, identification, registration, application, inventory check, change, assessment and disposal. It also provides the detailed requirements for each part of the procedures, and the use cases of how these processes can be applied are also provided in an appendix. + +## History\* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.743.25 | 2024-06-13 | 16 | 11.1002/1000/15905 | + +## Keywords + +Data asset, data asset management, management procedure. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2024 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|------------|-----------------------------------------------------------------------------------------|----| +| 1 | Scope ..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 1 | +| 4 | Abbreviations and acronyms ..... | 1 | +| 5 | Conventions ..... | 2 | +| 6 | Overview ..... | 2 | +| 7 | Procedures for multimedia data asset management..... | 3 | +| 7.1 | Flowchart for multimedia data asset management procedures..... | 3 | +| 7.2 | Procedure analysis for multimedia data asset management procedures..... | 3 | +| 7.3 | Relationship among each process in the multimedia data asset management procedures..... | 4 | +| 8 | Requirements for multimedia data asset management procedures..... | 6 | +| 8.1 | Requirements for MDA catalogue management ..... | 6 | +| 8.2 | Requirements for MDA identification..... | 7 | +| 8.3 | Requirements for MDA registration..... | 7 | +| 8.4 | Requirements for MDA application ..... | 7 | +| 8.5 | Requirements for MDA inventory check ..... | 7 | +| 8.6 | Requirements for MDA change..... | 7 | +| 8.7 | Requirements for MDA assessment ..... | 8 | +| 8.8 | Requirements for MDA disposal..... | 8 | +| Appendix I | – Use cases for multimedia data asset management procedures..... | 9 | +| I.1 | Use case for MDA identification..... | 9 | +| I.2 | Use case for MDA registration..... | 10 | +| I.3 | Use case for MDA application ..... | 10 | +| I.4 | Use case for MDA change..... | 11 | +| I.5 | Use case for MDA inventory check ..... | 12 | +| I.6 | Use cases for MDA assessment..... | 13 | +| I.7 | Use case for MDA disposal..... | 14 | +| | Bibliography..... | 15 | + + + +# Recommendation ITU-T F.743.25 + +## Procedures and requirements for multimedia data asset management + +# 1 Scope + +This Recommendation specifies the procedures and requirements for multimedia data asset management. The procedures mainly include catalogue management, identification, registration, application, inventory check, change, assessment and disposal of multimedia data assets. Detailed requirements and use cases for each part of the procedure are also provided in this Recommendation. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T F.743.21] Recommendation ITU-T F.743.21 (2020), *Framework for data asset management*. + +# 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 data asset** [b-ITU-T X.1750]: An electronically recorded data resource, owned or controlled by an organization. + +**3.1.2 data asset management** [ITU-T F.743.21]: A set of business functions for planning, controlling and providing data and information assets, including developing, executing and monitoring data related plans, policies, programs, projects, processes, methods and procedures, so as to control, protect, deliver and improve the value of data assets. Data asset management requires a full integration of business, technology, and management to ensure that the value of data assets is maintained and added. + +**3.1.3 multimedia information** [b-ITU-T Y.4551]: Digital information that uses multiple forms of information content and information processing, such as text, pictures, audio, video, three-dimensional panoramic pictures and digital maps, which informs or entertains users. + +### 3.2 Terms defined in this Recommendation + +None. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +DAM Data Asset Management + +DB Database + +ICT Information and Communication Technology + +| | | +|------|----------------------------------| +| IT | Information Technology | +| MDA | Multimedia Data Asset | +| MDAM | Multimedia Data Asset Management | +| NM | Network Management | + +## 5 Conventions + +In this Recommendation, the following conventions are used: + +- The keywords "**is required**" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The keywords "**is recommended**" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement needs not be present to claim conformance. + +# 6 Overview + +[ITU-T F.743.21] provides a high-level specification of a data asset management (DAM) framework with its corresponding objects, activities and supports. A standard multimedia data asset management procedure is needed in order to manage data assets effectively so that the value of multimedia data assets can be maintained and added. The procedures can instruct the owners to manage their multimedia data assets in a standard way, and help different enterprises to identify, recognize, share, exchange and apply multimedia data assets in a unified and standardized way. + +In the data asset management domain, there are some concepts that should be clearly distinguished. + +- Data: are usually collected and created by network equipment or information technology (IT) systems, and are usually stored in databases or other medias. The word data is used very widely, and there are many kinds of data types, for example, multimedia data, network traffic data, service data, network management data, user preference data, etc. Data can be either the original data, or data processed for some specific use. +- Data asset: is a kind of data resource which is owned or controlled by an organization or enterprise. These data assets are in electrical form and are also viewed as properties of the organization or enterprise, which usually can create added values. + +Based on the above description, not all "data" are "data asset". Only those that are specifically selected and recorded as property of the organization or enterprise can be classified as data asset. They are a kind of information not just in electronic form, but can also be viewed as assets, which can create added value for the owner, and can be registered, applied, checked, changed, assessed and disposed as a normal asset. + +This Recommendation focuses on multimedia data asset (MDA), which is a special case of data asset with the contents mainly stored in multimedia form (such as text, pictures, audio, video, three-dimensional panoramic pictures, digital maps etc.), and it is one of the most popular forms of data assets in the industry. The procedures for multimedia data asset management (MDAM) described in this Recommendation may include: MDA catalogue management, identification, registration, application, inventory check, change, assessment and disposal. The procedures and the detailed requirements for multimedia data asset management will be provided in the following clauses, and the example use cases for the above procedures are provided in Appendix I. + +# 7 Procedures for multimedia data asset management + +### 7.1 Flowchart for multimedia data asset management procedures + +Figure 1 shows the flowchart of the procedures for multimedia data asset management. There are mainly three phases for the life cycle management of an MDA. The first phase is the data asset non-existent phase, which means a data resource has not become a data asset yet, and it needs to be identified from an amount of data resources. The second phase is the data asset creating phase, during which the data asset record information is created, and after it is created and registered in the MDA catalogue, it becomes a formal data asset for the organization or enterprise. The last phase is the data asset existing phase, which starts when a data asset record is created and registered in the MDA catalogue; it is the most important phase for an MDA, during which multiple processes may occur for the data asset (such as application, inventory check, change, assessment, disposal, etc.). The detailed description of each process in the MDAM procedures are analysed in clause 7.2. + +![Flowchart for multimedia data asset management procedures](b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg) + +The flowchart illustrates the multimedia data asset management (MDAM) process, starting from 'Data resources' and moving through three main phases: 'Data asset non-existent phase', 'Data asset creating phase', and 'Data asset existing phase'. The 'Data asset non-existent phase' includes 'Data asset identification'. The 'Data asset creating phase' includes 'MDA registration'. The 'Data asset existing phase' includes 'MDA application', 'MDA inventory check', 'MDA change', 'MDA assessment', and 'MDA disposal'. All five processes in the existing phase are connected to a central 'MDA catalogue management' box at the top via double-headed arrows. A legend at the bottom identifies the phases by their background patterns: white for non-existent, diagonal lines for creating, and vertical lines for existing. The code 'F.743.25(24)' is located in the bottom right corner. + +Flowchart for multimedia data asset management procedures + +Figure 1 – Flowchart for multimedia data asset management procedures + +### 7.2 Procedure analysis for multimedia data asset management procedures + +#### 7.2.1 Multimedia data asset catalogue management + +Multimedia data asset catalogue is a set of information that describes the characteristics of data assets by classification and code. Data asset catalogue management contains a set of functions supporting the creation and maintenance of a catalogue that records all the identified data assets of an organization or enterprise. The purpose is to support the all the procedures of data asset management, including: identification, registration, application, inventory check, change, assessment, disposal, etc. + +#### 7.2.2 Multimedia data asset identification + +Multimedia data asset identification means the process of identifying data assets and their key information elements from existing data resources, according to some management goals. This process also needs to record the key information of the identified data assets. + +#### 7.2.3 Multimedia data asset registration + +Multimedia data asset registration means the process to register and confirm the ownership of a data asset in an organization or enterprise through technical means, so that it has the properties such as time, identity, content, etc. The purpose is to set up rights protection traceability during the application process of the data asset. + +#### **7.2.4 Multimedia data asset application** + +Multimedia data asset application means the process of increasing the added value of data assets through sharing, circulation and use, in order to meet the requirements of business scenarios and organizational development. + +#### **7.2.5 Multimedia data asset inventory check** + +Data asset inventory check means the process of checking the status of data assets to find inconsistencies between data assets and their information stored in the MDA catalogue. If found, the information in the catalogue should be updated to ensure consistency and integrity of data asset information. + +#### **7.2.6 Multimedia data asset change** + +Multimedia data asset change means the process of changing the information of the data asset catalogue through a controllable change process, so that the information in the catalogue is consistent with the actual data asset. + +#### **7.2.7 Multimedia data asset assessment** + +Multimedia data asset assessment means the process of quantitative assessment on the current status, quality and value of data assets in an organization or enterprise. + +#### **7.2.8 Multimedia data asset disposal** + +Multimedia data asset disposal means the process of destruction or transfer of data assets. The purpose is to optimize the management of data assets, reduce operation costs, and excavate the remaining utilization value. + +### **7.3 Relationship among each process in the multimedia data asset management procedures** + +Figure 2 shows the sequence diagram for the procedures of the MDAM, and it also illustrates the interactions between an MDA manager, the MDA catalogue and MDA records, where an MDA manager can be a person that manages or operates data assets. + +![Sequence diagram for the procedures and the interactions between MDA manager, MDA catalogue and MDA records. The diagram shows three lifelines: MDA manager (actor), MDA catalogue, and MDA Record. The interaction consists of 18 numbered steps. The MDA manager initiates the process by identifying an MDA (1) and registering it (2). The MDA catalogue creates a record (3). The manager then performs various operations (4, 7, 10, 13, 16) which are processed by the catalogue (5, 8, 11, 14, 17) and stored in the record (6, 9, 12, 15, 18). The MDA Record lifeline ends with an 'X' symbol. The diagram is labeled F.743.25(24) at the bottom right.](cfef993dcc8fb513de79eb1f93cf26ae_img.jpg) + +``` + +sequenceDiagram + actor MDA manager + participant MDA catalogue + participant MDA Record + Note left of MDA manager: 1. Identify an MDA from data resources + MDA manager->>MDA catalogue: 2. Register an MDA + Note left of MDA manager: 4. Perform MDA application + MDA catalogue->>MDA Record: 3. Create an MDA record + MDA manager->>MDA catalogue: 5. Apply an MDA + MDA catalogue->>MDA Record: 6. Store application information in the MDA record + Note left of MDA manager: 7. Perform MDA inventory check + MDA manager->>MDA catalogue: 8. Inventory check on an MDA + MDA catalogue->>MDA Record: 9. Store the inventory check result information in the MDA record + Note left of MDA manager: 10. MDA change is found + MDA manager->>MDA catalogue: 11. Update an MDA + MDA catalogue->>MDA Record: 12. Store the updated information in the MDA record + Note left of MDA manager: 13. Perform MDA assessment + MDA manager->>MDA catalogue: 14. Assess an MDA + MDA catalogue->>MDA Record: 15. Stores the assessment result information in the MDA record + Note left of MDA manager: 16. Perform MDA disposal + MDA manager->>MDA catalogue: 17. Dispose an MDA + MDA catalogue->>MDA Record: 18. Store the dispose result information in the MDA record + destroy MDA Record + Note right of MDA Record: F.743.25(24) + +``` + +Sequence diagram for the procedures and the interactions between MDA manager, MDA catalogue and MDA records. The diagram shows three lifelines: MDA manager (actor), MDA catalogue, and MDA Record. The interaction consists of 18 numbered steps. The MDA manager initiates the process by identifying an MDA (1) and registering it (2). The MDA catalogue creates a record (3). The manager then performs various operations (4, 7, 10, 13, 16) which are processed by the catalogue (5, 8, 11, 14, 17) and stored in the record (6, 9, 12, 15, 18). The MDA Record lifeline ends with an 'X' symbol. The diagram is labeled F.743.25(24) at the bottom right. + +**Figure 2 – Sequence diagram for the procedures and the interactions between MDA manager, MDA catalogue and MDA records** + +Before a multimedia data resource becomes a data asset, the MDA manager needs to perform some identification work, so that the key information of the data resource can be abstracted and specified as the characteristic information for the data asset. Then the MDA manager sends a request to the MDA catalogue with the identified information. After the MDA catalogue receives this request, it will create an instance of MDA record, and stores the related information in the record. Thus, an MDA starts to exist in the organization or enterprise. Steps 1-3 in Figure 2 shows the above procedure. + +When the application of an MDA is to be performed, the MDA manager will send a request to the MDA catalogue for the application of the MDA. The MDA catalogue will then store the application information in the corresponding MDA record. When the application is completed, the above steps will be performed again. Steps 4-6 in Figure 2 show the above procedure. + +When there is a need to perform an inventory check on an MDA, the MDA manager will send a request to the MDA catalogue for the inventory check, and the MDA catalogue will then store the inventory check information in the corresponding MDA record. When the inventory check is completed, the result information will also be stored in the MDA record. Steps 7-9 in Figure 2 show the above procedure. + +If some MDA information is found to be inconsistent with the real MDA during the inventory check process, or if the organization or enterprise finds changes regarding the MDA, the MDA manager will trigger the MDA change process. The change request will be sent to the MDA catalogue and the change result information will be stored in the corresponding MDA record. Steps 10-12 in Figure 2 show the above procedure. + +When there is the need to perform the assessment of an MDA, the MDA manager will send a request to the MDA catalogue for the assessment. The MDA catalogue will then store the assessment information in the corresponding MDA record. When the assessment process is completed, the assessment result information will also be stored in the MDA record. Steps 13-15 in Figure 2 show the above procedure. + +When there is the need to dispose of an MDA, the MDA manager will send a request to the MDA catalogue for the disposal. The MDA catalogue will then store the disposal result (destruction or transfer) information in the corresponding MDA record. An MDA record will be locked after the MDA is disposed of, which means the end of the life cycle of the MDA in this organization or enterprise. Steps 16-18 in Figure 2 show the above procedure. + +It should be noted that steps 4, 7, 10 and 13 in Figure 2 may occur in a random order, as they do not depend on each other in the procedures. + +## **8 Requirements for multimedia data asset management procedures** + +### **8.1 Requirements for MDA catalogue management** + +- 1) It is required to support the establishment of an MDA catalogue, which is used to store the information elements of multiple multimedia data assets. +- 2) It is required for the catalogue to support the creation of new MDA records, and when a new record is created, at least (but not limited to) the following information is required to be stored in the catalogue (it is mainly aligned with the information specified in [b-ITU-T Y.3603]): + - a) Unique code of the MDA, which can be used as the MDA identifier. + - b) MDA title. + - c) MDA category. + - d) MDA resource information. + - e) MDA domain information. + - f) Responsible party of the MDA (which provides the ownership information). + - g) MDA overview. +- 3) It is required to support the updating of the record information stored in an MDA catalogue. +- 4) It is required to support the deletion of the records in an MDA catalogue. +- 5) It is required to support the querying of the record information stored in an MDA catalogue. +- 6) It is required to support the establishment of the control mechanisms for permissions, versions and releases of the MDA catalogue management. +- 7) It is recommended to support the classification of MDA, with the dimensions including but not limited to entities, topics and businesses. + +- 8) It is required to add activity records associate to each MDA record, such as MDA application records, change records, inventory check records, change records, assessment records and disposal records. +- 9) It is required to maintain and update the information in MDA catalogue timely and effectively, in order to support other MDA management procedures. + +### **8.2 Requirements for MDA identification** + +- 1) It is required to support sorting out of existing data resources, and identifying MDAs and their information elements, including basic information, business information, management information and value information, based on business applications and market demands. +- 2) After the identification of an MDA is completed, it is required to store the MDA information in the MDA catalogue according to its classification, in order to support other MDA management processes, such as application, assessment, inventory check, etc. +- 3) It is required to support the identification of the changes in MDAs and trigger the MDA change process. + +### **8.3 Requirements for MDA registration** + +- 1) It is required to support the confirmation of the ownership of an MDA based on technical means, such as electronic authentication, block chain or others, under the joint authentication of a single or multiple institutions. +- 2) In the process of application and provision of an MDA, it is recommended to record the identity and time attributes of the MDA through digital signature, distributed ledger or other means. + +### **8.4 Requirements for MDA application** + +- 1) It is required to support the identification of the pathways and channels for data asset application, classify the application level based on business needs, management models, management strategies and data sensitivity, and to provide corresponding security measures. +- 2) It is required to establish mechanisms for MDA service assurance, benefit assessment and effectiveness assessment of the MDA application. +- 3) It is required to ensure the safe, controllable, compliant application process of MDAs. +- 4) It is required to completely record the behaviour of MDA applications to ensure traceability. + +### **8.5 Requirements for MDA inventory check** + +- 1) It is required to develop an MDA inventory check plan, clarifying the scope, requirements, procedures and timing of the inventory check. +- 2) It is required to assign a dedicated person to be responsible for MDA inventory check, and define the rights and responsibilities of inventory check personnel. +- 3) It is required for the inventory check personnel to verify the consistency and accuracy of the MDA catalogue record and the corresponding MDA, based on the inventory check plan, and record the inventory check results. +- 4) It is recommended to timely analyse and handle issues discovered during the inventory check process. + +### **8.6 Requirements for MDA change** + +- 1) It is required to establish an MDA change mechanism, clarify the triggering conditions for MDA changes, and effectively control the change process. + +- 2) It is required to review submitted data asset changes, including completeness of information, necessity of business, compliance with the requirements, scope of impact, ownership, etc. +- 3) It is recommended to analyse the impact of data asset changes and issue change impact notifications. +- 4) It is required to implement changes based on the review results of MDA changes and update the MDA catalogue accordingly. +- 5) It is required to record the change process and establish a continuous tracking, review, and improvement mechanism for MDA changes. + +### **8.7 Requirements for MDA assessment** + +- 1) It is required to establish an MDA assessment mechanism, clarify the purpose, scope, strategy and cycle of the assessment, and use internal or external assessment. +- 2) It is required to clarify the ownership, sensitive information and security compliance requirements of data assets, based on the purpose and scope of the data asset assessment. +- 3) It is required to assess the quality of the MDA, including accuracy, completeness, consistency, authenticity and timeliness of the data. The model and assessment methods of the assessment content should follow dedicated specifications defined elsewhere. +- 4) It is recommended to assess the economic and social value of the MDA. +- 5) It is required to timely update the assessment results to the MDA catalogue and ensure that the assessment process is recorded and traceable. + +### **8.8 Requirements for MDA disposal** + +- 1) It is required to establish an MDA disposal mechanism and effectively control the disposal process and risks. +- 2) It is required to design a disposal plan and a disposal scheme based on disposal requirements. +- 3) It is recommended to conduct risk assessment and impact analysis on the disposal scheme and conduct a review for the disposal scheme of the MDA. +- 4) It is required to execute the disposal and record the executed information according to the approved disposal scheme and set a retention time period for MDAs that need to be destroyed. + +## Appendix I + +## Use cases for multimedia data asset management procedures + +(This appendix does not form an integral part of this Recommendation.) + +This appendix introduces some use cases to support multimedia data asset management (MDAM). + +### I.1 Use case for MDA identification + +Suppose there are several categories of data resources being used and developed in an enterprise (e.g., *Company ABC*). For example, there may be enterprise management data, user private data and multimedia data for education. Among these data, some of them may provide added value to the enterprise. At the beginning, these running data are just used as normal data, but with the growing requirements and technologies, some of them can be shared with other organizations or enterprises, and such data can be selected and classified as a data asset. + +In this example, the enterprise management data is only used internally for this enterprise. The customer private information data are protected as they cannot be used for commercial purposes. These data are not selected as data asset. The multimedia data for education can be used not just for this enterprise, but for multiple education enterprises for both class teaching or self-learning by individual users. In this situation, the enterprise identifies the multimedia data for education as a data asset. + +In the use case of multimedia data asset identification, the following steps may be carried out. + +- 1) Select the multimedia education data from the potential data resources. +- 2) Identify the key information elements from the current existing multimedia education data resources, which may include but is not limited to the elements indicated in Table I.1. +- 3) The key information in Table I.1 of the identified multimedia education data asset should be stored in the data asset catalogue later. More detail is given in clause I.2. + +**Table I.1 – An example of key information elements for the identified MDA** + +| Information element | Example attribute value | +|---------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Title | Network management lessons for newcomers. | +| Category | Education | +| Domain | Information and communication technology (ICT), network management | +| Content summary | This is a series of lessons for newcomers of network management, and the contents contains the introduction of network management framework, concepts, functionalities of FCAPS (fault, configuration, accounting, performance, and security management), interface technologies and applications.
There are ten lessons in this series, and each lesson is about 45 minutes. | +| Key words | Information communication technology, network management, maintenance, operation. | +| Creation date | 20 January 2021 | +| Author | Mr Tom Smith | +| Media type | Video | +| Time length | 450 minutes | +| Media size | 1 500 Megabytes | + +### I.2 Use case for MDA registration + +In the above use case example, the MDA identification process is carried out, and the education multimedia data asset are identified as a data asset from several potential data resources. In the MDA registration process, the following steps may be carried out. + +- 1) Confirm the ownership of the multimedia data asset in the enterprise using some technical means, such as digital signature, encrypted water print or block chain technology, so that some asset related properties (e.g., identity, registration time) of the multimedia data asset are added to the data asset. +- 2) To register the MDA in the MDA catalogue, besides the key element information identified in the previous process, additional information for the catalogue record should also be provided in addition to that identified in Table I.1. An example is shown in Table I.2 for such additional data asset registration information. +- 3) The operator sends a request to register this new multimedia data asset in the MDA catalogue, with the information combined from Table I.1 and Table I.2. +- 4) The manager checks the information in the registration request. If the registration is approved, a new MDA record will be created in the MDA records. + +**Table I.2 – An example of registration information for the identified MDA** + +| Attribute name | Example attribute values | +|-------------------------------|---------------------------------------------------------------------------------| +| MDA identifier | Com-ABC-edu-NM-video-20230129-0001 | +| Registration date | 29 January 2023 | +| Registration operator | Mr Micheal Zhang | +| Identification technology | Digital signature | +| Responsible party (Ownership) | Company ABC | +| Contact point | Name: Mr Tom Williams
Email: Tom_Williams@abc.com
Telephone: +00 12345678 | +| Stored material | Database | +| Stored location | Central database, with the database (DB) table name: "NMEducationData-01" | + +### I.3 Use case for MDA application + +Multimedia data asset application means the process of increasing the added value of data assets through sharing, circulation and use, in order to meet the requirements of business scenarios and organizational development. + +An MDA can be applied to business scenarios after it is established. Suppose another enterprise *Company DEF* wants to rent the network management (NM) MDA for the education of their employees. In this case an application of the MDA will be triggered. In this process, the following steps may be carried out as an example. + +- 1) After receiving the application form from *Company DEF*, the MDA operator sends a request to start the application process with the required information. +- 2) The manager checks the application information in the request. If the request is approved, an application record will be created and stored in the catalogue. For example, the application record may include the information specified in Table I.3. + +- 3) After the application is approved and the record is created, the operator will follow the agreement between *Company DEF* and *Enterprise ABC*. The operator may copy the MDA from the database to a hard drive, create a one-month licence and forward the hard drive to the contact person in *Company DEF*. + +**Table I.3 – Example of application record information for a specified MDA** + +| Attribute name | Example attribute values | +|--------------------------|-------------------------------------------------------------| +| Application record ID | AR-20240329-001 | +| MDA identifier | Com-ABC-edu-NM-video-20230129-0001 | +| Application period | 1-30 April 2024 | +| Application organization | Company DEF
Contact person: Ms Jeny Wilson | +| Application description | Rent for education class | +| Application price | \$1 000 | +| Application method | Hard drive for MDA storage, with a one-month licence | +| Application operator | Mr Michel Brown | +| Application manager | Mr Tom Williams | + +### **I.4 Use case for MDA change** + +When there are some changes happened for the multimedia data asset, an MDA change process will be triggered. In this process, the following aspects may be carried out as an example. + +- 1) There are two more lessons added to this education multimedia data asset. +- 2) The operator stores these two updated lessons in the database. +- 3) The operator updates the information according to the actual changes of the specified multimedia data asset, sends a request to update the data asset catalogue to reflect these changes. +- 4) The manager checks the updated information in the request. If it is conformant, a change record will be created and stored in the catalogue. For example, the change record may include the following information, as shown in Table I.4. + +**Table I.4 – Example of change record information for the specified MDA** + +| Attribute name | Example attribute values | +|------------------|-------------------------------------------------------------| +| Change record ID | CR-20240129-001 | +| MDA identifier | Com-ABC-edu-NM-video-20230129-0001 | +| Change date | 30 January 2024 | +| Change operator | Mr Michel Brown | +| Change reason | Two more lessons for network management education are added | +| Change manager | Mr Tom Williams | + +**Table I.4 – Example of change record information for the specified MDA** + +| Attribute name | Example attribute values | +|------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Changed content | | +| Change attributes | New value | +| Content summary | This is a series of lessons for new comers of network management, and the contents contains the introduction of network management framework, concepts, functionalities of FCAPS (fault, configuration, accounting, performance, and security management), interface technologies and applications.
There are 12 lessons in this series, and each lesson is about 45 minutes, where the last two lessons related to network management applications are added on 29 January 2024. | +| Time length | 540 minutes | +| Media size | 1 750 Megabytes | + +### I.5 Use case for MDA inventory check + +Data asset inventory check means the process of checking the status of data assets to find inconsistencies between data assets and their catalogue. If inconsistencies are found, the information in the catalogue should be updated to ensure the consistency and integrity of the data asset information. + +In the MDA inventory check process, the following steps may be carried out as an example. + +- 1) In an ordinary inventory check, the data asset manager checks the status of the multimedia data asset and finds an inconsistency between the above-mentioned multimedia data asset and the information stored in the catalogue (for example, the media size is incorrect with the real media data size.) +- 2) If there is no inconsistency between the data asset and the stored information, the operator will send a request to create an inventory check record in the catalogue, indicating the related information for this check. The information may include the one shown in Table I.5. +- 3) If an inconsistency is found in this inventory check, the operator will send another request to record the information found, as shown in Table I.6, and then a corresponding change request will be triggered. + +**Table I.5 – Example of inventory check record information for the specified MDA without inconsistency found** + +| Attribute name | Example attribute values | +|---------------------------|------------------------------------| +| Inventory check record ID | IC-20240131-001 | +| MDA identifier | Com-ABC-edu-NM-video-20230129-0001 | +| Check date | 31 January 2024 | +| Check result | Everything is OK | +| Associated change record | – | + +**Table I.6 – Another example of inventory check record information for the specified MDA with inconsistency found** + +| Attribute Name | Example attribute values | +|---------------------------|-----------------------------------------------------------------------------------------------| +| Inventory check record ID | IC-20240129-002 | +| MDA Identifier | Com-ABC-edu-NM-video-20240129-0001 | +| Check date | January 31, 2024 | +| Check result | The media size is inconsistent, and the real data size should be recorded as 1 800 Megabytes. | +| Associated change record | CR-20240129-002 | + +The associated change record CR-20240129-002 will follow the same structure as shown in Table I.4, which is not specified in this example. + +### I.6 Use cases for MDA assessment + +After a multimedia data asset is created, the enterprise can assess the MDA at any time. Suppose the enterprise arranges an assessment of the above multimedia data asset. The data asset assessment process will be triggered. In this process, the following steps may be carried out as an example. + +- 1) Usually there will be a team for the assessment of the multimedia data asset. The information related to the NM education multimedia data asset will be collected by the assessment team (for example, the quality, cost, application, etc.), and based on that information, certain rules and assessment methods (the rules and the assessment methods will be given in other specifications), the team will provide the assessment result. +- 2) Based on the above result, the operator records the related assessment information, and an assessment record will be created and stored in the MDA catalogue. For example, the assessment record may include the information shown in Table I.7: + +**Table I.7 – An example of assessment record information of the specified MDA** + +| Attribute Name | Example attribute values | +|--------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Assessment record ID | AR-20240329-001 | +| MDA Identifier | Com-ABC-edu-NM-video-20230129-0001 | +| Assessment date | 29 March 2024 | +| Assessment leader | Mr Michel Brown | +| Assessment member | Mr John Zhao, Ms Martha Liu | +| Assessment result | | +| Quality | Good | +| Cost | \$20,000 | +| Application | This asset can be used for about five years, and will be applied about 100 times in the future. | +| Valuation | \$100,000 | +| Additional information | More detailed description about this assessment can be provided here, e.g., how the cost is calculated, and how the valuation of the data asset is calculated. | + +### I.7 Use case for MDA disposal + +The data asset disposal process will be triggered when a multimedia data asset is no longer valuable, will not be used, or will be transferred to another enterprise or organization. In this process, the following steps may be carried out as an example. + +- 1) Suppose the above-mentioned NM education multimedia data asset will not be used any more. The operator will send a request to start disposing of this multimedia data asset. +- 2) The manager checks the disposal information in the request. If confirmed, a disposal record will be created and stored in the catalogue. For example, the disposal record may include the information shown in Table I.8. +- 3) If the disposal request is approved by the manager, the operator can remove the MDA from the database, and this MDA asset becomes unavailable thereafter. + +**Table I.8 – An example of disposal record information for the specified MDA** + +| Attribute Name | Example attribute values | +|--------------------|-------------------------------------------------------------------------------| +| Disposal record ID | DR-20240529-001 | +| MDA Identifier | Com-ABC-edu-NM-video-20230129-0001 | +| Disposal date | 29 May 2024 | +| Disposal operator | Mr Michel Brown | +| Disposal reason | The lessons for network management are too old and will not be used any more. | +| Disposal manager | Mr Tom Williams | + +## Bibliography + +- [b-ITU-T X.1750] Recommendation ITU-T X.1750 (2020), *Guidelines on security of big data as a service for big data service providers.* +- [b-ITU-T Y.3603] Recommendation ITU-T Y.3603 (2023), *Big data – Requirements and conceptual model of metadata for data catalogue.* +- [b-ITU-T Y.4551] Recommendation ITU-T Y.4551/F.771 (2008), *Service description and requirements for multimedia information access triggered by tag-based identification.* + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.743.26-202406-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/F/T-REC-F.743.26-202406-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c1629394c02f697f60ae954ad36bf274fa8082cb --- /dev/null +++ b/marked/F/T-REC-F.743.26-202406-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3763acd7a0ed56aa30f3f90468c1fb0cb54573eb9045bca3eddf8976b4f0f41d +size 7219 diff --git a/marked/F/T-REC-F.743.26-202406-I_PDF-E/053f1077d592e6622cd21dc4bb4cb366_img.jpg b/marked/F/T-REC-F.743.26-202406-I_PDF-E/053f1077d592e6622cd21dc4bb4cb366_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..15332aa939249f3fb5cd70b89357c0f0818540e2 --- /dev/null +++ b/marked/F/T-REC-F.743.26-202406-I_PDF-E/053f1077d592e6622cd21dc4bb4cb366_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:202aaa0971ccc4e5f3a4f9062bf015cb735f823240efc1d3c0e99922d74c834f +size 69862 diff --git a/marked/F/T-REC-F.743.26-202406-I_PDF-E/7e670a2b556b53ea9002dfff3a420e08_img.jpg b/marked/F/T-REC-F.743.26-202406-I_PDF-E/7e670a2b556b53ea9002dfff3a420e08_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9875ba792bd922f6a773759c491994ef0dfd5a34 --- /dev/null +++ b/marked/F/T-REC-F.743.26-202406-I_PDF-E/7e670a2b556b53ea9002dfff3a420e08_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d1dd83d0ef904c5cce87684d21829ff6a2ddd41d1d7a3c2afc6e616259d56668 +size 48804 diff --git a/marked/F/T-REC-F.743.26-202406-I_PDF-E/a234352dfaccdc24745c88eef7724cc6_img.jpg b/marked/F/T-REC-F.743.26-202406-I_PDF-E/a234352dfaccdc24745c88eef7724cc6_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7a61d15d9e6eb6019e1aa56cab62c8b45effc005 --- /dev/null +++ b/marked/F/T-REC-F.743.26-202406-I_PDF-E/a234352dfaccdc24745c88eef7724cc6_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:278e52d08dfc159637f0ab0e20abdf14de7ca71788e7d66869699b0138716ed8 +size 98035 diff --git a/marked/F/T-REC-F.743.29-202503-I_PDF-E/raw.md b/marked/F/T-REC-F.743.29-202503-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..92be6637609d5112b6901ca50a7a74056138f5d0 --- /dev/null +++ b/marked/F/T-REC-F.743.29-202503-I_PDF-E/raw.md @@ -0,0 +1,524 @@ + + +# Recommendation **ITU-T F.743.29 (03/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +## **Requirements and framework of model generalization system in intelligent video surveillance** + +![ITU logo](6ed175c791b5e156d9c98a8dbcc3318c_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.743.29 + +# Requirements and framework of model generalization system in intelligent video surveillance + +## Summary + +Recommendation ITU-T F.743.29 addresses the issue of industry research and development progressing without a unified standard framework. This Recommendation provides an overview of the model generalization system in intelligent video surveillance and specifies the requirements and framework for model generalization system. + +With the accelerating integration of artificial intelligence (AI) and the real economy, video analysis capabilities based on AI algorithms have become a critical requirement for many vertical applications in video surveillance. + +With the popularity of artificial intelligence research, AI algorithms have been implemented in intelligent video surveillance (IVS) system. However, these AI models face several challenges when implemented in IVS systems: + +- 1) the generalization ability of AI models is weak, and their algorithm performance is poor across different video surveillance scenarios; +- 2) Data quality is crucial for the training and implementation of small AI models. + +There is a lack of high-quality labelled data in the IVS system. + +To better promote the application of IVS, ensure the application effect, and promote the research of IVS technology, many companies have launched IVS systems with high functional integration based on large-scale, pre-trained, generalized artificial intelligence models to meet the diverse needs of the IVS. This has led to a highly complex, heterogeneous, and fragile ecosystem. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.743.29 | 2025-03-16 | 21 | 11.1002/1000/16225 | + +## Keywords + +Fine-tune, intelligent video surveillance, model generalization system, model inference, pre-training. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2025 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|------------|---------------------------------------------------------------------------------------------------|----| +| 1 | Scope ..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 2 | +| 4 | Abbreviations and acronyms ..... | 2 | +| 5 | Conventions ..... | 2 | +| 6 | Overview of model generalization system in the intelligent video surveillance..... | 3 | +| 6.1 | Background and introduction of model generalization system in intelligent video surveillance..... | 3 | +| 7 | Requirements of model generalization system in intelligent video surveillance ..... | 3 | +| 7.1 | Requirements for data collection..... | 3 | +| 7.2 | Requirements for data analysis..... | 4 | +| 7.3 | Requirements for model pre-training, fine-tune, and inference ..... | 4 | +| 7.4 | Requirements for system evaluation and security ..... | 4 | +| 8 | Framework of model generalization system in intelligent video surveillance ..... | 4 | +| 8.1 | Internet data collection service (IDCS) ..... | 5 | +| 8.2 | Pre-training unit (PTU)..... | 6 | +| 8.3 | Fine tune unit (FTU)..... | 7 | +| 8.4 | Model inference service (MIS)..... | 7 | +| 8.5 | Data analysis unit (DAU) ..... | 8 | +| 8.6 | Data storage and service (DSS)..... | 8 | +| 8.7 | Pre-training management (PTM)..... | 9 | +| 8.8 | Fine tune management (FTM)..... | 9 | +| Appendix I | – Use cases of model generalization system in intelligent video surveillance ..... | 10 | +| I.1 | Use case of target recognition ..... | 10 | +| I.2 | Use case of traffic data analysis ..... | 10 | +| I.3 | Use case of image scene classification..... | 10 | +| | Bibliography..... | 12 | + + + +# Recommendation ITU-T F.743.29 + +## Requirements and framework of model generalization system in intelligent video surveillance + +# 1 Scope + +This Recommendation identifies the requirements and framework of model generalization system in intelligent video surveillance (IVS). + +The scope of this Recommendation includes: + +- Overview of model generalization system in IVS. +- Requirements of model generalization system in IVS. +- Framework of model generalization system in IVS. + +Any individual in the video surveillance capture must be anonymised to protect their identity. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +- [[ITU-T H.222.0](#)] Recommendation ITU-T H.222.0 v10 (2025), *Information technology - Generic coding of moving pictures and associated audio information: Systems*. +- [[ITU-T H.264](#)] Recommendation ITU-T H.264 (V15) (2024), *Advanced video coding for generic audiovisual services*. +- [[ITU-T H.265](#)] Recommendation ITU-T H.265 (V10) (2024), *High efficiency video coding*. +- [[ITU-T H.626](#)] Recommendation ITU-T H.626 (V2) (2019), *Architectural requirements for a video surveillance system*. +- [[ITU-T H.626.5](#)] Recommendation ITU-T H.626.5 (V2) (2022), *Architecture for intelligent video surveillance systems*. +- [[ITU-T T.871](#)] Recommendation ITU-T T.871 (2011), *Information technology - Digital compression and coding of continuous-tone still images: JPEG File Interchange Format (JFIF)*. + +# 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 intelligent video diagnosis** [b-ITU-T F.743.1]: Automatic recognition of video image failures, such as snowflakes, scrolling or fuzzy video, partial colour, picture freeze, gain imbalance, pan/tilt/zoom (PTZ) control problems, and video signal loss. + +**3.1.2 video and image information** [ITU-T H.626.5]: Video clips, images, related files and content description information. + +**3.1.3 video surveillance system** [ITU-T H.626]: A telecommunication service focusing on video (but including audio and image) application technology, which is used to remotely capture multimedia (such as audio, video, image, alarm signal.) and present them to the end user in a user friendly manner, based on managed broadband network with ensured quality, security and reliability. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following term: + +**3.2.1 model generalization system:** A system that includes large-scale, pre-trained, generalized artificial intelligence models. The model generalization system (MGS) is a video surveillance system based on the ability of large-scale, pre-trained, generalized artificial intelligence models, which can process both traditional images and artificial intelligence generated content (AIGC) images. The visual tasks that MGS can implement include image classification, video classification, object detection, semantic segmentation, instance segmentation, object tracking, retrieval, image or video understanding (such as questions and answering about image or video content), image editing (such as completion, style transfer, and extension). + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|-------------------------------------------| +| AI | Artificial Intelligence | +| AIGC | Artificial Intelligence Generated Content | +| DAU | Data Analysis Service Unit | +| DSS | Data Storage and Service | +| FTM | Fine Tune Management | +| FTS | Fine Tune Service | +| FTU | Fine Tune Unit | +| IAS | Intelligent Application Service | +| IDCS | Internet Data Collection Service | +| IVM | Intelligent Video Management | +| IVS | Intelligent Video Surveillance | +| MGS | Model Generalization System | +| MIS | Model Inference Service | +| PTM | Pre-training Management | +| PTS | Pre-training Service | +| PTU | Pre-training Unit | + +## 5 Conventions + +In this Recommendation, the following conventions apply: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted, if conformance to this Recommendation is to be claimed. +- The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. + +- The keywords "can optionally" indicate an optional requirement which is permissible, without implying any sense of being recommended. + +# **6 Overview of model generalization system in the intelligent video surveillance** + +## **6.1 Background and introduction of model generalization system in intelligent video surveillance** + +With the implementation of artificial intelligence algorithms in industry and business, many high-performance artificial intelligence (AI) algorithms have been implemented in intelligent video surveillance (IVS). However, it faces the following challenges: + +- 1) The amount of data in high-quality labelled datasets is insufficient; +- 2) Data labelling is a very expensive task; +- 3) The demand for AI experts is growing, but there is a shortage of them; +- 4) The application scenarios of the model are not broad. + +Based on the above reasons, the performance of the artificial intelligence algorithm model in IVS is currently suboptimal. Many companies have launched highly integrated IVS systems based on large-scale, pre-trained, generalized AI models to meet the performance requirements of IVS. This has resulted in a highly complex, heterogeneous, and fragile ecosystem. + +For example, in the application scenario, if a restaurant wants to detect smoke and flames in the kitchen to reduce the risk of fire, it can only customize the fire recognition and warning algorithm for the video surveillance application scenario. However, challenges such as "lack of algorithm training data", "no AI experts", and "high costs" make it difficult to implement an effective IVS system in this scenario. If an intelligent video surveillance system that supports model generalization is implemented, the restaurant does not need to solve the problems of data processing, algorithm training, and recognition accuracy, and can quickly implement flame recognition and detection. The model generalization system (MGS) in IVS includes data collection, data analysis, model pre-training, fine-tune and other capabilities, which enhance the generalization performance of the IVS system. The MGS can effectively address problems such as "lack of algorithm training data", "no AI experts" and "high costs". Thanks to these advantages, the IVS system has better performance. + +# **7 Requirements of model generalization system in intelligent video surveillance** + +- In this clause, requirements specific to the model generalization system (hereafter "the system") for intelligent video surveillance are identified. + +### **7.1 Requirements for data collection** + +The initial data is not only sourced from the intelligent video surveillance system but also requires a large amount of video data from the Internet or stored in other locations. Unstructured video data from application scenarios can be analysed and processed into standard formats. + +#### **7.1.1 General requirements** + +- It is recommended that the sample data comply with the required image and video data formats. + +#### **7.1.2 Requirements of data collection method** + +- The system is required to support video and image collection from the IVS, the collection method is defined in [ITU-T H.626.5]; +- The system is required to support single and batch sampling of local video and image files from the Internet or local storage servers. + +#### **7.1.3 Requirements of sample data quality** + +- The sample data is recommended to support the MGS, including but not limited to image resolution. + +#### **7.1.4 Requirements of sample data format** + +- The sample data format is required to support H.264 [ITU-T H.264] and H.265 [ITU-T H.265] encoded video data in MP4 [ITU-T H.222.0] format; +- The sample data format is required to support JPEG [ITU-T T.871] and the portable network graphics (PNG) format image data. + +#### **7.1.5 Requirements of data pre-processing** + +- The system is recommended to support cleaning and enhancement of video and image data. Data cleaning refers to the process of fixing or removing incorrect, corrupted, incorrectly formatted, duplicate, or incomplete data within a dataset. + +### **7.2 Requirements for data analysis** + +- The system is required to support for video and image data processing and analysis, including the extraction of useful feature information. +- The system is required to support data analysis of video and image data. + +### **7.3 Requirements for model pre-training, fine-tune, and inference** + +- The system is required to incorporate hardware components such as CPU and data storage. +- The system is recommended to support models with generalization capabilities, including image and video generation. +- The system is recommended to support data augmentation techniques to increase the diversity and richness of data. +- The system is recommended to support fusion of multiple models to improve the performance of video surveillance models. +- The system is recommended to support a variety of tasks, such as target detection, behaviour recognition, motion tracking, anomaly detection, and video retrieval. +- The system is recommended to support transfer learning, active learning and meta-learning. + +### **7.4 Requirements for system evaluation and security** + +- The system is required to support video quality evaluation, including assessment of video image resolution. +- The system is required to support video storage and protection mechanism. +- The system is recommended to support video transmission channels with security capabilities. +- The system is recommended to support data augmentation techniques to increase the diversity and richness of data. +- The system is required to support anomaly detection and alarm functions, as well as a mechanism for handling alarm events. + +## **8 Framework of model generalization system in intelligent video surveillance** + +Figure 1 shows the framework of the model generalization system and its relations with IVS. The intelligent video unit in IVS system, as defined in [ITU-T H.626.5], is updated to model generalization system in this Recommendation. + +The IVS system is a traditional system that processes conventional images and video. MGS is an updated system based on IVS that processes both traditional images and artificial intelligence generated content (AIGC) images. + +Centre management unit (CMU), service control unit (SCU), media distribution unit (MDU) and media storage unit (MSU) are defined in [ITU-T H.626]. + +Intelligent premises unit (IPU), intelligent customer unit (ICU), intelligent video management (IVM) and intelligent application service (IAS) are defined in [ITU-T H.626.5]. + +The system of model generalization is composed of Internet data collection service (IDCS), pre-training management (PTM), pre-training unit (PTU), fine tune management (FTM), fine tune unit (FTU), data analysis unit (DAU), data storage and service (DSS) and model inference service (MIS). + +![Figure 1 – System framework diagram showing the architecture of IVS system with MGS and VS system.](d26959f4514c26ca19c3d6f00da85956_img.jpg) + +The diagram illustrates the system framework, divided into two main components within a dashed purple border labeled "IVS system with MGS". + +- VS system (blue box):** Contains three sub-sections: + - CCS:** Centre management unit (CMU) and service control unit (SCU). + - MS:** Media distribution unit (MDU) and media storage unit (MSU). + - IPU and ICU:** Intelligent premises unit and intelligent customer unit, each with a PIV (Personal Identification Value) circle. +- Model generalization system (orange dashed box):** Contains: + - IAS (Intelligent application service) and IVM (Intelligent video management):** Located at the top. + - DAU (Data analysis unit) and DSS (Data storage and service):** Located on the left side. + - PTS (Pre-training system):** Contains PTM (Pre-training management) and PTU (Pre-training unit). + - FTS (Fine tune system):** Contains FTM (Fine tune management) and FTU (Fine tune unit). + - IDCS (Internet data collection service) and MIS (Model inference service):** Located at the bottom. + +F.743.29(25) + +Figure 1 – System framework diagram showing the architecture of IVS system with MGS and VS system. + +**Figure 1 – System framework** + +### 8.1 Internet data collection service (IDCS) + +Internet data collection service typically includes local data and Internet data collection module. Internet data collection module includes image/video capture function, encoding function, and decoding function. Internet data collection service can collect data from both the Internet and local data. The function modules of Internet data collection service are shown in Figure 2. + +![Figure 2 – Function modules of IDCS diagram showing the internal structure of the Internet data collection service.](c78c2eefd86269d1740ab85a916f24f2_img.jpg) + +The diagram shows the function modules of the Internet data collection service (IDCS) within a rectangular border. + +- Internet data collection module:** The top section, containing: + - Image/video capture function:** A central box. + - Encoding function and Decoding function:** Two boxes below the capture function. +- Local data collection module:** A box at the bottom of the IDCS structure. + +F.743.29(25) + +Figure 2 – Function modules of IDCS diagram showing the internal structure of the Internet data collection service. + +**Figure 2 – Function modules of IDCS** + +The image and video capture function has the ability to capture multimedia information such as images and videos from the Internet. The captured images and videos are then sent to the encoding function. + +The encoding function performs feature extraction on the encoded image to obtain a feature map, which includes multiple feature points. Based on the feature parameters of different regions in the feature map, key feature regions are determined from the feature map, where the feature parameters are used to represent the update of feature points. The key feature regions are then encoded to generate the bitstream, which is transmitted to the edge servers. Based on the feature parameters of different regions within the feature map, key feature regions are identified, and the feature map is divided into multiple feature blocks. According to the feature parameters of each feature block, key feature blocks are identified from these, and are considered equivalent to the key feature regions. For each feature block, the gradient value of each feature point in the feature block is determined, and the gradient value corresponding to the feature block is determined based on the gradient value of each feature point. This block-level gradient value is treated as the feature parameter of the feature block. + +The decoding function receives the bitstream and decodes it into a reconstructed video. The decoding function receives an encoded bitstream, which is generated by encoding the key feature regions in the feature map of the target image. The feature map is created through feature extraction from the target image, and the key feature regions are identified based on the feature parameters of different regions in the feature map. The feature parameters describe changes in feature points. The encoded bitstream is decoded to obtain a feature map of the target image. The encoding bitstream includes an encoding block and embedding position information, wherein the encoding block is obtained by encoding the key feature area, while the embedding position information is obtained by encoding the position information of the key feature area in the feature map. The encoding bitstream is decoded to obtain a feature map of the target image, and the encoding block and embedding position information are input into a pre-trained decoding model. Based on the decoding function, the encoding block and the embedding position information are decoded to obtain the feature map. + +### 8.2 Pre-training unit (PTU) + +Pre-training unit typically includes data processing module and machine learning module. The data processing module can preprocess various types of data, while the machine learning module is used to perform calculations using various machine learning models. + +PTU receives the training set, which includes sample images of multiple categories. Each sample image corresponds to a category, and the label of the sample image is the category corresponding to the sample image. + +Text is generated based on category information. A text encoder extracts text features for each text. These texts correspond one-to-one to the multiple categories, and each text includes the category corresponding to each text. + +For each image in the training sample set, the following operations are performed to determine its loss: + +- Each image is input into the network model to be trained; +- The network extracts image features from the region containing the target object in each image; +- Based on these image features and the text features of each text in the multiple texts, corresponding similarities are calculated; +- The loss function of each image is determined by multiple similarities corresponding to the image features. + +Using the losses from all images in the training sample set and a preset loss function, the parameters of the network model to be trained are optimized. + +In the loss function, the weight of the loss sub-function corresponding to the first pre-training sample subset is lower than the weight of the loss sub-function corresponding to at least one second sample subset. + +The loss function includes both: a loss sub-function for the pre-training first sample subset, and the loss sub-function for each second sample subset. + +### **8.3 Fine tune unit (FTU)** + +Fine tune unit typically includes machine learning module and task module. Machine learning module includes image encoder model, text encoder model and classifier model. The task module includes intelligent visual algorithms, i.e., it can perform target classification and detection. + +The text encoder model is configured to obtain the first text feature vector corresponding to each prompt template in the multiple prompt templates based on the input multiple prompt templates. Each prompt template includes text prompt information of an image category. + +The image encoder model is used to extract image feature vectors from the input image to be classified. + +The classifier model is configured to identify the category of the image to be classified based on the similarity between the image feature vector and the first text feature vector corresponding to each prompt template. + +Multiple prompt templates are input into the text encoder to generate a first text feature vector for each prompt template in multiple prompt templates. Each prompt template includes text prompt information for an image category. + +The image to be classified is input into an image encoder to obtain image feature vectors. The image category to which the image to be classified belongs is determined based on the similarity between the image feature vector and the first text feature vector corresponding to each prompt template. + +The image to be classified is the image frame in the video. The image categories include preset scene categories used to identify the location information of the image acquisition device for obtaining the video. + +Multiple prompt templates are input into the text encoder to obtain the first and second text feature vectors corresponding to each prompt template among the multiple prompt templates. Based on the statistical characteristics of probability distribution, the first text feature vector is sampled from the probability distribution followed by multiple components in the second text feature vector. + +The image category is identified based on the similarity between the image feature vector and the first text feature vector corresponding to each prompt template. The distance between the image feature vector and each first text feature vector is calculated. Based on the distance, the probability that the image to be classified belongs to each category is estimated. The image category with the highest probability is identified as the category of the image to be classified. + +### **8.4 Model inference service (MIS)** + +Model inference service typically includes data acquisition module and model inference module. + +The data acquisition module is used to obtain user input data, while the model inference module inputs user data and outputs inference results through the model. The category words are embedded in the pre-saved scene supplementary template to generate scene supplementary text. The scene supplementary text contains scene information of the category words of the object to be detected, and the text is updated using the scene supplementary text. + +### 8.5 Data analysis unit (DAU) + +The image of the location information to be determined and each preset prompt template are input into the location recognition model. Using the image encoder in the location recognition model, the image feature vector of the image is determined. Based on the text encoder in the location recognition model and multiple preset scene categories, the text embedding feature vectors corresponding to each preset scene category are identified, i.e., the text encoder can be based on transformer. + +The prompt learning module in the location recognition model processes the text embedding feature vectors to generate the scene feature vectors corresponding to each scene category, i.e., the prompt learning module can be based on the Gaussian distribution probability. Each preset prompt template includes corresponding location information. + +The location information of the camera corresponding to the image is confirmed by the similarity between the image feature vector and each prompt template feature vector. + +The similarity between the image feature vector and the feature vector of each sample scene is confirmed. + +The scene feature vector corresponding to the maximum similarity is determined as the target scene feature vector, and the target scene category corresponding to the target scene feature vector is identified as the location information of the camera corresponding to the image. + +Before inputting each preset prompt template into the location recognition model, for each prompt template, a preset number of input prompt words are obtained. The prompt word is the prompt information corresponding to the location information. Based on the preset number of prompt words and the preset basic prompt template, the corresponding prompt template is confirmed. + +### 8.6 Data storage and service (DSS) + +The data storage and service module in video surveillance is a component responsible for managing and providing video data storage, retrieval, and related services. + +This module manages data storage in the video surveillance system. It can include local storage devices, network storage, or cloud storage, etc. This module needs to handle operations such as writing, reading, deleting, and updating data, as well as managing the allocation and release of storage space. + +To ensure quick retrieval and querying of video data, the data storage and service module needs to establish an appropriate index and metadata management system. This includes indexing the timestamps, camera information, geographic location, target categories, etc. of video files for on-demand retrieval and querying of specific video segments. + +Based on multiple categories, multiple texts are generated. To generate text, at least one expansion field and the corresponding category for each text are required. The expanded field includes one or more of the following elements: the method of obtaining images in the training sample set, and the scene to which the images belong in the training sample set. The training sample set includes multiple subsets of samples. Multiple sample subsets consist of the pre-training first sample subset and at least one second sample subset. The acquisition method or scene of sample images in the pre-training first sample subset is different from that in the second sample subset. + +Moreover, the total number of sample images in the pre-training first sample subset is greater than that in at least one pre-training second sample subset. Multiple sample subsets include multiple pre-training second sample subsets. The scene of an image in any second sample subset is different from the scene of an image in any other second sample subset. + +Providing intelligent analysis and services on video surveillance data is an important function of the data storage and service module. It can be integrated with data processing and analysis modules to provide intelligent visual capabilities. + +### **8.7 Pre-training management (PTM)** + +Pre-training management (PTM) is a subsystem unit within the pre-training system (PTS). It functions when multiple pre-processed objects are present in the system. + +First, the PTM can obtain media, video and image information resource to generate a list. Then, according to specific rules, PTM can provide the input order for the PTU. The rules can be expert-defined rules or machine-generated rules. Finally, PTM can manage pre-training tasks, including data processing, machine learning, and post-processing tasks. Besides, the PTM also acts as a queue storage when there is an excess of data to be processed. + +### **8.8 Fine tune management (FTM)** + +Fine-tune management (FTM) is a subsystem unit within the fine tune system (FTS). It operates when multiple fine-tune objects are present in the system. + +First, the FTM can obtain fine-tune data lists, which including both local and Internet data. Then, based on specific rules, the FTM can provide the input order for the FTU. + +## Appendix I + +## Use cases of model generalization system in intelligent video surveillance + +(This appendix does not form an integral part of this Recommendation.) + +### I.1 Use case of target recognition + +In vehicle recognition application scenarios, the generalized model can be applied to areas such as campuses and factories to monitor vehicle entry and exit. + +In the fire recognition application scenario, the system supporting model generalization model in intelligent video surveillance (IVS) is applied to open fire detection in multiple parks. Once a fire is detected, an alarm is immediately activated to ensure the safety of the park. The use cases of target recognition, including smart city systems and fire recognition, are shown in Figure I.1. + +![Figure I.1: A collage of 14 images demonstrating target recognition. The top row shows five images of street scenes with red bounding boxes around vehicles. The bottom row shows nine images of fire scenes with purple bounding boxes around the flames.](a8e5c2ac336eb43cda4e333ea9c73237_img.jpg) + +Figure I.1: A collage of 14 images demonstrating target recognition. The top row shows five images of street scenes with red bounding boxes around vehicles. The bottom row shows nine images of fire scenes with purple bounding boxes around the flames. + +Figure I.1 – Use cases of target recognition including smart city systems and fire recognition + +### I.2 Use case of traffic data analysis + +Smart city infrastructure is the core foundation of the future city. As the "nerve centre" of the smart city, it can efficiently gather massive amounts of data. In the fields of traffic analysis and pedestrian flow analysis, the system supporting model generalization model in IVS can be used to build an information-driven city and enable cross-departmental coordination and command. The use case of traffic analysis is shown in Figure I.2. + +![Figure I.2: A screenshot of a traffic analysis interface. It shows a live video feed of a street with cars, overlaid with green and blue bounding boxes. A sidebar on the right displays a 'Traffic Status' panel with various data points and a list of vehicle types.](4a104be5f84f688417d8c222ec4ce4fa_img.jpg) + +Figure I.2: A screenshot of a traffic analysis interface. It shows a live video feed of a street with cars, overlaid with green and blue bounding boxes. A sidebar on the right displays a 'Traffic Status' panel with various data points and a list of vehicle types. + +Figure I.2 – The use case of traffic analysis + +### I.3 Use case of image scene classification + +Image scene classification aims to label entire images with a set of semantic categories based on their contents. It is a fundamental and challenging problem in computer vision. Due to big intraclass diversity, high interclass similarity, and large variance of object/scene scales, the scene classification + +algorithm has not yet achieved satisfactory accuracy. The use case of image scene classification is shown in Figure I.3. + +In the images scene classification application scenario, the generalized model can be applied to semantic tagging of surveillance camera images. For example, it can categorize surveillance camera images into residential, commercial, or industrial areas. + +![A 2x4 grid of eight images illustrating different scenes for image scene classification. The scenes are: Library, Dormitory, Mall, Gas station, Vegetable market, Swimming pool, Playground, and Road.](93afce28d7dec5b2202789b31b4ef8ab_img.jpg) + +The figure displays eight distinct scenes for image scene classification, arranged in a 2x4 grid. Each scene is represented by a photograph and a corresponding label below it: + +- Library:** A photograph showing rows of bookshelves filled with books. +- Dormitory:** A photograph of a room with two wooden bunk beds and a window. +- Mall:** A photograph of a modern, multi-level shopping mall interior with people and escalators. +- Gas station:** A photograph of a gas station at night, illuminated by its own lights. +- Vegetable market:** A photograph of a busy outdoor market with people selling various vegetables. +- Swimming pool:** A photograph of a swimming pool at night, surrounded by lounge chairs and palm trees. +- Playground:** A photograph of a large, modern playground with a green field and various play structures. +- Road:** A photograph of a wet road at night, with cars and streetlights visible. + +A 2x4 grid of eight images illustrating different scenes for image scene classification. The scenes are: Library, Dormitory, Mall, Gas station, Vegetable market, Swimming pool, Playground, and Road. + +F.743.29(25) + +**Figure I.3 – The use case of image scene classification** + +## Bibliography + +[b-ITU-T F.743.1] Recommendation ITU-T F.743.1 (2015), *Requirements for intelligent visual surveillance*. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.743.3-201607-I_PDF-E/raw.md b/marked/F/T-REC-F.743.3-201607-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..9176871ac1fd126e96fc9b2efb674f94cfea5c17 --- /dev/null +++ b/marked/F/T-REC-F.743.3-201607-I_PDF-E/raw.md @@ -0,0 +1,374 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.743.3** + +(07/2016) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Requirements for visual surveillance system +interworking** + +Recommendation ITU-T F.743.3 + +ITU-T + +![ITU logo: A globe with a red lightning bolt and the text 'ITU International Telecommunication Union'.](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with a red lightning bolt striking across it. To the right of the globe, the text "ITU" is written in a bold, blue, sans-serif font, with "International Telecommunication Union" in a smaller, blue, sans-serif font underneath it. + +ITU logo: A globe with a red lightning bolt and the text 'ITU International Telecommunication Union'. + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multidestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +## Recommendation ITU-T F.743.3 + +# Requirements for visual surveillance system interworking + +## Summary + +Recommendation ITU-T F.743.3 defines the interworking requirements for visual surveillance systems (VSSs). The visual surveillance system interworking (VSSI) mechanism can achieve cross-system scheduling of multimedia (such as video, audio and image) and enables resource and data sharing of different VSSs. Recommendation ITU-T F.743.3 provides the service scenarios and functional requirements for VSSI. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.743.3 | 2016-07-14 | 16 | 11.1002/1000/12896 | + +## Keywords + +Camera, interworking, playback, remote control, visual surveillance. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2016 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|---------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Scenarios..... | 3 | +| 6.1 Cross-system real-time surveillance..... | 3 | +| 6.2 Cross-system surveillance video download and playback ..... | 3 | +| 6.3 Alarm across different systems ..... | 4 | +| 7 Requirements for visual surveillance system interworking..... | 4 | +| 7.1 User requirements..... | 4 | +| 7.2 Service requirements ..... | 5 | +| 7.3 Security requirements ..... | 6 | +| 7.4 Management requirements ..... | 6 | +| 7.5 Quality of service and quality of experience requirements ..... | 7 | +| 7.6 Scalability requirements ..... | 7 | +| Bibliography..... | 8 | + + + +# Requirements for visual surveillance system interworking + +# 1 Scope + +This Recommendation describes the service scenarios, functional requirements and reference model for visual surveillance system interworking (VSSI), based on the requirements and architectures defined by [ITU-T F.743], [ITU-T H.626], [ITU-T H.626.1] and [ITU-T H.627]. + +A visual surveillance (VS) service is a telecommunication service focusing on video (and audio) application technology, which is used to remotely capture multimedia (such as audio, video, image and various alarm signals), and present them to end users in a friendly manner (including accessibility aspects), based on a broadband network with ensured quality, security and reliability. The VSSI mechanism can achieve cross-system scheduling of multimedia (such as video, audio and image) and enables the resource and data sharing of different visual surveillance systems (VSSs). + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +- [ITU-T F.743] Recommendation ITU-T F.743 (2009), *Requirements and service description for visual surveillance*. +- [ITU-T H.626] Recommendation ITU-T H.626 (2011), *Architectural requirements for visual surveillance*. +- [ITU-T H.626.1] Recommendation ITU-T H.626.1 (2013), *Architecture for mobile visual surveillance*. +- [ITU-T H.627] Recommendation ITU-T H.627 (2012), *Signalling and protocols for visual surveillance*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +- 3.1.1 application** [b-ITU-T Y.101]: A structured set of capabilities, which provide value-added functionality supported by one or more services. +- 3.1.2 customer** [b-ITU-T M.60]: A customer is an entity which receives services offered by a service provider based on a contractual relationship. It may include the role of a network user. +- 3.1.3 customer unit** [ITU-T H.626]: A device located at the customer part of a visual surveillance system and used to present multimedia information (such as audio, video, image, alarm signal, etc.) to the end user. + +**3.1.4 premises unit** [ITU-T H.626]: A device located at the remote part of a visual surveillance system and used to capture multimedia information (such as audio, video, image, alarm signal, etc.) from a surveilled object. + +**3.1.5 service** [b-ITU-T Y.101]: A structure set of capabilities intended to support applications. + +**3.1.6 visual surveillance** [ITU-T H.626]: A telecommunication service focusing on video (but including audio) application technology, which is used to remotely capture multimedia (such as audio, video, image, alarm signals, etc.) and present them to the end user in a user-friendly manner, based on a managed broadband network with ensured quality, security and reliability. + +## **3.2 Terms defined in this Recommendation** + +This Recommendation defines the following term: + +**3.2.1 visual surveillance system interworking platform**: A series of devices and subsystems located at the centred part of the interworking visual surveillance systems. The platform is used to integrate the capabilities of different visual surveillance systems and enable the resource and data sharing of different visual surveillance systems. + +# **4 Abbreviations and acronyms** + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|-----------------------------------------| +| CMU | Centre Management Unit | +| CU | Customer Unit | +| PTZ | Pan/Tilt/Zoom | +| PU | Premises Unit | +| QoE | Quality of Experience | +| QoS | Quality of Service | +| NAT | Network Address Translation | +| NVR | Network Video Recorder | +| PC | Personal Computer | +| SCU | Service Control Unit | +| SIG | System Interworking Gateway | +| VS | Visual Surveillance | +| VSS | Visual Surveillance System | +| VSSI | Visual Surveillance System Interworking | + +# **5 Conventions** + +In this Recommendation: + +- The keywords "is required to" indicate a requirement that must be strictly followed and from which no deviation is permitted if conformance to this recommendation is to be claimed. +- The keywords "is recommended" indicate a requirement that is recommended, but which is not absolutely required. Thus this requirement need not be present to claim conformance. + +# 6 Scenarios + +This clause describes typical service scenarios illustrating visual surveillance system interworking (VSSI) and deriving its service requirements. Each visual surveillance system (VSS) has its own centre management unit (CMU), and different VSSs have different CMUs. + +## 6.1 Cross-system real-time surveillance + +### *Case 1: Cross-system interoperation* + +Bus company A, traffic management department B and police department C have their own VSSs. The bus company installs its VSS in its buses and bus stations. The traffic management department installs its VSS on the main traffic roads of the city. The police department also installs its VSS on some streets of city. Those VSSs can be connected by the VSSI mechanism. Customer Li is a member of the police department. Li can directly view the real-time video of the VSSs of A, B and C so that he can find and handle any emergency by synthesizing a wide range of information. + +- Step 1: The VSSs of A, B and C are interconnected through a VSSI mechanism. The VSSs of A and B open the resource access privilege for the user from C. +- Step 2: Li logs into the VSS of C. There, he can view a list that displays detailed information for each camera, including the camera name and position of the VSSs of A, B and C. To track a suspected target trajectory, Li watches the multiple real-time videos from the three VSSs by clicking the corresponding cameras. +- Step 3: While watching a surveillance video, Li can make a screenshot of it or begin to record it at any time. +- Step 4: If Li is not satisfied with the positioning, he can click the direction buttons on the screen to pan/tilt/zoom (PTZ) the camera to meet his need. +- Step 5: If an abnormal event occurs in the monitoring region, the corresponding VSS can send an alert message to Li. Then Li can switch to that real-time video to observe the developing situation. + +### *Case 2: Multiple system unified access* + +The traffic management departments of city A, city B and city C have deployed their respective VSSs in the three cities. Administrator Bob of the provincial/state traffic management department wants to view the current traffic situations in all cities. + +- Step 1: The VSSs of A, B and C are connected to a VSSI platform, which provides a resource access privilege for the user of the platform. +- Step 2: Bob logs into the VSSI platform via his PC. There, he can view a list that displays detailed information for each camera of the three VSSs. To view the current traffic situations in the three cities, Bob simultaneously watches the multiple real-time videos from different VSSs by clicking the corresponding cameras. +- Step 3: If Bob is not satisfied with the positioning, he can click the direction buttons on the screen to pan/tilt/zoom (PTZ) the camera to meet his need. + +## 6.2 Cross-system surveillance video download and playback + +When a burglary happens in a community, the police officer will fetch the recorded surveillance video of the community and the nearby roads to find a clue about the suspicion. + +- Step 1: Li logs into the VSS of the police department via his computer. +- Step 2: The VSS provides multiple query conditions for retrieving video recordings, e.g., camera position, video starting time and video duration. Li searches the related video recordings made before and after the incident time in the VSSs of the traffic management department and police department, and retrieves some relevant results. Li replays these video recordings + +to checking for any suspicious behaviour. Li can play these recordings at normal speed or fast forward/backward, pause, search a specified position, stop, etc. + +Step 3: To further investigate the evidence, Li downloads some key video records from the VSS of the traffic management department and send the video copies to other case investigation departments. + +## **6.3 Alarm across different systems** + +Administrator Bob in system A needs to receive alarm signals when premises unit (PU) of system B is triggered by some kind of event, but the customer of system A and the PU of system B are not within the same system, this scenario can be realized with the interaction of the system interworking gateway (SIG). The simplified procedure for alarm across different systems is as follows (see clause 6.1, Case 2): + +Step 1: Bob connects the service control unit (SCU) in system A to require an alarm signal when the PU of system B is triggered by some predefined event. + +Step 2: The SCU in system A performs lookup process (with the interaction of the SIG) and finds the PU of system B. + +Step 3: Bob sends a request to the SCU in system A to subscribe to the alarm service, the SCU in system A transports this request to the SCU in system B with the interaction of the SIG in system B, which returns a success response. + +Step 4: The PU of system B sends an alarm signal to the SCU in system B when it detects a predefined event. + +Step 5: The SCU in system B notifies Bob of the alarm event with the interaction of the SIG, SPU and SCU of system A, if necessary. + +# **7 Requirements for visual surveillance system interworking** + +## **7.1 User requirements** + +There are two types of user. One is the interworking VSS type who can login to one of the VSSs connected to the VSSI platform. The other is the VSSI platform user who can login to the VSSI platform, but cannot directly login any VSS connected to the VSSI platform. + +### **7.1.1 User level requirements** + +- USR-001: A VSSI platform is recommended to classify users into several levels. Users of different levels have different operational rights; high-level users have more operation privileges than low-level ones. Low-level users can do basic operations that meet the following basic requirements, and high-level users can do advanced operations that meet the following advanced requirements. For example, users of level 1 can only watch the cross-system real-time video, while users of level 2 can control the PUs of other system, as well as watch the cross-system real-time video. + +### **7.1.2 Basic interworking visual surveillance system user requirements** + +- USR-002: An interworking VSS is required to support the VSS user view of information about a PU of other interworking VSS, including name, status, capabilities, location and other information about the PU. +- USR-003: An interworking VSS is required to support the presentation of real-time video captured by the PU of other interworking VSSs for its users. +- USR-004: An interworking VSS is required to support VSS users to order and cancel different services of other interworking VSS, including basic and value-added services. + +### **7.1.3 Basic visual surveillance system interworking platform user requirements** + +- USR-005: A VSSI platform is required to support registration and de-registration of the platform user through the interface provided by the platform, while the platform user can view and modify personal information. +- USR-006: A VSSI platform is required to support platform user login and logout from the VSSI platform conveniently. A user name and password are required when a platform user logs on to the platform. +- USR-007: A VSSI platform is recommended to support a platform user view of the user access logs or other platform logs. +- USR-008: A VSSI platform is required to support a platform user view of information about the PU of an interworking VSS, including name, status, capabilities, location and other information about the PU. +- USR-009: A VSSI platform is required to support the presentation of real-time video captured by the PU of interworking VSSs for platform users. +- USR-010: A VSSI platform is required to support platform users to order and cancel different services, including basic and value-added services. The VSSI platform is also required to support platform users to view personal account information and operation logs conveniently. + +### **7.1.4 Advanced interworking visual surveillance system user requirements** + +- USR-011: An interworking VSS is recommended to simultaneously present video coverage from different channels captured by different PUs of other interworking VSSs. When watching video, VSS users can optionally choose to pause or continue watching. +- USR-012: An interworking VSS is recommended to support PTZ in the PUs of other interworking VSSs by the end user, with the restriction that a PU can only be controlled by one end user at a time. +- USR-013: An interworking VSS is recommended to support search and retrieval of recorded video coverage from other interworking VSSs by diverse means and to control various playing operations by the end user. +- USR-014: An interworking VSS is recommended to support sending the alarm messages of other interworking VSS to end users in cases of alarm events. + +### **7.1.5 Advanced visual surveillance system interworking platform user requirements** + +- USR-015: A VSSI platform is recommended to simultaneously present video coverage from different channels captured in different PUs of interworking VSSs. When watching video, VSSI platform users can optionally choose to pause or continue watching. +- USR-016: A VSSI platform is recommended to support PTZ in the PUs of interworking VSSs by the end user, with the restriction that only one PU can be controlled by one end user at a time. +- USR-017: A VSSI platform is recommended to support search and retrieval of recorded video coverage from interworking VSSs by diverse means and to control various playing operations by the end user. +- USR-018: A VSSI platform is recommended to support forwarding of any alarm messages from other interworking VSSs to end users. + +## **7.2 Service requirements** + +- SRV-001: A VSSI platform is required to support real-time video transfer from PUs to customer units (CUs) of interworking VSSs. +- SRV-002: A VSSI platform is required to support status and alarm information transfer from PUs to CUs of interworking VSSs. + +- SRV-003: A VSSI platform is required to support remote PTZ control over PU by end users via a CU of interworking VSSs. +- SRV-004: A VSSI platform is recommended to support audio communication between a PU and the CU of interworking VSSs. +- SRV-005: A VSSI platform is required to provide the functionality for playing records, and to provide control functions, such as fast playing, fast rewinding, pausing, location of specific points, stopping and time-shift playing. +- SRV-006: A VSSI platform is recommended to provide electronic maps on which PUs are marked and displayed. +- SRV-007: A VSSI platform is recommended to provide intelligent video analysis functions by calling the related services of interworking VSSs. +- SRV-008: A VSSI platform is recommended to support searching, retrieval and management of recorded and stored images and other video metadata. + +## **7.3 Security requirements** + +### **7.3.1 Authentication security requirements** + +- SEC-001: A VSSI platform is required to provide mechanisms for authentication and authorization; it is required to restrict access to the platform and use applications to authorized users only. A VSSI platform is required to forbid an unauthorized user to handle any resource on the platform. + +### **7.3.2 Access security requirements** + +- SEC-002: A VSSI platform is required to ensure security when accessing and controlling PUs of interworking VSSs. Each PU has access attributes. Only authorized users can access PUs. The types of access include real-time surveillance, replaying video recordings and remote control. End users can access a PU according to their access privileges to it. +- SEC-003: A VSSI platform is required to be able to operate in environments where network address translation (NAT) or firewall devices are present at either or both ends between two communicating entities. It is recommended that the specified firewall, gatekeeper and other network devices be utilized to ensure security to access some special visual surveillance (VS) services. + +### **7.3.3 System security requirements** + +- SEC-004: A VSSI platform is required to have the capability of resisting malicious attacks. +- SEC-005: A VSSI platform is required to provide mechanisms for troubleshooting and data backup. It is required that a structural single-node problem be avoided (i.e., a problem at one node should not cause failure of the whole platform). + +## **7.4 Management requirements** + +- MAN-001: A VSSI platform is required to manage registration/de-registration, authentication, connection and troubleshooting of interworking VSS. +- MAN-002: A VSSI platform is required to support remote configuration management, capability management, status detection, trouble alarm management and security management for the various kinds of devices in the network. The management privilege is required to be divided into different levels. +- MAN-003: A VSSI platform is required to support platform user management. + +## **7.5 Quality of service and quality of experience requirements** + +- QoS-001: A VSSI platform is required to comply with the quality of service (QoS) and quality of experience (QoE) requirements in [ITU-T F.743]. + +## **7.6 Scalability requirements** + +- SCA-001: A VSSI platform is required to support the scalability of the number of VSSs. A VSSI platform provides an efficient system interworking mechanism. A VSS can easily connect to the VSSI platform via the interworking device, and interoperate with other interworking VSSs. +- SCA-002: A VSSI platform is required to support the scalability of each VSS. Each VSS can optionally increase the number of cameras and network video recorders (NVRs), etc. +- SCA-003: A VSSI platform is required to support scalability of number of users, including interworking VSS users and VSSI platform users. A VSSI platform can ensure QoS with the increase in the number of users. + +# Bibliography + +- [b-ITU-T M.60] Recommendation ITU-T M.60 (1993), *Maintenance terminology and definitions.* +- [b-ITU-T Y.101] Recommendation ITU-T Y.101 (2000), *Global Information Infrastructure terminology: Terms and definitions.* + + + +# SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | General tariff principles | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Terminals and subjective and objective assessment methods | +| Series Q | Switching and signalling | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects and next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.743.32-202503-I_PDF-E/raw.md b/marked/F/T-REC-F.743.32-202503-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..31eb17b5b85fb35ba0b2b31046ce9da64f6a8447 --- /dev/null +++ b/marked/F/T-REC-F.743.32-202503-I_PDF-E/raw.md @@ -0,0 +1,449 @@ + + +Recommendation + +**ITU-T F.743.32 (03/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +**Framework for multimedia data asset valuation** + +![ITU logo](84a1d09fb489061482111515543b60dc_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +# ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +## Recommendation ITU-T F.743.32 + +# Framework for multimedia data asset valuation + +## Summary + +Recommendation ITU-T F.743.32 specifies a framework for multimedia data asset valuation, which contains valuation preparation, indicator configuration, measurement approach selection, valuation implementation and valuation guarantee. In valuation preparation, the objectives, principles, objects and team of the valuation are clarified. In indicator configuration, indicator dimensions of quality, cost, application and circulation are identified. Measurement approaches include quality assessment approach, cost-based approach, income-based approach, market-based approach and comprehensive approach. In valuation implementation, the minimum level of indicators and weights of indicators are recommended to be defined based on specific scenarios. + +## History\* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.743.32 | 2025-03-16 | 21 | 11.1002/1000/16228 | + +## Keywords + +Data asset, data valuation, data value, multimedia data, valuation framework. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2025 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|--------------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Overview ..... | 2 | +| 7 Valuation preparation ..... | 3 | +| 7.1 Identify valuation objectives ..... | 3 | +| 7.2 Formulate valuation principles ..... | 4 | +| 7.3 Confirm valuation objects ..... | 4 | +| 7.4 Form valuation team..... | 5 | +| 8 Indicator dimensions configuration and measurement approaches selection..... | 5 | +| 8.1 Configure indicator dimensions ..... | 5 | +| 8.2 Select measurement approaches ..... | 6 | +| 9 Valuation implementation ..... | 8 | +| 9.1 Configure minimization indicators..... | 8 | +| 9.2 Allocate indicators weights ..... | 8 | +| 9.3 Measure results..... | 8 | +| 9.4 Issue a report..... | 9 | +| 10 Valuation guarantee..... | 9 | +| Bibliography..... | 10 | + + + +## Recommendation ITU-T F.743.32 + +## Framework for multimedia data asset valuation + +# 1 Scope + +This Recommendation specifies the framework for multimedia data asset valuation. The scope of this Recommendation includes: + +- Valuation preparation; +- Indicator configuration and measurement approach selection; +- Valuation implementation; +- Valuation guarantee. + +Detailed valuation indicators are out of the scope of this Recommendation. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[[ITU-T F.743.21](#)] Recommendation ITU-T F.743.21 (2020), *Framework for data asset management*. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 data asset** [b-ITU-T X.1750]: An electronically recorded data resource, owned or controlled by an organization. + +**3.1.2 data asset management** [ITU-T F.743.21]: A set of business functions for planning, controlling and providing data and information assets, including developing, executing and monitoring data related plans, policies, programs, projects, processes, methods and procedures, so as to control, protect, deliver and improve the value of data assets. Data asset management requires a full integration of business, technology, and management to ensure that the value of data assets is maintained and added. + +**3.1.3 data quality** [ITU-T F.743.21]: The integrity, normalization, consistency, accuracy, uniqueness, and relevance of data, which is the premise and guarantee for data mining. + +**3.1.4 multimedia services** [b-ITU-T E.800]: A telecommunications service that supports the simultaneous use of multiple media types (e.g., voice, data, video). + +**3.1.5 multimedia information** [b-ITU-T Y.4551]: Digital information that uses multiple forms of information content and information processing, such as text, pictures, audio, video, three-dimensional panoramic pictures and digital maps, which informs or entertains users. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following term: + +**3.2.1 data asset valuation:** The activity of evaluating the value of data assets, which is usually carried out by valuation agencies according to the entrusted tasks, and data asset valuation reports will be issued as the output. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +DAM Data Asset Management + +MDA Multimedia Data Asset + +MDAV Multimedia Data Asset Valuation + +## 5 Conventions + +In this Recommendation: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted, if conformance to this Recommendation is to be claimed. + +## 6 Overview + +[ITU-T F.743.21] specifies a data asset management (DAM) framework with its corresponding objects, activities and supports at a high level, where data valuation is one of the activities in the framework for DAM. Data valuation management is a measurement of the intrinsic value of data, including the cost and application value of data. Multimedia data asset (MDA) is a special case of data asset containing multimedia information that is mainly stored in multimedia form (such as text, pictures, audio, video, three-dimensional panoramic pictures, digital maps), and it is one of the most popular forms of data assets in the industry. This Recommendation provides a more detailed framework for multimedia data asset valuation (MDAV), and can be considered an extension and refinement of [ITU-T F.743.21]. + +The framework for MDAV is shown in Figure 1. The valuation process consists of three main stages: valuation preparation stage, indicator configuration and measurement approach selection stage and valuation implementation stage. Where: + +- **Valuation preparation:** In this stage, the valuation objectives, principles, objects and team are defined. +- **Indicator configuration and measurement approach selection:** This stage includes two parts. In the indicator configuration part, four dimensions need to be considered, which are quality, cost, application and circulation. In the measurement approaches selection part, there is one general approach for data quality assessment and four main classifications of measurement approaches for data valuation, which are cost-based approach, income-based approach, market-based approach and comprehensive approach. +- **Valuation implementation:** In this stage, there are four steps involved, which are configuring minimization indicators, allocating indicators weights, measuring results and issuing report. +- **Valuation guarantee:** It runs through the previous three stages to support the safety, compliance and effectiveness of the valuation. + +![Figure 1 – Framework for multimedia data asset valuation. The diagram shows three main stages: Valuation preparation, Indicator dimensions configuration and measurement approaches selection, and Valuation implementation. A vertical bar on the right labeled 'Valuation guarantee' spans all stages. A legend at the bottom defines the box types: Valuation stage (grey), Valuation action (white), Action contents (white), Sub-contents (grey), and Valuation support (dark grey).](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +The diagram illustrates the framework for multimedia data asset valuation, organized into three main stages connected by downward arrows, with a vertical 'Valuation guarantee' bar on the right. + +- Valuation preparation** (Valuation stage): + - Identify valuation objectives (Valuation action) + - Formulate valuation principles (Valuation action) + - Confirm valuation objects (Valuation action) + - Form valuation team (Valuation action) +- Indicator dimensions configuration and measurement approaches selection** (Valuation stage): + - Configure indicator dimensions** (Valuation action): + - Quality dimension (Action contents) + - Cost dimension (Action contents) + - Application dimension (Action contents) + - Circulation dimension (Action contents) + - Select measurement approaches** (Valuation action): + - Data quality measurement approach (Action contents) + - Data value measurement approaches (Action contents): + - Cost-based approach (Sub-contents) + - Market-based approach (Sub-contents) + - Income-based approach (Sub-contents) + - Comprehensive approach (Sub-contents) +- Valuation implementation** (Valuation stage): + - Configure minimum indicators (Valuation action) + - Allocate indicators weights (Valuation action) + - Measure results (Valuation action) + - Issue a report (Valuation action) + +**Legend:** + +- Valuation stage (Grey box) +- Valuation action (White box) +- Action contents (White box) +- Sub-contents (Grey box) +- Valuation support (Dark grey box) + +F.743.32(25) + +Figure 1 – Framework for multimedia data asset valuation. The diagram shows three main stages: Valuation preparation, Indicator dimensions configuration and measurement approaches selection, and Valuation implementation. A vertical bar on the right labeled 'Valuation guarantee' spans all stages. A legend at the bottom defines the box types: Valuation stage (grey), Valuation action (white), Action contents (white), Sub-contents (grey), and Valuation support (dark grey). + +**Figure 1 – Framework for multimedia data asset valuation** + +In Figure 1, the different boxes can be described as follows: + +- **Valuation stage:** Indicates one of the stages that form the whole process of data asset valuation. +- **Valuation action:** Indicates one of the actions (activities) performed under a specific valuation stage, in order to make progress on data asset valuation. +- **Action contents:** Indicates the detailed contents that a valuation action includes or is performed on. +- **Sub-contains:** Indicates the decomposition of an action content at more detailed level. +- **Valuation support:** Indicates the means to support the valuation process. + +## 7 Valuation preparation + +### 7.1 Identify valuation objectives + +The valuation objectives include: + +- Evaluating the degree of value release of MDA from the perspective of DAM, as the basis for allocating data-related resources, to optimize the overall effectiveness of data application in the enterprise or organization. +- Evaluating the value of data assets involved in data transactions from the perspective of data transaction subjects, as a reference for data asset pricing assessment. +- Objectively evaluating the value of MDA from the perspective of industry management, in order to promote and standardize the development of industry data applications. + +### 7.2 Formulate valuation principles + +The valuation principles include: + +- **Independence:** The valuation agency and personnel are required to conduct the valuation independently according to the valuation requirements, without being influenced by external interference or the intention of the valuation enterprise or organization, and keeping independence and fairness. +- **Objectivity:** The valuated party (the enterprise or organization whose data asset is to be evaluated) provides all documents required for valuation requests truthfully and accurately to ensure the authenticity and accuracy of the valuation. The valuation personnel of the valuation agency are required to objectively and accurately evaluate the value of the data asset of the valuated party. +- **Feasibility:** Participants in the data asset valuation, especially those from the valuation team, are required to adopt scientific and feasible valuation methods and valuation plans during the valuation process to ensure that the data asset valuation can be completed. +- **Professionalism:** The valuation personnel in the valuation process are required to accurately explain the meaning of valuation indicators, timely discover problems in the valuation process, follow the scientific decision-making process, use scientific thinking methods for valuation, and write objective and accurate valuation reports. +- **Traceability:** Complete document records are required for the valuation process, and the valuation personnel of the valuation agency are required to archive supporting documents for the valuation results and to file them with the valuation agency to ensure traceability of the relevant results. +- **Safety:** The valuation agency and the valuated party are required to strictly comply with confidentiality agreements and safeguard the related materials involved in the valuation process to ensure the safety of the relevant materials. + +### 7.3 Confirm valuation objects + +#### 7.3.1 Scope and classification of valuation objects + +Standardized valuation objects include all data assets within the scope of multimedia services and circulation. Valuation objects can be judged from multiple dimensions: + +- **According to the source of data:** Valuation objects can include externally obtained data assets and internally collected and generated data assets. +- **According to business ownership:** Valuation objects can include production-related data assets, management-related data assets and business analysis-related data assets. +- **According to structural characteristics:** Valuation objects can include structured data assets, unstructured data assets and semi-structured data assets. +- **According to circulation types:** Valuation objects can include directly tradable data, indirectly tradable data and non-tradable data. + +#### 7.3.2 Granularity of valuation objects + +The granularity of valuation objects can be divided according to the hierarchical structure of the data classification directory. + +The entire data under a certain level of the data classification directory can be selected as the valuation object. For example, select externally obtained data assets for valuation according to the source of data. + +Alternatively, two parallel levels can be selected for valuation and comparison. For example, select two parallel levels of production-related and business analysis-related data assets according to the ownership of the business for valuation. + +### 7.4 Form valuation team + +Form valuation team refers to clarifying the main responsible party and the cooperating party for each output in the valuation process of valuation preparation, indicator dimensions configuration and measurement approaches selection, and valuation implementation. A valuation team is formed jointly by the valuation agency and the valued party. + +## 8 Indicator dimensions configuration and measurement approaches selection + +### 8.1 Configure indicator dimensions + +#### 8.1.1 Quality dimension + +The quality dimension refers to the intrinsic quality of the data asset, which directly impacts the usability and reliability of the data. The indicators in the quality dimension include, but are not limited to, normality, accuracy, completeness, consistency, uniqueness, timeliness and validity. + +- **Normality:** The extent to which data conform to data standards, data models, business rules, metadata or authoritative reference data, which includes but is not limited to, data type compliance (such as date format checking), data length compliance and data association relationship normality (such as "the sum of itemized outgoing revenues equals the total outgoing revenues"). +- **Accuracy:** The extent to which the data accurately represents the true value of the real entity (actual object) it describes, which includes but is not limited to, the correctness of the data content, the accuracy of the data range and whether the data values conform to the business logic constraints. +- **Completeness:** The extent to which data elements are assigned values as required by data rules, including data record coverage ratios and null value data validation. +- **Consistency:** The extent to which data do not contradict data used in other specific contexts, including identical data consistency and correlated data consistency. +- **Uniqueness:** The degree of redundancy with which data items with unique feature sets are recorded. That is, each unique entity is recorded only once in the same data set. +- **Timeliness:** The degree to which data represents reality at the required point in time, including correctness based on time periods, validity based on specific point in time and accuracy of ordering based on time. +- **Validity:** The extent to which data can be effectively accessed. + +#### 8.1.2 Cost dimension + +The cost dimension refers to the financial expenditures associated with the generation and maintenance of the data asset. The indicators in the cost dimension include, but are not limited to, acquisition costs, construction costs, management costs, labour costs and other costs. + +- **Acquisition costs:** The costs invested in obtaining data from internal and external sources. Depending on the form of acquisition. It can be further divided into: + - a) **External data procurement costs:** Direct expenses incurred from purchasing data. This includes the purchase price, related taxes and fees, insurance premiums and any other expenses required to reach the desired state of use. + - b) **Internal data collection costs:** Other expenses incurred from directly collecting data within the enterprise or organization, excluding labour costs. + +- **Construction costs:** The expenses incurred for cleaning, transforming, analyzing, mining, and other operations on data based on market analysis, business optimization, risk management, and the like, excluding labour costs. +- **Management costs:** Other expenses incurred during the governance, maintenance and management of data assets, excluding labour costs. +- **Labour costs:** Employee salaries and training expenses involved in various processes throughout the life-cycle of data products and in the maintenance of data assets. +- **Other costs:** Expenses incurred during the generation of data apart from the above-mentioned costs. + +#### 8.1.3 Application dimension + +The application dimension refers to the scope and effectiveness of the application of data asset in practical business scenarios. The indicators in the application dimension include, but are not limited to, scope of use, application scenarios, frequency of use, effect of use, data relevance and application risks. + +- **Scope of use:** The enterprises or organizations that use the data asset, the scope of application and the platforms covered. +- **Application scenarios:** The usage methods of the data asset, the degree of openness and the number of business processes covered. +- **Frequency of use:** The internal and external access frequency of the data asset. +- **Effect of use:** The metrics such as marketing conversion rate, marketing revenue and user churn rate. +- **Data relevance:** This includes but is not limited to, user relevance, data interrelationships and business relevance of the data asset. +- **Application risks:** This includes but is not limited to, management risks, circulation risks, data security risks, failure risks, sensitivity risks and regulatory risks associated with the data asset. + +#### 8.1.4 Circulation dimension + +The circulation dimension refers to the transaction attributes of data asset in the market circulation and the influencing factors of market value fluctuations. The indicators in the circulation dimension include, but are not limited to, supply-demand relationship and historical transaction situation. + +- **Supply-demand relationship:** The dynamic connection between the supply quantity and demand quantity of similar data assets in the market, such as scarcity and market scale. +- **Historical transaction situation:** Detailed information regarding data assets during past market transactions, such as transaction prices, trading volumes, transaction times and frequencies. + +### 8.2 Select measurement approaches + +#### 8.2.1 Data quality measurement approach + +There is a general procedure for data quality measurement, as shown in Figure 2: + +![Flowchart of the procedure for data quality assessment. The steps are: Begin, Determining the dimensions of data quality assessment, Defining data quality rules, Building a data quality rule base, Managing data quality tasks, Generating data quality report, Meet the requirement? (Decision), Rectification (if No), and End (if Yes).](af7916c89a458fdab6c3f443217388ae_img.jpg) + +``` + +graph TD + Begin([Begin]) --> Step1[Determining the dimensions of data quality assessment] + Step1 --> Step2[Defining data quality rules] + Step2 --> Step3[Building a data quality rule base] + Step3 --> Step4[Managing data quality tasks] + Step4 --> Step5[Generating data quality report] + Step5 --> Decision{Meet the requirement?} + Decision -- No --> Rectification[Rectification] + Rectification --> Step4 + Decision -- Yes --> End([End]) + +``` + +F.743.32(25) + +Flowchart of the procedure for data quality assessment. The steps are: Begin, Determining the dimensions of data quality assessment, Defining data quality rules, Building a data quality rule base, Managing data quality tasks, Generating data quality report, Meet the requirement? (Decision), Rectification (if No), and End (if Yes). + +**Figure 2 – Procedure for data quality assessment** + +The descriptions of the above procedure are listed below: + +- **Determining the dimensions of data quality assessment:** It is the basis for evaluating data quality standards and defining constraint rules, measuring data quality from different aspects. It mainly includes characteristics such as normality, accuracy, completeness, consistency, uniqueness, timeliness and validity as specified in clause 8.1.1. According to the requirements of the assessment service, the above indicators can be selected and combined with assessment methods. +- **Defining data quality rules:** Data rules are a way of constraining data knowledge and service scope through semantics, syntax, and other restrictive methods. The definition of data quality rules mainly provides detailed explanations for the identification and classification of data defects based on quality assessment dimensions, including the determination of assessment indicators and types; defining rule names, types, and descriptions; and developing quality rule algorithms and statements. +- **Building a data quality rule base:** The rule base refers to the formation of targeted rule sets based on the characteristics of data assessment entities, in order to conduct different quality assessment on different data sets. The rules in the data quality rule base can be reused for other data assessment tasks. +- **Managing data quality task:** A quality task refers to a task of conducting data quality assessment on data entities in order to facilitate the configuration and management of assessment tasks. The management of quality tasks mainly involves controlling aspects such as data assessment objects, data assessment frequency, data assessment time and data assessment methods. +- **Generating data quality report:** The completion of the quality assessment task is generally presented through the generation of a quality report that contains the assessment results, including the execution status of rules, assessment result data and comprehensive assessment. + +#### 8.2.2 Data value measurement approaches + +The data value measurement approaches refer to the technical methods used to examine and analyze the value of data assets, mainly including the following approaches. + +- **Cost-based approach:** It is a method used to evaluate the value of data assets based on their acquisition, processing, management and maintenance costs. This approach is typically suitable for datasets that are difficult to directly measure their economic value, such as internally generated data or non-market traded data. The cost-based approach considers the value of a data asset as the sum of all the costs required for its production, maintenance and management. +- **Income-based approach:** It is a method used to evaluate the value of data assets by analyzing the future economic benefits they can bring. This approach is typically suitable for data sets that can be directly measured for their economic value, such as market-traded data or data with clear commercial use cases. The income-based approach considers the value of a data asset as the present value sum of the net income it can generate over its expected usage period. +- **Market-based approach:** It is a method used to evaluate the value of data assets by selecting recent or past transaction reference frame prices as references, and adjusting for specific and personalized factors, in order to obtain the valuation of the data asset, under the premise of an open and active trading market. This approach is typically suitable for data sets having similar contents and quality as the data assets that have already be valued or have commercial prices on the market. The market-based approach mainly considers the value of a data asset as a weighted average of the following factors: price, quality, supply-demand relationship, time point, and quantity of each referenced data set. +- **Comprehensive approach:** It is a method used to evaluate the value of data assets by using the combination of the income-based approach, the cost-based approach, and the market-based approach. This approach is typically suitable for data sets that can be measured by the above mentioned three approaches, and a more precise valuation is expected than just using one of the above approaches. The comprehensive approach obtains the value of a data asset by weighted summation of the above three approaches. + +## 9 Valuation implementation + +### 9.1 Configure minimization indicators + +According to the specific valuation objectives and objects, select appropriate minimum-level valuation indicators, which can be dynamically adjusted and expanded according to actual needs and scenarios. + +### 9.2 Allocate indicators weights + +Configure the weights of the primary dimensions, the secondary dimensions and specific indicators. The weight settings of the primary and secondary dimensions can be adjusted according to business characteristics and needs. + +### 9.3 Measure results + +Verify and evaluate the value of data assets according to the configured rules and methods. Convert the valuation results of each rule into a grade or score band. According to the evaluation framework and specific evaluation model, calculate the secondary dimension score from the evaluation index through the average value. The secondary dimension score is weighted to obtain the primary dimension score, and the primary dimension score is summed according to the weight to finally obtain the overall value. + +### 9.4 Issue a report + +Summarize, organize and analyze the results of MDAV to conduct and generate a valuation report. The content of the report should include, but is not limited to, the organization and data profile, valuation objectives, valuation objects, valuation criteria, valuation benchmark date, valuation indicators and weights, valuation implementation process and the conclusion of the valuation. + +## 10 Valuation guarantee + +Valuation guarantee refers to the support system that ensures the safety, compliance and effectiveness of MDAV. It specifies the resource conditions guarantee for valuation activities, including technology, platform and institutional aspects. + +- **Technical guarantee:** Integrate key technologies and algorithmic models from the fields of asset valuation and information technology to build a comprehensive technical system with cross-disciplinary innovation and strong scalability. +- **Platform guarantee:** Solidify, implement and validate the data asset valuation framework, valuation methods and valuation processes through software systems. The platform should support the MDAV work, with functions including valuation process management, valuation tool management, valuation model management and valuation report management. +- **Institutional guarantee:** Standardize MDAV behaviours through the construction of relevant standard specifications, institutional processes and personnel management systems. + +## Bibliography + +- [b-ITU-T E.800] Recommendation ITU-T E.800 (2008), *Definitions of terms related to quality of service*. +- [b-ITU-T X.1750] Recommendation ITU-T X.1750 (2020), *Guidelines on security of big data as a service for big data service providers*. +- [b-ITU-T Y.4551] Recommendation ITU-T Y.4551/F.771 (2008), *Service description and requirements for multimedia information access triggered by tag-based identification*. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.743.33-202512-I_PDF-E/raw.md b/marked/F/T-REC-F.743.33-202512-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..5966c7cdcd129b3e270a707bcfe9d681785d661e --- /dev/null +++ b/marked/F/T-REC-F.743.33-202512-I_PDF-E/raw.md @@ -0,0 +1,383 @@ + + +# Recommendation **ITU-T F.743.33 (12/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +## **Requirements and framework for computer vision-based anomaly detection service in wind farms** + +![ITU logo](6ed175c791b5e156d9c98a8dbcc3318c_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with white lines representing latitude and longitude, and the letters 'ITU' in a bold, blue, sans-serif font overlaid on the globe. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.743.33 + +# Requirements and framework for computer vision-based anomaly detection service in wind farms + +## Summary + +Traditional video surveillance systems are usually employed to discover anomalies by physical observation of videos and cannot meet the requirements for automatic anomaly detection in wind farms. With the development of machine learning algorithms, computer vision can achieve automatic intelligent processing such as image enhancement, image restoration, feature extraction, object detection, object recognition, object tracking, pose estimation, motion estimation, event detection and anomaly detection. Therefore, an intelligent system based on computer vision can be deployed on a wind farm for automatic anomaly detection. + +The computer vision-based anomaly detection service can be an effective way to automatically detect general anomalies such as vessel trespass, small flames and abnormal presences in the wind farm. It can address the limitations of manually detecting general anomalies through video observation, thereby preventing anomalies from being overlooked due to human oversight. + +Recommendation ITU-T F.743.33 addresses the essential requirements and framework for computer vision-based anomaly detection service in the wind farm. + +## History\* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.743.33 | 2025-12-14 | 21 | 11.1002/1000/16574 | + +## Keywords + +Anomaly detection, computer vision, wind farm. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2026 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-----------------------------------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 Reference ..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Overview of CVADS in wind farms ..... | 2 | +| 7 Requirements for CVADS in wind farm ..... | 3 | +| 7.1 General requirements..... | 3 | +| 7.2 Functional requirements ..... | 3 | +| 7.3 Performance requirements..... | 4 | +| 7.4 Device requirements..... | 4 | +| 7.5 Network requirements ..... | 5 | +| 8 Reference framework for CVADS in the wind farm..... | 5 | +| 8.1 Device layer..... | 5 | +| 8.2 Network layer ..... | 6 | +| 8.3 Service layer ..... | 6 | +| 8.4 Application layer ..... | 7 | +| Appendix I – Use cases of computer vision-based anomaly detection service in a wind farm..... | 8 | +| I.1 Vessel trespass detection service for offshore wind farms..... | 8 | +| I.2 Detection service for the blade crack of the wind turbine..... | 8 | +| I.3 Detection service for the icing on the blades of the wind turbine..... | 9 | +| Bibliography..... | 10 | + + + +# Recommendation ITU-T F.743.33 + +## Requirements and framework for computer vision-based anomaly detection service in wind farms + +# 1 Scope + +This Recommendation addresses the requirements and framework for computer vision-based anomaly detection service (CVADS) in the wind farm, including the following: + +- Overview of CVADS in the wind farm. +- Requirements specific to CVADS in the wind farm. +- Reference framework for CVADS in the wind farm. + +# 2 Reference + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T F.743] Recommendation ITU-T F.743 (V2) (2019), *Requirements and service description for video surveillance*. + +[ITU-T F.747.11] Recommendation ITU-T F.747.11 (2022), *Requirements for intelligent surface-defect detection service in industrial production lines*. + +[ITU-T F.748.16] Recommendation ITU-T F.748.16 (2022), *Requirements for applications and services in smart manufacturing based on machine vision*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 algorithm** [b-ISO/IEC 11557]: A set of rules for transforming the logical representation of data. + +**3.1.2 application** [b-ITU-T Y.101]: A structured set of capabilities, which provide value-added functionality supported by one or more services. + +**3.1.3 video surveillance system** [b-ITU-T H.626]: A telecommunication service focusing on video (including audio and image) application technology, which is used to remotely capture multimedia (such as audio, video, image, alarm signal, etc.) and present them to the end user in a user-friendly manner, based on a managed broadband network with ensured quality, security and reliability. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following term: + +**3.2.1 computer vision-based anomaly detection (CVADS):** A service capability that utilizes computer vision techniques and algorithms to automatically identify and flag deviations from normal patterns or behaviours in visual data for the purpose of detecting anomalies in specific scenarios. + +# **4 Abbreviations and acronyms** + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-------|-------------------------------------------------| +| CVADS | Computer Vision-based Anomaly Detection Service | +| GPU | Graphics Processing Unit | +| GUI | Graphical User Interface | +| UI | User Interface | + +# **5 Conventions** + +In this Recommendation: + +- The expression "is required" indicates a requirement which must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The expression "is recommended" indicates a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance with this Recommendation. +- The expression "may" indicates an optional requirement which is permissible, without implying any sense of being recommended. This term is not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with this Recommendation. + +# **6 Overview of CVADS in wind farms** + +Computer vision applied to anomaly detection in wind farms includes methods for acquiring, processing, analysing and understanding digital images or videos from the physical wind farm. This understanding can be seen as discerning symbolic information from image/video data using algorithms constructed with the aid of geometry, statistics and learning theory. Sub-domains of computer vision include image enhancement, image restoration, feature extraction, object detection, object recognition, object tracking, pose estimation, motion estimation, event detection, etc. + +Figure 1 shows the overall conceptual diagram for CVADS in the wind farm. Here, CVADS mainly includes server, client, the surveillance cameras in the wind farm, and the broadband network for transmitting data. The server is located in the centre of CVADS and communicates with the cameras and client through the network devices. The server provides functions including management, detection, storage, etc. The surveillance cameras are installed in some critical positions of the wind farm, and their presence is required to support joint anomaly detection based on computer vision. The videos captured by the cameras are transferred to the server for anomaly detection. The client (such as an application on a personal computer or mobile phone) provides the user interface (UI) for user presentation, configuration and search. In addition, the server provides the interface to connect with other alarm systems and triggers an alarm when an anomaly is detected. The broadband network is required to support real-time video transmission. + +![Conceptual diagram for CVADS in wind farms showing the flow of data from cameras to a central server and then to clients and other alarm systems.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +The diagram illustrates the architecture of a CVADS (Video Content Analysis and Detection System) in a wind farm. On the left, a box labeled 'Cameras in wind farm' contains a collage of images showing various surveillance cameras (PTZ, dome, etc.) installed in a wind farm environment. A dashed blue line labeled 'Network' connects this box to the central 'CVADS server' box. The 'CVADS server' box contains a stack of server units and a switch. Inside the server box, there are labels for 'Management database' and 'GPU detection' (in red). Below the server box, a dashed blue line labeled 'Network' connects it to a box labeled 'Other alarm systems', which contains icons for a laptop and a smartphone. To the right of the 'CVADS server', another dashed blue line labeled 'Network' connects it to a box labeled 'CVADS client (Such as Windows/Android application)'. This box contains an image of a person working at a desk with multiple computer monitors displaying video feeds from the cameras. Additionally, there are 'Interface' labels on the left and right sides of the 'CVADS server' box, and a 'Network' label at the bottom of the server box. + +Conceptual diagram for CVADS in wind farms showing the flow of data from cameras to a central server and then to clients and other alarm systems. + +F.743.33(25) + +**Figure 1 – Conceptual diagram for CVADS in wind farms** + +# 7 Requirements for CVADS in wind farm + +## 7.1 General requirements + +- GR-01: All devices and materials for CVADS are required to suit the natural environment in the wind farm. +- GR-02: CVADS is required to support the availability, compatibility, scalability and controllability requirements. +- GR-03: CVADS is recommended to support the security requirements mentioned in [ITU-T F.748.16]. + +## 7.2 Functional requirements + +### 7.2.1 General functions + +- FR-GF-01: CVADS is required to provide interface management, database management, configuration management, video dispatch and parallel detection for multiple anomalies. +- FR-GF-02: CVADS is required to provide the function of alarm processing. +- FR-GF-03: CVADS is required to generate detection log files in real time. +- FR-GF-04: CVADS is required to provide the function of remotely configuring the general parameters of the server. +- FR-GF-05: CVADS is required to provide users with visual human-computer interaction services: browsing, querying, anomaly presenting, and counting the anomalies that have occurred. +- FR-GF-06: CVADS may support wind farms' actual needs for adding new anomaly detection devices. + +### 7.2.2 Detection functions + +- FR-DF-01: CVADS is required to automatically detect vessel trespass in the offshore wind farm. +- FR-DF-02: CVADS is required to automatically detect small flames in the wind farm. +- FR-DF-03: CVADS is required to automatically detect blade cracks of the wind turbine. + +- FR-DF-04: CVADS is required to automatically detect icing on the blades of the wind turbines. +- FR-DF-05: CVADS is required to automatically detect oil leakage from the wind turbine gearbox. +- FR-DF-06: CVADS is required to automatically detect the breakage of cable protection sleeves. +- FR-DF-07: CVADS is required to automatically detect corrosion of the ladder in the wind farm. + +### 7.2.3 Supporting algorithms + +- FR-SA-01: CVADS is recommended to support frame sampling algorithms that select specific frames from a surveillance video. +- FR-SA-02: CVADS is recommended to support image pre-processing algorithms that realize image scaling, image denoising, image normalization, image enhancement, etc. +- FR-SA-03: CVADS is recommended to support semantic segmentation algorithms that assign semantic labels to each pixel in an image. +- FR-SA-04: CVADS is recommended to support feature extraction algorithms that extract specific features from images for an anomaly detection service. + +## 7.3 Performance requirements + +- PR-01: The recall rate is recommended to be higher than 99% for detecting vessel trespass in the wind farm. +- PR-02: The recall rate is recommended to be higher than 99.9%, and the false alarm rate not higher than 1% for detecting the small flame in the wind farm. +- PR-03: The recall rate is recommended to be higher than 95%, and the false alarm rate not higher than 5% for detecting a blade crack of the wind turbine. +- PR-04: The recall rate is recommended to be higher than 90%, and the false alarm rate not higher than 5% for detecting icing on the blades of the wind turbine. +- PR-05: The recall rate is recommended to be higher than 99%, and the false alarm rate not higher than 10% for detecting oil leakage from the wind turbine gearbox. +- PR-06: The false alarm rate is recommended to be not higher than 5% for detecting the breakage of the cable protection sleeve. +- PR-07: The recall rate is recommended to be higher than 95% for detecting the corrosion of ladders in the wind farm. +- PR-08: CVADS is required to meet the needs of real-time and periodic anomaly detection in the wind farm. +- PR-09: The anomaly detection results are required to be delivered with low latency to an external alarm system [ITU-T F.747.11]. + +## 7.4 Device requirements + +- DR-01: The surveillance cameras are recommended to support high-quality videos (the resolution is recommended to be higher than 6 million pixels) for anomaly detection based on computer vision in the wind farm. +- DR-02: The hardware of the CVADS server is recommended to adopt graphics processing units (GPUs) for parallel multiple anomaly detections in the wind farm. +- DR-03: The hardware of the CVADS client is recommended to support the running of general graphical user interface (GUI) applications. + +- DR04: The communication devices are recommended to support the requirement for real-time anomaly detection of CVADS. + +## **7.5 Network requirements** + +- NR-01: The network bandwidth is required to support the real-time video transmission of CVADS in the wind farm. +- NR-02: The network is required to provide low jitter and low packet loss rate to transmit the video stream [ITU-T F.743]. +- NR-03: The network is required to support the general data transmission between the surveillance cameras and CVADS server, CVADS server and CVADS client, CVADS server and other alarm systems. + +# **8 Reference framework for CVADS in the wind farm** + +Figure 2 shows the reference framework for CVADS in the wind farm. It consists of four layers: the device layer, the network layer, the service layer and the application layer. + +## **8.1 Device layer** + +The device layer represents the devices adopted for video collecting and processing in CVADS, including surveillance cameras, server devices, client devices, communication devices, etc. Some surveillance cameras are required to be installed at the designated locations in wind farms according to the requirement of specific anomaly detection. The performance of server devices and client devices must meet the minimum requirements for anomaly detection in wind farms. Communication devices are responsible for real-time video transmission. + +![Reference framework for CVADS in a wind farm diagram showing four layers: Application, Service, Network, and Device.](d4af765160d04ecef538e5066006dc77_img.jpg) + +The diagram illustrates the reference framework for CVADS in a wind farm, organized into four hierarchical layers: + +- Application layer:** Contains five functional blocks: Graphical user interface, Browsing, Querying, Counting, and Configuring general parameters. +- Service layer:** + - Top section: Interface management and Video dispatch module. + - Middle section: Multiple parallel detection services (service 1, service 2, ..., service N). Each service contains a vertical stack of sub-modules: Frame sampling, Image preprocessing, Semantic segmentation, Feature extracting, Vessel trespass detecting, and Anomaly storage. + - Bottom section: Anomaly presentation, Database management, Configuration management, and Alarm processing. +- Network layer:** Contains four functional blocks: Network connection, Data transmission, Real-time video transmission, and Network security. +- Device layer:** Contains five functional blocks: Surveillance cameras, Server devices, Client devices, Communication devices, and ... (others). + +Reference framework for CVADS in a wind farm diagram showing four layers: Application, Service, Network, and Device. + +F.743.33(25) + +**Figure 2 – Reference framework for CVADS in a wind farm** + +## 8.2 Network layer + +The network layer provides network connection, general data transmission, real-time video transmission and network security. Its bandwidth is required to support the real-time video transmission of CVADS in the wind farm. + +## 8.3 Service layer + +The service layer provides interface management to accept the videos uploaded from the network. The video dispatch module distributes the received video to the corresponding anomaly detection sub-module according to different anomaly detection types. Certainly, the service layer is required to perform multiple detection sub-modules in parallel. A detection sub-module may normally include the following functions: Frame sampling, image pre-processing, semantic segmentation, feature extracting, anomaly detecting, and anomaly storage. According to the actual needs of the wind farm, anomaly detecting includes vessel trespass detection, small flame detection, abnormal appearance detection, etc. Abnormal appearance refers to the phenomena such as blade crack, oil leakage from the gearbox, breakage of the cable protection sleeve and corrosion of the ladder. When the detection service detects an anomaly, the anomaly presentation module will package the anomaly data and send it to CVADS client through the interface. Meanwhile, the alarm processing module will also trigger other alarm systems. The requirements for alarm systems are not within the scope of this + +Recommendation. When an anomaly occurs, CVADS simply uses an external alarm system. In addition, the service layer should provide configuration management and database management. Configuration management is primarily responsible for the unified management and configuration of various system parameters, detection rules and resource allocation. Database management is used to store and manage data related to anomaly detection, ensuring the integrity and traceability of system data. + +## **8.4 Application layer** + +The application layer provides visual human-computer interaction service to users in the form of applications such as an application on a personal computer or a mobile phone. Through the application layer, users can browse, query and count the anomalies that have occurred and can also configure the general parameters of the server. + +## Appendix I + +## Use cases of computer vision-based anomaly detection service in a wind farm + +(This appendix does not form an integral part of this Recommendation.) + +### I.1 Vessel trespass detection service for offshore wind farms + +Vessel trespass in offshore wind farms poses potential hazards to submarine cables and wind turbines. To ensure the safe and stable operation of the wind farm, CVADS is required to provide an automatic real-time detection service for vessel trespass into offshore wind farms. CVADS is also required to trigger an alarm system when any vessel intrusion is detected. Figure I.1 illustrates that CVADS detects vessel trespass in the wind farm via a deep neural network. + +![Figure I.1: Illustration of vessel trespass detection for offshore wind farms. The figure consists of two vertically stacked images of an offshore wind farm. The top image shows several wind turbines in the ocean with a few small boats nearby. The bottom image shows the same scene but with red rectangular bounding boxes drawn around the boats, labeled 'boat', indicating they have been detected by the CVADS system.](472b4b1952940f919fbd8584e83daa36_img.jpg) + +Figure I.1: Illustration of vessel trespass detection for offshore wind farms. The figure consists of two vertically stacked images of an offshore wind farm. The top image shows several wind turbines in the ocean with a few small boats nearby. The bottom image shows the same scene but with red rectangular bounding boxes drawn around the boats, labeled 'boat', indicating they have been detected by the CVADS system. + +Figure I.1 – Illustration of vessel trespass detection for offshore wind farms + +### I.2 Detection service for the blade crack of the wind turbine + +The detection of blade cracks in wind turbines poses significant risks to their operational integrity and efficiency. To maintain the continued safe and reliable operation of wind turbines, CVADS is required to detect blade cracks with high precision, leveraging state-of-the-art technology for accurate and timely identification of potential faults. By providing immediate alerts upon detection, CVADS can facilitate prompt maintenance actions, reducing the risk of further damage and minimizing operational downtime. Figure I.2 shows that CVADS detects the blade crack of the wind turbine via a convolutional neural network. + +![Two side-by-side images of a wind turbine blade. The left image shows a clean blade. The right image shows the same blade with two red rectangular boxes highlighting cracks on its surface.](5f2c99ae08864cf2d5c949947bac2b98_img.jpg) + +Two side-by-side images of a wind turbine blade. The left image shows a clean blade. The right image shows the same blade with two red rectangular boxes highlighting cracks on its surface. + +**Figure I.2 – Illustration of blade crack detection for wind turbines** + +### **I.3 Detection service for the icing on the blades of the wind turbine** + +Icing on the blades of wind turbines can severely affect their aerodynamic performance and operational safety. To guarantee the normal and efficient operation of wind turbines, CVADS is required to offer a real-time detection service for icing on the blades. It is also necessary for CVADS to send out early warnings once icing is detected. Figure I.3 illustrates that CVADS detects the icing on a blade of the wind turbine via a deep neural network. + +![Two side-by-side images of a wind turbine blade. The left image shows a blade covered in ice. The right image shows the same blade with a red rectangular box highlighting the icing.](33809b11cc711711ebb7be1282fcd4b7_img.jpg) + +Two side-by-side images of a wind turbine blade. The left image shows a blade covered in ice. The right image shows the same blade with a red rectangular box highlighting the icing. + +**Figure I.3 – Illustration of the icing detection on a blade of wind turbine** + +## Bibliography + +- [b-ITU-T H.626] Recommendation ITU-T H.626 (V2) (2019), *Architectural requirements for a video surveillance system.* +- [b-ITU-T Y.101] Recommendation ITU-T Y.101 (2000), *Global Information Infrastructure terminology: Terms and definitions.* +- [b-ISO/IEC 11557] ISO/IEC 11557:1992, *Information technology – 3,81 mm wide magnetic tape cartridge for information interchange – Helical scan recording – DDS-DC format using 60 m and 90 m length tapes.* + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.743.34-202512-I_PDF-E/raw.md b/marked/F/T-REC-F.743.34-202512-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..f20a61ba4cba0a723708c69c8254c50166c5855f --- /dev/null +++ b/marked/F/T-REC-F.743.34-202512-I_PDF-E/raw.md @@ -0,0 +1,507 @@ + + +# Recommendation + +## **ITU-T F.743.34 (12/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +**Requirements and reference framework for a +service platform to support industry visual +services** + +![ITU logo](84a1d09fb489061482111515543b60dc_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with white lines representing latitude and longitude, and the letters 'ITU' in a bold, blue, sans-serif font superimposed on the globe. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.743.34 + +# Requirements and reference framework for a service platform to support industry visual services + +## Summary + +There are many demands of industry visual computing service requirements from industry video surveillance, smart grid, industrial automation, intelligent traffic and remote medical, etc. In the ecosystem of visual computing service provision and consumption, visual service consumers need to quickly find service providers and choose suitable services in order to save cost and complete tasks conveniently and quickly. On the other hand, many visual service providers wait for consumers to find and select their services to share their service resources and achieve the maximization of resource utilization. Therefore, it is necessary to establish a service platform to meet the communication needs between visual service providers and consumers and to manage them. + +Recommendation ITU-T F.743.34 focuses on visual computing services and provides the requirements and reference framework of a service platform to support industry visual services. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.743.34 | 2025-12-14 | 21 | 11.1002/1000/16575 | + +## Keywords + +Industry visual service, reference framework, requirements, service platform. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2026 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|-----|---------------------------------------------------------------------------------------------|----| +| 1 | Scope..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 2 | +| 4 | Abbreviations and acronyms ..... | 2 | +| 5 | Conventions ..... | 2 | +| 6 | Industry visual service ecosystem ..... | 2 | +| 6.1 | Computing service providers..... | 3 | +| 6.2 | Industry visual service and application support platform..... | 3 | +| 6.3 | Computing service consumers..... | 3 | +| 6.4 | Network providers ..... | 4 | +| 7 | Reference framework of service platform to support industry visual services ..... | 4 | +| 8 | Functional requirements for the service platform..... | 5 | +| 8.1 | Functional requirements for the application layer..... | 5 | +| 8.2 | Functional requirements for the industry visual service and application support layer ..... | 5 | +| 8.3 | Functional requirements for the network layer ..... | 7 | +| 8.4 | Functional requirements for the computing service provision layer ..... | 7 | +| | Appendix I – Use cases of a service platform to support industry visual services ..... | 8 | +| | Bibliography..... | 11 | + + + +# Requirements and reference framework for a service platform to support industry visual services + +## 1 Scope + +This Recommendation provides the requirements and reference framework of a service platform to support industry visual services, including the functional requirements of visual computing service provision, service and application support, network and visual computing service consumption. + +The scope of this Recommendation includes: + +- Reference framework of a service platform to support industry visual services; +- Functional requirements for visual computing service provision, service and application support, network and application; +- Use cases of a service platform to support industry visual services. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 algorithm** [b-ISO/IEC 11557]: A set of rules for transforming the logical representation of data. + +**3.1.2 application** [b-ITU-T Y.101]: A structured set of capabilities, which provide value-added functionality supported by one or more services. + +**3.1.3 cloud computing** [b-ITU-T Y.3500]: Paradigm for enabling network access to a scalable and elastic pool of shareable physical or virtual resources with self-service provisioning and administration on-demand. + +NOTE – Examples of resources include servers, operating systems, networks, software, applications and storage equipment. + +**3.1.4 cloud service** [b-ITU-T Y.3500]: One or more capabilities offered via cloud computing invoked using a defined interface. + +**3.1.5 edge computing** [b-ITU-T Y.3073]: This refers to a strategy to deploy processing capability at network edge where end terminals are connected, and to perform the processing of data which is derived from and fed to the end terminals. + +**3.1.6 edge computing service** [b-ITU-T Y.3540]: One or more capabilities provided on edge computing infrastructure. + +**3.1.7 model** [b-ITU-T F.748.16]: An output created by an algorithm running on training data that can generate an inference or prediction, based on input data. + +**3.1.8 service** [b-ITU-T Y.101]: A structure[d] set of capabilities intended to support applications. + +**3.1.9 service resource** [b-ITU-T X.1046]: A set of network devices in the telecom operator's network, which work together to deliver services to the telecom operator's customers. + +**3.1.10 scheduling** [b-ITU-T X.746]: The method of controlling the timing of the execution of a scheduled activity within or represented by a managed object. + +### **3.2 Terms defined in this Recommendation** + +This Recommendation defines the following term: + +**3.2.1 computing service:** One or more capabilities used to output specific results after computing and processing data through related computing software, chips and device resources. + +## **4 Abbreviations and acronyms** + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|----------|---------------------------------------------| +| AI | Artificial Intelligence | +| AGV | Automated Guided Vehicle | +| API | Application Programming Interface | +| CPU | Central Processing Unit | +| GPU | Graphics Processing Unit | +| IMT-2020 | International Mobile Telecommunication-2020 | +| NPU | Neural network Processing Unit | +| VPN | Virtual Private Network | +| WLAN | Wireless Local Area Network | + +## **5 Conventions** + +In this Recommendation: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. + +## **6 Industry visual service ecosystem** + +There are a large number of industry visual service requirements from industry video surveillance, smart grid, industrial automation, intelligent traffic and remote medical, etc. On the other hand, there are also a large number of computing service providers (such as cloud computing service providers, edge computing service providers and job computing service providers). From the perspective of computing service consumers, they need to quickly find service providers in order to save cost and complete tasks conveniently and quickly. From the perspective of computing service providers, it is necessary to demonstrate their service capabilities so that consumers can quickly find and choose their services, in order to maximize the utilization of service resources. Therefore, it is necessary to establish a service platform to meet the communication needs between computing service providers and consumers and to manage service providers and consumers. Figure 1 depicts business roles that are relevant in an industry visual service ecosystem and relationships between them. + +![Diagram of business roles in an industry visual service ecosystem. On the left, three stacked boxes represent 'Computing service consumers', 'Industry visual service and application support platform', and 'Computing service providers'. On the right, a single box represents 'Network providers'. Lines connect each of the three left boxes to the 'Network providers' box. Below the diagram is the text 'F.743.34(25)'.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +``` + +graph LR + subgraph Left [ ] + direction TB + A[Computing service consumers] + B[Industry visual service and application support platform] + C[Computing service providers] + end + D[Network providers] + A --- D + B --- D + C --- D + style Left fill:none,stroke:none + style D fill:#f9b591,stroke:#333,stroke-width:1px + style A fill:#f9b591,stroke:#333,stroke-width:1px + style B fill:#f9b591,stroke:#333,stroke-width:1px + style C fill:#f9b591,stroke:#333,stroke-width:1px + style A --- D stroke:none + style B --- D stroke:none + style C --- D stroke:none + style Left --- D stroke:none + +``` + +F.743.34(25) + +Diagram of business roles in an industry visual service ecosystem. On the left, three stacked boxes represent 'Computing service consumers', 'Industry visual service and application support platform', and 'Computing service providers'. On the right, a single box represents 'Network providers'. Lines connect each of the three left boxes to the 'Network providers' box. Below the diagram is the text 'F.743.34(25)'. + +**Figure 1 – Business roles in an industry visual service ecosystem** + +### 6.1 Computing service providers + +The computing service provider is the role responsible for providing cloud computing services, edge computing services and job computing services. It registers with the industry visual service and application support platform and exposes its computing service capabilities and types so that the platform can classify, manage and schedule them. Specifically, the types of computing services include AI computing services, function computing services, job computing services, application hosting and cloud infrastructure services. Using one or more computing resources, these computing services are exposed in the form of systems, devices or software, such as a single chip (e.g., central processing unit (CPU), graphics processing unit (GPU), neural network processing unit (NPU), etc.) server, or a server or server cluster that integrates multiple chips (e.g., CPU plus GPU plus NPU), as well as software running on the server (including virtual machines, containers and application programming interfaces (APIs)), etc. + +Service providers have business relationship with industry visual service and application support platform and network providers. + +### 6.2 Industry visual service and application support platform + +The industry visual service and application support platform is the role responsible for managing and scheduling the computing services provided by service providers, including registration and full lifecycle management of computing services, along with unified scheduling, authentication and billing of computing services. + +The following service functions are identified (but not limited to): + +- Computing service presentation; +- Computing service classification; +- Computing service management. + +The industry visual service and application support platform has a business relationship with computing service providers, computing service consumers and network providers. + +### 6.3 Computing service consumers + +The computing service consumer is the role responsible for consuming or using computing services provided by the computing service providers, and paying after the service is completed. Before that, service consumers first register on the industry visual service and application support platform and become authorized users of the system. + +The computing service consumers have business relationship with the industry visual service and application support platform and network providers. + +### 6.4 Network providers + +The network provider is the role responsible for providing channels for computing service providers and service consumers to establish connections. Based on the service requirements of service consumers, it provides network services such as IMT-2020, WLAN, optical fibre and VPNs to meet the needs of service providers and consumers. + +The network providers have a business relationship with the computing service providers, industry visual service and application support platform and computing service consumers. + +## 7 Reference framework of service platform to support industry visual services + +Figure 2 shows the reference framework of service platform to support industry visual services. It is composed of four layers, namely the application layer, service and application support layer, network layer and computing service provision layer. + +![Figure 2: Reference framework of service platform to support industry visual services. The diagram shows four stacked layers. The top layer is the 'Application layer' containing boxes for 'Smart grid visual application', 'Intelligent traffic visual application', 'Industrial automation visual application', 'Remote medical visual application', and an ellipsis. Below these is a wide box for 'Computing service browsing or client agents'. The second layer is the 'Industry visual service and application support layer' with three boxes: 'Computing service presentation' (top), 'Computing service classification' (bottom left), and 'Computing service management' (bottom right). The third layer is the 'Network layer' with a box for 'Networking and transport (e.g., IMT-2020, WLAN, optical fibre, VPN, etc.)'. The bottom layer is the 'Computing service provision layer' with three boxes: 'Cloud computing services', 'Edge computing services', and 'Job computing services'.](b8661c6c54f72ecc7ff6cb05e47b2891_img.jpg) + +| | | | | | | +|-------------------------------------------------------|---------------------------------------------------------------------------|----------------------------------------|------------------------------------------|-----------------------------------|------------------------| +| Application layer | Smart grid visual application | Intelligent traffic visual application | Industrial automation visual application | Remote medical visual application | ... | +| | Computing service browsing or client agents | | | | | +| Industry visual service and application support layer | Computing service presentation | | | | | +| | Computing service classification | | Computing service management | | | +| Network layer | Networking and transport (e.g., IMT-2020, WLAN, optical fibre, VPN, etc.) | | | | | +| Computing service provision layer | Cloud computing services | | Edge computing services | | Job computing services | + +Figure 2: Reference framework of service platform to support industry visual services. The diagram shows four stacked layers. The top layer is the 'Application layer' containing boxes for 'Smart grid visual application', 'Intelligent traffic visual application', 'Industrial automation visual application', 'Remote medical visual application', and an ellipsis. Below these is a wide box for 'Computing service browsing or client agents'. The second layer is the 'Industry visual service and application support layer' with three boxes: 'Computing service presentation' (top), 'Computing service classification' (bottom left), and 'Computing service management' (bottom right). The third layer is the 'Network layer' with a box for 'Networking and transport (e.g., IMT-2020, WLAN, optical fibre, VPN, etc.)'. The bottom layer is the 'Computing service provision layer' with three boxes: 'Cloud computing services', 'Edge computing services', and 'Job computing services'. + +F.743.34(25) + +Figure 2 – Reference framework of service platform to support industry visual services + +**Application layer.** At the application layer, consumers choose the type of computing service they need to complete various tasks, such as industry video surveillance, smart grid visual application, industrial automation visual application, intelligent traffic visual application and remote medical visual application. An application layer is required to provide browsing or client agent functions for users to choose and operate. + +**Industry visual service and application support layer.** It includes three functional modules: computing service presentation module, computing service classification module and computing service management module. Computing service management completes the registration and authentication management of computing service providers and consumers/users. Computing service classification and computing service presentation modules complete the classification of computing service capabilities and provides a method to present various service types and their type description information for service consumers/users to select. + +**Network layer.** The network layer provides networking and transport to establish channels for service providers and consumers. + +**Computing service provision layer.** The computing service provision layer provides cloud computing services, edge computing services, job computing services. The computing service + +providers is required to register with the industry visual service and application support functional modules and expose their computing service capabilities and types to the functional modules for service classification and management. + +## **8 Functional requirements for the service platform** + +Based on the above reference framework, the functional requirements for the service platform are described in clauses 8.1-8.4. + +### **8.1 Functional requirements for the application layer** + +At the application layer, consumers choose the type of computing service they need to complete various tasks, such as industry video surveillance, smart grid visual application, industrial automation visual application, intelligent traffic visual application and remote medical visual application. + +- It is required that service consumers first register on the industry visual service and application support functional modules and become authorized users. +- It is required that the application layer provide browsing or client agent functions for users to choose and operate, or that users input their demand information (e.g., using demand template) and the application layer returns the appropriate service type recommended by the service and application support functional modules. +- It is recommended that the application layer dynamically update the list of computing services and the resource description information (such as central or edge nodes, types and quantities of chips, single server or server clusters, virtual machines, containers and APIs), allowing users to preview and select appropriate services. + +### **8.2 Functional requirements for the industry visual service and application support layer** + +The industry visual service and application support layer includes three functional modules: the computing service presentation module, computing service classification module and computing service management module. Computing service management completes the registration and authentication management of computing service providers and consumers/users, as well as the management of the provided service capabilities and types, and schedules service requests from users to the service nodes. Computing service classification and computing service presentation modules complete the classification of computing service capabilities and provide a method to present various service type and their type description information for service consumers/users to select. The computing service presentation module supports recommending appropriate service types to consumers/users based on their demand information. + +#### **8.2.1 Requirements for computing service classification** + +The computing service classification module needs to support the abstraction of computing resources provided by computing service providers, and the subdivision and quantification of cloud computing, edge computing and job computing resources according to users' service and application needs. It is recommended that heterogeneous computing service types be divided into the following five categories: + +- Job computing. Based on the user's job files and job requirements, allocate high-performance computing resources and provide feedback results after the job is completed. +- AI computing. Allocate inference models or training tasks to appropriate one or more computing service nodes to achieve on-demand AI computing services. +- Application hosting. Allocate application mirror files to appropriate computing service nodes and configure reasonable virtual machine and container resources to provide hosting services. + +- Cloud infrastructure. Based on service requirements, provide virtual machines, a bare machine and container resources to users, intelligently allocate and reserve computing resources deployed centrally or in close proximity to the user or at the network edge. +- Function computing. Event driven hosted serverless computing. Schedule the business logic code submitted by users to appropriate computing nodes, allocate computing resources and run the code in a flexible and reliable manner. + +#### **8.2.2 Requirements for computing service management** + +Computing service management completes the registration and authentication management of service providers and consumers/users, and realizes functions such as the registration of computing services, full lifecycle management and monitoring and unified scheduling of computing services, as well as authentication of computing services. + +- It is required to support registration, query, modification, deletion and other operations for all service nodes, as well as the list of all computing resources, operation and maintenance management and log management for all nodes. +- It is recommended to support methods such as job scheduling, virtual machine scheduling and container scheduling and distribute encapsulated mirror files or jobs to designated computing service nodes. +- It is recommended to support dynamic scaling and migration of hosted applications. +- It is recommended to support multiple scheduling strategies and their configurations, including but not limited to differentiated scheduling based on use location, task type, chip type and quantity, service quality and other factors. +- It is recommended to support differentiated scheduling based on reasoning time consuming, training time consuming and other factors, and support scheduling to edge computing nodes. +- It is recommended to support unified monitoring of all accessed computing service resources and obtaining the available computing resources and resource usage of each service node. +- It is recommended to support full lifecycle management of service nodes, including identification information, access time, operation status, access location, online/offline and management information of each computing resource. +- It is recommended to support accurately verifying the identity and integrity of computing service providers and their resources, as well as putting security measures in place to ensure resistance against external attacks and internal tampering. + +#### **8.2.3 Requirements for computing service presentation** + +Computing service presentation is a portal that provides users with various abstract computing resources, service type description and service orchestration, and supports measuring and recommending appropriate service types based on their demand information. + +- It is required to provide service descriptions and lists of computing service resources, and dynamically update the list. +- It is recommended to provide computing service supply templates and demand templates, as well as trial and feedback mechanisms. +- It is recommended that suitable computing service types and products be provided based on user needs, helping users choose the optimal service solution. +- It is recommended to provide a unified service interface for users to facilitate integration and calling by other systems. +- It is recommended to help users orchestrating multiple service applications, supporting mixed orchestration of job computing, AI computing, function computing and other hosted applications to assist users in generating one or more workflows. + +### **8.3 Functional requirements for the network layer** + +Network layer provides networking and transport to establish channels for service providers and consumers. Based on the service requirements of service consumers, it provides network services such as IMT-2020 and WLAN, optical fibre, VPNs or network slicing to meet the needs of service providers and consumers. + +- It is required that network provides high-performance, low latency and highly reliable connection channels to ensure real-time service priority. +- It is recommended that network can handle high concurrency service requests and user authentication requests, ensuring stable operation when used by multiple users. + +### **8.4 Functional requirements for the computing service provision layer** + +The computing service provision layer provides cloud computing services, edge computing services and job computing services. In order to facilitate users to better choose the computing services they need, computing service providers are required to further decompose and abstract their computing service resources, quantify computing service resources as much as possible and expose them in the form of systems, devices or software, such as a single chip (e.g., CPU/GPU/NPU) server, or a server or server cluster that integrates multiple chips (e.g., CPU plus GPU plus NPU), as well as software running on the server (including virtual machines, containers and APIs), etc. + +- It is required that computing service providers register with the industry visual service and application functional modules and expose their computing service capabilities and types to the service and application functional modules for service classification and management. + +## Appendix I + +## Use cases of a service platform to support industry visual services + +(This appendix does not form an integral part of this Recommendation.) + +### Use case 1: Industrial automation production visual services + +Industry visual system needs to be deployed at each work unit for object/material recognition, component code reading, positioning and assembly, and automatic inspection and detection, as well as AI monitoring and production process management. For example, mechanical arms, combined with industrial cameras, can intelligently sort/grab objects, intelligently recognize and locate to complete operations of dispensing/welding, screw tightening and stacking. Industrial cameras can detect the quality of products on the production line in real time, including appearance defect detection, size detection, pattern detection, etc. Automated guided vehicles (AGVs) and robots can use machine vision and other sensors to complete loading and unloading, handling and assembly. In addition, AI video monitoring is required to monitor the behaviour of production line equipment and operators and report any abnormalities in a timely manner. These processes will generate a large number of images and data. Therefore, it is necessary to have strong computing resources and capabilities (such as cloud infrastructure, application hosting and AI computing) deployed in close proximity to a production line or at the network edge, as well as high reliability, low latency and high speed communication networks, in order to quickly respond to service requests from various work units and accurately complete various operations and processes. + +Based on the demand for visual services in the industrial automation production activities mentioned above, Figure I.1 provides an overview of a computing resource scheduling process that supports visual services. Firstly, the computing service presentation module in the industry visual service and application support layer provides service descriptions and lists of computing service resources. Industrial automation production visual service users can select the service provider and computing service types they need through computing service browser or client agents, based on their service requirements. The services include visual detection, visual measurement, visual positioning and navigation, video surveillance and on-site inference, as well as algorithm and model training for the applications. Secondly, the computing service management function module in the industry visual service and application support layer completes service orchestration, supporting mixed orchestration of cloud infrastructure at the network edge, application hosting, and AI computing based on users' needs, and then schedules users' service requests to appropriate computing service nodes to complete various operations and processes. + +![Figure I.1: An overview of computing resource scheduling that supports visual services. The diagram shows a flow from 'Industrial automation production visual service demands' to 'Visual services', which then branch into 'Visual detection', 'Visual measurement', and 'Algorithm/Models training'. These further lead to 'Visual positioning and navigation', 'On-site inference', and 'Video monitoring'. These services are supported by 'Cloud infrastructure (Network edge)', 'Application hosting', and 'AI computing'. A 'Service platform' on the right contains 'Service presentation', 'Service orchestration', and 'Service scheduling' modules that interact with the demands and the cloud infrastructure.](33ed1f9b27c7c21c797aa928b0f06851_img.jpg) + +``` + +graph TD + subgraph SP [Service platform] + SP1[Service presentation] + SP2[Service orchestration] + SP3[Service scheduling] + SP1 --- SP2 --- SP3 + end + IAP[Industrial automation production visual service demands] <--> SP1 + IAP --> VS[Visual services] + VS --> VD[Visual detection] + VS --> VM[Visual measurement] + VS --> AMT[Algorithm/Models training] + VD --> VPN[Visual positioning and navigation] + VM --> OI[On-site inference] + AMT --> VMON[Video monitoring] + VPN --- OI --- VMON + OI --> CI[Cloud infrastructure Network edge] + OI --> AH[Application hosting] + OI --> AC[AI computing] + CI --- AH --- AC + SP3 <--> AC + F7433425[F.743.34(25)] + +``` + +Figure I.1: An overview of computing resource scheduling that supports visual services. The diagram shows a flow from 'Industrial automation production visual service demands' to 'Visual services', which then branch into 'Visual detection', 'Visual measurement', and 'Algorithm/Models training'. These further lead to 'Visual positioning and navigation', 'On-site inference', and 'Video monitoring'. These services are supported by 'Cloud infrastructure (Network edge)', 'Application hosting', and 'AI computing'. A 'Service platform' on the right contains 'Service presentation', 'Service orchestration', and 'Service scheduling' modules that interact with the demands and the cloud infrastructure. + +**Figure I.1 – An overview of computing resource scheduling that supports visual services** + +### Use case 2: Intelligent traffic visual services + +Intelligent traffic involves intelligent motor vehicles and smart roads as well as traffic management systems. Motor vehicles are equipped with various sensors, such as high-definition cameras, laser radars, millimetre wave radars and other devices, which can obtain information about themselves and the surrounding environment, and transmit this information to the surrounding vehicles, edge computing nodes and cloud platforms for processing. There are various perception devices deployed on smart roads, including laser radar, high-definition cameras and millimetre wave radar, which can accurately detect the status of driving vehicles and real-time road conditions, providing support for intelligent motor vehicles and management systems. For remote driving situations, a remote cockpit can display 360 degree video/images from the vehicle carrying equipment by front view, rear view and look around using high-definition cameras, as well as displaying the road environment images obtained by roadside cameras, so as to ensure that the remote driver has the same field of vision as the local driver. All of these data, images and videos are transmitted to the traffic management system to achieve remote takeover, intelligent scheduling, data analysis, perception calculation, fault monitoring, path planning, joint decision-making and collaborative control, as well as data mining, deep learning, algorithm and model training, etc.. It is necessary for the service platform to schedule cloud infrastructure, application hosting, AI computing and function computing to complete processing. + +Based on the demand for visual services in intelligent traffic mentioned above, Figure I.2 provides an overview of a computing resource scheduling process that supports visual services. Firstly, the computing service presentation module in the industry visual service and application support layer provides service descriptions and lists of computing service resources. Intelligent traffic visual service users can select the service provider and computing service type they need through computing service browser or client agents, based on their service requirements. The services include road environment detection, visual monitoring, visual fusion, path planning, decision-making and scheduling, as well as algorithm and models training for the applications. Secondly, the computing service management function module in the industry visual service and application support layer completes service orchestration, supporting mixed orchestration of cloud infrastructure, application hosting, AI computing and function computing based on users' needs, and then schedules users' service requests to appropriate computing service nodes to complete various tasks. + +![A flowchart titled 'An overview of computing resource scheduling that supports visual services'. It shows a top-down flow from 'Intelligent traffic service demands' to 'Visual services', which then branches into a group of six services: 'Road environment detection', 'Decision-making and scheduling', 'Algorithm/Models training', 'Visual fusion', 'Visual monitoring', and 'Path planning'. These services are supported by a bottom group of four resources: 'Cloud infrastructure', 'Application hosting', 'AI computing', and 'Function computing'. A 'Service platform' on the right contains 'Service presentation', 'Service orchestration', and 'Service scheduling', which interacts with the demands and resources.](fa859e4e468bfb2710a94527f2c504af_img.jpg) + +``` + +graph TD + subgraph Service_platform [Service platform] + SP[Service presentation] + SO[Service orchestration] + SS[Service scheduling] + SP --- SO + SO --- SS + end + + ITSD[Intelligent traffic service demands] <--> SP + ITSD --> VS[Visual services] + VS --> RES1[Road environment detection] + VS --> RES2[Decision-making and scheduling] + VS --> RES3[Algorithm/Models training] + RES1 --> RES4[Visual fusion] + RES2 --> RES5[Visual monitoring] + RES3 --> RES6[Path planning] + + subgraph Resources [ ] + direction TB + R1[Cloud infrastructure] + R2[Application hosting] + R3[AI computing] + R4[Function computing] + end + + Resources --> RES2 + SS --> R4 + + style Resources fill:none,stroke-dasharray: 5 5 + style Service_platform fill:none,stroke-dasharray: 5 5 + +``` + +F.743.34(25) + +A flowchart titled 'An overview of computing resource scheduling that supports visual services'. It shows a top-down flow from 'Intelligent traffic service demands' to 'Visual services', which then branches into a group of six services: 'Road environment detection', 'Decision-making and scheduling', 'Algorithm/Models training', 'Visual fusion', 'Visual monitoring', and 'Path planning'. These services are supported by a bottom group of four resources: 'Cloud infrastructure', 'Application hosting', 'AI computing', and 'Function computing'. A 'Service platform' on the right contains 'Service presentation', 'Service orchestration', and 'Service scheduling', which interacts with the demands and resources. + +**Figure I.2 – An overview of computing resource scheduling that supports visual services** + +## Bibliography + +- [b-ITU-T F.748.16] Recommendation ITU-T F.748.16 (2022), *Requirements for applications and services in smart manufacturing based on machine vision.* +- [b-ITU-T X.746] Recommendation ITU-T X.746 (2000), *Information technology – Open Systems Interconnection – Systems Management: Scheduling function.* +- [b-ITU-T X.1046] Recommendation ITU-T X.1046 (2020), *Framework of software-defined security in software-defined networks/network functions virtualization networks.* +- [b-ITU-T Y.101] Recommendation ITU-T Y.101 (2000), *Global Information Infrastructure terminology: Terms and definitions.* +- [b-ITU-T Y.3073] Recommendation ITU-T Y.3073 (2024), *Framework for service function chaining in information-centric networking.* +- [b-ITU-T Y.3500] Recommendation ITU-T Y.3500 (2014) ISO/IEC 17788:2014, *Information technology – Cloud computing – Overview and vocabulary.* +- [b-ITU-T Y.3540] Recommendation ITU-T Y.3540 (2023), *Edge computing – Overview and high-level requirements.* +- [b-ISO/IEC 11557] ISO/IEC 11557:1992, *Information technology – 3,81 mm wide magnetic tape cartridge for information interchange – Helical scan recording – DDS-DC format using 60 m and 90 m length tapes.* + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.743.35-202512-I_PDF-E/raw.md b/marked/F/T-REC-F.743.35-202512-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..6cc5ace5754432118d97ff22b154aa02e6c9a50d --- /dev/null +++ b/marked/F/T-REC-F.743.35-202512-I_PDF-E/raw.md @@ -0,0 +1,573 @@ + + +# Recommendation**ITU-T F.743.35 (12/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +### **Requirements and functional architecture for a machine-vision-based intelligent farm inspection system** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.743.35 + +# Requirements and functional architecture for a machine-vision-based intelligent farm inspection system + +## Summary + +The machine-vision-based intelligent farm inspection system is based on agricultural autonomous vehicles and can complete the inspection task using machine vision under complex farm conditions. This system has the functions of data collection, video processing, object detection, data encapsulation, video transmission and remote control. The system can evaluate crop growth and detect pests more efficiently than can be done with manual work. + +Recommendation ITU-T F.743.35 specifies requirements and functional architecture for the machine-vision-based intelligent farm inspection system and provides use cases. It is intended to guide the design and development of machine-vision-based intelligent farm inspection systems. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.743.35 | 2025-12-14 | 21 | 11.1002/1000/16576 | + +## Keywords + +Agricultural autonomous vehicle, inspection system, intelligent video surveillance, machine vision. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2026 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|--------------|-----------------------------------------------------------------------------------|----| +| 1 | Scope..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 1 | +| 4 | Abbreviations and acronyms ..... | 2 | +| 5 | Conventions ..... | 2 | +| 6 | Overview..... | 2 | +| 7 | Requirements ..... | 3 | +| 7.1 | Requirements for data collection..... | 3 | +| 7.2 | Requirements for data analysis..... | 4 | +| 7.3 | Requirements for data encapsulation..... | 4 | +| 7.4 | Requirements for data storage ..... | 5 | +| 7.5 | Requirements for human–machine interaction..... | 5 | +| 7.6 | Requirements for security and personally identifiable information protection..... | 5 | +| 8 | Functional architecture ..... | 5 | +| 8.1 | Agricultural autonomous vehicle ..... | 6 | +| 8.2 | Local server ..... | 7 | +| 8.3 | Cloud service centre ..... | 8 | +| 8.4 | Mobile terminal ..... | 8 | +| 9 | Reference points ..... | 8 | +| 9.1 | Reference point Ca/Ca': DCU–DAU..... | 8 | +| 9.2 | Reference point Ce/Ce': DCU–DEU ..... | 8 | +| 9.3 | Reference point Ae: DAU–DEU ..... | 8 | +| 9.4 | Reference point Et/Et': DEU–DTUA ..... | 8 | +| 9.5 | Reference point Ts: DTUA–CSC ..... | 9 | +| 9.6 | Reference point St: CSC–mobile-terminal..... | 9 | +| 9.7 | Reference point Tb/Tb': DTUS–DSU ..... | 9 | +| 9.8 | Reference point Bs/Bs': DSU–DDU..... | 9 | +| 9.9 | Reference point Mt: HMI–DTUS..... | 9 | +| 9.10 | Reference point Tt/Tt'/Tt'': DTUA–DTUS ..... | 9 | +| 9.11 | Reference point Tp: DTUA – Vehicle/camera operation unit..... | 9 | +| Appendix I | – Use cases of machine-vision-based intelligent farm inspection system ..... | 10 | +| I.1 | Use case of pest detection of fungus in greenhouse ..... | 10 | +| I.2 | Use case of cotton growth status evaluation..... | 10 | +| Bibliography | ..... | 12 | + + + +# Recommendation ITU-T F.743.35 + +# Requirements and functional architecture for a machine-vision-based intelligent farm inspection system + +# 1 Scope + +This Recommendation identifies the requirements and functional architecture for the machine-vision-based intelligent farm inspection system and provides use cases. + +The scope of this Recommendation includes: + +- Overview; +- Requirements; +- Functional architecture; +- Use cases. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +# 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 machine vision** [b-ITU-T F.748.16]: Signal processing to acquire, process and interpret an image or video to support visual analysis for applications, such as automatic inspection, process control and guidance. + +**3.1.2 personally identifiable information** [b-ITU-T Y.2720]: The information pertaining to any living person, which makes it possible to identify such individual (including the information capable of identifying a person when combined with other information, even if the information does not clearly identify the person). + +**3.1.3 video and image information** [b-ITU-T H.626.5]: Video clips, images, related files and content description information. + +**3.1.4 video surveillance system** [b-ITU-T H.626]: A telecommunication service focusing on video (including audio and image) application technology, which is used to remotely capture multimedia (such as audio, video, image, alarm signal, etc.) and present them to the end user in a user-friendly manner, based on a managed broadband network with ensured quality, security and reliability. + +### 3.2 Terms defined in this Recommendation + +None. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|----------------------------------------| +| AAV | Autonomous Agricultural Vehicle | +| CDN | Content Delivery Network | +| CSC | Cloud Service Centre | +| DAU | Data Analysis Unit | +| DCU | Data Collection Unit | +| DDU | Data Display Unit | +| DEU | Data Encapsulation Unit | +| DSU | Data Storage Unit | +| DTUA | Data Transmission Unit in AAV | +| DTUS | Data Transmission Unit in Local Server | +| HMI | Human–Machine Interface | +| IVS | Intelligent Video Surveillance | +| PII | Personally Identifiable Information | +| RTMP | Real-Time Messaging Protocol | +| RTSP | Real-Time Streaming Protocol | +| TCP | Transmission Control Protocol | +| UAV | Unmanned Aerial Vehicle | + +# 5 Conventions + +In this Recommendation, the following conventions apply: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "is recommended" indicate a specification which is recommended but which is not absolutely required. Thus this specification need not be present to claim conformance. +- The keyword 'YUV' describes a colour model that denotes luma as the Y component, blue luminance as the U component and red luminance as the V component. + +# 6 Overview + +Crop growth and pest monitoring are important issues in agriculture. An intelligent video surveillance system can complete the inspection of crop growth and pests more efficiently with the replacement of manual work. At present, traditional intelligent video surveillance (IVS), which requires a large number of frame-based or pillar-based fixed cameras, is costly and it is not easy to cover the entire area. Unmanned aerial vehicles (UAVs) are hardly used for IVS in greenhouses, because mechanical failures may occur during low altitude inspection operations. Neither traditional IVS nor UAV-IVS can collect the complete information of one crop. + +A machine-vision-based intelligent farm inspection system can solve the above problems, and represents a new type of IVS for agricultural scenarios. The system mainly consists of autonomous agricultural vehicles (AAVs), a local server, a cloud service centre (CSC) and mobile terminals, as shown in Figure 1. Utilizing the cameras and sensors installed on the AAVs, crop video and environment information are collected comprehensively, and real-time evaluation of crop growth + +status and pest detection are conducted on the vehicle end. The status/analysis result data are sent to the local server. Video data are transmitted to the local server and the CSC. Mobile terminals can show videos from the CSC. Under poor communication conditions, video transmission can be delayed, and only environmental and alarm information can be transmitted to the local server. The system provides comprehensive monitoring services that can cover every plant in farms with vast areas and can work under poor communication conditions as well as in greenhouses. Human staff can issue commands through the local server to control the AAVs and the cameras mounted on them. + +![Figure 1 – Overview of machine-vision-based intelligent farm inspection system. The diagram shows four main components in dashed boxes: Mobile terminals (top left), CSC (top center), Local server (top right), and AAVs (bottom center). Arrows indicate data flow: Video (black) from AAVs to CSC to Mobile terminals; Control (red) from Local server to AAVs; Status/analysis data (green) from AAVs to Local server. A legend at the bottom defines the arrow colors: Video (black), Control (red), Status/analysis data (green). The reference F.743.35(25) is noted at the bottom right.](acfc53eca625d62b38aa2563efa95c3e_img.jpg) + +F.743.35(25) + +Figure 1 – Overview of machine-vision-based intelligent farm inspection system. The diagram shows four main components in dashed boxes: Mobile terminals (top left), CSC (top center), Local server (top right), and AAVs (bottom center). Arrows indicate data flow: Video (black) from AAVs to CSC to Mobile terminals; Control (red) from Local server to AAVs; Status/analysis data (green) from AAVs to Local server. A legend at the bottom defines the arrow colors: Video (black), Control (red), Status/analysis data (green). The reference F.743.35(25) is noted at the bottom right. + +**Figure 1 – Overview of machine-vision-based intelligent farm inspection system** + +# 7 Requirements + +The requirements for machine-vision-based intelligent farm inspection system can be classified into requirements for data collection, requirements for data analysis, requirements for data encapsulation, requirements for data storage, requirements for human-machine interaction and requirements for security and personally identifiable information (PII) protection. + +### 7.1 Requirements for data collection + +#### 7.1.1 Environmental data collection + +- The system is recommended to use environmental sensors with high accuracy and high precision. +- The system is recommended to use stable and reliable environmental sensors. +- The system is recommended to support several types of environmental data, such as temperature, humidity, light and carbon dioxide content. +- The system is required to comply with a predetermined collection strategy, such as collecting data at predetermined time intervals. + +#### 7.1.2 Autonomous vehicle status data collection + +- The system is required to have sufficient autonomous vehicle status data types to ensure that human or computers can analyse and judge whether the autonomous vehicle is operating in a safe state, including power, position, speed, acceleration, angle, heading and attitude angle. +- The system is required to collect autonomous vehicle status data at high frequency to ensure that human or computers can analyse and judge whether the unmanned vehicle is operating in a safe state. + +#### 7.1.3 Video data collection + +- The system is recommended to use YUV format without any transformation or processing for visible light video data collection. + +- The system is recommended to have video resolution at 1280\*720 or above, and high resolution ensures subsequent object detection and video encoding tasks. + +### **7.2 Requirements for data analysis** + +#### **7.2.1 Environmental data analysis** + +- The system is required to support environmental anomaly detection, including temperature anomalies, humidity anomalies and light anomalies. +- The system is recommended to support environmental anomaly detection at a low false alarm rate. + +#### **7.2.2 Video data analysis for disease and pest detection** + +- The system is required to support pest and disease detection of different crops for the specific scenarios in which the system works. +- The system is required to extract image features from detected pest and disease targets. +- The system is required to identify the types of pests and diseases. +- The system is required to store local images of disease and pest occurrences separately. +- The system is recommended to evaluate the severity of the occurrence of diseases and pests. +- The system is recommended to estimate the time of occurrence of pests and diseases. +- The system is recommended to include field weed identification function. +- The system is recommended to have high detection accuracy with low false alarm rate. + +#### **7.2.3 Video data analysis for crop growth status evaluation** + +- The system is recommended to support growth status evaluation of different crops. +- The system is recommended to segment the crop image area to be evaluated from the video image. +- The system is recommended to extract the growth status characteristics of crops. + +### **7.3 Requirements for data encapsulation** + +#### **7.3.1 Status and analysis of data encapsulation** + +- The system is recommended to use transmission control protocol (TCP) to complete the transmission of status data, analysis data and control instructions between the AAV and local server. +- The system is recommended to use reasonable communication frame formats for the three types of data, including status data, analysis data and control instructions, to facilitate mutual recognition of the data. +- The system is required to encapsulate the status and analysis data at the AAV end and then transmit it to the local server. +- The system is required to encapsulate the control instruction at the local server and then transmit it to the AAV. + +#### **7.3.2 Video encapsulation** + +- The system is recommended to encode the video data using ITU-T H.264, ITU-T H.265, and MPEG4. +- The system is recommended to encapsulate video using real-time messaging protocol (RTMP) [b-Parmar] or real-time streaming protocol (RTSP) [b-IETF RFC 7826] protocol. + +### **7.4 Requirements for data storage** + +- The system is recommended to store multiple types of data and generate integrated data. +- The system is recommended to support long-term storage of integrated data and raw video data. +- The system is recommended to support big data storage technology to ensure the efficiency of external retrieval and query. + +### **7.5 Requirements for human–machine interaction** + +- The system is required to provide a convenient human–machine interaction interface for a human to control AAVs and cameras. + +### **7.6 Requirements for security and personally identifiable information protection** + +- The system is recommended to support verification mechanisms, encryption mechanisms and access control to ensure the integrity of specific data during the collection, storage and transmission processes. +- The system is recommended to perform PII authentication on users to ensure data security. + +# **8 Functional architecture** + +The machine-vision-based intelligent farm inspection system is a kind of video surveillance system that combines video processing, object detection and state evaluation technologies. This system can analyse, organize and manage video data, video and image information, sensor information, detection results and control instructions effectively. The system mainly consists of four parts: the AAV subsystem, local server subsystem, CSC and mobile terminals. Figure 2 shows the composition of the system. + +![Functional architecture of machine-vision-based intelligent farm inspection system diagram. The diagram shows a hierarchical system of units: Local server (containing DDU, HMI, DSU), DTUS, DTUA, DEU, DAU, DCU, and AAV. It also includes external components like Mobile terminal and CSC. Arrows indicate data flow: Video (black), Control (red), and Status/analysis data (green).](d4af765160d04ecef538e5066006dc77_img.jpg) + +The diagram illustrates the functional architecture of a machine-vision-based intelligent farm inspection system. It is organized into several functional units and external components: + +- Local server:** Contains DDU, HMI, and DSU. DDU and HMI send control signals (red arrows, labeled $M_t$ ) to the Data sending unit in DTUS. DSU contains Status/detection result and Video modules. Status/detection result sends status/analysis data (green arrows, labeled $T_b'$ ) to the Data receiving unit in DTUS. Video sends video data (black arrows, labeled $T_b$ ) to the Video receiving unit in DTUS. +- DTUS:** Contains Data sending, Data receiving, and Video receiving units. Data sending receives control signals (red arrows, labeled $T_t''$ ) from the Local server and sends status/analysis data (green arrows, labeled $T_t'$ ) to the DTUA. Data receiving receives status/analysis data (green arrows, labeled $T_t'$ ) from the DTUA and sends control signals (red arrows, labeled $T_t''$ ) to the Local server. Video receiving receives video data (black arrows, labeled $T_t$ ) from the DTUA. +- DTUA:** Contains Data receiving, Data sending, and Video sending units. Data receiving receives status/analysis data (green arrows, labeled $T_t'$ ) from the DTUS and sends control signals (red arrows, labeled $T_p$ ) to the Vehicle/camera operation unit. Data sending receives control signals (red arrows, labeled $T_p$ ) from the Vehicle/camera operation unit and sends status/analysis data (green arrows, labeled $E_t'$ ) to the DEU. Video sending receives video data (black arrows, labeled $T_t$ ) from the DTUS and sends video data (black arrows, labeled $T_s$ ) to the CSC. +- DEU:** Contains Status/analysis data encapsulation and Video encapsulation units. Status/analysis data encapsulation receives status/analysis data (green arrows, labeled $E_t'$ ) from the DTUA and sends status/analysis data (green arrows, labeled $A_e$ ) to the DAU. Video encapsulation receives video data (black arrows, labeled $T_t$ ) from the DTUA and sends video data (black arrows, labeled $C_e$ ) to the DAU. +- DAU:** Contains Environmental data analysis and Video data analysis modules. Environmental data analysis receives status/analysis data (green arrows, labeled $A_e$ ) from the DEU and sends status/analysis data (green arrows, labeled $C_a'$ ) to the DCU. Video data analysis receives video data (black arrows, labeled $C_e$ ) from the DEU and sends video data (black arrows, labeled $C_a$ ) to the DCU. +- DCU:** Contains Vehicle status data collection, Environmental data collection, and Video collection modules. Vehicle status data collection receives status/analysis data (green arrows, labeled $C_a'$ ) from the DAU and sends status/analysis data (green arrows, labeled $C_e'$ ) to the DEU. Environmental data collection receives status/analysis data (green arrows, labeled $C_a$ ) from the DAU and sends status/analysis data (green arrows, labeled $C_e'$ ) to the DEU. Video collection receives video data (black arrows, labeled $C_a$ ) from the DAU and sends video data (black arrows, labeled $C_e$ ) to the DEU. +- AAV:** Contains the Vehicle/camera operation unit, which receives control signals (red arrows, labeled $T_p$ ) from the DTUA and sends control signals (red arrows, labeled $T_p$ ) to the DTUA. +- External components:** Mobile terminal and CSC. Mobile terminal sends status/analysis data (green arrows, labeled $S_t$ ) to the CSC. CSC receives video data (black arrows, labeled $T_s$ ) from the DTUA. + +**Legend:** + +- Video: Black arrow +- Control: Red arrow +- Status/analysis data: Green arrow + +F.743.35(25) + +Functional architecture of machine-vision-based intelligent farm inspection system diagram. The diagram shows a hierarchical system of units: Local server (containing DDU, HMI, DSU), DTUS, DTUA, DEU, DAU, DCU, and AAV. It also includes external components like Mobile terminal and CSC. Arrows indicate data flow: Video (black), Control (red), and Status/analysis data (green). + +**Figure 2 – Functional architecture of machine-vision-based intelligent farm inspection system** + +### 8.1 Agricultural autonomous vehicle + +The AAV subsystem is mainly composed of five functional units: the data collection unit (DCU), data analysis unit (DAU), data encapsulation unit (DEU), data transmission unit (DTUA) and vehicle/camera operation unit. + +#### 8.1.1 Data collection unit + +The DCU consists of three functional modules: vehicle status data collection module, environmental data collection module and video collection module. + +The vehicle status data collection module obtains the status information of the autonomous vehicle itself, including power, location, direction, angle and velocity. + +The environmental data collection module obtains farm environmental information, including temperature, humidity, light and carbon dioxide content. + +The video collection module obtains videos of crops, including visible light video data and infrared video data. + +#### **8.1.2 Data analysis unit** + +The DAU consists of two functional modules: the environmental data analysis module and video data analysis module. + +Environmental data analysis module is responsible for analysing the collected environmental data, such as temperature, humidity, light and carbon dioxide content. If values exceed thresholds, an environmental anomaly status alarm signal is issued. + +Video data analysis module is responsible for pest and disease detection and crop growth status evaluation. Pest and disease detection refers to using target detection algorithms on collected video frames. If pests or diseases are detected, an alarm signal for pest or disease is issued and the corresponding pest or disease analysis result data are generated. Crop growth status evaluation refers to processing the collected video frames, evaluating the growth status of crops and generating the crop growth status result data. + +#### **8.1.3 Data encapsulation unit** + +The DEU consists of two functional modules: the status/analysis data encapsulation module and video encapsulation module. + +The status/analysis data encapsulation module is responsible for encapsulating all data except video data, including collected environmental data, environmental abnormal state alarm signals, pest or disease status alarm signals, pest or disease analysis data, and crop growth status data. + +The video encapsulation module is responsible for encoding raw video data first, followed by encapsulation. For example, H.264 can be used for video encoding, and the RTMP [b-Parmar] protocol can be used to encapsulate video streams. + +#### **8.1.4 Data transmission unit in AAV** + +The DTUA consists of three functional modules: the data sending module, data receiving module and video sending module. + +The data sending module is responsible for sending status/analysis data packets to the local server. + +The data receiving module is responsible for receiving autonomous vehicle control instructions from the local server. + +The video sending module is responsible for delivering encapsulated video streams to the local server and the CSC. + +#### **8.1.5 Vehicle/camera operation unit** + +The vehicle/camera operation unit is responsible for executing control instructions transmitted from the local server. + +### **8.2 Local server** + +The local server is composed of four functional unit: the data transmission unit (DTUS), data storage unit (DSU), data display unit (DDU) and human–machine interface (HMI). + +#### **8.2.1 Data transmission unit in local server** + +The DTU in the local server consists of three functional modules: data sending module, data receiving module and video receiving module. + +The data sending module is responsible for sending control instructions received from the HMI for controlling the autonomous vehicle or camera in the AAV subsystem. + +The data receiving module is responsible for receiving status/analysis data from the AAV subsystem. + +The video receiving module is responsible for receiving videos from the AAV subsystem. + +#### **8.2.2 Data storage unit** + +The DSU stores video data and status/analysis data separately. Integrated data can be generated through techniques such as database table association and knowledge graph. + +#### **8.2.3 Data display unit** + +The DDU displays video, environmental information, autonomous vehicle status, environmental alarm signals, pest detection and analysis results, and crop status evaluation results on the screen of the local server. + +#### **8.2.4 Human–machine interface** + +The HMI provides a human–machine interaction interface for staff to control devices such as autonomous vehicles and cameras. + +### **8.3 Cloud service centre** + +The CSC can perform the live streaming, storage and playback functions of videos. The media server of the CSC obtains the video stream captured in the farm. The media server cooperates with the scheduling strategy server and mixed stream server to transcode the video stream and then push it to the content delivery network (CDN) network. Mobile terminals pull streams from the CDN network to watch farm videos. + +### **8.4 Mobile terminal** + +Mobile terminals, such as smartphones and tablet computers, can obtain videos from the CSC to observe the farm. + +## **9 Reference points** + +### **9.1 Reference point Ca/Ca': DCU–DAU** + +The reference points Ca and Ca' are between the DCU and the DAU. Ca is used to acquire video stream from the DCU and send it to the DAU. + +Ca' is used to acquire environmental information data from the DCU and send them to the DAU. + +### **9.2 Reference point Ce/Ce': DCU–DEU** + +The reference points Ce and Ce' are between the DCU and the DEU. Ce is used to acquire video stream from the DCU and send it to the DEU. + +Ce' is used to acquire vehicle status data from the DCU and send it to DEU. + +### **9.3 Reference point Ae: DAU–DEU** + +The reference point Ae is between the DAU and the DEU. Ae is used to acquire three types of data from DAU, including the environmental information data, the analysis results of environmental data and the analysis results of video data, and send them to the DEU. + +### **9.4 Reference point Et/Et': DEU–DTUA** + +The reference points Et and Et' are between the DEU and the DTUA. Et is used to acquire encapsulated video stream from the DEU and send it to the DTUA. + +Et' is used to acquire encapsulated status/analysis data from the DEU and send them to the DTUA. + +### **9.5 Reference point Ts: DTUA–CSC** + +The reference point Ts is between the DTUA and the CSC. Ts is used to acquire encapsulated video stream from the DTUA and send it to the CSC. + +### **9.6 Reference point St: CSC–mobile-terminal** + +The reference point St is between the CSC and the Mobile terminal. St is used to acquire video stream from the CSC and send it to the mobile terminal. + +### **9.7 Reference point Tb/Tb': DTUS–DSU** + +The reference points Tb and Tb' are between the DTUS and the DSU. Tb is used to acquire the video stream from the DTUS and send it to the DSU. + +Tb' is used to acquire encapsulated status/analysis data from the DTUS and send them to the DSU. + +### **9.8 Reference point Bs/Bs': DSU–DDU** + +The reference points Bs and Bs' are between the DSU and the DDU. Bs is used to acquire the video stream from the DSU and send it to the DDU. + +Bs' is used to acquire the vehicle status information, environmental status information and analysis results from the DSU and send them to the DDU. + +### **9.9 Reference point Mt: HMI–DTUS** + +The reference point Mt is between the HMI and the DTUS. Mt is used to acquire the control instructions of the vehicle or camera from the HMI and send it to the DTUS. + +### **9.10 Reference point Tt/Tt'/Tt'': DTUA–DTUS** + +The reference points Tt, Tt' and Tt'' are between the DTUA and the DTUS. Tt is used to acquire encapsulated video stream from the DTUA and send it to the DTUS. + +Tt' is used to acquire the status information and analysis results from the DTUA and send them to the DTUS. + +Tt'' is used to acquire the control instructions of vehicles or cameras from the DTUS and send them to the DTUA. + +### **9.11 Reference point Tp: DTUA – Vehicle/camera operation unit** + +The reference point Tp is between the DTUA and the vehicle/camera operation unit. Tp is used to acquire the control instructions and send them to the vehicle/camera operation unit. + +## Appendix I + +## Use cases of machine-vision-based intelligent farm inspection system + +(This appendix does not form an integral part of this Recommendation.) + +### I.1 Use case of pest detection of fungus in greenhouse + +A typical use case of machine-vision-based intelligent farm inspection system is shown in Figure I.1. The application scenario is pest detection of fungus in a greenhouse. Figure I.1(a) shows an agricultural autonomous vehicle (AAV) working in the greenhouse. The camera on the AAV can capture the status of crop leaves and stems closely. Videos can be live streamed and uploaded to the cloud live centre. The target recognition algorithms and pest detection algorithms stored in the AAV can help detect fungus, and the vehicle can send out alarms. Figure I.1(b) shows the pest detection results using the algorithms. The temperature, humidity and light intensity sensors configured on the AAV monitor greenhouse conditions in real time and provide environmental background information on the occurrence of pests and diseases. The AAV has GPS or positioning functions to ensure that pest information can be bound to the corresponding geographical coordinates, facilitating statistics and management. Both the automatic control algorithm stored in the local server or the human–machine interface can adjust the distance and angle of the camera on the AAV automatically based on the detected pests and diseases or manually. + +![Agricultural autonomous vehicle (AAV) working in a greenhouse.](a4d009d5dd6a4d83759d6d6538188e23_img.jpg) + +A photograph of a white and black agricultural autonomous vehicle (AAV) with four wheels and a camera sensor mounted on top. The AAV is positioned in a narrow aisle between two long, low tables in a greenhouse. The tables are covered with numerous small, light-colored objects, likely mushroom spawn bags, arranged in rows. The greenhouse structure, with its translucent panels and metal framework, is visible in the background. + +Agricultural autonomous vehicle (AAV) working in a greenhouse. + +a) Agricultural autonomous vehicle working in the greenhouse + +![Pest detection results showing multiple frames with bounding boxes.](0656422bf374a8a7bcc6fe99adc48599_img.jpg) + +A composite image consisting of six smaller frames arranged in a 2x3 grid. Each frame shows a close-up view of a plant, specifically a mushroom, with green and red rectangular bounding boxes overlaid on it. These boxes highlight specific areas of the plant, likely indicating detected pests or diseases. The background of each frame shows the greenhouse environment with rows of plants. + +Pest detection results showing multiple frames with bounding boxes. + +b) Pest detection results + +F.743.35(25) + +**Figure I.1 – The use case of detection of fungus in greenhouse** + +### I.2 Use case of cotton growth status evaluation + +Monitoring the growth status of crops is an important issue in agricultural production. Figure I.2(a) shows an AAV working in a cotton field. The AAV can approach every plant closely, take videos + +from all directions and monitor its growth status, which enables people to obtain complete and detailed crop growth data. Figure I.2(b) shows the interface of a cotton inspection system. + +![Agricultural autonomous vehicle working in cotton fields](7b1cea8c45c49887a750926f2135fa84_img.jpg) + +A photograph of a white, box-shaped autonomous agricultural vehicle with four large black wheels, positioned in a field of young cotton plants. The vehicle is equipped with sensors and cameras. In the background, there are more rows of cotton plants and a clear sky. To the right of the vehicle, there is a small inset image showing a computer screen displaying a live video feed from the vehicle's camera, along with various data overlays. + +Agricultural autonomous vehicle working in cotton fields + +a) Agricultural autonomous vehicle working in cotton fields + +![Cotton field inspection system interface](fed39b841ae2dce01088b84bfc1e2789_img.jpg) + +A screenshot of a digital user interface for a cotton field inspection system. The interface is titled "INSPECTION SYSTEM" and shows various data panels. On the left, there are "Environmental monitoring" gauges for temperature (46.8°C), humidity (30.6%rh), light intensity (65537.0lux), and CO2 levels (401.0ppm). Below this is an "Environmental statistics" section with a bar chart. The central part of the interface features a "Surveillance camera" feed showing a close-up of cotton plants. To the right is a "Map" section with a location pin. Below the camera feed is a "System warnings" section with two entries: "2024-06-16 16:03:55 [66537] 60000lux" and "2024-06-16 16:03:49 [66536] 40000lux". At the bottom is a "Panoramic camera" feed. On the far right, there is a "Device management" section with a "Start" button and status indicators for battery (18%), speed (0.0m/s), and distance (0.7m). The top right corner shows the time "16:03:57" and temperature "16°C-26°C". + +Cotton field inspection system interface + +b) Cotton field inspection system interface + +F.743.35(25) + +**Figure I.2 – The use case of monitoring cotton growth status** + +## Bibliography + +- [b-ITU-T F.748.16] Recommendation ITU-T F.748.16 (2022), *Requirements for applications and services in smart manufacturing based on machine vision.* +- [b-ITU-T H.626] Recommendation ITU-T H.626 (V2) (2019), *Architectural requirements for a video surveillance system.* +- [b-ITU-T H.626.5] Recommendation ITU-T H.626.5 (V2) (2022), *Architecture for intelligent video surveillance systems.* +- [b-ITU-T Y.2720] Recommendation ITU-T Y.2720 (2009), *NGN identity management framework.* +- [b-IETF RFC 7826] IETF RFC 7826 (2016), *Real-Time Streaming Protocol Version 2.0.* +<> +- [b-Parmar] Hardeep S. Parmar and Michael C. Thornburgh (2012), *Adobe's Real Time Messaging Protocol.* +<[https://rtmp.veriskope.com/pdf/rtmp\\_specification\\_1.0.pdf](https://rtmp.veriskope.com/pdf/rtmp_specification_1.0.pdf)> + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.743.4-201703-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.743.4-201703-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9002ee68e2b9ebf63d27d39ba8c73bdafb0e6be6 --- /dev/null +++ b/marked/F/T-REC-F.743.4-201703-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8df701536c25ab8246d2868de3c20c8a4e25168bb52c5eb9078278ab1b5b4fd8 +size 6859 diff --git a/marked/F/T-REC-F.743.4-201703-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg b/marked/F/T-REC-F.743.4-201703-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..06b572774f98f7b2872d13d1adc945b6aac86a3d --- /dev/null +++ b/marked/F/T-REC-F.743.4-201703-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c01c130aee062ac8820095ccf11d29179e0fbeb74899d30d6769d6d529bce37a +size 74829 diff --git a/marked/F/T-REC-F.743.4-201703-I_PDF-E/9c6461e1e94afae4dec455e69a2ce152_img.jpg b/marked/F/T-REC-F.743.4-201703-I_PDF-E/9c6461e1e94afae4dec455e69a2ce152_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..2d03676128be3aee3242ef0f039ceb96a89a099e --- /dev/null +++ b/marked/F/T-REC-F.743.4-201703-I_PDF-E/9c6461e1e94afae4dec455e69a2ce152_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f85e565b17a88ff0e036adb7963a149c930d289dd61c75e0be59506e005452b0 +size 77752 diff --git a/marked/F/T-REC-F.743.4-201703-I_PDF-E/dbe553cf16dd14073b89a8263a428664_img.jpg b/marked/F/T-REC-F.743.4-201703-I_PDF-E/dbe553cf16dd14073b89a8263a428664_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..b4d85d320234f6f78ab1768b422afaf3aa7733a2 --- /dev/null +++ b/marked/F/T-REC-F.743.4-201703-I_PDF-E/dbe553cf16dd14073b89a8263a428664_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b84409570950de1dbbb7bd825f5457fb5eaec8111cc01b305da95a5b94b71a7d +size 136108 diff --git a/marked/F/T-REC-F.743.6-201808-I_PDF-E/raw.md b/marked/F/T-REC-F.743.6-201808-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..3aad911750805196647cb57b11dcddd14fbe374c --- /dev/null +++ b/marked/F/T-REC-F.743.6-201808-I_PDF-E/raw.md @@ -0,0 +1,432 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.743.6** + +(08/2018) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Service requirements for next generation +content delivery networks** + +Recommendation ITU-T F.743.6 + +ITU-T + +![ITU logo](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) features a globe with a red lightning bolt striking across it. To the right of the globe, the text "International Telecommunication Union" is written in blue, with the acronym "ITU" in a larger, bold blue font below it. + +ITU logo + +International +Telecommunication +Union + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.743.6 + +# Service requirements for next generation content delivery networks + +## Summary + +Recommendation ITU-T F.743.6 describes use cases, service requirements and challenges of the next generation content delivery network (CDN). A reference architecture named access CDN for meeting these requirements and challenges, which decouples the common functions and customized services, is also defined. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.743.6 | 2018-08-29 | 16 | 11.1002/1000/13657 | + +## Keywords + +Content delivery network (CDN), next generation content delivery network (NG CDN). + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2018 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-----------------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 1 | +| 5 Conventions ..... | 2 | +| 6 Next generation content delivery network use cases and requirements ..... | 2 | +| 6.1 4K/8K/VR/AR..... | 2 | +| 6.2 Live streaming ..... | 3 | +| 6.3 IoT/edge computing..... | 3 | +| 7 NG CDN requirements summary ..... | 4 | +| 8 Challenges of NG CDN ..... | 5 | +| 9 Reference architecture to meet service requirements ..... | 6 | +| Appendix I – Existing modes for bringing CDN to the edge..... | 8 | + +## Introduction + +With the popularity of video services such as 4K/8K television (TV), virtual reality/augmented reality (VR/AR) and live sports events, demands on network throughput are increasing by orders of magnitude along with the associated requirements for ultra-low latency. These bring new challenges for the current content delivery network (CDN), resulting in a trade-off between exponential increases in the network cost and the need for smooth growth in end-user access price. It will be difficult to cover the throughput magnitude gap by solely relying on a pure underlay network; thus the next generation CDN should solve backbone traffic convergence and improve end-user experience to resolve this gap, and accelerate the introduction of emerging services. + +# Recommendation ITU-T F.743.6 + +## Service requirements for next generation content delivery networks + +# 1 Scope + +This Recommendation describes use cases as the driving force for service requirements, and challenges for the next generation content delivery network (NG CDN). It also provides a reference architecture to meet these service requirements. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T Y.2019] Recommendation ITU-T Y.2019 (2010), *Content delivery functional architecture in NGN*. + +# 3 Definitions + +### 3.1 Terms defined elsewhere + +None. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 access CDN (A-CDN):** A reference architecture for a highly distributed CDN. The architecture horizontally decouples the current vertically integrated CDN system into the content delivery network control part (CDN-CP) for end-to-end (E2E) service control and the content delivery network delivery part (CDN-DP) for distributed deployment. + +**3.2.2 CDN control part (CDN-CP):** Provides customized services including dynamic content acceleration to meet customers' requirements, which is provided by the content delivery network service provider (CDN SP) and Internet content provider (ICP). + +**3.2.3 CDN delivery part (CDN-DP):** Provides a generic content delivery function and open network capability, including real time multi-cast, to support content delivery network service provider (CDN SP) and Internet content provider (ICP) basic requirements provided by operators. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|--------|----------------------------------------| +| A-CDN | Access CDN | +| AR | Augmented Reality | +| CDN | Content Delivery Network | +| CDN-CP | Content Delivery Network Control Part | +| CDN-DP | Content Delivery Network Delivery Part | + +| | | +|--------|-------------------------------------------| +| CDN-SP | Content Delivery Network Service Provider | +| E2E | End-to-End | +| HD | High Definition | +| ICP | Internet Content Provider | +| IoT | Internet of Things | +| ISP | Internet Service Provider | +| IXP | Internet Exchange Point | +| NG CDN | Next Generation of CDN | +| OTT | Over the Top | +| QoE | Quality of Experience | +| QoS | Quality of Service | +| ROI | Range of Interest | +| SD | Standard Definition | +| TV | Television | +| VR | Virtual Reality | + +## 5 Conventions + +None. + +# 6 Next generation content delivery network use cases and requirements + +The existing hot services and merging of new technologies and services such as 4K/8K television (TV), virtual reality/augmented reality (VR/AR), live streaming, and Internet of things (IoT) drive the next generation of CDN (NG CDN) with requirements entering a new decade of development. As a driver, some of the typical use cases of NG CDN are briefly discussed below. + +### 6.1 4K/8K/VR/AR + +For VR/AR applications, latency is the most important performance parameter. VR/AR is latency sensitive because human perception requires accurate and smooth movements in vision. Increased latency can lead to a detached VR experience and contribute to a motion-sickness sensation. VR/AR industries agree that application latency should be less than 20 ms for the motion-to-photon latency to be imperceptible. Range of interest (ROI) needs to be regenerated once a user's head turns, with latency less than 20ms. Content cannot be downloaded wholly with limited access bandwidth, while the terminal requires content delivery at a dynamic ROI zone. + +The network bandwidth for VR/AR is the throughput required to stream 360 degree video to a user, which is another critical parameter for VR/AR applications. 8K quality and above is necessary for VR, as 4K VR 360 degree video is only equivalent to 240p on TV. VR 360 degree video experience may evolve through the stages listed in Table 1, requiring a throughput of 400 Mbit/s and above, more than 100 times higher than the current throughput of high definition (HD) video services + +**Table 1 – VR network requirements (bandwidth and latency)** + +| VR resolution | | Equivalent TV resolution | Bandwidth | Latency | +|--------------------------|-------------------------------------------|--------------------------|-----------------------------------------------------|---------| +| Early stage VR (current) | 1K*1K@visual field
2D_30fps_8bit_4K | 240p | 25 Mbit/s | 40 ms | +| Entry-Level VR | 2K*2K@visual field
2D_30fps_8bit_8K | SD | 100 Mbit/s | 30 ms | +| Advanced VR | 4K*4K@visual field
2D_60fps_10bit_12K | HD | 400 Mbit/s | 20 ms | +| Extreme VR | 8K*8K@visual field
3D_120fps_12bit_24K | 4K | 1 Gbit/s (smooth play)
2.35 Gbit/s (interactive) | 10 ms | + +Due to the acute traffic increase, change in convergence model and high requirement of VR videos on bandwidth, latency and packet loss, traditional high-convergence networks face the following challenges: + +- low network efficiency: Expansion of end-to-end (E2E) network equipment is needed as video traffic increases. CDN is deployed in a higher position and away from the end user; service traffic travels through more network equipment, which therefore occupies a greater number of network resources. In addition, more equipment increases the probability of bottlenecks, blocking and E2E latency; +- poor user experience: With multi-service concurrency, the network utilization ratio increases and packet loss and delay also increase. + +High resolution and bitrates of 4K/8K/VR cause a rapid rise of delivery cost, with standard definition (SD) delivery already occupying 40% - 60% of the cost, which becomes the biggest obstacle for users to enjoy 4K/VR. NG CDN requires low cost with high performance high bit-rate content delivery. + +### 6.2 Live streaming + +Live video brings a much higher demand on concurrency, continuity and low latency. Taking user generated content (UGC) mode for example, hosts are distributed; a traditional centralized CDN could cause extra latency and packet loss while content is uploading. Future live streaming emphasizes liveness and interaction, with video lag ratio under 2%, and latency less than 3s. Spiky unicast traffic of live events creates congestion in transit, core and aggregation. Traditional centralized CDN nodes cannot meet the requirement of live streaming service, since: + +- CDN is a static multi-level cache system with circuitous traffic. It does not take network and topology status into considerations, making it difficult to meet the real time and interactive demand of live streaming; +- live streaming only requires forward instead of traditional CDN storage-forward mode. Further, the traditional server architecture has inferior performance for live streaming. + +Thus, a new architecture is required for CDN services with high performance and edge live streaming as a service. + +### 6.3 IoT/edge computing + +The rapid emergence of IoT will connect numerous devices to the network, generating high concurrency data analysis, storage and processing in the network edge. The core router will be heavily loaded and will become a bottleneck if all traffic flows completely back to the data centre. Distribution of CDN nodes enables and strengthens IoT in data pre-processing, storage and + +forwarding that is specific to large scale thing-to-thing communication. Data could be filtered, merged and lightweight computed in the CDN edge node and then sent back to the source centre. + +From an end-user perspective, users are more and more willing to pay for a better experience. With the development of IoT, NG CDN needs to guarantee user experience in not only traffic and speed, but also in numerous devices, and large numbers of computing and learning tasks. + +CDN faces the following challenges when adapting to IoT: + +- capacity at the edge: Edge computing enables IoT due to the lack of bandwidth infrastructure. Points of presence placed closer to geographic location allows users to interact more quickly with content; +- technology to move information faster: Data has to be moved reliably and on time consistently. CDN should be able to securely accelerate enterprise applications; +- data security: With growing IoT units, the number of distributed denial of service (DDoS) attacks is expected to increase exponentially in the coming years. Research shows that 85% of enterprises around the world are deploying or intend to deploy IoT, while only 10% feel confident in securing devices. + +# 7 NG CDN requirements summary + +As HD/ultra-high definition (UHD) and VR/AR video, live streaming become basic services of operators, network infrastructure needs to support these high-quality services. Thus, the CDN caching, storage and computing abilities should be taken into consideration. For a video service-centric featured future network, CDN service capabilities will become a basic infrastructure for operators, thus becoming a key feature for network reconfiguration and redesign. It is possible that "pay by traffic mode" will enable CDN capabilities to become an operators' main income source (rather than bandwidth offering paid per month). These capabilities are illustrated in Figure 7-1. + +![Figure 7-1: A diagram illustrating the evolution of CDN capabilities. On the left, under 'Traditional OTT CDN', are RTSP/RTP, RTMP, and Multicast IPTV. An arrow points to the right, under 'Infrastructure of basic CDN capability', which includes HTTP streaming, 4K/8K, and AR/VR.](053f1077d592e6622cd21dc4bb4cb366_img.jpg) + +The diagram is a horizontal arrow pointing from left to right, divided into two main sections. The left section is labeled 'Traditional OTT CDN' and contains the text 'RTSP/RTP', 'RTMP', and 'Multicast IPTV'. The right section is labeled 'Infrastructure of basic CDN capability' and contains the text 'HTTP streaming', '4K/8K', and 'AR/VR'. A dashed arrow points from the left section to the right section, indicating a transition or evolution. + +Figure 7-1: A diagram illustrating the evolution of CDN capabilities. On the left, under 'Traditional OTT CDN', are RTSP/RTP, RTMP, and Multicast IPTV. An arrow points to the right, under 'Infrastructure of basic CDN capability', which includes HTTP streaming, 4K/8K, and AR/VR. + +F.743.6(18)\_F7-1 + +**Figure 7-1 – CDN capabilities becomes basic infrastructure of video service-centric featured future network** + +The emergence of new technologies and services such as high-definition television (HDTV), VR/AR and IoT, require NG CDN to be able to meet the requirements of ultra-low latency, high throughput, multi-direction traffic, and experience lossless content delivery. These considerations, as illustrated in Figure 7-2, include: + +- ultra-low latency, improving from seconds to milliseconds, meeting challenges brought by live streaming, live events, network gaming, IoT, etc.; +- ultra-high throughput, improving from kilobits/s or megabits/s to gigabits/s meeting challenges brought by 4K/8K TV, VR/AR, etc.; +- multi-direction delivery from a few sources in the source-tree to everyone as source in a source mesh; +- user experience of lossless delivery to guarantee the content and latency in high reliability. + +![Figure 7-2 – NG CDN requirements. A diagram showing four key requirements for Next Generation CDNs: Ultra low latency, Ultra high throughput, Multi-direction traffic, and Lossless delivery. Each requirement is in a separate box with icons and text. 'Ultra low latency' includes live interaction, VR/AR, network gaming, IoT, and s/m to ms latency. 'Ultra high throughput' includes 4KTV, VR/AR, light field, hologram, and k/Mbps to Gbps. 'Multi-direction traffic' shows a tree network becoming a mesh network where everyone can be a source. 'Lossless delivery' shows a process of loss and delay followed by lossless delivery using adaptive FEC and network prediction.](cfef993dcc8fb513de79eb1f93cf26ae_img.jpg) + +Figure 7-2 – NG CDN requirements. A diagram showing four key requirements for Next Generation CDNs: Ultra low latency, Ultra high throughput, Multi-direction traffic, and Lossless delivery. Each requirement is in a separate box with icons and text. 'Ultra low latency' includes live interaction, VR/AR, network gaming, IoT, and s/m to ms latency. 'Ultra high throughput' includes 4KTV, VR/AR, light field, hologram, and k/Mbps to Gbps. 'Multi-direction traffic' shows a tree network becoming a mesh network where everyone can be a source. 'Lossless delivery' shows a process of loss and delay followed by lossless delivery using adaptive FEC and network prediction. + +F.743.6(18)\_F7-2 + +**Figure 7-2 – NG CDN requirements** + +# 8 Challenges of NG CDN + +With the emergence of new technologies and services such as VR/AR, IoT and holographic interaction, CDN faces issues for accelerating and delivering dynamic and interactive content instead of traditional web services. Challenges of current CDN include: + +- the backbone/metro network construction and expansion cannot meet the requirements from the increase of video and new emerging services. Current CDN nodes are often highly placed at the backbone/metro network. This type of CDN network architecture cannot fit these new services well since it poses a great challenge on the backbone/metro network capacity and the E2E delivery performance; +- the traditional storage and forwarding models have almost no considerations for dynamic routing to guarantee high bandwidth and low latency experiences, and no mechanisms to optimize traffic direction and mesh network capabilities; +- the single node of current CDN is usually located in the backbone/metro network, and in low-cost regions geographically far from users. This makes it hard for current CDN to cover high-rate/high burst traffic and strongly interactive services; +- the lack of data sharing and CDN interworking between CDN/over-the-top (OTT) and network operators does not fully take advantage of the potential of the access network; that is, the important "last mile" role in the CDN value chain; +- disorderly traffic flow and out-of-order traffic causes high network transmission costs; this is undesirable from the perspective of network operators; +- storage capacity and throughput are the main contradictions which CDN should handle. There is a gap between the requirement of CDN edge nodes and source nodes. How the most suitable storage media for different storage capacity and throughput requirements of different content types should be considered. Table 2 provides examples of storage capacity and throughput for different content types. + +**Table 2 – Examples of storage capacity and throughput for different content types** + +| Content type | Storage capacity | Throughput | +|---------------------------|------------------|------------| +| Image | 3.4 PB | 7 Tbit/s | +| Standard definition video | 403 TB | 30 Tbit/s | +| 4K | 24 PB | 1.8 Pbit/s | +| VR | 2.4 EB | 180 Pbit/s | + +# 9 Reference architecture to meet service requirements + +As the background analysis in Appendix I shows, even though major players in the CDN industry, including cloud providers, CDN service providers (SPs) and operators, are tending to bring CDN nodes to the edge in different ways, there are still issues facing the requirements of NG CDN. + +The root cause is that CDN, as a typical overlay network, cannot perceive the underlay network status. An overlay network that ignores the properties of the underlay infrastructure can lead to congestion and reliability issues. + +- some CDN SPs are calling for operators for open network capabilities such as topology information and dynamic running data for targeting performance optimization, quality of service (QoS) guarantees for insuring differentiated services and E2E service ability via tight coordination with underlay network; +- from an operators' perspective, they believe it is not appropriate to open the network to provide this key information. However, operators are suffering from pressures brought by rapid growing bandwidth, and also disordered traffic. + +The existing ITU-T CDN Recommendation [ITU-T Y.2019], published in 2010, describes the content delivery functional architecture in NGN. It has no consideration for the trend of bringing CDN nodes to the edge, which is driven by modern emerging services such as 4K, 8K, AR, VR etc., and with the feature of ultra-high throughput and ultra-low latency. + +Thus, there is a demand for an updated solution to meet both operators' and CDN SP's requirements. This new technology and architecture should be considered through tight coordination between overlay and underlay networks to improve the delivery quality while not bring backbone network heavy investment pressure to the operators. + +A reference architecture for highly distributed CDN, termed access CDN (A-CDN) is illustrated in Figure 9-1. The architecture horizontally decouples the current vertically integrated CDN system into the content delivery network control part (CDN-CP) for E2E service control and the content delivery network delivery part (CDN-DP) for distributed deployment. The key features of this architecture are: + +- CDN-CP: Provides customized services including dynamic content acceleration to meet customers' requirements, which is provided by a CDN SP and Internet content provider (ICP); +- CDN-DP: Provides a generic content delivery function and open network capability, including real time multicast to support the CDN SP and ICP's basic requirement, which is provided by operators; +- A-CDN underlay network: Enhances the underlay network capability for a CDN service, which is provided by operators; +- both CDN-DP and A-CDN underlay networks combine to be multi-telecom A-CDN, an open platform for distributed CDN. Open application programming interfaces (APIs) are provided by the platform via which network capabilities such as multicast can be opened to the CDN SP and ICP. + +This architecture can utilize each role's strength. For SPs, they are strong at CDN service customization. For operators, they have rich central office resources and can provide QoS guarantees through coordination with the underlay. + +The benefits for SPs are footprint extension and user equipment (UE) enhancement. The benefits for operators are high throughput video traffic offload and localization. + +![Figure 9-1: Access-CDN architecture diagram showing User terminal, CDN-CP (control part), ICP, and a dashed box containing CDN-DP (Delivery part) and A-CDN underlay network.](af7916c89a458fdab6c3f443217388ae_img.jpg) + +The diagram illustrates the Access-CDN architecture. At the top, three boxes are arranged horizontally: 'User terminal' on the left, 'CDN-CP (control part)' in the center, and 'ICP' on the right. The 'CDN-CP' box contains the text: '(customization service by **CDN SP, Telecom self-constructed CDN, ICP self-constructed CDN**)'. Below the 'CDN-CP' box is a large dashed rectangle containing two stacked boxes. The top box is 'CDN-DP (Delivery part)' with the text: '(General content delivery function, open network capability including real time multi-cast)'. The bottom box is 'A-CDN underlay network' with the text: '(enhancement underlay network capability for CDN service)'. Below these two boxes, inside the dashed rectangle, is the text: '**Multi-Telecom A-CDN as a general open platform for distributed CDN**'. The identifier 'F.743.6(18)\_F9-1' is located at the bottom right of the dashed box. + +Figure 9-1: Access-CDN architecture diagram showing User terminal, CDN-CP (control part), ICP, and a dashed box containing CDN-DP (Delivery part) and A-CDN underlay network. + +**Figure 9-1 – Access-CDN - a recommended architecture for highly distributed CDN** + +The values generated through coordination between an operator A-CDN platform and the underlying network are illustrated in Figure 9-2. Some of these values include: + +- overlay and underlay coordination for more appropriate content placement (to ensure the hit rate and reuse rate); +- routing, distribution and scheduling considering the underlay network topology (local requests and local scheduling based on network status); +- within the same administration domain inside one telecom provider, the operator can deliver all types of services by opening the underlay capability, such as multicast services, QoS guaranteed services, security guaranteed services or other personalized services. + +![Figure 9-2: Coordination between access-CDN platform and the underlay network diagram. It shows the same components as Figure 9-1 but adds a cloud representing the 'Underlay network' with nodes. Arrows point from the underlay network to various service labels: 'QoS guaranteed', 'Lossless network', 'Personalized service', 'And more...', and 'Traffic localization'. The text 'Within the same administration domain inside one telecom provider' is at the bottom.](cab0834804fb031b43865554cc8d06ab_img.jpg) + +This diagram builds on Figure 9-1. It shows the same 'User terminal', 'CDN-CP', and 'ICP' components. The 'CDN-CP' box now contains: '(customization service by **CDN SP, Telecom self-constructed CDN, ICP self-constructed CDN**)'. Below it, the 'CDN-DP' and 'A-CDN underlay network' boxes are present. Below the 'A-CDN underlay network' box is a cloud labeled 'Underlay network' containing five nodes connected by dashed lines. Two large curved arrows originate from the 'Underlay network' cloud. The left arrow points to a label 'Traffic localization'. The right arrow branches out to three labels: 'Lossless network', 'Personalized service', and 'And more...'. On the left side, there is a label 'QoS guaranteed'. At the bottom of the dashed box, the text '**Within the same administration domain inside one telecom provider**' is present. The identifier 'F.743.6(18)\_F9-2' is at the bottom right. + +Figure 9-2: Coordination between access-CDN platform and the underlay network diagram. It shows the same components as Figure 9-1 but adds a cloud representing the 'Underlay network' with nodes. Arrows point from the underlay network to various service labels: 'QoS guaranteed', 'Lossless network', 'Personalized service', 'And more...', and 'Traffic localization'. The text 'Within the same administration domain inside one telecom provider' is at the bottom. + +**Figure 9-2 – Coordination between access-CDN platform and the underlay network** + +## Appendix I + +## Existing modes for bringing CDN to the edge + +(This appendix does not form an integral part of this Recommendation.) + +CDN nodes are shifting from the backbone network (the traditional CDN deployment position) to the edge, which implies a metro network architecture as illustrated in Figure I.1, with lower convergence ratio; this has become common among CDN SPs and operators. + +![Diagram illustrating the shift of CDN nodes from the backbone network to the metro network, closer to the border network gateway (BNG).](bd671b21db63e6fdb2196e9b18502aac_img.jpg) + +The diagram illustrates a network architecture with three layers: Backbone, Metro, and Access. The Backbone layer (top) is labeled 'Backbone higher convergence ratio' and contains a cloud labeled 'Backbone' with two routers. A callout box points to this layer with the text 'Current CDN deployment'. The Metro layer (middle) is labeled 'Metro high convergence ratio' and contains two Core Routers (CR), a Border Router (BR), and two Routers (R). A callout box points to this layer with the text 'CDN shifts to low convergence metro, close to BNG (rich metro CO resource for operators)'. The Access layer (bottom) is labeled 'Access low convergence ratio' and contains two Border Network Gateways (BNG), a Packet Gateway (PGW), and a Base station. Below the BNGs are Access LSWs, OLTs, ONTs, MDUs, and various user devices including Phones, PCs, and STBs. A mobile phone is also shown connected to the Base station. The diagram shows the shift of CDN nodes from the backbone to the metro network, closer to the BNG. A legend at the bottom left defines CR as Core router and BR as Border router. The diagram is labeled F.743.6(18)\_FI.1. + +Diagram illustrating the shift of CDN nodes from the backbone network to the metro network, closer to the border network gateway (BNG). + +**Figure I.1 – CDN nodes shift from backbone network to metro network, which is close to the border network gateway (BNG)** + +There are several benefits. First, significant saving on backbone bandwidth for the Internet service provider (ISP). Second, quality of experience (QoE) of ISP's customers is improved with a distributed architecture closer to the user. + +An analysis of current mainstream modes to bring CDN to the edge are described below. + +Mode 1 – peering (some CDN service providers and also operators who build global CDN): + +- for footprint extension, some CDN SPs are peering with thousands of Internet exchange points (IXP) points of presence to interconnect with ISPs that provide direct paths to access networks; +- for better user experience, some CDN SPs also deploy servers in ISP network utilizing operators' site and bandwidth for lower latency and packet loss rate; +- they have deployed servers in networks across countries, able to manage routing and delivery quality to geographically diverse users. Servers are typically placed in tier 2, and tier 1 networks; +- some operators' global CDN construction also achieve broad coverage by peering mode, by interconnecting with thousands of global ISPs via IXP mode; +- pros: Low cost and easy deploying for inter-region, inter-network coverage; + +- cons: Requires heavy investment on backbone for dozens of times bandwidth growth. But operators are unwilling since there is no profit model. Thus, peering mode will cause packet loss due to congestion and latency due to long distances from users. Thus, the future requirements cannot be met. 4K/8K/VR video requires high throughput as bandwidths move from Mbit/s to Gbit/s for a single user. + +Mode 2 – self-build CDN + ISP cooperation mode (global OTT service providers): + +- some large OTT service providers are running their own CDN nodes for the following reasons: + - grow faster: traditional CDN SP cannot meet the requirement of fast growing CDN bandwidth; + - reduced cost: especially when it comes to a certain scale; + - enhanced video QoE, including customized services; +- deployed CDN nodes down and enhance service quality by cooperation with ISPs for better user experience: + - OTT service providers provide CDN server and software, ISP provide site and bandwidth. No need to go to the traffic bottlenecks such as backbone network and IXP nodes when ISP user is accessing their service; + - although OTT service providers spend almost no money on bandwidth, CDN server costs are very high in order to offload up to 80% of traffic according to ISP's requirement; +- pros: Solving the problem of ISP backbone bandwidth expansion; +- cons: Since there are numerous ISPs worldwide, CDN node number will be huge to deploy down to the edge. Furthermore, massive storage should be deployed to offload 80% traffic to meet the requirement of ISP, which causes the storage cost to be very high. + +Mode 3 – self-build CDN (some CDN SPs and OTT service providers in China): + +- China CDN SPs and OTT service providers have already brought CDN to the edge (e.g., city Internet data center (IDC)). Yet disordered traffic still exists since the focus is on usage rate of CDN servers to reduce storage cost rather than localized scheduling. Statistics by China Telecom in 2015 indicate that less than 20% of traffic flows are localized. +- cons: Disordered traffic causes high network transmission costs to the operator; also no experience guarantee is provided: + - insufficient number of nodes, limited coverage, with no reliable user experience; + - dispatching service off-site means inter-network traffic settlement cost is high; + - CDN service scheduling is not aware of the underlying network status. Traffic flow with no reasonable guidance and experience is not guaranteed. + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound-programme and television-transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.743.7-201905-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.743.7-201905-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9002ee68e2b9ebf63d27d39ba8c73bdafb0e6be6 --- /dev/null +++ b/marked/F/T-REC-F.743.7-201905-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8df701536c25ab8246d2868de3c20c8a4e25168bb52c5eb9078278ab1b5b4fd8 +size 6859 diff --git a/marked/F/T-REC-F.743.8-201905-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.743.8-201905-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..22dc7ff0fb30f976cc1e576c45d2b7e65fa5ddc7 --- /dev/null +++ b/marked/F/T-REC-F.743.8-201905-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aa14933e776f7d0be8b2d60b81cbb6c1ee49deb5080e0281da05296f48189055 +size 6859 diff --git a/marked/F/T-REC-F.743.9-201905-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.743.9-201905-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9002ee68e2b9ebf63d27d39ba8c73bdafb0e6be6 --- /dev/null +++ b/marked/F/T-REC-F.743.9-201905-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8df701536c25ab8246d2868de3c20c8a4e25168bb52c5eb9078278ab1b5b4fd8 +size 6859 diff --git a/marked/F/T-REC-F.743.9-201905-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.743.9-201905-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..07fb1bfdb119a6547f6da7c242bb8017a0b6d048 --- /dev/null +++ b/marked/F/T-REC-F.743.9-201905-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a5a97a9b0418ded146eb08a23910d1b67d174f0cd27b7e3e7e0fe996249d0db6 +size 46652 diff --git a/marked/F/T-REC-F.746-201206-I_PDF-E/08a978a124d3ed6cf1a3d0cfd89418d0_img.jpg b/marked/F/T-REC-F.746-201206-I_PDF-E/08a978a124d3ed6cf1a3d0cfd89418d0_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..44a98b5fc84c8a18d86d4b224f151d13f1c59553 --- /dev/null +++ b/marked/F/T-REC-F.746-201206-I_PDF-E/08a978a124d3ed6cf1a3d0cfd89418d0_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2b550ad78ea9bdd7b35b4a7e75f478ccbb26fe7cf894ad417226e4d66c344484 +size 77928 diff --git a/marked/F/T-REC-F.746-201206-I_PDF-E/365b54f616aff249b4e6c0edafdcb9b3_img.jpg b/marked/F/T-REC-F.746-201206-I_PDF-E/365b54f616aff249b4e6c0edafdcb9b3_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..b78ed900248484f4861d68d415a297aec380e384 --- /dev/null +++ b/marked/F/T-REC-F.746-201206-I_PDF-E/365b54f616aff249b4e6c0edafdcb9b3_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:97f5a4db56e25d85d04f4164a2177884e7fc907445334d755df37b94c473acfb +size 117327 diff --git a/marked/F/T-REC-F.746-201206-I_PDF-E/4e4be0bd8b235167902f2c03e41da651_img.jpg b/marked/F/T-REC-F.746-201206-I_PDF-E/4e4be0bd8b235167902f2c03e41da651_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c02cc3bfd9527af2fb9e321678c8c1138ae41b61 --- /dev/null +++ b/marked/F/T-REC-F.746-201206-I_PDF-E/4e4be0bd8b235167902f2c03e41da651_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:97437f317f39a882f2dc88569037a7254ceb1962532f629e50789d53d3a961e2 +size 97000 diff --git a/marked/F/T-REC-F.746-201206-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg b/marked/F/T-REC-F.746-201206-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..cd540561531aeca328d3417697c7dddfd557ef06 --- /dev/null +++ b/marked/F/T-REC-F.746-201206-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bc5647f4411405b69ad5a8beab1506e0a8ed7b1a9194db3f9c9bb2d79c3d0887 +size 56037 diff --git a/marked/F/T-REC-F.746-201206-I_PDF-E/7efae06af3af43ffe5d4b956a679cf54_img.jpg b/marked/F/T-REC-F.746-201206-I_PDF-E/7efae06af3af43ffe5d4b956a679cf54_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..8a7f246a427483d6c97023afd7ac31a3a6800a05 --- /dev/null +++ b/marked/F/T-REC-F.746-201206-I_PDF-E/7efae06af3af43ffe5d4b956a679cf54_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6539f8c8d51fcbea7cb01a7f8238f06275e4230c7575e6f71a2d8c5843f9eab7 +size 34087 diff --git a/marked/F/T-REC-F.746-201206-I_PDF-E/81a4cbf0b3c4cbc065efdf8f800dadde_img.jpg b/marked/F/T-REC-F.746-201206-I_PDF-E/81a4cbf0b3c4cbc065efdf8f800dadde_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..329f91a624301e6d460534028541fa4020dae709 --- /dev/null +++ b/marked/F/T-REC-F.746-201206-I_PDF-E/81a4cbf0b3c4cbc065efdf8f800dadde_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3e9d11e3171209776fa48be65bd745110a54e0149f960c24cd83b922d3d67523 +size 118440 diff --git a/marked/F/T-REC-F.746-201206-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.746-201206-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..62a2e98629da5a6328a0ac2bd6ddfe35692322ae --- /dev/null +++ b/marked/F/T-REC-F.746-201206-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:704894a92b416ba01d0084281c9f236cf8f10581b6beb515db7a33ac54f5b95d +size 6625 diff --git a/marked/F/T-REC-F.746-201206-I_PDF-E/8fa679f79a1bb1f527cba9f29e784e89_img.jpg b/marked/F/T-REC-F.746-201206-I_PDF-E/8fa679f79a1bb1f527cba9f29e784e89_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..fa17619aa1f78bd07e91c57f2a494002f5201a03 --- /dev/null +++ b/marked/F/T-REC-F.746-201206-I_PDF-E/8fa679f79a1bb1f527cba9f29e784e89_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5db4a18174fd264e63cc71e670b9e3e0098a44462fb3dd1802dc04cf0b7784cc +size 58629 diff --git a/marked/F/T-REC-F.746-201206-I_PDF-E/acfc53eca625d62b38aa2563efa95c3e_img.jpg b/marked/F/T-REC-F.746-201206-I_PDF-E/acfc53eca625d62b38aa2563efa95c3e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7c2d1f5a4889c857ee5a277779baf577bb918b5d --- /dev/null +++ b/marked/F/T-REC-F.746-201206-I_PDF-E/acfc53eca625d62b38aa2563efa95c3e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:182034160df615ec10a942a06691d9969f5e0b960fc80d0ffcdd53e3dbef5daa +size 55773 diff --git a/marked/F/T-REC-F.746-201206-I_PDF-E/bd671b21db63e6fdb2196e9b18502aac_img.jpg b/marked/F/T-REC-F.746-201206-I_PDF-E/bd671b21db63e6fdb2196e9b18502aac_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..93c43bb9802097da28bf73d68807d89444fca6e8 --- /dev/null +++ b/marked/F/T-REC-F.746-201206-I_PDF-E/bd671b21db63e6fdb2196e9b18502aac_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d5ede96ccc0465cca1aae30a6dd8614371c43f0cb9900eb3a3b681ac70a38239 +size 38959 diff --git a/marked/F/T-REC-F.746.1-201410-I_PDF-E/raw.md b/marked/F/T-REC-F.746.1-201410-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..0da9133288d367f6485c1c85f73e0535a0540d4d --- /dev/null +++ b/marked/F/T-REC-F.746.1-201410-I_PDF-E/raw.md @@ -0,0 +1,716 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.746.1** + +(10/2014) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Audiovisual services + +--- + +**Requirements for low-latency interactive +multimedia streaming** + +Recommendation ITU-T F.746.1 + +ITU-T + +![ITU logo: A globe with a red lightning bolt and the text 'ITU International Telecommunication Union'.](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a stylized globe with a red lightning bolt striking across it. To the right of the globe, the text "ITU" is written in a bold, blue font, and below it, "International Telecommunication Union" is written in a smaller, blue font. + +ITU logo: A globe with a red lightning bolt and the text 'ITU International Telecommunication Union'. + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| AUDIOVISUAL SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +## Recommendation ITU-T F.746.1 + +## Requirements for low-latency interactive multimedia streaming + +## Summary + +Recommendation ITU-T F.746.1 describes the major functions and their interactions for low-latency interactive multimedia streaming (LIMS). The scenarios and requirements are also identified in this Recommendation. + +Desktop and client-server applications can be provided to users through LIMS. The desktop application software and client software of client-server applications are deployed in the LIMS platform (server) and the outputs of the applications are delivered back to users as media streams. + +In such a way, the traditional data storing, computing and graphic processing tasks are offloaded from user devices to LIMS servers. Provided the interaction delay is kept low, LIMS can provide the same experience to end users with less powerful devices. + +The output of one instance of the application can be encoded into different audio and video formats, which can be sent to one or more user devices simultaneously. That means the application can be used continuously when the user switches from a powerful device such as a high-end computer to a less powerful device such as a mobile phone or vice versa. + +Since the software is no longer installed in the user device, easier deployment, maintenance and upgrade of the software is possible in the centrally-managed LIMS servers are much easier than before. If needed, it is easy to provide centralized management capability to fulfil the requirements of enterprises, such as single sign-on and access control. + +It is very difficult to hack an application provided through LIMS. Furthermore, software reverse engineering is no longer possible, which is useful for the protection of intellectual property rights. + +Since the delay of interaction with the application has a major impact on the user experience, it is critical for LIMS to keep the interaction delay in a range that the end user is willing to tolerate. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.746.1 | 2014-10-14 | 16 | 11.1002/1000/12225 | + +## Keywords + +Interactive multimedia streaming, low-latency, service provider. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2015 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|--------------------------------------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Introduction..... | 3 | +| 7 Scenarios for LIMS..... | 5 | +| 7.1 LIMS for enterprise application delivery ..... | 5 | +| 7.2 LIMS for application delivery service..... | 6 | +| 7.3 LIMS for family application delivery..... | 6 | +| 8 Requirements for LIMS ..... | 7 | +| 8.1 General requirements..... | 7 | +| 8.2 Client-side requirements..... | 8 | +| 8.3 Platform-side requirements ..... | 8 | +| 8.4 Requirements for the service provider side ..... | 9 | +| 8.5 Codec requirements ..... | 9 | +| 8.6 Security requirements ..... | 9 | +| Appendix I – An experiment for measuring the delay tolerance level of various applications ..... | 10 | +| I.1 Testing environment..... | 10 | +| I.2 Extra delay imposed and applications tested..... | 10 | +| I.3 Test solutions..... | 11 | +| I.4 Test results..... | 12 | +| Bibliography..... | 14 | + + + +# Recommendation ITU-T F.746.1 + +## Requirements for low-latency interactive multimedia streaming + +# 1 Scope + +This Recommendation identifies the functions, scenarios and requirements for low-latency interactive multimedia streaming (LIMS). + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +- [ITU-T H.263] Recommendation ITU-T H.263 (2005), *Video coding for low bit rate communication*. +- [ITU-T H.264] Recommendation ITU-T H.264 (2013), *Advanced video coding for generic audiovisual services*. +- [ITU-T H.265] Recommendation ITU-T H.265 (2013), *High efficiency video coding*. +- [ITU-T T.81] Recommendation ITU-T T.81 (1992), *Information technology – Digital compression and coding of continuous-tone still images – Requirements and guidelines*. +- [ITU-T T.87] Recommendation ITU-T T.87 (1998), *Information technology – Lossless and near-lossless compression of continuous-tone still images – Baseline*. +- [ITU-T T.800] Recommendation ITU-T T.800 (2002) | ISO/IEC 15444-1:2004, *Information technology – JPEG 2000 image coding system: core coding system*. +- [ISO/IEC 14496-3] ISO/IEC 14496-3 (2009), *Information technology – Coding of audio-visual objects – Part 3: Audio*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +None. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 application server:** A device composed of the server software of a client-server application as well as the hardware and software environment to support it. + +**3.2.2 low-latency interactive multimedia streaming (LIMS):** A service used to provide the applications to users as interactive media streams. + +**3.2.3 low-latency interactive multimedia streaming (LIMS) client:** A device used to collect the user commands and send them to the LIMS server as well as display the media stream from the LIMS server. + +**3.2.4 low-latency interactive multimedia streaming (LIMS) platform communication module:** A module used to receive user commands from the LIMS client and send them to the LIMS proxy service module as the input of application logic. + +**3.2.5 low-latency interactive multimedia streaming (LIMS) proxy service module:** A module used to provide the software environment for desktop application software or the client software of client-server application. It is also responsible for forwarding the user commands from the LIMS platform communication module to the application, and sending the output of the application to the LIMS visualization and compression module. + +**3.2.6 low-latency interactive multimedia streaming (LIMS) server:** A device composed of the LIMS platform communication module (LPC), LIMS proxy service module (LPS), LIMS visualization and compression module (LVC) and LIMS stream service module (LSS). + +**3.2.7 low-latency interactive multimedia streaming (LIMS) stream service module:** A module used to receive audio and video streams from the LIMS visualization and compression module and stream the mixed media stream back to the LIMS client. + +**3.2.8 low-latency interactive multimedia streaming (LIMS) visualization and compression module:** A module used to convert the output of application into images and audio data and encode them into video streams and audio streams. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|----------------------------------------------| +| AAC | Advanced Audio Coding | +| ALS | Audio Lossless Coding | +| CPU | Central Processing Unit | +| DDoS | Distributed Denial of Service | +| GPU | Graphical Processing Unit | +| HID | Human Interface Device | +| LIMS | Low-latency Interactive Multimedia Streaming | +| LPC | LIMS Platform Communication | +| LPS | LIMS Proxy Service | +| LSS | LIMS Streaming Service | +| LVC | LIMS Visualization and Compression | +| MPEG | Motion Picture Experts Group | +| PC | Personal Computer | +| TV | Television | +| USB | Universal Serial Bus | +| VoD | Video on Demand | + +# 5 Conventions + +In this Recommendation the following conventions are used: + +- "Shall" indicates a mandatory requirement. +- "Should" indicates a suggested but optional course of action. + +- "May" indicates an optional course of action rather than a recommendation that something take place. + +# 6 Introduction + +In a low-latency interactive multimedia streaming (LIMS) system, a user's experience of one application is the display of an interactive media stream customized for him/her. + +The various functions and modules required by a LIMS system are shown in Figure 6-1. + +The LIMS client collects the user inputs from devices such as a keyboard, pointing device or any other input device and transfers them to the LIMS server as user commands. + +The LIMS platform communication (LPC) module of the LIMS server relays the user commands to LIMS proxy service (LPS), which are used as the input of the application software. Besides providing the input and output channels, the LPS is also responsible for providing all the necessary software environment and resources for the application software, which might be the software of desktop applications or the client software of client-server applications. + +![Figure 6-1 – Framework of a LIMS system. The diagram shows the interaction between a LIMS client, a LIMS platform/server, and a Service provider application server. The LIMS client contains 'User interaction' and 'Media decoder'. The LIMS platform/server contains 'LPC', 'LPS', 'LVC', and 'LSS'. The Service provider application server contains 'Server software of application'. Arrows show the flow of 'User commands', 'Application input', 'Request', 'Response', 'Application output', 'Media streams', 'Encoded video stream', and 'Encoded audio stream'.](dbe553cf16dd14073b89a8263a428664_img.jpg) + +``` + +graph LR + subgraph LIMS_client [LIMS client] + UI[User interaction] + MD[Media decoder] + end + subgraph LIMS_platform_server [LIMS platform/server] + subgraph LPC [LPC] + LCI[LIMS client interaction] + end + subgraph LPS [LPS] + SE[Software environment] + DAC[Desktop application/client software of application] + end + subgraph LVC [LVC] + DC[Display controller] + VE[Video encoder] + AE[Audio encoder] + end + subgraph LSS [LSS] + MS[Media streaming] + end + end + subgraph Service_provider_application_server [Service provider application server] + SSA[Server software of application] + end + + UI -- "User commands" --> LCI + LCI -- "Application input" --> DAC + DAC -- "Request" --> SSA + SSA -- "Response" --> DAC + DAC -- "Application output" --> DC + DC -- "Images" --> VE + DC -- "Audio data" --> AE + VE -- "Encoded video stream" --> MS + AE -- "Encoded audio stream" --> MS + MS -- "Media streams" --> MD + +``` + +Figure 6-1 – Framework of a LIMS system. The diagram shows the interaction between a LIMS client, a LIMS platform/server, and a Service provider application server. The LIMS client contains 'User interaction' and 'Media decoder'. The LIMS platform/server contains 'LPC', 'LPS', 'LVC', and 'LSS'. The Service provider application server contains 'Server software of application'. Arrows show the flow of 'User commands', 'Application input', 'Request', 'Response', 'Application output', 'Media streams', 'Encoded video stream', and 'Encoded audio stream'. + +F.746.1(14)\_F6-1 + +**Figure 6-1 – Framework of a LIMS system** + +The commands are then processed by the application logic, which is part of the standalone desktop program or the server software of a client-server application. + +The output is then processed by the display controller. Usually a graphical processing unit (GPU), of LIMS visualization and compression (LVC) module shall convert the visual information into images and an audio driver shall convert the audio information into audio data. The images and audio data shall be compressed by the video encoder and audio encoder respectively. The resulted audio stream and video stream shall be sent to LIMS streaming service (LSS), which delivers the mixed media stream back to the LIMS client. Finally, the LIMS client decodes the media stream and displays the audio and video contents to the user. + +The software including a standalone program of a desktop application and the client part of a client-server application is not installed in the user devices as before. That means a user device will not execute the application logic as well as most of the graphic processing tasks. The tasks left to the user device are collecting the user inputs/commands from the keyboard, mouse as well as other input + +devices and forwarding them to the LIMS server, and displaying the media streams which is the output of the application sent back from the LIMS server. + +The software is deployed in the LIMS servers and the output of application logic is sent to the display controller for further processing, in which case GPU rendering and audio processing might be involved. The output images from display controller might be encoded or compressed into different formats, such as lossy and lossless JPEG formats [ITU-T T.81], [ITU-T T.87] or JPEG 2000 format [ITU-T T.800]. The output images are further encoded into video sequences. Many video codec standards are available, such as [ITU-T H.263], [ITU-T H.264], [ITU-T H.265]. Both lossy digital audio compression and lossless audio compression codec standards are available, such as advanced audio coding (AAC) and MPEG-4 audio lossless coding (ALS) [ISO/IEC 14496-3]. + +There are several protocol stacks can be used to provide media content streaming service, such as the RTSP/RTP/RTCP [b-IETF RFC 2326], [b-IETF RFC 3550] stack. + +The audio and video codecs, as well as the streaming service, should be chosen according to the processing power of the LIMS server, the bandwidth of network as well as the processing resources of the LIMS client. + +Since computing and graphic processing tasks are offloaded from user devices to the LIMS server, the LIMS client could be much less powerful than an off-the-shelf PC with a multiple-core CPU, multiple-gigabytes memory and multiple-gigabytes storage. Even with a less powerful CPU and lower memory frequency, the LIMS client still could have the similar experience if the interaction delay is low or is tolerable for the user experience. + +Since all the graphic processing and video encoding tasks are fulfilled in the LIMS servers, the application output could be encoded into multiple audio and video streams with different formats and sent to different LIMS clients (user devices) simultaneously. It is easy to support multi-cast applications such as videoconferencing and multi-player games. + +Besides the benefits brought to end users, service providers could also benefit from LIMS. First of all, rather than being deployed on all kinds of user devices, software programs can now be deployed in the centrally managed LIMS servers. It is much easier to deploy new software or upgrade and maintain the installed software. + +In LIMS, all the applications are provided by the LIMS server. If needed, it is easy to provide a centralized management capability to fulfil the requirements of the enterprise, such as single sign-on and access control. A user is allowed to log in to the LIMS platform once and be authenticated to access to all the applications and resources without being prompted to log in again at each of them, i.e., the using of applications and resources does not require further authentication processes. Furthermore, all the data and transaction logs are commonly saved in the LIMS platform. This builds a centralized data backup, which can be shared among the enterprise under a particular security policy. + +The LIMS platform provides enhanced security protection for the applications. When trying to hack an application delivered by LIMS, a hacker has to analyse the media stream. Most information about the application is unrecoverable based on the LIMS media stream. Hacking of the application is extremely difficult, if not impossible. Even if some sensitive information, such as usernames, might be recoverable, all the information sent from the LIMS platform is encoded by the video and audio codec. Since it is very hard to analyse such media streams, it will be extremely difficult to hack a password by a brute force attack. Compared with protecting many personal computers against distributed denial of service (DDoS) attacks, the deployment of DDoS protection on the LIMS platform is much easier and more effective. As the user receives only a media stream, it is relatively easy to keep sensitive information readable and keep it from being copied. Furthermore, software reverse engineering is no longer possible which is useful for the protection of intellectual property rights. + +A LIMS system shall collect the user inputs, forward them to the LIMS server, execute the application logic, accomplish audio and graphic processing, encode the audio and video streams and stream the mixed audio/video media stream back to the user. To ensure a good user experience, the sequential execution of these operations, one after another, should be completed very quickly. The interaction delay shall be low enough so that it is tolerable. Different applications have different thresholds for maximum tolerable delay. Appendix I shows experimental results of the tolerable delay for average users of several applications. Further experimental results may be available in the future. + +# 7 Scenarios for LIMS + +## 7.1 LIMS for enterprise application delivery + +An enterprise can provide all of its business applications to its employees or customers via LIMS, no matter what kind of application they are. + +By doing so, the following LIMS characterized use cases can be provided. + +![Diagram illustrating the scenario of LIMS for enterprise application delivery. On the left, three user terminals (PC, Phone, TV) are connected to a central cloud labeled 'INTERNET'. The 'INTERNET' cloud is connected to a 'Gateway' server. The 'Gateway' server is connected to a large cloud labeled 'Enterprise inner network'. Inside this network, there is a cloud labeled 'LIMS platform' which is connected to three 'Enterprise application servers' (represented by server icons). Below the 'LIMS platform' cloud, three user terminals (PC, Phone, TV) are shown, indicating they are connected to the LIMS platform within the enterprise network. A small label 'F.746.1(14)_F7-1' is visible in the bottom right corner of the diagram area.](7e670a2b556b53ea9002dfff3a420e08_img.jpg) + +Diagram illustrating the scenario of LIMS for enterprise application delivery. On the left, three user terminals (PC, Phone, TV) are connected to a central cloud labeled 'INTERNET'. The 'INTERNET' cloud is connected to a 'Gateway' server. The 'Gateway' server is connected to a large cloud labeled 'Enterprise inner network'. Inside this network, there is a cloud labeled 'LIMS platform' which is connected to three 'Enterprise application servers' (represented by server icons). Below the 'LIMS platform' cloud, three user terminals (PC, Phone, TV) are shown, indicating they are connected to the LIMS platform within the enterprise network. A small label 'F.746.1(14)\_F7-1' is visible in the bottom right corner of the diagram area. + +Figure 7-1 – Scenario of LIMS for enterprise application delivery + +### 7.1.1 One platform to deliver all applications + +The LIMS platform can support the delivery of any application. + +When the enterprise deploys a new application, it just needs to deploy it to the application server(s), connect the new servers to the LIMS platform, and create the appropriate configuration and test the application on the LIMS platform before release. + +### 7.1.2 Multiple terminal supporting for an application + +For applications allowing simultaneous access to multiple users, each user accessing the service may use a different kind of terminal. + +When developing an application, the developer does not need to consider the adaptation of different terminals. The terminal adaptation problem can be solved by the LIMS platform. Once the application is deployed on the platform, it can be used seamlessly across multiple devices, such as personal + +computers (PCs), phones and television (TV) sets. The usage of any application is not limited by the computing or storage resource on the terminal, except that all LIMS-enabled terminals shall support media stream playback. + +If a user suspends an application on a terminal, e.g., a notebook, the application can be resumed from any terminal, starting right from the stop point. + +### 7.1.3 Customizable unified client + +A unified client is provided to the users, either employees or customers. With this client, a personal portal that can be customized by the user is provided. This portal provides users with consistent service experience. In other words, no matter what kind of terminal is used, the service remains unchanged from the view point of the users. + +Through this portal, the user can access all the applications and resources provided by the enterprise. The addition of new applications, the removal of old applications or the software upgrade of existing applications requires no modification to the unified client. + +## 7.2 LIMS for application delivery service + +By deploying the LIMS platform in the Internet, commercial LIMS application delivery service can be provided. Besides the use cases listed in clause 7.1, in this scenario, the use cases listed in the following subclause can be supported. + +![Diagram illustrating the scenario of LIMS for application delivery service. A central cloud labeled 'INTERNET' is connected to a 'LIMS platform' (represented by three server icons) on the left, and two service groups, 'Service A' and 'Service B' (each represented by three server icons), on the right. Below the 'INTERNET' cloud, three terminals are connected: 'PC' (laptop icon), 'Phone' (mobile phone icon), and 'TV' (television icon). The diagram is labeled F.746.1(14)_F7-2.](853f59c89931a666c07903b31d098277_img.jpg) + +The diagram shows a central cloud labeled 'INTERNET'. To its left, a 'LIMS platform' consisting of three server icons is connected. To its right, two service groups, 'Service A' and 'Service B', are connected. Each service group is enclosed in an oval and contains three server icons. Below the 'INTERNET' cloud, three terminals are connected: a 'PC' (laptop icon), a 'Phone' (mobile phone icon), and a 'TV' (television icon). The diagram is labeled F.746.1(14)\_F7-2. + +Diagram illustrating the scenario of LIMS for application delivery service. A central cloud labeled 'INTERNET' is connected to a 'LIMS platform' (represented by three server icons) on the left, and two service groups, 'Service A' and 'Service B' (each represented by three server icons), on the right. Below the 'INTERNET' cloud, three terminals are connected: 'PC' (laptop icon), 'Phone' (mobile phone icon), and 'TV' (television icon). The diagram is labeled F.746.1(14)\_F7-2. + +**Figure 7-2 – Scenario of LIMS for application delivery service** + +### 7.2.1 Trial service + +As the application delivered by the LIMS platform can be used without download and installation process, it is convenient for an application to be released for trial. At the beginning, a user can, for example, use a free 30-day trial of this application. At the conclusion of the trial period, the user is then required to pay for further usage. As there is no need to download and install software, such trial service is more convenient for end users. + +## 7.3 LIMS for family application delivery + +A service provider for family entertainments can provide video interaction applications to its clients based on LIMS, as shown in Figure 7-3. + +By doing so, besides the use cases listed in the previous clauses, the following LIMS characterized use cases can be provided. + +![Diagram illustrating the scenario of LIMS for family application delivery. At the top, an oval labeled 'Application servers' contains icons for 'Video chat server', 'Game server', 'File server', and ellipsis dots. Below this, a 'LIMS platform' icon (three server racks) is connected to a central cloud labeled 'INTERNET'. The 'INTERNET' cloud is connected to three user scenarios: 1) A 'PC' icon with a person, 2) A 'TV' icon with two people, and 3) A 'Phone' icon with a person. A small label 'F.746.1(14)_F7-3' is in the bottom right corner of the diagram area.](af7916c89a458fdab6c3f443217388ae_img.jpg) + +Diagram illustrating the scenario of LIMS for family application delivery. At the top, an oval labeled 'Application servers' contains icons for 'Video chat server', 'Game server', 'File server', and ellipsis dots. Below this, a 'LIMS platform' icon (three server racks) is connected to a central cloud labeled 'INTERNET'. The 'INTERNET' cloud is connected to three user scenarios: 1) A 'PC' icon with a person, 2) A 'TV' icon with two people, and 3) A 'Phone' icon with a person. A small label 'F.746.1(14)\_F7-3' is in the bottom right corner of the diagram area. + +**Figure 7-3 – Scenario of LIMS for family application delivery** + +### 7.3.1 Family entertainment service + +The LIMS platform provides entertainment applications for family users, for example the online game service. Individual users in a family can play any game application deployed in the LIMS platform. Multiple users either within the same house or staying in a different location can play the same game with different terminals simultaneously. When a user plays the game, the game audio and video can be recorded and transformed to media streaming for video on demand (VoD), to be watched by other people. + +### 7.3.2 Family multimedia communication service + +LIMS enables novel multimedia communication for its users: + +- On the one hand, users can transmit their portraits, along with their voices and text messages, to a chat application server deployed in the LIMS platform. All audio, video and text are combined to create media streams that are then distributed to all users; +- On the other hand, users can simultaneously run more than one application deployed in LIMS, i.e., game and drawing applications, where the views from the applications can also be mixed with the user videos and text messages. + +# 8 Requirements for LIMS + +This clause addresses the requirements for LIMS services. The description of the following requirements takes the framework shown in Figure 1 as a reference. + +## 8.1 General requirements + +**GEN-01:** A LIMS system shall have one or multiple LIMS clients and one LIMS platform. It may have a set of interfaces to the service provider/application server. + +**GEN-02:** A LIMS system should provide applications to its users transparently. That is, the adoption of the LIMS technique should introduce no change to the user experience of an application, such as its user interface and process. + +**GEN-03:** Messages for management shall be exchanged between the LIMS client and the LIMS platform for supporting LIMS client registration, end user registration, application discovery, end user authorization as well as codec selection. + +**GEN-04:** There shall be enough bandwidth between one LIMS client and the LIMS platform to satisfy the requirements of the media stream being sent from LIMS platform to the LIMS client. The interaction delay shall be sufficiently low to guarantee a similar user experience no matter what kind of terminal or network is used. + +**GEN-05:** A LIMS system may adopt either a centralized or a distributed architecture. For example, a local company may deploy LIMS centralized, and a multinational company may need distributed LIMS deployment. + +## **8.2 Client-side requirements** + +**CLI-01:** A LIMS client shall decode the mixed audio/video streams sent from the LIMS platform. + +**CLI-02:** A LIMS client shall collect the user inputs and transfer them to the LIMS platform. + +**CLI-03:** A LIMS client shall register itself to the LIMS platform. + +**CLI-04:** A LIMS client shall register the end user which is the owner of the client to the LIMS platform. + +**CLI-05:** A LIMS client shall have the capability of discovering the applications installed in the LIMS platform. + +**CLI-06:** A LIMS client shall have the capability of choosing the audio and video codecs. + +## **8.3 Platform-side requirements** + +**PLT-01:** A LIMS platform shall provide the software environment required by application software. The software environment for a particular application may include the operating system, the database system, specific development tools or compiler. The installation, uninstallation and upgrade of application software shall be supported. + +**PLT-02:** A LIMS platform shall relay the user commands from the LIMS client to the application logic. + +**PLT-03:** A LIMS platform shall process the output of the application logic and encode the resulting images and audio data into video and audio streams according to the codec requirements of the LIMS client and the available bandwidth between them. + +**PLT-04:** A LIMS platform shall deliver a mixed media stream to the LIMS client. + +**PLT-05:** A LIMS platform should support the registration of the LIMS client. + +**PLT-06:** A LIMS platform should support the registration of the end user. + +**PLT-07:** A LIMS platform should provide the information of the applications to the LIMS clients. + +**PLT-08:** A LIMS platform should authorize or reject the requests for using applications from LIMS clients. + +**PLT-09:** A LIMS platform should provide the capability of logging all the instructions issued by a user as well as the output of the application logic. + +**PLT-10:** A LIMS platform may support single-sign-on. After login to the LIMS platform, the access to every application should not require further login. + +**PLT-11:** A LIMS platform should support application management. It should allow the loading, updating, suspending and removing of an application. It should support collecting the usage statistics for an application. + +**PLT-12:** A LIMS platform should support multiple service providers concurrently. It shall support the communication between service providers and the client parts of client-server applications installed in the platform. + +## **8.4 Requirements for the service provider side** + +**SP-01:** LIMS should support applications from different service providers (application server). No extra development of the server software of client-server applications should be required. + +**SP-02:** Service providers may require the LIMS system to provide usage information about their applications, such as the number of online users and their usage, etc. + +## **8.5 Codec requirements** + +**COD-01:** The LIMS implementation shall encode both audio and video streams in a very short time. Several issues affect the encoding delay limitation, such as the applications supported, the service coverage of the system and the underlying bandwidth conditions, etc. The time limitation varies among different use cases. The experimental results of the tolerable delay for average users of several kinds of applications are shown in Appendix I, which may be used as a reference to set the time limitation of the audio and video codecs used. + +**COD-02:** To avoid unpredictable congestion caused by instantaneous high traffic load, the LIMS implementation should choose the codecs which generate stable bitrate traffic load. + +**COD-03:** To adapt to the changing network conditions, the codec system shall generate streams with different rates. Each quality class of the mixed audio/video stream is optimized for a specified range of available network bandwidth. The LIMS implementation changes the audio/video rate (quality) according to the available network bandwidth and the requirement of the LIMS client. + +**COD-04:** To guarantee low latency required by interactive applications, the codec system may adjust the number of data buffered. Generally, the more data are buffered, the more the delay introduced by a codec is. + +**COD-05:** The LIMS platform shall support multiple users simultaneously. Interference between different streams is strictly prohibited. + +## **8.6 Security requirements** + +**SEC-01:** A LIMS system shall protect its users' privacy. As all information of a user is being stored on the platform, appropriate methods shall be deployed on the platform to protect users' privacy. At the same time, some authentication and confidentiality methods should be deployed between a LIMS client and the platform. + +**SEC-02:** Appropriate authorization methods should be deployed to prevent unauthorized access to user data and multimedia streams. + +**SEC-03:** Appropriate data recovery methods should be deployed to prevent a LIMS system from data loss caused by unpredictable environmental disaster. For the multimedia streams generated by the LIMS platform, an appropriate backup and recovery scheme should also be adopted. + +**SEC-04:** Appropriate methods should be deployed to prevent attacks targeted at the LIMS platform. + +**SEC-05:** Appropriate high availability architecture may be adopted to provide reliable service to the users. + +# Appendix I + +## An experiment for measuring the delay tolerance level of various applications + +(This appendix does not form an integral part of this Recommendation.) + +Subjective evaluations were carried out to evaluate how users felt about multiple applications under different delay conditions. A standardized opinion questionnaire was used throughout the evaluation. + +### I.1 Testing environment + +**Server (APP-HOST):** One PC with Windows XP was used as server, with universal serial bus (USB) driver and human interface device (HID) server. Multiple applications were running on the server. One delay toolset was installed to introduce extra delay to the running applications. + +**Client:** One PC, with USB client delay toolset, mouse and keyboard installed, was used for the Client. + +**Connectivity:** Two computers were connected using a crossover cable. + +**Interactions:** The motion of the mouse of the client was sent to the server and translated into the motion of a pointer on the display of the server, and other operations of the mouse, such as clicking or hovering were then used as the input of the running application. The input of the keyboard of the client was also sent to the server and used as the input to the running application. + +**Aim:** Before being sent to the application on the server, the input message of the mouse and keyboard from the client was delayed according to pre-set parameters. Users evaluated the experience of multiple applications under different added delays. + +![Diagram of the test environment showing a Client PC connected to a Server PC via a crossover cable. A user is shown interacting with the Client PC, and a yellow lightning bolt labeled 'Watching the monitor' points from the user to the Server PC's monitor. The diagram is labeled F.746.1(14)_FI.1.](f57a881fcbcaf494a7dedc4f61224991_img.jpg) + +The diagram illustrates the experimental setup. On the left, a 'Client' computer (monitor, keyboard, and mouse) is shown with a user icon next to it. On the right, a 'Server' computer (tower and monitor) is shown. A double-headed arrow labeled 'Crossover cable' connects the Client and Server. A yellow lightning bolt labeled 'Watching the monitor' points from the user to the Server's monitor. The text 'F.746.1(14)\_FI.1' is in the bottom right corner. + +Diagram of the test environment showing a Client PC connected to a Server PC via a crossover cable. A user is shown interacting with the Client PC, and a yellow lightning bolt labeled 'Watching the monitor' points from the user to the Server PC's monitor. The diagram is labeled F.746.1(14)\_FI.1. + +Figure I.1 – Test environment + +### I.2 Extra delay imposed and applications tested + +#### I.2.1 Extra delay imposed + +The extra delay added in test was in the range of 0 ms to 320 ms. + +The maximum step size of the added delay was 40 ms. + +#### I.2.2 Applications tested + +To be impartial to all applications, the actual names of the tested applications were removed: + +- 1) Casual games: CG1, CG2; +- 2) Word processing: WP1, WP2; +- 3) Video-on-demand: VoD1, VoD2, VoD3; +- 4) Web browsing: WB1, WB2, WB3; +- 5) Real-time strategy games: RTS1, RTS2, RTS3; +- 6) Role-Playing game: RPG1, RPG2; +- 7) Racing game: RG1, RG2; + +- 8) Fighting game: FG1. + +### I.3 Test solutions + +Fifty users who attended the test were given standardized questionnaires to fill their subjective evaluation for each application under different added delay conditions. There were three options in the questionnaire: significant delay (unacceptable), delayed but acceptable, timely (satisfactory). The operational procedures for different applications are given as follows. + +1. Casual games: CG1, CG2 + - Log in CG1 or CG2 with your own account and password. + - Play your favourite game. +2. Word processing: WP1, WP2 + - Create a new file. + - Input a paragraph of 100 or more words. + - Edit the paragraph. +3. Video-on-demand: VoD1, VoD2, VoD3 + - Click the shortcut to the program. + - Search for your favourite movies. + - Watch one movie for at least 5 minutes. +4. Web browsing: WB1, WB2, WB3 + - Click the shortcut to the web browser. + - Input a URL in the address bar. + - Browse the opened website for at least 5 minutes. +5. Real-time strategy games: RTS1, RTS2, RTS3 + - Play the game for at least 10 minutes. +6. Role-playing game: RPG1, RPG2 + - Play the game for at least 10 minutes. +7. Racing game: RG1, RG2 + - Play the game for at least 10 minutes. +8. Fighting game: FG1 + - Play the game for at least 10 minutes. + +**Table I.1 – Extra delays imposed in the group of tests** + +| # of Test | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | +|------------------------|---|----|----|-----|-----|-----|-----|-----|-----| +| Extra delay added (ms) | 0 | 40 | 80 | 120 | 160 | 200 | 240 | 280 | 320 | + +NOTE – When the extra delay is set to zero, the original delay of the input message, i.e., from the time of clicking of the mouse to the time of clicking message arriving at the application, was about 30 to 40 ms. + +### I.4 Test results + +In the following tables, for a given extra delay, the number in the second row represents the number of users who chose *significant delay (unacceptable)* in the questionnaire of the corresponding test, the number in the third row represents the number of users who chose *delayed but acceptable*, and the number in the fourth row represents the number of users who chose *timely (satisfactory)*. + +**Table I.2 – Test results for casual games (CG1 and CG2)** + +| | | | | | | | | | | +|----------------------------------|----|----|----|-----|-----|-----|-----|-----|-----| +| Extra delay added (ms) | 0 | 40 | 80 | 120 | 160 | 200 | 240 | 280 | 320 | +| Significant delay (unacceptable) | 0 | 0 | 0 | 0 | 9 | 35 | 50 | 50 | 50 | +| Delayed but acceptable | 0 | 17 | 38 | 48 | 41 | 15 | 0 | 0 | 0 | +| Timely (satisfactory) | 50 | 33 | 12 | 2 | 0 | 0 | 0 | 0 | 0 | + +**Table I.3 – Test results for word processing (WP1 and WP2)** + +| | | | | | | | | | | +|----------------------------------|----|----|----|-----|-----|-----|-----|-----|-----| +| Extra delay added (ms) | 0 | 40 | 80 | 120 | 160 | 200 | 240 | 280 | 320 | +| Significant delay (unacceptable) | 0 | 0 | 0 | 0 | 3 | 32 | 39 | 42 | 49 | +| Delayed but acceptable | 0 | 0 | 19 | 37 | 46 | 18 | 11 | 8 | 1 | +| Timely (satisfactory) | 50 | 50 | 31 | 13 | 1 | 0 | 0 | 0 | 0 | + +**Table I.4 – Test results for video on demand (VoD1, VoD2 and VoD3)** + +| | | | | | | | | | | +|----------------------------------|----|----|----|-----|-----|-----|-----|-----|-----| +| Extra delay added (ms) | 0 | 40 | 80 | 120 | 160 | 200 | 240 | 280 | 320 | +| Significant delay (unacceptable) | 0 | 0 | 0 | 0 | 7 | 32 | 39 | 47 | 50 | +| Delayed but acceptable | 0 | 2 | 7 | 37 | 39 | 18 | 11 | 3 | 0 | +| Timely (satisfactory) | 50 | 48 | 43 | 13 | 4 | 0 | 0 | 0 | 0 | + +**Table I.5 – Test results for web browsing (WB1, WB2 and WB3)** + +| | | | | | | | | | | +|----------------------------------|----|----|----|-----|-----|-----|-----|-----|-----| +| Extra delay added (ms) | 0 | 40 | 80 | 120 | 160 | 200 | 240 | 280 | 320 | +| Significant delay (unacceptable) | 0 | 0 | 3 | 4 | 8 | 36 | 45 | 49 | 50 | +| Delayed but acceptable | 0 | 2 | 5 | 37 | 40 | 14 | 5 | 1 | 0 | +| Timely (satisfactory) | 50 | 48 | 42 | 9 | 2 | 0 | 0 | 0 | 0 | + +**Table I.6 – Test results for RTS1, a real-time strategy game** + +| | | | | | | | | | | +|----------------------------------|----|----|----|-----|-----|-----|-----|-----|-----| +| Extra delay added (ms) | 0 | 40 | 80 | 120 | 160 | 200 | 240 | 280 | 320 | +| Significant delay (unacceptable) | 0 | 0 | 0 | 10 | 13 | 46 | 50 | 50 | 50 | +| Delayed but acceptable | 0 | 0 | 34 | 38 | 37 | 4 | 0 | 0 | 0 | +| Timely (satisfactory) | 50 | 50 | 16 | 2 | 0 | 0 | 0 | 0 | 0 | + +**Table I.7 – Test results for RTS2, a real-time strategy game** + +| | | | | | | | | | | +|----------------------------------|----|----|----|-----|-----|-----|-----|-----|-----| +| Extra delay added (ms) | 0 | 40 | 80 | 120 | 160 | 200 | 240 | 280 | 320 | +| Significant delay (unacceptable) | 0 | 0 | 0 | 10 | 15 | 8 | 49 | 50 | 50 | +| Delayed but acceptable | 0 | 0 | 30 | 36 | 34 | 42 | 1 | 0 | 0 | +| Timely (satisfactory) | 50 | 50 | 20 | 4 | 1 | 0 | 0 | 0 | 0 | + +**Table I.8 – Test results for RTS3, a real-time strategy game** + +| | | | | | | | | | | +|----------------------------------|----|----|----|-----|-----|-----|-----|-----|-----| +| Extra delay added (ms) | 0 | 40 | 80 | 120 | 160 | 200 | 240 | 280 | 320 | +| Significant delay (unacceptable) | 0 | 0 | 0 | 4 | 4 | 25 | 37 | 45 | 49 | +| Delayed but acceptable | 0 | 6 | 10 | 21 | 37 | 23 | 13 | 5 | 1 | +| Timely (satisfactory) | 50 | 44 | 40 | 25 | 9 | 2 | 0 | 0 | 0 | + +**Table I.9 – Test results for RPG1, a role-playing game** + +| | | | | | | | | | | +|----------------------------------|----|----|----|-----|-----|-----|-----|-----|-----| +| Extra delay added (ms) | 0 | 40 | 80 | 120 | 160 | 200 | 240 | 280 | 320 | +| Significant delay (unacceptable) | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | +| Delayed but acceptable | 0 | 0 | 0 | 0 | 2 | 5 | 6 | 5 | 30 | +| Timely (satisfactory) | 50 | 50 | 50 | 50 | 48 | 45 | 44 | 35 | 20 | + +**Table I.10 – Test results for RPG2, a role-playing game** + +| | | | | | | | | | | +|----------------------------------|----|----|----|-----|-----|-----|-----|-----|-----| +| Extra delay added (ms) | 0 | 40 | 80 | 120 | 160 | 200 | 240 | 280 | 320 | +| Significant delay (unacceptable) | 0 | 0 | 0 | 0 | 3 | 5 | 22 | 40 | 50 | +| Delayed but acceptable | 0 | 0 | 2 | 39 | 38 | 42 | 28 | 10 | 0 | +| Timely (satisfactory) | 50 | 50 | 48 | 11 | 9 | 3 | 0 | 0 | 0 | + +**Table I.11 – Test results for RG1, a racing game** + +| | | | | | | | | | | +|----------------------------------|----|----|----|-----|-----|-----|-----|-----|-----| +| Extra delay added (ms) | 0 | 40 | 80 | 120 | 160 | 200 | 240 | 280 | 320 | +| Significant delay (unacceptable) | 0 | 0 | 0 | 0 | 4 | 5 | 12 | 25 | 38 | +| Delayed but acceptable | 0 | 1 | 5 | 6 | 19 | 33 | 30 | 23 | 12 | +| Timely (satisfactory) | 50 | 49 | 45 | 44 | 27 | 12 | 8 | 2 | 0 | + +**Table I.12 – Test results for RG2, a racing game** + +| | | | | | | | | | | +|----------------------------------|----|----|----|-----|-----|-----|-----|-----|-----| +| Extra delay added (ms) | 0 | 40 | 80 | 120 | 160 | 200 | 240 | 280 | 320 | +| Significant delay (unacceptable) | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | +| Delayed but acceptable | 0 | 0 | 0 | 0 | 0 | 4 | 5 | 15 | 26 | +| Timely (satisfactory) | 50 | 50 | 50 | 50 | 50 | 46 | 45 | 35 | 24 | + +**Table I.13 – Test results for FG1, a fighting game** + +| | | | | | | | | | | +|----------------------------------|----|----|----|-----|-----|-----|-----|-----|-----| +| Extra delay added (ms) | 0 | 40 | 80 | 120 | 160 | 200 | 240 | 280 | 320 | +| Significant delay (unacceptable) | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 28 | +| Delayed but acceptable | 0 | 0 | 0 | 0 | 0 | 7 | 17 | 33 | 18 | +| Timely (satisfactory) | 50 | 50 | 50 | 50 | 50 | 43 | 33 | 15 | 4 | + +# Bibliography + +[b-IETF RFC 2326] IETF RFC 2326 (1998), *Real Time Streaming Protocol (RTSP)*. + +[b-IETF RFC 3550] IETF RFC 3550 (2003), *RTP: A Transport Protocol for Real-Time Applications*. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|---------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | General tariff principles | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Terminals and subjective and objective assessment methods | +| Series Q | Switching and signalling | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects and next-generation networks | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.746.10-202008-I_PDF-E/raw.md b/marked/F/T-REC-F.746.10-202008-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..af75ce3917d720dd17d6f5ae6adb08ced001bd69 --- /dev/null +++ b/marked/F/T-REC-F.746.10-202008-I_PDF-E/raw.md @@ -0,0 +1,433 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.746.10** + +(08/2020) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Architecture for spontaneous dialogue +processing system for language learning** + +Recommendation ITU-T F.746.10 + +**ITU-T** + +![ITU logo](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) features a globe with a red lightning bolt striking across it. To the right of the globe, the text "International Telecommunication Union" is written in a blue, sans-serif font, with "ITU" in a larger, bold font above it. + +ITU logo + +International +Telecommunication +Union + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.746.10 + +## Architecture for spontaneous dialogue processing system for language learning + +## Summary + +Recommendation ITU-T F.746.10 describes the architecture, functional entities, and interfaces for a spontaneous dialogue processing system for language learning. The scope of this Recommendation is focused on describing the architecture with different functional components in a spontaneous dialogue processing system, which are: input/output management, dialogue understanding, dual dialogue management and generation, dialogue knowledge management, incremental dialogue knowledge learning, unstructured spontaneous speech recognition management and language learning function. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.746.10 | 2020-08-13 | 16 | 11.1002/1000/14327 | + +## Keywords + +Dialogue processing, dialogue system, language learning framework, spontaneous speech. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2020 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|---------------------------------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 1 | +| 5 Conventions ..... | 2 | +| 6 Functional architectures for spontaneous dialogue processing system for language learning ..... | 2 | +| 6.1 Architectural framework ..... | 2 | +| 6.2 Functional entities of spontaneous dialogue processing system for language learning..... | 3 | +| 6.3 Input/output management module ..... | 4 | +| 6.4 Dialogue understanding module..... | 4 | +| 6.5 Dual dialogue management and generation module ..... | 5 | +| 6.6 Dialogue knowledge management module ..... | 5 | +| 6.7 Incremental dialogue knowledge learning module..... | 5 | +| 6.8 Unstructured spontaneous speech recognition management module ..... | 6 | +| 6.9 Language learning module ..... | 7 | +| Bibliography..... | 10 | + + + +# Recommendation ITU-T F.746.10 + +## Architecture for spontaneous dialogue processing system for language learning + +## 1 Scope + +This Recommendation describes the architecture, functional entities, and interfaces for a spontaneous dialogue processing system for language learning. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T F.746.5] Recommendation ITU-T F.746.5 (2017), *Framework for language learning system based on speech and natural language processing (NLP) technology*. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 dialogue act** [ITU-T F.746.5]: The user's intention or purpose of the utterances in a dialogue. Example: request for information, command for action, agreement. + +**3.1.2 natural language processing** [b-ITU-T F.746.3]: A method that analyses text in natural languages through several processes such as part-of-speech recognition, syntactic analysis and semantic analysis. + +**3.1.3 speech recognition** [b-ITU-T H.703]: A kind of user interface for translation of spoken words into text. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following term: + +**3.2.1 natural language understanding**: A method that analyses text in natural languages to extract information through several processes such as part-of-speech recognition, syntactic analysis and semantic analysis. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-----|------------------------------| +| ASR | Automatic Speech Recognition | +| DB | Database | +| IR | Information retrieval | +| LM | Language Model | + +| | | +|-----|--------------------| +| PLU | Phone-Like Unit | +| POS | Part of Speech | +| QA | Question Answering | +| TTS | Text to Speech | + +## 5 Conventions + +In this Recommendation: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The keywords "is prohibited from" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. +- The keywords "is not recommended" indicate a requirement which is not recommended but which is not specifically prohibited. Thus, conformance with this specification can still be claimed even if this requirement is present. +- The keywords "can optionally" indicate an optional requirement which is permissible, without implying any sense of being recommended. This term is not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +## 6 Functional architectures for spontaneous dialogue processing system for language learning + +### 6.1 Architectural framework + +Figure 1 describes a general architecture of a spoken dialogue system which comprises the following modules: + +- Speech recognition module to generate the user utterance sentence by transcribing the user utterance (speech to text), +- Natural language understanding module to generate several candidates for user utterance intention, which analyses the sentences spoken by a user to codify a representation of its meaning, +- System dialogue managing module to search for a system utterance intention and a pattern by referring to the dialogue knowledge database (DB) for meaning representation of several user utterance intention candidates, +- User dialogue managing module to search the dialogue knowledge DB for the subsequent several candidates for user utterance intention after the system dialogue managing module generates the system utterance intention and the pattern after a current user utterance, +- Dialogue generation module to search the dialogue knowledge DB for the system utterance intention and the pattern selected by the system dialogue managing module or the user dialogue managing module and a dialogue pattern with respect to the several candidates for user utterance intention, and to generate a system utterance text using the found dialogue pattern; + +- Speech synthesis (or TTS (text to speech)) module to output the generated dialogue text using speech. + +![Figure 1 – General architecture of spoken dialogue system. The diagram shows a flow from a user icon on the left through several modules: 'Speech recognition (ASR)' (receiving 'User input (Speech)' and sending 'User input (Text)'), 'Natural language understanding' (sending 'User input (Semantics)'), 'Dialogue management (User/system dialogue mng.)' (connected to 'Search (IR/QA)' and sending 'System utter (Semantics)'), 'Dialogue generation' (sending 'System utter (Text)'), and 'Speech synthesis (TTS)' (outputting 'System utter (Speech)' back to the user).](acfc53eca625d62b38aa2563efa95c3e_img.jpg) + +``` + +graph LR + User((User)) -- "User input (Speech)" --> ASR[Speech recognition (ASR)] + ASR -- "User input (Text)" --> NLU[Natural language understanding] + NLU -- "User input (Semantics)" --> DM[Dialogue management (User/system dialogue mng.)] + DM <--> Search[Search (IR/QA)] + DM -- "System utter (Semantics)" --> DG[Dialogue generation] + DG -- "System utter (Text)" --> TTS[Speech synthesis (TTS)] + TTS -- "System utter (Speech)" --> User + +``` + +F.746.10(20)\_F01 + +Figure 1 – General architecture of spoken dialogue system. The diagram shows a flow from a user icon on the left through several modules: 'Speech recognition (ASR)' (receiving 'User input (Speech)' and sending 'User input (Text)'), 'Natural language understanding' (sending 'User input (Semantics)'), 'Dialogue management (User/system dialogue mng.)' (connected to 'Search (IR/QA)' and sending 'System utter (Semantics)'), 'Dialogue generation' (sending 'System utter (Text)'), and 'Speech synthesis (TTS)' (outputting 'System utter (Speech)' back to the user). + +**Figure 1 – General architecture of spoken dialogue system** + +### 6.2 Functional entities of spontaneous dialogue processing system for language learning + +The functional entities that comprise the spontaneous dialogue processing system for language learning are described in Figure 2 as follows: + +- Input/output management module, +- Dialogue understanding module, +- Dual dialogue management and generation module, +- Dialogue knowledge management module, +- Incremental dialogue knowledge learning module, +- Unstructured spontaneous speech recognition management module, +- Language learning module. + +![Functional entities of spontaneous dialogue system for language learning diagram](a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg) + +The diagram illustrates the functional entities of a spontaneous dialogue system for language learning. It shows the flow of information between various modules and external sources. At the top left, the 'Informal natural language recognition management module' contains 'Multiple decoding integration', 'Backtracking decision', and 'Rescoring'. It receives 'Speech recognition result' from the 'Dialogue understanding module' and sends it to the 'Input/output management module'. The 'Input/output management module' includes 'Input processing', 'Output processing', and 'Speech synthesis', interacting with a user (represented by a stick figure). The 'Dialogue understanding module' contains 'Sentence boundary recognition', 'Topic intention analysis', and 'Non-topic feature analysis'. It receives input from the 'Input/output management module' and sends 'Log data' to the 'Dual dialogue management and generation module'. The 'Dual dialogue management and generation module' includes 'System dialogue management and generation' and 'User dialogue management and generation', which in turn sends 'System response education information' to the 'Input/output management module'. The 'Dialogue knowledge management module' contains 'Self learning dialogue model increase' and 'Topic dialogue model management', receiving 'Dialogue learning corpus' from the 'Incremental dialogue knowledge learning module' and 'Dialogue DBs'. The 'Incremental dialogue knowledge learning module' receives 'Large dialogue corpus from outside' and contains 'Incremental dialogue knowledge acq.' and 'Dialogue scenario extraction'. The 'Language learning module' contains 'Dialogue expression learning' and 'Grammar error correction', receiving 'Log data' from the 'Dual dialogue management and generation module' and sending 'Dialogue management and generation knowledge' to the 'Dialogue knowledge management module'. + +Functional entities of spontaneous dialogue system for language learning diagram + +F.746.10(20)\_F02 + +**Figure 2 – Functional entities of spontaneous dialogue system for language learning** + +### 6.3 Input/output management module + +The Input/output management function provides the audio input interface function which processes the speech data input from the microphone and the output interface function which recognizes speech data and sends back the recognition result to the user. The input/output management module receives the information from an unstructured spontaneous speech recognition management module, a dialogue understanding module as well as a dual dialogue management and generation module. + +### 6.4 Dialogue understanding module + +The dialogue understanding module performs dialogue act understanding function to express the dialogue intention/act of the utterances by analysing user's utterance strings and understanding their meaning. This module also performs dialogue act training function to automatically train dialogue patterns and to provide statistical classification of training information based on the domain dialogue corpus annotated with the dialogue acts. The goal of these module activities is to understand the participating user's intention. + +There are three sub-modules in the dialogue understanding module as follows: + +- Sentence boundary recognition sub-module: in this sub-module the sentences are processed so that the boundary of each sentence is recognized and sent to the next processing sub-module, +- Topic intention analysis sub-module: from the preceding sentence boundary recognition sub-module, each sentence arrives at this module and the topic intention is analysed using deep neural network. +- Non-topic feature analysis: in this sub-module, features other than the topic related ones such as dialogue acts are analysed, and the resulting information is produced for more analysis. + +### 6.5 Dual dialogue management and generation module + +Dual dialogue management function is to analyse both system dialogues and user dialogues for n-best dialogue acts and meaning expressions to decide the best dialogue act to generate the relevant system dialogue and manage the dialogue situations. Dialogue generation function selects the correct system response template for the user's intention and decides meaningful information values to fill in the slots of the selected template. Finally, it generates the appropriate system dialogue sentences. + +Dual dialogue management and generation function provide the following sub functions: + +- User intention analysis function to understand the user's intention correctly for the current topic, +- Auxiliary topic handling function to provide responses if the topic is outside the main topic and to guide the dialogue towards the appropriate topic flow, +- Topic management knowledge search function to find relevant response patterns, slots, or task knowledge to achieve the objective of the current topic, +- Dialogue history management function to store changing tasks, slots and intentions and monitor the intention flows of the dialogue, +- Next utterance recommendation function to suggest the next intention and utterances of the user according to the ongoing dialogue flow, +- System reply generation function to provide responses according to the user's intention, +- Dialogue management log record function to record the dialogue flow, the history changes or search conditions. This log information is used to check the errors in the system and dialogue knowledge. + +### 6.6 Dialogue knowledge management module + +Dialogue knowledge management module performs the function of storing the dialogue knowledge required to achieve the dialogue goal of a domain task in the dialogue knowledge DB. The dialogue knowledge management module includes the function of updating the dialogue knowledge DB, based on hierarchical task, with the subtask dialogue knowledge that is constructed or newly generated by a dialogue service designer. Dual dialogue management function in a spoken dialogue system uses a hierarchical dialogue task library for a system response suitable for a user utterance input. A spoken dialogue processing system performs a dialogue with a user by repeatedly performing the process of recognizing a user utterance as a user utterance sentence generating a system utterance text referring to the dialogue knowledge DB in order to achieve the dialogue goal of the domain task, and thereby producing as speech output the generated system utterance text. + +### 6.7 Incremental dialogue knowledge learning module + +The incremental dialogue knowledge learning module performs the function of adding new dialogue knowledge in the dialogue management database by pseudo situation simulation and incremental learning of situation states. The sub-functions are automatic extraction sub-function of conversation scenarios and incremental dialogue learning sub-functions. + +Automatic extraction of conversation scenarios sub-function is dedicated to collect big data dialogues among people from the scenarios of movies and dramas and to perform the speaker boundary recognition, relevance assessment and topic classification. + +Incremental dialogue learning sub-function is dedicated to recognize the topic of the dialogue scenarios, augment with the slot information, intention and subjects according to the topic, and to produce automatically or by human validation of the dialogue knowledge that can be used for the dialogue system. + +### 6.8 Unstructured spontaneous speech recognition management module + +Unstructured spontaneous speech includes speech errors, stuttering, murmuring hesitation, such as "hmm" and it is more difficult to recognize it. Unstructured spontaneous speech recognition management module integrates multiple decoding results, deciding on the partial backtracking section and rescoring of the selected section to get the best speech recognition results. When the multiple decoding results are integrated, the following decoding and detecting results are aligned and integrated according to the time frames (see Figure 3): + +- Spontaneous speech recognition results from multi-model-based decoding module, +- Multiple detector results from multi-model-based decoding module, +- Prosodic feature-based speech detection results from multi-model-based decoding module, +- Frame unit utterance verification results from phone-like unit-based (PLU-based) decoding module, +- PLU detector results from PLU-based decoding module, +- Symbol recognition results per time from symbol unit recognition module. + +![Diagram of sub-functions for unstructured spontaneous speech recognition management. The diagram shows five input boxes at the bottom: 'Speech decoding results', 'Speech detector results', 'PLU decoding results', 'PLU detector results', and 'Symbol decoding results'. These inputs feed into a central box labeled 'Multiple decoding integration'. From this box, an arrow points up to a box labeled 'Backtracking decision'. From 'Backtracking decision', an arrow points right to a box labeled 'Rescoring'. Finally, an arrow points up from 'Rescoring' to the output 'Speech recognition text'. The entire process is enclosed in a light blue border. A label 'F.746.10(20)_F03' is at the bottom right of the diagram.](27b06ec9f42b5d727a2630f61a5f1861_img.jpg) + +``` +graph TD; A1[Speech decoding results] --> MDI[Multiple decoding integration]; A2[Speech detector results] --> MDI; A3[PLU decoding results] --> MDI; A4[PLU detector results] --> MDI; A5[Symbol decoding results] --> MDI; MDI --> BD[Backtracking decision]; BD --> R[Rescoring]; R --> SRT[Speech recognition text]; +``` + +F.746.10(20)\_F03 + +Diagram of sub-functions for unstructured spontaneous speech recognition management. The diagram shows five input boxes at the bottom: 'Speech decoding results', 'Speech detector results', 'PLU decoding results', 'PLU detector results', and 'Symbol decoding results'. These inputs feed into a central box labeled 'Multiple decoding integration'. From this box, an arrow points up to a box labeled 'Backtracking decision'. From 'Backtracking decision', an arrow points right to a box labeled 'Rescoring'. Finally, an arrow points up from 'Rescoring' to the output 'Speech recognition text'. The entire process is enclosed in a light blue border. A label 'F.746.10(20)\_F03' is at the bottom right of the diagram. + +**Figure 3 – Sub-functions for unstructured spontaneous speech recognition management** + +In the multiple decoding integrating function, the decoding results are selectively chosen for integration if the decoding results are to contribute to the speech recognition performance. In addition, the main speech recognition results from the spontaneous model-based decoding are managed separately for integration from those results of the sub-decoding functions. + +The function of deciding on the partial backtracking section is to predict the error section of spontaneous speech recognition using the integrated multiple decoding results based on the time frame and then to decide reconsidering of the decoding results which does not overlap with the error section. The sub-recognition results are aligned to the time frames first and then the main recognition results are aligned to be used for the backtracking decision. + +Rescoring function serves to reconsider speech recognition decoding results. Then the new recognition candidates are searched for the word boundary that is not in the error section. The target of the rescoring function constitutes the main recognition results. For this purpose, the main recognition results are stored by time frame and are subject to rescoring the words in the section designated by the backtracking function. + +### 6.9 Language learning module + +As described in [ITU-T F.746.5], the language learning module (see Figure 4) takes the user's speech as input and automatically detects grammatical errors, corrects them and then sends the corrected sentence to the user as a feedback to improve his competence of foreign languages. The language learning module consists of a grammar error correction function and a dialogue expression learning function. The grammar error correction function detects errors included in the user's spoken sentences and corrects them, while the dialogue expression learning function provides feedback of the grammar errors and suggests better expressions to the user. + +- Grammar error detection sub-function is dedicated to detecting grammatical errors in the spoken sentences of the users. The grammatical errors are limited to those defined in the system and multiple errors can be detected in one sentence. Various error detection models are used for the grammar error detection. +- Grammar error correction sub-function is dedicated to providing the correction for the detected grammar errors. The error correction function produces the grammar error correction information by using the part of speech (POS) of the error words and context information around the words. +- Error selection and correction sub-function is dedicated to selecting the best error candidate among those from different error detection models and to generate error correction information on the selected candidate error. +- Error feedback providing sub-function informs the user of the grammar errors and suggests better expressions. The explicit/implicit error feedbacks are used accordingly depending on the context in which errors occur. + +![Figure 4 – Language learning module [ITU-T F.746.5] flowchart. The diagram shows a flow from 'Input' through various error detection and correction stages. The stages are grouped into 'Grammar error detection', 'Grammar error correction', and 'Grammar error feedback'. 'Grammar error detection' includes 'Error detection by answer candidates' (linked to 'Answer set based error correction KB'), 'Error detection by correction memory' (linked to 'Error correction memory'), 'Dialogue base error detection' (linked to 'Dialogue context error correction knowledge'), 'Error detection by N-gram' (linked to 'N-gram data'), and 'Error detection by machine learning' (linked to 'Grammar error detection model'). 'Grammar error correction' includes 'Correction candidate generation' and 'Best candidate selector' (linked to 'LM based verification' and 'Language model DB'). 'Grammar error feedback' includes 'Correction feedback generation' (linked to 'Feedback controller' and 'Grammar error detection model').](4801720824e4b5e2361a5564f91cfb70_img.jpg) + +``` + +graph TD + Input --> EDA[Error detection by answer candidates] + Input --> EN[Error detection by N-gram] + + EDA --> ASK[Answer set based error correction KB] + EDA --> GED[Grammar error detection] + + ECM[Error detection by correction memory] --> EM[Error correction memory] + ECM --> GED + + DBED[Dialogue base error detection] --> DCK[Dialogue context error correction knowledge] + DBED --> GED + + EN --> ND[N-gram data] + EN --> GED + + EML[Error detection by machine learning] --> GEDM[Grammar error detection model] + EML --> GED + + GEDM --> CCG[Correction candidate generation] + CCG --> BCS((Best candidate selector)) + BCS --> GEC[Grammar error correction] + + BCS <--> LMV[LM based verification] + LMV --> LMD[Language model DB] + LMD --> GEF[Grammar error feedback] + + CCG --> CFG[Correction feedback generation] + CFG <--> FC((Feedback controller)) + CFG --> GEDM + GEDM --> GEF + +``` + +Figure 4 – Language learning module [ITU-T F.746.5] flowchart. The diagram shows a flow from 'Input' through various error detection and correction stages. The stages are grouped into 'Grammar error detection', 'Grammar error correction', and 'Grammar error feedback'. 'Grammar error detection' includes 'Error detection by answer candidates' (linked to 'Answer set based error correction KB'), 'Error detection by correction memory' (linked to 'Error correction memory'), 'Dialogue base error detection' (linked to 'Dialogue context error correction knowledge'), 'Error detection by N-gram' (linked to 'N-gram data'), and 'Error detection by machine learning' (linked to 'Grammar error detection model'). 'Grammar error correction' includes 'Correction candidate generation' and 'Best candidate selector' (linked to 'LM based verification' and 'Language model DB'). 'Grammar error feedback' includes 'Correction feedback generation' (linked to 'Feedback controller' and 'Grammar error detection model'). + +F.746.10(20)\_F04 + +**Figure 4 – Language learning module [ITU-T F.746.5]** + +Dialogue expression learning function as shown in Figure 5 consists of the following functional blocks: an input recognizer, a speech recognizer, a native language intent analyser, a learning target language intent analyser, a translator and a dialogue processor that comprise a two-language free dialogue system for language learning. When the input form recognized by the input recognizer is a speech, the speech recognizer converts the speech into a text by speech recognition function in the input/output management module. The native language intent analyser analyses a dialogue intent from the recognized native language when the recognized type of the input language is a native language. + +Meanwhile, the learning target language intent analyser analyses a dialogue intent from the recognized learning target language when the recognized type of the input language is a learning target language. This intent analysis function is performed through the topic intention analysis function in the dialogue understanding module by interaction among involved modules. When a dialogue intent of the user is recognized as a translation request through the native language intent analyser, the translator translates a translation target into the learning target language. The dialogue processor provides a system response according to the results of an analysis of the dialogue intent performed by the learning target language intent analyser. The system response is also made based on the result of processing the dialogue of a user native utterance translated into the learning target language through the translator. + +![Block diagram of the Dialogue expression learning module showing the flow from Input recognizer to Dialogue processor via Speech recognizer, Learning target language intent analyser, Native language intent analyser, and Translator.](33ed1f9b27c7c21c797aa928b0f06851_img.jpg) + +``` +graph TD; subgraph Module [Dialogue expression learning module]; IR[Input recognizer] --> SR[Speech recognizer]; SR --> LTLIA[Learning target language intent analyser]; SR --> NLI[Native language intent analyser]; LTLIA --> DP[Dialogue processor]; NLI --> T[Translator]; T --> DP; end +``` + +Block diagram of the Dialogue expression learning module showing the flow from Input recognizer to Dialogue processor via Speech recognizer, Learning target language intent analyser, Native language intent analyser, and Translator. + +F.746.10(20)\_F05 + +**Figure 5 – Dialogue expression learning module** + +## Bibliography + +- [b-ITU-T F.746.3] Recommendation ITU-T F.746.3 (2015), *Intelligent question answering service framework*. +- [b-ITU-T H.703] Recommendation ITU-T H.703 (2016), *Enhanced user interface framework for IPTV terminal devices*. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.746.11-202008-I_PDF-E/raw.md b/marked/F/T-REC-F.746.11-202008-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..63e5465e0b33ee2817d2abcfe3adeda8b837f82c --- /dev/null +++ b/marked/F/T-REC-F.746.11-202008-I_PDF-E/raw.md @@ -0,0 +1,1174 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.746.11** + +(08/2020) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Interfaces for intelligent question answering +system** + +Recommendation ITU-T F.746.11 + +ITU-T + +![ITU logo](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) features a globe with a red lightning bolt striking across it. To the right of the globe, the text "International Telecommunication Union" is written in blue, with the acronym "ITU" in a larger, bold blue font below it. + +ITU logo + +International +Telecommunication +Union + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.746.11 + +## Interfaces for intelligent question answering system + +## Summary + +Recommendation ITU-T F.746.11 describes interfaces for the intelligent question answering service framework (Recommendation ITU-T F.746.3). This Recommendation also defines the interfaces among functional modules to support the intelligent question answering service, which provides advanced functions to generate answers for the user's question in a natural language. The scope of this Recommendation is focused on describing interfaces and functional features for natural language processing function, question analysis function, candidate answer generation function, and answer inference/generation function of intelligent question answering system. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.746.11 | 2020-08-13 | 16 | 11.1002/1000/14328 | + +### Keywords + +Natural language processing, QA intelligent system, QA interfaces QA metadata, question answering. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2020 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Overview ..... | 2 | +| 7 Functional components of intelligent question answering service ..... | 3 | +| 8 Interfaces for natural language processing ..... | 4 | +| 9 Interfaces for question analysis ..... | 7 | +| 10 Interfaces for candidate answer generation ..... | 9 | +| 11 Interfaces for answer inference/generation..... | 10 | +| Annex A – Class API for intelligent question answering system ..... | 12 | +| A.1 Class API for question analysis ..... | 12 | +| A.2 Class API for candidate answer generation..... | 22 | +| A.3 Class API for answer inference/generation ..... | 28 | + + + +# Recommendation ITU-T F.746.11 + +## Interfaces for intelligent question answering system + +## 1 Scope + +This Recommendation addresses the descriptions for interfaces among modules and functions related to intelligent question answering systems. In particular, the scope of this Recommendation includes interfaces and functional features for the following modules of the intelligent question answering system: + +- Natural language processing, +- Question analysis, +- Candidate answer generation, +- Answer inference/generation. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T F.746.3] Recommendation ITU-T F.746.3 (2015), *Intelligent question answering service framework*. + +[ITU-T F.746.7] Recommendation ITU-T F.746.7 (2018), *Metadata for an intelligent question answering service*. + +[ITU-T H.703] Recommendation ITU-T H.703 (2016), *Enhanced user interface framework for IPTV terminal devices*. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 named entity recognition** [ITU-T F.746.3]: A function that recognizes named entities such as PLO which are people, locations and organizations from the sentences. The PLO can be decomposed into more specific named entities depending on the applications. + +**3.1.2 natural language processing** [ITU-T F.746.3]: A method that analyses text in natural languages through several processes such as part-of- speech recognition, syntactic analysis and semantic analysis. + +**3.1.3 semantic analysis** [ITU-T F.746.3]: A function that recognizes the semantic relations among the words around predicates that exist in the same sentence. The semantic analysis function then generates a semantic predicate argument structure (PAS). + +**3.1.4 speech** [ITU-T H.703]: Speech is the vocalized form of human communication. + +**3.1.5 syntactic analysis** [ITU-T F.746.3]: A function that analyses sentence structures and generates dependency relations among words based on dependency grammars. + +**3.1.6 knowledge base** [ITU-T F.746.7]: A collection of knowledge resources that consist of structured and unstructured data. The knowledge base is used to provide information to the various applications that are related to information provisioning such as QA systems and search systems. + +**3.1.7 question answering** [ITU-T F.746.7]: A system that provides answers in a natural language to questions which are in the natural language form by analysing the questions and all the knowledge resources that are available to the system. + +### **3.2 Terms defined in this Recommendation** + +None. + +## **4 Abbreviations and acronyms** + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|--------------------------------------------------| +| DB | Data Base | +| ID | Identifier | +| IQAS | Intelligent Question Answering Service | +| IR | Information Retrieval | +| KB | Knowledge Base | +| LAT | Lexical Answer Type | +| NE | Named Entity | +| NLP | Natural Language Processing | +| NNG | Noun General | +| PAS | Predicate Argument Structure | +| PLO | People, Location, Organization | +| POS | Part of Speech | +| QA | Question Answering | +| SAT | Semantic Answer Type | +| UIMA | Unstructured Information Management Architecture | +| VV | Verb | +| XML | extensible Mark-up Language | + +## **5 Conventions** + +None. + +## **6 Overview** + +Intelligent question answering (QA) system is an advanced function to generate answers for the user's question in a natural language. + +Figure 1 presents the exemplary QA architecture. The QA system consists of several functional blocks: Natural language processing, question analysis, candidate answer generation, answer inference and generation functional blocks. + +![Figure 1 – Example of a QA architecture [ITU-T F.746.3]. The diagram illustrates a sequential process for intelligent question answering. It starts with a 'Question' input, which undergoes 'Question analysis' (containing 'Question understanding'). This is followed by 'Question decomposition', which is supported by 'Natural language processing'. The next stage is 'Candidate answer generation', which involves 'Multiple information search' (illustrated with icons of a globe, a book, and a network). This is followed by 'Answer inference' (containing 'Confidence calculation') and 'Answer generation' (containing 'Best answer generation'), leading to the final 'Answer' output. A reference code 'F.746.11(20)_F01' is present in the bottom right corner.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +Figure 1 – Example of a QA architecture [ITU-T F.746.3]. The diagram illustrates a sequential process for intelligent question answering. It starts with a 'Question' input, which undergoes 'Question analysis' (containing 'Question understanding'). This is followed by 'Question decomposition', which is supported by 'Natural language processing'. The next stage is 'Candidate answer generation', which involves 'Multiple information search' (illustrated with icons of a globe, a book, and a network). This is followed by 'Answer inference' (containing 'Confidence calculation') and 'Answer generation' (containing 'Best answer generation'), leading to the final 'Answer' output. A reference code 'F.746.11(20)\_F01' is present in the bottom right corner. + +**Figure 1 – Example of a QA architecture [ITU-T F.746.3]** + +This Recommendation addresses interfaces among modules and functions related to intelligent question answering services to present details of data and operation functions. [ITU-T F.746.3] and [ITU-T F.746.7] have specified some elements and metadata that are applicable to intelligent question answering services. + +This Recommendation selects basic interfaces among modules and functions from these specifications that are applicable to intelligent question answering services. Names of elements/attributes and functional entities are quoted as they are in the specifications, to distinguish the relationship among the standards. + +## 7 Functional components of intelligent question answering service + +[ITU-T F.746.7] gives the definition for major functional components for intelligent question answering service (IQAS). IQAS functional components are basically composed of natural language processing, question analysis, candidate answer generation, answer inference and answer generation functional blocks as follows: + +- **Natural language processing functional block**, which supports natural language processing on the QA server, +- **Question analysis functional block**, which supports question analysis on the terminal, +- **Candidate answer generation functional block**, which supports candidate answer generation by searching information in various data bases (DBs), +- **Answer inference/generation functional block**, which supports answer inference based on feature normalization and ranking of candidate answers and best answer generation on the terminal. + +![Flowchart of the IQAS system functional blocks. The process starts with 'Question', followed by 'Natural language understanding', 'Question entailment', 'Question decomposition', and 'Question analysis'. A green shaded area encompasses these five steps. Below this, the flow splits into 'KB based candidate retrieval' and 'IR-based passage retrieval', which merge into 'Answer candidates generation'. This is followed by 'Answer candidates reduction', 'Answer verification', and 'Answer confidence reasoning'. The final output is 'Answer, confidence, source'. Brackets on the left group 'Answer candidates reduction' with 'Answer type axiom proving', and 'Answer verification' and 'Answer confidence reasoning' with 'Answer constraints axiom proving'. A dashed line connects 'Question decomposition' to 'Question analysis'. A label 'F.746.11(20)_F02' is at the bottom right.](a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg) + +``` + +graph TD + Question --> NLU[Natural language understanding] + NLU --> QE[Question entailment] + QE --> QD[Question decomposition] + QD --> QA[Question analysis] + QA --> KBCR[KB based candidate retrieval] + QA --> IBPR[IR-based passage retrieval] + KBCR --> ACG[Answer candidates generation] + IBPR --> ACG + ACG --> ACR[Answer candidates reduction] + ACR --> AV[Answer verification] + AV --> ACOR[Answer confidence reasoning] + ACOR --> ACS[Answer, confidence, source] + ATAP[Answer type axiom proving] -.-> ACR + ACAP[Answer constraints axiom proving] -.-> AV + ACAP -.-> ACOR + QD -.-> QA + +``` + +F.746.11(20)\_F02 + +Flowchart of the IQAS system functional blocks. The process starts with 'Question', followed by 'Natural language understanding', 'Question entailment', 'Question decomposition', and 'Question analysis'. A green shaded area encompasses these five steps. Below this, the flow splits into 'KB based candidate retrieval' and 'IR-based passage retrieval', which merge into 'Answer candidates generation'. This is followed by 'Answer candidates reduction', 'Answer verification', and 'Answer confidence reasoning'. The final output is 'Answer, confidence, source'. Brackets on the left group 'Answer candidates reduction' with 'Answer type axiom proving', and 'Answer verification' and 'Answer confidence reasoning' with 'Answer constraints axiom proving'. A dashed line connects 'Question decomposition' to 'Question analysis'. A label 'F.746.11(20)\_F02' is at the bottom right. + +**Figure 2 – Information flow of functional blocks in IQAS system [ITU-T F.746.7]** + +## 8 Interfaces for natural language processing + +This clause specifies interfaces for natural language processing (NLP) function. The NLP module consists of part of speech (POS) analysis, named entity (NE) analysis, dependency analysis, semantic role labelling and ellipsis recovery. The natural language processing function analyses user's questions as well as target documents for knowledge extraction using the semantic and syntactic analysis technology. + +The information flow of natural language processing is described in Figure 3 as sequentially connected submodules which analyse input documents to generate one best main result for each processing sub-module. + +![Flowchart of the natural language processing information flow. It shows five sequential submodules: 'POS analysis', 'NE analysis', 'Dependency analysis', 'Semantic role labelling', and 'Ellipsis recovery'. A label 'F.746.11(20)_F03' is at the bottom right.](b8661c6c54f72ecc7ff6cb05e47b2891_img.jpg) + +``` + +graph LR + PA[POS analysis] --> NEA[NE analysis] + NEA --> DA[Dependency analysis] + DA --> SRL[Semantic role labelling] + SRL --> ER[Ellipsis recovery] + +``` + +F.746.11(20)\_F03 + +Flowchart of the natural language processing information flow. It shows five sequential submodules: 'POS analysis', 'NE analysis', 'Dependency analysis', 'Semantic role labelling', and 'Ellipsis recovery'. A label 'F.746.11(20)\_F03' is at the bottom right. + +**Figure 3 – Information flow of natural language processing** + +Table 8-1 illustrates the functional description and summarizes the input/output of each submodule for the natural language processing module. + +**Table 8-1 – Interfaces of functional blocks for natural language processing module** + +| Functional blocks | Function description | Input | Output | +|-------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------|--------------------------------------|-------------------------------------------------------------| +| POS analysis | To recognize parts of speech in the sentences and assign relevant POS tags considering contextual meaning of the target sentences. | Input sentence | POS tagged sentence in N-doc structure | +| NE analysis | To recognize named entities such as people, locations, organizations (PLO) and others from the sentences. | POS tagged sentence in N-doc | NE tagged sentence in N-doc structure | +| Dependency analysis | To analyse sentence structures and generate dependency relations among words based on dependency grammars. | POS/NE tagged sentence in N-doc | Dependency marked sentence components in N-doc structure | +| Semantic role labelling | To recognize the semantic relations among the words around predicates that exist in the same sentence and assign roles to the analysed semantic entities. | N-doc marked with POS/NE/Dependency | Semantic roles for each component stored in N-doc structure | +| Ellipsis recovery | To detect the ellipsis in a sentence and replace them with relevant noun phrases for the input sentences. | N-doc with previous analysis results | N-doc filled with recovered entities as a list | + +The boxes below present example class API for the POS analysis submodule. The input is a structure called "N\_doc structure" of the input sentence. The output of this module is the unit of words with assigned POS tags. + +The input of the POS analysis submodule is exemplified in the following box. + +``` +{ + "sentence": [ + { + "id": 0, + "text": "Olympic Tug of war is an Olympic program that pits two teams against each other in the Olympic games ", + "morp": [], + "word": [] + } + ] +} +``` + +Output of the POS analysis is exemplified in the following box. The output is the N-Doc structure which stores the words and results of the POS analysis. + +``` +{ + "sentence": [ + { + "id": 0, + "text": "Tug of war is an Olympic program that pits two teams against each other in the Olympic games.", + "morp": [ +``` + +``` + +{"id": 0, "lemma": "Olympic", "type": "NNG", "position": 0, +"weight": 0.9}, +{"id": 1, "lemma": "Tug of war", "type": "NNG", "position": +10, "weight": 0.0701606}, +{"id": 2, "lemma": "in", "type": "JKB", "position": 22, +"weight": 0.0332677}, +{"id": 3, "lemma": " Olympic", "type": "NNG", "position": 26, +"weight": 0.9}, +{"id": 4, "lemma": "an", "type": "JKB", "position": 35, +"weight": 0.153407}, +{"id": 5, "lemma": "program", "type": "NNG", "position": 42, +"weight": 0.0789692}, +{"id": 6, "lemma": "pits", "type": "VV", "position": 54, +"weight": 0.0822907}, +{"id": 7, "lemma": "teams", "type": "NNG", "position": 58, +"weight": 0.152575}, +{"id": 8, "lemma": "against", "type": "JKO", "position": 64, +"weight": 0.137686}, +{"id": 9, "lemma": "each", "type": "VV", "position": 68, +"weight": 0.9}, +{"id": 10, "lemma": "other", "type": "ETM", "position": 74, +"weight": 0.184941}, +{"id": 11, "lemma": "Olympic", "type": "NNG", "position": 78, +"weight": 0.9}, +{"id": 12, "lemma": "games", "type": "NNG", "position": 88, +"weight": 0.135556}, ], +"word": [ +{"id": 0, "text": "Olympic""type": "", "begin": 0, "end": 0}, +{"id": 1, "text": " Tug of war", "type": "", "begin": 1, +"end": 2}, +{"id": 2, "text": "Olympic", "type": "", "begin": 3, "end": +4}, +{"id": 3, "text": "Program", "type": "", "begin": 5, "end": +6}, +{"id": 4, "text": "pits", "type": "", "begin": 7, "end": 8}, +{"id": 5, "text": "teams", "type": "", "begin": 9, "end": 10}, +{"id": 6, "text": "Olympic", "type": "", "begin": 11, "end": +11}, +{"id": 7, "text": " games", "type": "", "begin": 12, "end": +12}, +] +} +] +} + +``` + +## 9 Interfaces for question analysis + +This clause specifies interfaces for a user question analysis function. The question analysis module as shown in Figure 4 analyses questions which are input in a natural language by the user, understands the user's intention, and recognizes various information on the answers that should be presented as the output of the intelligent question answering system. + +![Flowchart of sub-modules for question analysis. The process starts with 'Question' input to 'Question entailment'. 'Question entailment' feeds into 'Question decomposition' and 'Complex questions'. 'Question decomposition' includes 'Sentence based question split', 'Sub-question type classification', 'Sub-question relation recognition', and 'QIF based question split'. 'QIF based question split' leads to 'Question relation type classification' (outputting 'Axioms of individual facts') and 'QIF generation' (with input {QD, QC, QF, AT, QT, AC, QIF}). 'Sub-questions' from 'Sub-question relation recognition' feed into 'QDomain classification', 'QFocus detection', and 'QIF generation'. 'QDomain classification' leads to 'QClass classification' (with sub-blocks QC^AForm, QC^Syntactic, QC^Semantic). 'QFocus detection' leads to 'AType recognition' (with sub-blocks LAT detection, SAT classification) and 'QTopic detection'. 'AType recognition' and 'QTopic detection' both feed into 'AConstraint recognition' (with sub-blocks: FRAME^Syntactic, Comparison, Definition, Negation, Apposition, Time, Language, Location). 'AConstraint recognition' feeds into 'QIF generation'.](eefe19c5e14dc4d6c316b7f7fbb7d7d7_img.jpg) + +F.746.11(20)\_F04 + +Flowchart of sub-modules for question analysis. The process starts with 'Question' input to 'Question entailment'. 'Question entailment' feeds into 'Question decomposition' and 'Complex questions'. 'Question decomposition' includes 'Sentence based question split', 'Sub-question type classification', 'Sub-question relation recognition', and 'QIF based question split'. 'QIF based question split' leads to 'Question relation type classification' (outputting 'Axioms of individual facts') and 'QIF generation' (with input {QD, QC, QF, AT, QT, AC, QIF}). 'Sub-questions' from 'Sub-question relation recognition' feed into 'QDomain classification', 'QFocus detection', and 'QIF generation'. 'QDomain classification' leads to 'QClass classification' (with sub-blocks QC^AForm, QC^Syntactic, QC^Semantic). 'QFocus detection' leads to 'AType recognition' (with sub-blocks LAT detection, SAT classification) and 'QTopic detection'. 'AType recognition' and 'QTopic detection' both feed into 'AConstraint recognition' (with sub-blocks: FRAME^Syntactic, Comparison, Definition, Negation, Apposition, Time, Language, Location). 'AConstraint recognition' feeds into 'QIF generation'. + +**Figure 4 – Sub-modules for question analysis [ITU-T F.746.7]** + +The question analysis module consists of the following functional blocks: + +- Question decomposition and recognition of sub-question relations, +- Question type classification, +- Recognition of question focus, answer type, and reliability, +- Question topic detection. + +Table 9-1 describes the functional description and summarises the input/output of each functional block for the question analysis module. + +**Table 9-1 – Interfaces functional blocks for question analysis module** + +| Functional blocks | Function description | Input | Output | +|------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-----------------------------------------------------------------------|---------------------------------------------------------| +| Question decomposition and recognition of sub-question relations | To decompose a question based on its sentence structure and meaning.
To recognize the decomposed sub-question types and relations among them. | Question focus, lexical answer type (LAT), NLP-analysed question | Sub question information and relation information | +| Question type classification | To define question types based on QA strategies depending on the domain.
To classify the question types. | Question, NLP-analysed question | Pre-defined question types | +| Recognition of question focus, answer type, and reliability | To recognize question focus for the words or phrases which can be replaced by answer candidates.
To assign the required answer types for lexical answer types (LAT) and semantic answer types (SAT). | NLP-analysed question, rule dictionary for answer type recognition | Recognized question focus and answer type (LAT and SAT) | +| Question topic detection | To recognize major entities in the question sentence and detect the core topic of the question. | Question, NLP-analysed question, Title list of on-line encyclopaedias | Core topic of question | + +The functional block "Question decomposition and recognition of sub-question relations" divides a question based on its sentence structure and meaning and does the function of recognizing sub-question types and relations among them. To do this decomposition, question decomposition types are pre-defined as shown in Table 9-2. + +**Table 9-2 – Question decomposition types** + +| Decomposition types | Description | +|----------------------------|---------------------------------------------------------------------------------------| +| QD_Pa | Sub-questions are in parallel relations. | +| QD_Pa_Se | Sub-questions are separate and in parallel relations. | +| QD_Ne | Sub-questions are in nested relations.
One question is nested in another question. | +| QD_Ne_Se | Sub-questions are in nested relations.
They are separate questions. | +| QD_None | The question is not decomposable. | +| QD_None_Se | The non-decomposable questions are separated. | + +The following text presents example class API for a functional block "Question decomposition and recognition of sub-question relations". The input is N\_doc structure of user's question and LAT vector of the original question. The output of this block is the sub-question relation vector and question decomposition information vector. + +Class API for question analysis module is presented in clause A.1. + +## 10 Interfaces for candidate answer generation + +This clause specifies interfaces for a candidate answer generation function. A candidate answer generation module performs the index and search functions on the document collections and generates candidate answers from the input query using search results and various databases such as a structured/unstructured knowledge base as described in Figures 5 and 6. + +The answer candidate generation block generates all possible answer candidates from the structured/unstructured resources based on the question and question division information. + +The answer linguistic axiom prover block selects candidate answers for the evidence collection target through answer type and answer constraint axiom proving. + +The answer evidence retrieval and contextual axiom prover collects evidences for the answers and verifies the axiom for the contexts. + +Class API for candidate answer generation is found in clause A.2. + +![Diagram of sub-blocks for the candidate answer generation module](c67d21fb3d9042e88cdc669f071b4e7c_img.jpg) + +The diagram illustrates the flow of the candidate answer generation module. It starts with 'Question understanding' on the left, which is represented by the variable $Q^{Understanding}$ . This input is processed through a series of four sub-blocks enclosed in a dashed red box: 'Answer candidate generation' (producing $A^{candidate}$ ), 'Answer linguistic axiom prover' (producing $A^{Reduction}$ ), and 'Answer evidence retrieval and contextual axiom prover' (producing $A^{Verification}$ ). The output of the third sub-block is then processed by 'Answer confidence reasoning' (producing $A^{Confidence}$ ), which is outside the dashed box. The final output is 'Best answer'. The diagram is labeled F.746.11(20)\_F05 at the bottom right. + +Diagram of sub-blocks for the candidate answer generation module + +Figure 5 – Sub-blocks for the candidate answer generation module + +![Figure 6: Candidate answer generation module in QA system. This is a complex architectural diagram showing the flow of information from a client through various processing modules to generate answers. On the left, a 'WiseQA client' sends a 'Question' and receives an 'Answer, confidence'. The central 'WiseQA controller' contains an 'Answer confidence reasoner' and a 'Merger'. The 'Question analyzer' sends 'Natural language question' and 'Question analysis Info (Question axioms)' to the controller. The 'Answer candidate generator' (with IR-based, Entity, property-based, and QIF-based components) sends 'Question analysis Info' and 'Answer candidates' to the controller. The 'Linguistic axiom prover' (with inductive/deductive answer type and constraint axioms) sends 'Question axioms, answer candidates' and 'Scored answer candidates' to the controller. The 'Answer evidence retriever' sends 'Question axioms feasible answer candidates' and 'Evidence passages' to the controller. The 'Contextual axiom prover' (with inductive, abductive, and deductive components) sends 'Question axioms feasible answer candidates evidence passages' and 'Scored evidence passage' to the controller. The controller also interacts with several external resources: 'Document resource', 'WiseKB', 'Answer resource' (connected via WWN), and 'Evidence resource'.](fa859e4e468bfb2710a94527f2c504af_img.jpg) + +Figure 6: Candidate answer generation module in QA system. This is a complex architectural diagram showing the flow of information from a client through various processing modules to generate answers. On the left, a 'WiseQA client' sends a 'Question' and receives an 'Answer, confidence'. The central 'WiseQA controller' contains an 'Answer confidence reasoner' and a 'Merger'. The 'Question analyzer' sends 'Natural language question' and 'Question analysis Info (Question axioms)' to the controller. The 'Answer candidate generator' (with IR-based, Entity, property-based, and QIF-based components) sends 'Question analysis Info' and 'Answer candidates' to the controller. The 'Linguistic axiom prover' (with inductive/deductive answer type and constraint axioms) sends 'Question axioms, answer candidates' and 'Scored answer candidates' to the controller. The 'Answer evidence retriever' sends 'Question axioms feasible answer candidates' and 'Evidence passages' to the controller. The 'Contextual axiom prover' (with inductive, abductive, and deductive components) sends 'Question axioms feasible answer candidates evidence passages' and 'Scored evidence passage' to the controller. The controller also interacts with several external resources: 'Document resource', 'WiseKB', 'Answer resource' (connected via WWN), and 'Evidence resource'. + +F.746.11(20)\_F06 + +Figure 6 – Candidate answer generation module in QA system + +## 11 Interfaces for answer inference/generation + +This clause specifies interfaces for candidate answer generation function. The function of answer inference and the generation module is to decide and generate the best answer by measuring reliabilities of the answer candidates using the query axiom, the filtered answer candidates, and the reasoned answer candidates as features to determine ranks of the answer candidates, based on the calculated reliability. The answer candidates are filtered and reasoned out based on a similarity between the query axiom and the answer candidates and by using an inductive, deductive or an abductive reasoning. + +The inference and generation modules compare a threshold value with a reliability ratio of a best answer candidate to the second-best answer candidate, readjusting the determined ranks according to the result of the comparison, and detecting the best answer candidate, determined through the readjustment, as a final answer of the question answering service. + +Class APIs for answer inference/generation are presented in clause A.3. + +**Table 11-1 – Interfaces of functional blocks for answer inference/generation module** + +| Functional blocks | Function description | Input | Output | +|--------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------------------------------------|----------------------------------------------------------------------------| +| Reliability measuring | To measure reliabilities of the answer candidates using the query axiom, the filtered answer candidates, the reasoned answer candidates.
To determine ranks of the answer candidates, based on the calculated reliability. | Top N answer candidates
Evidence feature vector | Ranked list of Top N answer candidates
Confidence
Source information | +| Best answer generation | To compare a threshold value of a best answer candidate to the second-best answer candidate.
To readjust the determined ranks according to a result of the comparison.
To generate the best answer candidate as the final answer. | Ranked list of Top N answer candidates
Confidence
Source information
Data structure of the answer candidates | If threshold met, best answer produced, otherwise no answer | + +## Annex A + +### Class API for intelligent question answering system + +(This annex forms an integral part of this Recommendation.) + +This annex describes class API for intelligent question answering system for each module. + +### A.1 Class API for question analysis + +#### A.1.1 Class API for QDecomposition + +This clause provides information for the class API for QDecomposition. + +| | | +|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--| +| Class Name | | +| QDecomposition | | +| Description | | +| Natural language question decomposition, recognition of decomposition type, sub-question type, sub-question relations | | +| Include files | | +| #include "QAnalHeader/QDecomposition.h"
#include "QAnalHeader/NDocUtil.h"
#include "QAnalHeader/RegEx.h"
#include "QAnalHeader/QStruct.h"
#include "QAnalHeader/QAnalRSCMng.h" | | +| Member Variables | | +| /**
Question Decomposition Type)
*/
typedef enum _Question_Decomposition_Type_{
QD_Pa, /**questions in parallel*/
QD_Ne, /**nested questions*/
QD_None, /**} eQDType;

/**
SubQuestion Type
*/
typedef enum _SubQuestion_Type_{
SQT_Fact, /**SQT_RelFact, /**SQT_Question, /**SQT_InnerQ, /**SQT_OuterQ /**} eSQType;

/**
Relation of the sub-questions
*/
typedef enum _SubQ_Relations_
{
SQR_And, /**SQR_Dep, /**SQR_None, /**SQR_SUPPORT /**}eSQRelation;

/** | | + +``` + +doc structure to store information about the sub-questions +*/ +typedef struct _QDecomposition_Info_ +{ + unsigned int iID; /** vSubQRelation; /** vSubQInfo; /** Split_Question(sQAnal_Unit &sQAnal_Info); + /** &vSubQs, sQAnal_Unit + sQAnal_Info); /** vSubQs, + vector &vSubQRel); /** GetSubQSentIDs(sQAnal_Unit sQAnal_Info); + bool IsQuesitonByRule(string strTaggedSent); /** GetQFsInSubQ(QDecomp_Info sSubQ, vector + vOrgQFInfo); /** GetLATsInSubQ(QDecomp_Info sSubQ, vector + vOrgLATInfo); /**#include "QAnalHeader/ Classifier_ML.h"
#include "QAnalHeader/ QAnalRSCMng.h" | | +| Member Variables | | +| protected:
CQAnalRSCMng *pRscMng; /**/** fuction for registering Resource manager handle */
virtual void SetRscMng(CQAnalRSCMng * pRscManager) = 0;

/**function for recognizing question types */
virtual vector QClassifier(string strFeatures, string strTaggedQ, bool bIsMC, bool bIsCB);

/**feature extraction function for machine learning Classification */
virtual string ExtractFeatures(string strQ, N_Doc ndoc);

protected:
/**Function for a hybrid Lexico-Semantic Rule and machine learning result */
vector HybridCQT(vector vRuleCQTs, vector vMLCQTs); | | + +| | | +|-----------------------------------------------------------------------------------------------------------------------------------------------------------------|--| +| Class Name | | +| CQClassifier_Rules | | +| Description | | +| Core class for classifying questions by rule-based method | | +| Include files | | +| #include "QAnalHeader/Classifier_Rules.h"
#include "QAnalHeader/QAnalRSCMng.h"
#include "QAnalHeader/RegEx.h" | | +| Member Variables | | +| private:
CQAnalRSCMng *pRscMng;
vector vRules; /**rule dictionary for classification*/ | | +| Member Functions | | +| public:
CClassifier_Rules(void);
~CClassifier_Rules(void);

protected:

/** Registration for the resource handler and classification rules */ | | + +``` + + void SetRscMng_Rules(CQAnalRSCMng *pRscManager, vector +&vClassifier_Rules); + + /** classification function for rule-based function */ + virtual vector Classifier_Rules(string strTaggedQ, bool +bIsMC, bool bIsCB); + +private: + /** function to find question classification candidates based on rules */ + vector GetCQTCandidates(string strTaggedQ, bool bIsMC, bool +bIsCB); + + /** function to integrate classified question types based on rules */ + void UnifyingCandidates(vector &vCandidates); + +``` + +| | | +|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--| +| Class Name | | +| CQClassifier_ML | | +| Description | | +| Core class for classifying questions by machine learning method | | +| Include files | | +| #include "QAnalHeader/Classifier_ML.h"
#include "QAnalHeader/QAnalRSCMng.h" | | +| Member Variables | | +| private:
CQAnalRSCMng *pRscMng;
CRF_MODEL *pCRFHandle;/**CClassifier_ML(void);
~CClassifier_ML(void);

protected:
/** Registration for the resource handler and classification ML */
void SetRscMng_ML(CQAnalRSCMng * pRscManager, CRF_MODEL *crf_model);

/** classification function based on ML */
virtual vector Classifier_ML(string strFeatures);

public:
/** function for CQT classification based on features */
void ClassifyCQT(string strFeature, vector &vReturnCQT);

/** function to extract QT features for ML */
string ExtractQTFeature(N_Doc ndoc);
string ExtractBiGramFeature(N_Doc ndoc);
string ExtractSymbolFeature(N_Doc ndoc);
string ExtractLastWordFeature(N_Doc ndoc); | | + +| | | +|------------------------------------------------------------|--| +| Class Name | | +| CQClassifier_AForm | | +| Description | | +| Question classification class according to the answer type | | +| Include files | | +| #include "QAnalHeader/QClassifier.h" | | + +| | | +|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--| +| #include "QAnalHeader/QAnalRSCMng.h"
#include "QAnalHeader/QClassifier_AFrom.h" | | +| Member Variables | | +| Member Functions | | +| public:
CQClassifier_AForm(void);
~CQClassifier_AForm(void);

void SetRscMng(CQAnalRSCMng * pRscManager);

/** function to recognize CQT */
sQAnal_CQT QClassifier(string strQ, N_Doc ndoc, string strTaggedQ, bool bIsMC, bool bIsCB); | | + +| | | +|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--| +| Class Name | | +| CQClassifier_Sem | | +| Description | | +| Question classification class based on semantic feature of the question | | +| Include files | | +| #include "QAnalHeader/QClassifier.h"
#include "QAnalHeader/QAnalRSCMng.h"
#include "QAnalHeader/QClassifier_sem.h" | | +| Member Variables | | +| Member Functions | | +| public:
CQClassifier_Sem(void);
~CQClassifier_Sem(void);

void SetRscMng(CQAnalRSCMng * pRscManager);

/** function to recognize question types (CQT) for questions */
vector QClassifier(string strQ, N_Doc ndoc, string strTaggedQ, bool bIsMC, bool bIsCB);

/** function to map the question classification results onto string */
void MappingStrQType(vector &vResult); | | + +#### A.1.3 Class\_API for recognition of question focus, answer type and reliability + +This clause provides information for the Class\_API for recognition of question focus, answer type and reliability. + +| | | +|-------------------------------------------------------------------------------------------------------------------------|--| +| Class Name | | +| CQAnalLAT | | +| Description | | +| Lexical answer type recognition module Class | | +| Include files | | +| #include "QAnalHeader/QAnalRSCMng.h"
#include "QAnalHeader/QAnalLAT_Rules.h"
#include "QAnalHeader/QAnalLAT_ML.h" | | +| Member Variables | | +| CQAnalRSCMng *pRscMng;

//SAT recognition based on rules-module handle | | + +| | | +|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--| +| 
CQAnalLAT_Rules *pRuleLAT;  // SAT recognition based on ML-module handle CQAnalLAT_ML *pMLLAT;
| | +| Member Functions | | +| 
    void SetRscMng(CQAnalRSCMng * pRscManager);      /** function to recognize lexical answer types for a question */     vector<sLAT_Info> RecognizeLAT(string strQ, N_Doc ndoc, vector<sQT_Info> vQTs, string strTaggedQ);      /** to get the LAT analysis handle based on rule-based method */     CQAnalLAT_ML *GetQAnalLAT_ML_Handle() { return this->pMLLAT; };  protected:      /** function to hybrid Lexico-Semantic Rule and ML results */     vector<sLAT_Info> HybridLAT(vector<sLAT_Info> vRuleLATs, vector<sLAT_Info> vMLLATs);  private:     void EraseStopWordInLATs(vector<sLAT_Info> &vReturnLAT);
| | +| Function Name | | +| RecognizeLAT | | +| Class Name | | +| CQAnalLAT | | +| Description | | +| Function to recognize lexical answer type for a given question | | +| Syntax | | +| vector RecognizeLAT(string strQ, N_Doc ndoc, vector vQTs, string strTaggedQ); | | +| Return Value | | +| vector – Recognized LAT candidates vector | | +| Parameters | | +| @param string strQ – original question string
@param N_Doc d – language analysis result of the question
@param vector vQTs – question type information recognized in a question
@param string strTaggedQ – input string for rule matching | | + +| | | +|----------------------------------------------------|--| +| Function Name | | +| GetQAnalLAT_ML_Handle | | +| Class Name | | +| CQAnalLAT | | +| Description | | +| Function to get the handle of ML based LAT results | | +| Syntax | | +| CQAnalLAT_ML *GetQAnalLAT_ML_Handle(); | | +| Return Value | | +| CQAnalLAT_ML * | | +| Parameters | | + +| | | +|---------------------------------------------------------------------------------------------------------------------|--| +| Function Name | | +| HybridLAT | | +| Class Name | | +| CQAnalLAT | | +| Description | | +| Hybrid function of Lexico-Semantic Rule and ML results | | +| Syntax | | +| vector<sLAT_Info> HybridLAT(vector<sLAT_Info> vRuleLATs, vector<sLAT_Info> vMLLATs); | | +| Return Value | | +| vector<sLAT_Info> | | +| Parameters | | +| @param vector<sLAT_Info> vRuleLATs – rule-based LAT result | | +| @param vector<sLAT_Info> vMLLATs – ML based LAT result | | + +| | | +|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--| +| Class Name | | +| CQAnalSAT | | +| Description | | +| SAT recognition class | | +| Include files | | +| #include "QAnalHeader/QAnalRSCMng.h"
#include "QAnalHeader/QAnalSAT_Rules.h"
#include "QAnalHeader/QAnalSAT_ML.h" | | +| Member Variables | | +| CQAnalRSCMng *pRscMng;
CQAnalSAT_Rules *pRuleSAT;
CQAnalSAT_ML *pMLSAT; | | +| Member Functions | | +| 
 void SetRscMng(CQAnalRSCMng * pRscManager);      /** function to recognize Semantic answer type for a question*/     vector<sSAT_Info> RecognizeSAT(string strQ, N_Doc ndoc, vector<sQT_Info> vQTs, string strTaggedQ);      /** function to get SAT analysis handle based on ML */     CQAnalSAT_ML *GetQAnalSAT_ML_Handle() { return this->pMLSAT; };      /** function to get SAT analysis handle based on rules */     CQAnalSAT_Rules *GetQAnalSAT_Rule_Handle() { return this->pRuleSAT; };  protected:     /** function to hybrid Lexico-Semantic Rule and ML results */     vector<sSAT_Info> HybridSAT(vector<sSAT_Info> vRuleSATs, vector<sSAT_Info> vMLSATs);      /**function to expand the recognized SAT */     void ExpansionSAT(vector<sSAT_Info> &vSATs);
| | + +| | | +|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--| +| Function Name | | +| Class Name | | +| CQAnalSAT | | +| Description | | +| Function to recognize SAT for a question | | +| Syntax | | +| vector<sSAT_Info> RecognizeSAT(string strQ, N_Doc ndoc, vector<sQT_Info> vQTs, string strTaggedQ); | | +| Return Value | | +| vector<sSAT_Info> - recognized SAT results | | +| Parameters | | +| @param string strQ – question string
@param N_Doc d – language analysis result of the question
@param vector vQTs – answer type information recognized in the question
@param string strTaggedQ | | + +| | | +|------------------------------------------------------------------------------------------------------------------------------------------|--| +| Function Name | | +| HybridSAT( | | +| Class Name | | +| CQAnalSAT | | +| Description | | +| Hybrid function of Lexico-Semantic Rule and ML results | | +| Syntax | | +| vector<sSAT_Info> HybridSAT(vector<sSAT_Info> vRuleSATs, vector<sSAT_Info> vMLSATs); | | +| Return Value | | +| vector<sSAT_Info> - Hybrid SAT candidate result bector | | +| Parameters | | +| @param vector vRuleSATs – SAT results by rule-based mehod
@param vector vMLSATs – SAT results by ML-based mehod | | + +| | | +|---------------------------------------------------------------------|--| +| Function Name | | +| ExpansionSAT | | +| Class Name | | +| CQAnalSAT | | +| Description | | +| This function is to expand the SAT which was recognized. | | +| Syntax | | +| void ExpansionSAT(vector<sSAT_Info> &vSATs); | | +| Return Value | | +| void | | +| Parameters | | +| @param vector &vSATs – Results of the recognized SAT | | + +#### A.1.4 Class API for question topic detection + +This clause provides information for the class API for question topic detection. + +| | | +|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--| +| Class Name | | +| CQAnalKeywords | | +| Description | | +| Function to recognize main keywords in a question | | +| Include files | | +| #include "QAnalHeader/QAnalRSCMng.h" | | +| Member Variables | | +| CQAnalRSCMng *pRscMng; | | +| Member Functions | | +| 
void SetRscMng(CQAnalRSCMng * pRscManager) { this->pRscMng =
pRscManager; };
CQAnalRSCMng *GetRscMng() { return this->pRscMng; };

/** function to get NE tagged objects from the language analysis results
*/
map<string, sTitle_Info> GetNEWords(N_Doc ndoc);

/** function to get chunk information from the language analysis results
*/
map<string, sTitle_Info> GetChunkWords(N_Doc ndoc, vector<string>
vStrType);
| | +| Function Name | | +| GetNEWords | | +| Class Name | | +| CQAnalKeywords | | +| Description | | +| function to get NE tagged objects from the language analysis results | | +| Syntax | | +| map GetNEWords(N_Doc ndoc); | | +| Return Value | | +| map - recognized object information (key : text_morp
begin) | | +| Parameters | | +| @param N_Doc ndoc – language analysis results | | + +| | | +|-----------------------------------------------------------------------------|--| +| Function Name | | +| GetChunkWords | | +| Class Name | | +| CQAnalKeywords | | +| Description | | +| function to get chunk information from the language analysis results | | +| Syntax | | +| map GetChunkWords(N_Doc ndoc, vector vStrType) | | + +| | | +|----------------------------------------------------------------------------------|--| +| Return Value | | +| map - recognized object information (key : text_morp begin) | | +| Parameters | | +| @param N_Doc ndoc - language analysis results | | +| @param vector vStrChunkTypes - chunk types to get | | + +| | | +|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--| +| Class Name | | +| CQAnalWikiTitle | | +| Description | | +| Entity Linking class to map the keywords in the question into the WIKI titles to resolve the ambiguity | | +| Include files | | +| #include "QAnalHeader/QAnalRSCMng.h"
#include "QAnalHeader/QAnalKeywords.h" | | +| Member Variables | | +| CQAnalRSCMng *pRscMng; | | +| Member Functions | | +| /** function to recognize WIKI titles */
vector GetWikiTitles(string strQ, N_Doc ndoc);

private:

/** function to recognize candidates for WIKI titles */
vector GetWikiTitleCandidates(N_Doc ndoc);

/** function to resolve the WIKI title ambiguity */
void DisambiguationTitle(string strQ, vector &vTitles);

/** function to find WIKI title from the candidates */
void FindWikiTitles(sTitle_Info sCandidate, vector &vTitles);

/** function to search for the string if it is in the WIKI title dictionary */
void LookupWikiDic(string title, vector &vEntities); | | + +| | | +|-------------------------------------------------------------|--| +| Function Name | | +| GetWikiTitles | | +| Class Name | | +| CQAnalWikiTitle | | +| Description | | +| Function to recognize WIKI titles | | +| Syntax | | +| vector GetWikiTitles(string strQ, N_Doc ndoc); | | +| Return Value | | +| vector - Wikititle information | | +| Parameters | | +| @param string strQ - Question | | +| @param N_Doc ndoc - language analysis results | | + +### A.2 Class API for candidate answer generation + +#### A.2.1 Class\_API for candidate answer index and search + +This clause provides information for the Class\_API for candidate answer index and search. + +| | | +|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--| +| Class Name | | +| DIndexingMultiThread | | +| Description | | +| Unstructured indexing: top class | | +| Include files | | +| none | | +| Member Variables | | +| private Thread t;
private String threadName;
private String filePath;
String syntacticLexical;
String syntacticRelation;
String surficialRelation;
long start,end;
String hbaseTableName;
CloudSolrServer solrcloud; | | +| Member Functions | | +| DIndexingMultiThread(String name, String hbaseTableName, CloudSolrServer solrcloud, String filePath)
public void run()
public void start()
public String
makeTerm2StringWithWhiteSpace(ArrayList>>
sentenceList) | | + +| | | +|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--| +| Class Name | | +| PrimarySearch.cpp | | +| Description | | +| Unstructured indexing: top class | | +| Include files | | +| typedef struct {
string docid; // document ID
string domain; // wiki, dictionary
string type; // document, section, definition, passage
string rowkey; // hbase key (for fetching language analysis results)
string page_struct; // doc title
string description; // information for the origin of doc
double weight; // doc weight
int ranking; // doc ranking
int s_sentid; // start sentence ID
int e_sentid; // end sentence ID
vector topic; // topic of the doc
vector syntacticLexical; // keyword info matched in the
search
vector syntacticRelation; // keyword info matched in the
search
vector surficialSemanticRelation; // keyword info
matched in the search
vector semanticRelation; // keyword info matched in the
search
} | | + +| | | +|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------| +| 
 } DOC_RESULT_STRUCT;  //Primary Search integration typedef struct {     string query;     //      // save the search result     //      vector<DOC_RESULT_STRUCT> doc_result_list;      // to store doc search results     vector<DOC_RESULT_STRUCT> def_result_list;      // to store definition search results     vector<DOC_RESULT_STRUCT> sec_result_list;      // to store section search results     vector<DOC_RESULT_STRUCT> psg_result_list;      // to store paragraph search results } doc_result;
| | +| Member Variables | //Skipped// | +| Member Functions | | +| 
 void hexconvert( char *text, unsigned char bytes[] ); string replaceAll(string str, string pattern, string replace); string* strSplit(string strOrigin, string strTok); PrimarySearch(string solrIP, string hbaseIP, string jarPath); string p_q_generation(string collectionName, N_Doc ndoc); string p_q_reGeneration(string collectionName, N_Doc ndoc, vector<string> candidateAnswer); string p_q_reGeneration(string collectionName, N_Doc ndoc, string docid_sentid); vector<string> assignPassageRange(string docid_sentid, int range); PrimarySearch(); ~PrimarySearch(); string p_search( string collectionName, string query, int max_cnt); string p_search( string collectionName, string query, string title, int max_cnt); string getLang(string collectionName, string rowkey); string getRowkey(string docid);
| | + +#### A.2.2 Class\_API for candidate answer generation sub-block + +This clause provides information for the Class\_API for candidate answer generation sub-block. + +| | | +|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--| +| Class Name | | +| AnswerUnit | | +| Description | | +| Answer candidates and structure of representing evidences | | +| Include files | | +| 
 class AnswerUnit { public:      string answer;     string answer_key;      //     double tot_weight;      //engine matching weight (normalized reliability)      int total_sent;
| | + +``` + + list answer_sent; + CIFeature ci_feat; //context independent features +}; + +class AnswerFeature +{ +public: + + AnswerFeature(void); + ~AnswerFeature(void); + + double TyCor_score; //TyCor feature + double Ln_score; //IR+ voc similarity feature + double SEM_score; //syntax and meaning feature + double CONS_score; //question constraint feature + + double IR_score; //IRweight + + string TE; //integration + + double confidence; //answer confidence + + string src_ENGINE; // source of engine +}; + +// answer candidate sentences for Answer Evidence +class AnswerSent +{ +public: + string answer; + double weight; + + string docID; //answer doc ID + int sentID; //answer sentence ID + QADocType doc_type; // + + //canGen results + string source; // source (Content, Title, ...) + double DIRweight; // doc search weight + int DIRsrcType; // 0=Doc, 1=section, 2=sentence + + string title; + string sent; + string url; + string cpname; + unsigned long long date=0; +}; + +//context independent features +class CIFeature +{ +public: + string type; // types + string NEType; // NE type + double inQ = 0; // if included in question + + double CIweight; //after feature engineering + double TyCor_vec[TyCor_vec_SIZE]; + + string CI_feature; //CI feature for training str +}; + +``` + +| | | +|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--| +| Class Name | | +| CandidateGenerator | | +| Description | | +| Top class for answer candidate generation | | +| Include files | | +| 
 #include "DR/DR.h" #include "QStruct.h" #include "DBSolution/DBSolution.h" #include "GetNDocHeader/GetNDoc.h" #include "TyCor/TyCor.h" #include "LingCor/LingCor.h" #include "SpaTempCor/SpaTempCor.h" #include "set"  typedef struct CandidateAnswer {     string text;    // text of candidate answers     string type;    // type of candidate answers (NE, NP, ...)     string NEType;  // NE type     string source;  // source (Content, Title, ...)     string sentenceText;    // text of the sentence     string docID;    // extracted document ID     string paraID;    // extracted paragraph ID     int sentenceID = -1;    // extracted sentence ID     int beginMorpID = -1;    // starting morpheme ID     int endMorpID = -1;    // last morpheme ID     int beginWordID = -1;    // starting word ID     int endWordID = -1;    // last word ID     LATScore LATScr;    // LAT similarity     SATScore SATScr;    // SAT similarity     SpaceScore spaceScr;    // space similarity     TimeScore timeScr;    // time similarity     LingScore lingScr;    // context similarity     string dependency;    // used in TE } CandidateAnswer;  typedef struct CandidateAnswerConfig {     bool SUPPORT_PASSAGE_SEARCH=false;     bool EXTRACT_NE=false;     bool EXTRACT_NP=false;     bool USE_LAT_SCORE=false;     bool USE_SAT_SCORE=false;     bool USE_SPACE_SCORE=false;     bool USE_TIME_SCORE=false;     bool USE_WORD_SCORE=false;     bool USE_WORD_BI_SCORE=false;     bool USE_WORD_ORDER_SCORE=false;     bool USE_SYN_SCORE=false;     bool VERBOSE_Q_LAT=false;     bool VERBOSE_Q_DEPENDENCY=false;     bool VERBOSE_DOCS_DEPENDENCY=false;     bool VERBOSE_SYN_SCORE=false; } CandidateAnswerConfig;  typedef pair<int,int> CandidateAnswerId;
| | +| Member Variables | | +| 
 DBSolution *dbSol; string dbPath; CGetNDoc getNDoc; TyCor tyCor; LingCor lingCor;
| | + +| | +|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| 
 SpaTempCor spaTempCor; CandidateAnswerConfig config;
| +| Member Functions | +| 
 CandidateGenerator(); bool initialize( string rscDir, DBSolution *dbSol, string dbPath, Searcher *kornetSearcher=NULL ); vector<CandidateAnswer> generateCandAns( QDecomp_Info& q, doc_result& dr ); static void printCandidateAnswers( vector<CandidateAnswer>& cands ); static void printCandidateAnswer( CandidateAnswer& cand, int index=-1 );
| + +| | +|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Class Name | +| AnswerHypo | +| Description | +| Extraction of answer candidate and hypothesis generation according to the questions | +| Include files | +| 
 #include "math.h"  #include "QAConfig.h" #include "AnswerType.h"  //=====Question analysis #include "QStruct.h" #include "QAnalHeader/QAnalyzer.h" #include "QAJsonHeader/QAnalJsonRW.h" //===== document search #include "PrimarySearch/PrimarySearch.h" #include "QAJsonHeader/DocResultParser.h" //===== candidate answer generation #include "CanGen/CanGenerator.h" #include "CI/TypeCor.h" /*===== KB search #include "KBInterface/KBInterface.h" //===== real time extraction of candidate answers #include "CanGen/CandidateGenerator.h" /**/ /*****/  //for multi-thread extern CQAnalRSCMng * qResourceMng; // extern Dictionary * gkornet; //lexical semantic concept network  extern map<int, string> *ExAtMap; extern map<string, int> *rExAtMap;  extern QAConfig QAConfig_core;
| +| Member Variables | +| 
 QAnal_Result Q; //question analysis result structure CQAnalJsonRW * QJsonRW; // question analyzer JsonRW  //Primary Search----- PrimarySearch * ps; DocResultParser *DocJsonRW; double DIR_top_weight; //anwer candidate extraction----- float CONF_CUT_OFF; CanGenerator * CanGen; TypeCor * CI;
| + +##### **Member Functions** + +``` + + //real time candidate answer extraction + int get_answerCandidate_fromIR(string Q_json, int subQ_idx, doc_result + lvResult, list & RES); + //subQ question merge + bool merge_answerCandidate_fromSubQ(vector + sub_AnsCan_vec, int subQ_idx, map & AnsUnit_map); + //pre_softfilter + int pre_soft_filter(CandidateAnswer & AnsCan, int subQ_idx); + +``` + +###### **Function Name** + +get\_answerCandidate\_fromIR + +Class Name + +AnswerHypo + +Description + +Candidate answer generation based on document search + +Syntax + +``` + +int get_answerCandidate_fromIR(string Q_json, int subQ_idx, doc_result +lvResult, list & RES); + +``` + +Return Value + +``` + +return int; // number of candidate answers +list RES : //candidate answer list basically ordered + +``` + +Parameters + +``` + +string input_Json //result json (with "||" division) +int subQ_idx: // processed question index (-1: original question) + +``` + +##### **Function Name** + +pre\_soft\_filter + +Class Name + +AnswerHypo + +Description + +Filtering function for the answer candidates as the doc search results + +Syntax + +``` + +int pre_soft_filter(CandidateAnswer & AnsCan, int subQ_idx); + +``` + +Return Value + +``` + +return int; //if filtering applied +0: not object for filtering +1: if more than 1, filtering should be applied + +``` + +Parameters + +``` + +CandidateAnswer & AnsCan // individual answer candidate +int subQ_idx: (processed question) index (-1: original question) + +``` + +### A.3 Class API for answer inference/generation + +This clause provides information for the class API for answer inference/generation. + +| | | +|------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Class Name | | +| AnswerConfidence | | +| Description | Among answer candidates from different answer generation module, to select the answer candidate with the highest reliability and to verify the answer to the question. | +| Include files | 
#if !defined(AFX_ANSWERRANK_H__C4797D53_04A5_4E4E_BE8F_841D4BC76032__INCLUDED_) #define AFX_ANSWERRANK_H__C4797D53_04A5_4E4E_BE8F_841D4BC76032__INCLUDED_  #if _MSC_VER > 1000 #pragma once #endif // _MSC_VER > 1000  #define INDI_MAX 30  #include "QAConfig.h"
| +| Member Variables | 
xmlrpc_c::clientSimple myClient;     /// Object pointer of QAConfiging for answer inference resource loading and     reliability model loading     QAConfig *qconfig;      Merge * AnsMerge;      /// answer filtering vector     vector<string> except_answer;      /// answer candidates integration exception rules     vector<string> except_is_redirect_rule;      /// answer filtering rule map     map<string, string> filter_map;      /// answer filtering rule map using LAT and SAT     map<string, string> latsat_filter_map;
| +| Member Functions | 
public:     AnswerRank();     virtual ~AnswerRank();     void ARank_THREAD_init();     string soft_filter_4UIMA(string json_in);     string answer_merge_4UIMA(string json_in); private:     xmlrpc_c::clientSimple myClient;     QAConfig *qconfig;     /// Merge object pointer for answer candidates integration     Merge * AnsMerge;     /// answer filtering vector     vector<string> except_answer;     /// space information NE vector     vector<string> space;     /// time information NE vector     vector<string> time;     map<string, string> filter_map;     map<string, string> latsat_filter_map;
| + +``` + + bool latsat_filter(string NE_Type, map & latsat_filter_map, +vector latVec, string first_SAT, double KB_score, bool q_type); + bool answer_filter(string NE_Type, map & filter_map, string +SAT_Type); + bool answer_filter_lat(string NE_Type, vector latVec, bool +q_type); + bool soft_filter_lat(string candidate, string NE_Type, string SAT_Type, +vector latVec, double KB_score, bool q_type, string first_SAT); + bool soft_filter_sat(string candidate, string NE_Type, string SAT_Type, +vector latVec, double KB_score, bool q_type, string first_SAT); + bool soft_filter_string(string candidate, string NE_Type, vector +latVec, string question, double KB_score, bool q_type); + bool update_resource(); + bool except_one_char_answer(string & answer, string & taggedQ, QAnal_Result +Q); + string make_feature(list::iterator lpAu, int rank); + double conf_wiseqa(int blank_size, string answer_type, int q_exam_size, +list::iterator lpAu, bool lat_flag, bool sat_flag, string q_type, int +rank, string SAT_CONF_TYPE); + double conf_all(int blank_size, string answer_type, int q_exam_size, +list::iterator lpAu, string q_type, int rank, string SAT_CONF_TYPE); + /** Reliability calculation */ + bool overlap_question(vector vec, string taggedQ, string question, +QAnal_Result Q); + /** final answer inference */ + int re_ranking_answerCandidate(list & RES, int mode, bool +select, int select_size, bool negation, bool lat_flag, bool sat_flag, string +qstring, string SAT_Type, QAnal_Result Q, string q_type, string SAT_CONF_TYPE, +string first_SAT); + bool exist_kb_AnswerUnit(list & RES, string json); + double thre_type(string conf_type, bool lat, bool sat); //threshold measure + +``` + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.746.12-202203-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.746.12-202203-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1e32f9b8f1141bed243c16e02a6cd1411cbb87f1 --- /dev/null +++ b/marked/F/T-REC-F.746.12-202203-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c395c4bbde83b1e232652a43f51538278c116f51f710c37e20a073f7333726c8 +size 6134 diff --git a/marked/F/T-REC-F.746.12-202203-I_PDF-E/410562339ce067fdc6fa41940c118658_img.jpg b/marked/F/T-REC-F.746.12-202203-I_PDF-E/410562339ce067fdc6fa41940c118658_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..e295910408dfb68378bffa4d5ee02c90074f2833 --- /dev/null +++ b/marked/F/T-REC-F.746.12-202203-I_PDF-E/410562339ce067fdc6fa41940c118658_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d70cfd1f4da1825453f5fac612bfab0627cb9bcb9d14d6ff1bb23cd8dc918108 +size 71330 diff --git a/marked/F/T-REC-F.746.12-202203-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg b/marked/F/T-REC-F.746.12-202203-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..414efe10f3029db162962acd2152e58c98bd8e12 --- /dev/null +++ b/marked/F/T-REC-F.746.12-202203-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2d3d08658e8a0ea92786136f7c1a5f977278eafa2a2de80d29063987496d9da9 +size 53306 diff --git a/marked/F/T-REC-F.746.12-202203-I_PDF-E/724c7777b608e53be38b12b6fb3c43bc_img.jpg b/marked/F/T-REC-F.746.12-202203-I_PDF-E/724c7777b608e53be38b12b6fb3c43bc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..79d5003c5ed90bba89982c8e982b8ab3fb78949c --- /dev/null +++ b/marked/F/T-REC-F.746.12-202203-I_PDF-E/724c7777b608e53be38b12b6fb3c43bc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fe8f544fbf36e91bb7c7fd764446ca6557c3dcd940d06d9f27bd191b9f56c4be +size 90671 diff --git a/marked/F/T-REC-F.746.12-202203-I_PDF-E/83926108dd52e3998fda7a99a91cdf3b_img.jpg b/marked/F/T-REC-F.746.12-202203-I_PDF-E/83926108dd52e3998fda7a99a91cdf3b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..dadb391dcf1885aa2cc38e17d81e88fe4f2d8c22 --- /dev/null +++ b/marked/F/T-REC-F.746.12-202203-I_PDF-E/83926108dd52e3998fda7a99a91cdf3b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3576cf8db2694ddb29dd3217ac86a983aee96d08c47d563fde034709d582d8f1 +size 41327 diff --git a/marked/F/T-REC-F.746.12-202203-I_PDF-E/9ae17964ddd9b814c7d905b1af2fddf2_img.jpg b/marked/F/T-REC-F.746.12-202203-I_PDF-E/9ae17964ddd9b814c7d905b1af2fddf2_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a007080da70f75f5a14d1061db0c60d410490ab8 --- /dev/null +++ b/marked/F/T-REC-F.746.12-202203-I_PDF-E/9ae17964ddd9b814c7d905b1af2fddf2_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9953359cc76ed987c1ff584b8a553a6e1910d24fb9637be69ce5b6809e598fe2 +size 44449 diff --git a/marked/F/T-REC-F.746.12-202203-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg b/marked/F/T-REC-F.746.12-202203-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..4dbefbc48e84240c20a444f8e8b439f41e8c4a77 --- /dev/null +++ b/marked/F/T-REC-F.746.12-202203-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:474b495b4d7267dbe37c9debe44f0d111e738d0abcdd8dd917a901236f25c5c9 +size 89548 diff --git a/marked/F/T-REC-F.746.12-202203-I_PDF-E/bd671b21db63e6fdb2196e9b18502aac_img.jpg b/marked/F/T-REC-F.746.12-202203-I_PDF-E/bd671b21db63e6fdb2196e9b18502aac_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ab93d13dac7961f3f78a2891de158143058215ec --- /dev/null +++ b/marked/F/T-REC-F.746.12-202203-I_PDF-E/bd671b21db63e6fdb2196e9b18502aac_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:beec4a4669915b57ac7ecafd05a253f9c6a7e7782245f42d2821b7c031e5faa3 +size 49692 diff --git a/marked/F/T-REC-F.746.12-202203-I_PDF-E/raw.md b/marked/F/T-REC-F.746.12-202203-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..452f72ced8cabed3399678be76077572ee049bf0 --- /dev/null +++ b/marked/F/T-REC-F.746.12-202203-I_PDF-E/raw.md @@ -0,0 +1,521 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.746.12** + +(03/2022) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Requirements for a real-time interactive +multimedia service under poor network +conditions** + +Recommendation ITU-T F.746.12 + +ITU-T + +![ITU logo](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with a white grid pattern, overlaid by a stylized 'ITU' text in blue. + +ITU logo + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.746.12 + +# Requirements for a real-time interactive multimedia service under poor network conditions + +## Summary + +Recommendation ITU-T F.746.12 describes the scenarios, general framework and requirements for a real-time interactive multimedia service (RIMS) under poor network conditions. The RIMS system plays an important role in many scenarios and situations, e.g., emergency relief, remote education and emergency communication. The RIMS requires provision of measures for dynamic adjustment of coding parameters, including those for video and audio coding, and it requires setting maximal priority of audio communication under low-speed network conditions and configuring usage attributes to ensure the appropriate priority from high to low of service uses. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.746.12 | 2022-03-16 | 16 | 11.1002/1000/14959 | + +## Keywords + +Multimedia service, poor network conditions, real-time. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2022 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|------------|--------------------------------------------------------------------------------------|----| +| 1 | Scope ..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 1 | +| 4 | Abbreviations..... | 2 | +| 5 | Convention..... | 2 | +| 6 | Overview ..... | 2 | +| 7 | Key components of a real-time interactive multimedia service system ..... | 3 | +| 8 | Requirements for real-time interactive multimedia service system ..... | 4 | +| 8.1 | General requirements..... | 4 | +| 8.2 | Requirements for application service unit..... | 4 | +| 8.3 | Requirements for application client..... | 5 | +| 8.4 | Requirements for call control service unit..... | 5 | +| 8.5 | Requirements for media processing service unit..... | 5 | +| 8.6 | Requirements for media storage service unit ..... | 5 | +| 8.7 | Requirements for directory server ..... | 5 | +| 8.8 | Requirements for encoder..... | 6 | +| 8.9 | Requirements for decoder..... | 6 | +| 9 | Security considerations..... | 6 | +| Appendix I | – Example use cases of real-time interactive multimedia service ..... | 7 | +| I.1 | Real-time interactive multimedia service in highly rural areas..... | 7 | +| I.2 | Real-time interactive multimedia service in maritime communication ..... | 7 | +| I.3 | Real-time interactive multimedia service in disaster relief ..... | 8 | +| I.4 | Real-time interactive multimedia service for managing critical events
safety..... | 9 | +| I.5 | Real-time interactive multimedia service in emergency rescue ..... | 9 | +| I.6 | Real-time interactive multimedia service in remote technical support ..... | 10 | +| | Bibliography..... | 11 | + + + +# Recommendation ITU-T F.746.12 + +# Requirements for a real-time interactive multimedia service under poor network conditions + +## 1 Scope + +This Recommendation specifies requirements for the framework to provide real-time interactive multimedia service (RIMS) under poor network conditions, and corresponding application scenarios. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +- [ITU-T X.500] Recommendation ITU-T X.500 (2019), *Information technology – Open Systems Interconnection – The Directory: Overview of concepts, models and services.* +- [ITU-T X.511] Recommendation ITU-T X.511 (2019), *Information technology – Open Systems Interconnection – The Directory: Abstract service definition.* +- [ITU-T Y.2201] Recommendation ITU-T Y.2201 (2009), *Requirements and capabilities for ITU-T NGN.* +- [ITU-T Y.2701] Recommendation ITU-T Y.2701 (2007), *Security requirements for NGN release 1.* + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 application service** [b-ITU-T H.360]: A network-based service involving the transmission and/or processing of multimedia information. + +**3.1.2 call control** [b-ITU-T G.8081]: Call control is a signalling association between one or more user applications and the network to control the set-up, release, modification and maintenance of sets of connections. + +**3.1.3 directory server** [b-ITU-T X.518]: A DSA or an LDAP server. + +**3.1.4 decoder** [b-ITU-T H.265]: An embodiment of a decoding process. + +**3.1.5 encoder** [b-ITU-T H.265]: An embodiment of an encoding process. + +**3.1.6 multimedia service** [b-ITU-T Q.1743]: Services that handle several types of media such as audio and video in a synchronized way from the user's point of view. A multimedia service may involve multiple parties, multiple connections, and the addition or deletion of resources and users within a single communication session. + +### 3.2 Terms defined in this Recommendation + +None. + +## 4 Abbreviations + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|------------------------------------------| +| AC | Application Client | +| ASU | Application Service Unit | +| CCSU | Call Control Service Unit | +| DS | Directory Server | +| DSA | Directory System Agent | +| FEC | Forward Error Correction | +| LDAP | Lightweight Directory Access Protocol | +| MPSU | Media Processing Service Unit | +| MSSU | Media Storage Service Unit | +| RIMS | Real-time Interactive Multimedia Service | +| SAN | Storage Area Networking | +| VACU | Video and Audio Capture Unit | +| VARU | Video and Audio Render Unit | +| VSAT | Very Small Aperture Terminal | + +## 5 Convention + +In this Recommendation: + +- The phrase "is required to" indicates a requirement that must be strictly followed and from which no deviation is permitted if conformity to this Recommendation is to be claimed. +- The phrase "is recommended" indicates a requirement that is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformity. +- The phrase "can optionally" indicates an optional requirement that is permissible, without implying any sense of being recommended. This term is not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator or service provider. Rather, it means the vendor may optionally provide the feature and still claim conformity with the specification. + +## 6 Overview + +Real-time multimedia service plays an increasingly important role in many scenarios and situations with poor network conditions, e.g., emergency relief, medical emergency, remote education, emergency communication, maritime communication, remote technical support and vehicle communication [b-Martini] [b-Tsou]. + +In most of these scenarios, the communications infrastructure is broken (in disaster scenarios) or even non-existent (in remote rural areas). Thus, the underlying network conditions are very poor. Packets often suffer a high bit error rate (burst packet loss), significant delay jitter or long transmission delay. The wireless environment changes rapidly, and the transmission power of nodes is often adjusted, which leads to frequent changes of network topology [b-Ni]. + +There are several methods deployed, such as flexible macroblock ordering [b-ITU-T H.264], slice segment and key frames re-synchronization, to provide resilient multimedia services over the Internet. However, under poor network conditions, the transmission issues previously mentioned cannot be solved by these methods [b-Wang] [b-Zhang]. + +Forward error correction (FEC) performs well with random error, but cannot deal with burst errors. In the situations previously mentioned, encoding every packet with FEC will overburden the network. An automatic repeat request sends data repeatedly when the network condition goes badly and makes congestion worse. When some links are instable or with high latency (> 200 ms), e.g., satellite links, the general signalling control mechanism, such as that specified in [b-ITU-T H.323], may lead to frequent re-connection. + +To overcome these issues, many vendors have proposed proprietary solutions to provide better performance, which leads to the problem of interoperability [b-Nguyen]. This Recommendation focuses on the requirements of RIMS under poor network conditions. + +## 7 Key components of a real-time interactive multimedia service system + +This clause presents a reference framework for a RIMS system, which is shown in Figure 7-1. It consists of an application service unit (ASU), application client (AC), call control service unit (CCSU), media processing service unit (MPSU), media storage service unit (MSSU), directory server (DS), encoder, decoder, video and audio capture unit (VACU) and a video and audio render unit (VARU). + +Unless explicitly stated in the text, the framework depicted in the figures and functions in this Recommendation are logical and thus do not necessarily represent a particular physical implementation. + +The ASU, CCSU, MPSU, MSSU and DS are deployed in a RIMS service node. The ASU provides specific application service(s) to RIMS users. Its main functions include access registration, access authentication, identification, authorization, presence, business processing, information management and target media serving function selection. The CCSU is a signalling association unit between service node equipment and user equipment to control the set-up, release, modification and maintenance of sets of connections. The MPSU provides video and audio processing services. Its main functions include media reception, media mixing, media transcoding, media routing, media transmission, media forwarding and media replication. The MSSU is used to retrieve, store media and provide media serving capability. Its main functions include media storage, media serving, media indexing and media downloading. The DS is an entity that stores information about objects as records and provides records to other units. It can return directory information when receiving a request over the directory access protocol or lightweight directory access protocol (LDAP). + +RIMS user equipment consists of five units: AC; encoder; decoder; VACU; and VARU. The AC is an agent that initiates requests and receives ASU responses. The encoder is an embodiment of an encoding process. The decoder is an embodiment of a decoding process. The VACU captures video and audio media streams, such as a camera or microphone. The VARU feeds media streams to devices, such as a screen or loudspeaker. + +![Figure 7-1: Reference framework for real-time interactive multimedia service system. The diagram shows the interaction between RIMS user equipment and RIMS service node equipment. RIMS user equipment (yellow box) contains an Application client, Video and audio capture unit, Encoder, Decoder, and Video and audio render unit. RIMS service node equipment (green box) contains an Application service unit, Call control service unit, Media processing service unit, Media storage service unit, and Directory server. Arrows indicate the flow of User commands, Call control signalling, and Encoded media packets. A legend at the bottom defines the packet types and equipment types.](a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg) + +The diagram illustrates the reference framework for a real-time interactive multimedia service system. It is divided into two main components: RIMS user equipment and RIMS service node equipment. + +**RIMS user equipment (Yellow Box):** + +- Application client +- Video and audio capture unit +- Encoder +- Decoder +- Video and audio render unit + +**RIMS service node equipment (Green Box):** + +- Application service unit +- Call control service unit +- Media processing service unit +- Media storage service unit +- Directory server + +**Interactions:** + +- User commands: From Application client to Application service unit. +- Call control signalling: Bidirectional between Application client and Call control service unit. +- Encoded media packets: From Encoder to Media processing service unit, and from Media processing service unit to Decoder. +- Directory packets (dashed lines): Between Application service unit and Directory server, and between Media storage service unit and Directory server. + +**Legend:** + +- Signalling packet: Thin solid line +- Media packet: Thick solid line +- Directory packet: Dashed line +- User equipment: Yellow box +- Service node equipment: Green box + +F.746.12(22) + +Figure 7-1: Reference framework for real-time interactive multimedia service system. The diagram shows the interaction between RIMS user equipment and RIMS service node equipment. RIMS user equipment (yellow box) contains an Application client, Video and audio capture unit, Encoder, Decoder, and Video and audio render unit. RIMS service node equipment (green box) contains an Application service unit, Call control service unit, Media processing service unit, Media storage service unit, and Directory server. Arrows indicate the flow of User commands, Call control signalling, and Encoded media packets. A legend at the bottom defines the packet types and equipment types. + +**Figure 7-1 – Reference framework for real-time interactive multimedia service system** + +# 8 Requirements for real-time interactive multimedia service system + +This clause presents the requirements for a RIMS system. + +### 8.1 General requirements + +GEN-01: The system is required to configure routing policies between different service nodes directly, based on emergency communication requirements, to realize fast-forwarding of signalling and media packets. + +GEN-02: The system is required to provide mechanisms for adjusting video coding parameters, including video resolution, frame rate, bit rate and network abstract layer (NAL) size. + +GEN-03: The system is required to provide mechanisms for adjusting audio coding parameters including audio sampling frequency, sample accuracy, bit rate and sound channel numbers. + +GEN-04: The system can optionally provide protecting mechanisms for important information including video sequence head, image frame head, strip head, motion vector information and I-frame. + +GEN-05: The system is required to set maximal priority of audio communication under low-speed network conditions. + +GEN-06: The system can optionally deliver media packets using the network multicast protocol. + +GEN-07: The system is required to ensure that high-priority uses have priority of service over low-priority uses. + +### 8.2 Requirements for application service unit + +ASU-01: ASU is required to support the authentication of users and devices. + +ASU-02: ASU is required to support the authorization of allowed uses. + +ASU-03: ASU is required to respond to AC operation requests and convert them into commands for other functional units. + +ASU-04: ASU is required to provide business processing functions including video calling, video conferencing and video surveillance. + +ASU-05: ASU is required to collect information about system status including user device online status and session status. + +### **8.3 Requirements for application client** + +AC-01: AC is required to provide services configuration function. + +AC-02: AC is required to provide system status display function. + +AC-03: AC is required to provide system configuration function. + +AC-04: AC is recommended to provide shortcuts for control services. + +### **8.4 Requirements for call control service unit** + +CCSU-01: CCSU is required to negotiate the media parameters for the two parties during a session establishment stage. + +CCSU-02: CCSU is recommended to ensure effective connection of signalling in a network environment with long time delay and frequent interruptions. + +CCSU-03: CCSU is required to control the creation or closure of media transfer channels. + +CCSU-04: CCSU is recommended to update I-frame using a signalling protocol to speed up the synchronization of I-frames. + +CCSU-05: CCSU can optionally configure the interval of I-frames using a signalling protocol. + +### **8.5 Requirements for media processing service unit** + +MPSU-01: MPSU is required to provide media delivery. + +MPSU-02: MPSU is required to provide video compositing. + +MPSU-03: MPSU is required to provide audio mixing. + +MPSU-04: MPSU is required to provide media transcoding. + +MPSU-05: MPSU is recommended to support constant bit rate flow control to avoid bursts-induced traffic congestion. + +MPSU-06: MPSU can optionally smooth traffic through setting cache size, I-frame interval, I-frame size and key message retransmission frequency. + +MPSU-07: MPSU is required to unify the transport packaging format and related parameters including package length and timestamp. + +### **8.6 Requirements for media storage service unit** + +MSSU-01: MSSU is required to support media storage and streaming. + +MSSU-02: MSSU can optionally receive media packets through network storage facilities including storage area networking and network attached storage. + +### **8.7 Requirements for directory server** + +DS-01: DS is required to support storage and synchronization of static information. + +DS-02: DS is required to synchronize static information with an ASU. + +DS-03: Each request may be accompanied by information in support of security mechanisms for protecting directory information. These definitions are contained in [ITU-T X.500]. + +DS-04: The DS is required to support interrogation, retrieval and modification of directory information. This service is specified in [ITU-T X.511]. + +### **8.8 Requirements for encoder** + +En-01: Encoder is required to implement media coding according to the strategy that carries out coding adjustment with respect to the network state feedback information. + +En-02: Encoder is recommended to provide type identifiers for its underlying channels (e.g., fixed network, satellite network and mobile network) in the registration message. + +En-03: Encoder is recommended to contain the description of coding capability in the registration message to improve negotiating efficiency. + +En-04: Encoder is recommended to support constant bit rate flow control to bursts-induced traffic congestion. + +En-05: Encoder can optionally assist in managing sending traffic load through setting cache size, I-frame interval, I-frame size, key message retransmission frequency, etc. + +En-06: Encoder is required to unify the transport packaging format and related parameters including package length and timestamp. + +### **8.9 Requirements for decoder** + +De-01: Decoder is required to implement media decoding according to the strategy. + +De-02: Decoder is required to support adaptive decoding function, within the capacity range of media negotiation, to automatically adapt to media flow. + +De-03: Decoder can optionally provide adjusting mechanisms of buffer threshold to adapt and eliminate out-of-order packets, duplication and jitter in network transmission, and to ensure the synchronization of video and audio media. + +## **9 Security considerations** + +It is recommended that the security requirements of [ITU-T Y.2201], [ITU-T Y.2701], and applicable ITU-T X, ITU-T Y and ITU-T M series security Recommendations be taken into consideration, including access control, authentication, data confidentiality, communications security, data integrity, availability and privacy. + +## Appendix I + +## Example use cases of real-time interactive multimedia service + +(This appendix does not form an integral part of this Recommendation.) + +### I.1 Real-time interactive multimedia service in highly rural areas + +Figure I.1 shows that, given their poor communication infrastructures, satellites can provide worldwide Internet access for some highly rural areas. However, the long latency associated with satellite channels ( $> 200$ ms) often triggers session timeout. RIMSs, such as education and medical diagnosis delivered remotely, are expected to be tolerant of long latency. This is one RIMS scenario. + +![Diagram of real-time interactive multimedia service in highly rural areas](410562339ce067fdc6fa41940c118658_img.jpg) + +The diagram illustrates a network architecture for real-time interactive multimedia services in rural areas. At the top center is a 'Satellite'. It has wireless links (yellow lightning bolts) to a 'Satellite master station' and a 'VSAT'. The 'Satellite master station' is connected via 'Cable' to a 'Fixed network' cloud. This cloud connects to a 'RIMS user' and a 'RIMS service node' (containing four server icons) via 'Fixed network' lines. The 'VSAT' is connected via 'Cable' to a 'Communication station' tower. This tower has a wireless link to a 'RIMS user' and another wireless link to a 'Communication vehicle' (a truck). The 'Communication vehicle' further connects wirelessly to another 'RIMS user'. All rural 'RIMS user' icons are enclosed in dashed boxes. The bottom right area is labeled 'Rural areas' and includes the code 'F.746.12(22)'. + +Diagram of real-time interactive multimedia service in highly rural areas + +**Figure I.1 – Real-time interactive multimedia service in highly rural areas** + +### I.2 Real-time interactive multimedia service in maritime communication + +Figure I.2 shows that, during an ocean voyage, uses on ships mainly rely on satellite and other wireless means to communicate with remote users in real-time multimedia. In this scenario, the channel state, such as end-to-end delay and the available bandwidth, changes dynamically. RIMS system can adapt to the coding strategy of the sender (e.g., resolution, frame rate, code rate and NAL size) and the flow shaping strategy (e.g., I-frame interval and cache setting) according to the changing status of the channel. + +![Diagram of real-time interactive multimedia service in maritime communication. A RIMS user on land is connected via a fixed network to a cloud labeled 'Fixed network'. This cloud is connected to another 'Fixed network' cloud, which in turn connects to a 'RIMS service node' (four server icons). The second 'Fixed network' cloud is also connected to a 'Satellite master station' (ground station). The 'Satellite master station' has a bidirectional link to a 'Satellite' in space. The 'Satellite' has a bidirectional link to a ship at sea. The ship is connected to another 'RIMS user' on board. The diagram is labeled F.746.12(22).](4801720824e4b5e2361a5564f91cfb70_img.jpg) + +Diagram of real-time interactive multimedia service in maritime communication. A RIMS user on land is connected via a fixed network to a cloud labeled 'Fixed network'. This cloud is connected to another 'Fixed network' cloud, which in turn connects to a 'RIMS service node' (four server icons). The second 'Fixed network' cloud is also connected to a 'Satellite master station' (ground station). The 'Satellite master station' has a bidirectional link to a 'Satellite' in space. The 'Satellite' has a bidirectional link to a ship at sea. The ship is connected to another 'RIMS user' on board. The diagram is labeled F.746.12(22). + +**Figure I.2 – Real-time interactive multimedia service in maritime communication** + +### I.3 Real-time interactive multimedia service in disaster relief + +RIMS during disaster relief operations is crucial. In this scenario, the communication infrastructure is broken. As shown in Figure I.3, temporary *ad hoc* networks are set up to provide local network connections, and satellite networks can provide remote connections. The RIMS system, which is on top of these networks, can provide interactive multimedia services for disaster medicine and disaster management. However, the quality of wireless links is poor and varies rapidly. The network topology is also unstable. The service is expected to perform at its best in these hostile conditions with limited resources. + +![Diagram of real-time interactive multimedia service in disaster relief. A RIMS user on land is connected via a fixed network to a 'Satellite master station' (ground station). The 'Satellite master station' has a bidirectional link to a 'Satellite' in space. The 'Satellite' has a bidirectional link to a 'Local network' (dashed box). Inside the 'Local network', there are 'On-board equipment' (a car), 'Unmanned aerial vehicles' (a drone), a 'Helicopter', and three 'User' icons. All these components are interconnected by red dashed double-headed arrows, forming a mesh network. The diagram is labeled F.746.12(22).](bd671b21db63e6fdb2196e9b18502aac_img.jpg) + +Diagram of real-time interactive multimedia service in disaster relief. A RIMS user on land is connected via a fixed network to a 'Satellite master station' (ground station). The 'Satellite master station' has a bidirectional link to a 'Satellite' in space. The 'Satellite' has a bidirectional link to a 'Local network' (dashed box). Inside the 'Local network', there are 'On-board equipment' (a car), 'Unmanned aerial vehicles' (a drone), a 'Helicopter', and three 'User' icons. All these components are interconnected by red dashed double-headed arrows, forming a mesh network. The diagram is labeled F.746.12(22). + +**Figure I.3 – Real-time interactive multimedia service in disaster relief** + +### I.4 Real-time interactive multimedia service for managing critical events safety + +A large number of terminals need to access the network in some major events, such as celebrations, conferences and exhibitions. Traffic demands can easily overwhelm the capacities of mobile networks. Figure I.4 shows that the RIMS system can be deployed to provide real-time monitoring and management of critical parts of these activities. + +![Diagram of RIMS system for managing critical events safety.](724c7777b608e53be38b12b6fb3c43bc_img.jpg) + +This diagram illustrates a real-time interactive multimedia service for managing critical events safety. It features a central 'Headquarters' unit, represented by a small inset image of a control room, connected by red dashed lines to four 'RIMS user' units. Each 'RIMS user' is depicted with a person icon. The background is an aerial view of a large stadium or arena with surrounding infrastructure, showing the deployment of the system for event management. The text 'F.746.12(22)' is visible in the bottom right corner. + +Diagram of RIMS system for managing critical events safety. + +Figure I.4 – Real-time interactive multimedia service for managing critical events safety + +### I.5 Real-time interactive multimedia service in emergency rescue + +Visual command and remote diagnostics are often needed for emergency rescue in underground situations such as mass transit railways, garages and tunnels. Figure I.5 shows the application of a RIMS system to an underground railway scenario. + +![Diagram of RIMS system in an underground railway scenario.](9ae17964ddd9b814c7d905b1af2fddf2_img.jpg) + +This diagram shows the application of a RIMS system in an underground railway scenario. A 'Headquarters' unit, shown in an inset, is connected by blue dashed lines to three 'RIMS user' units. The 'RIMS user' units are positioned at different levels of the railway: one on the surface, one on an escalator, and one inside a tunnel. Each 'RIMS user' is represented by a person icon. The background depicts a 3D model of a subway station and tunnel with trains, illustrating the system's use in emergency rescue operations. The text 'F.746.12(22)' is visible in the bottom right corner. + +Diagram of RIMS system in an underground railway scenario. + +Figure I.5 – Real-time interactive multimedia service in emergency rescue + +### I.6 Real-time interactive multimedia service in remote technical support + +To conduct tasks in highly dangerous situations, such as offshore platform operation, high-voltage cable inspection and fire-fighting operation, high-definition image backhaul and communication provided by RIMS system in real time are critical for technical support. Figure I.6 shows this scenario. + +![Diagram illustrating the RIMS system for remote technical support. A worker on an offshore platform (RIMS user) is shown in a circular inset, with an orange arrow pointing from the worker to a satellite in space. Another orange arrow points from the satellite to a person at a computer workstation (Headquarters). The background shows an offshore oil rig and a cloudy sky.](83926108dd52e3998fda7a99a91cdf3b_img.jpg) + +RIMS user + +Headquarters + +F.746.12(22) + +Diagram illustrating the RIMS system for remote technical support. A worker on an offshore platform (RIMS user) is shown in a circular inset, with an orange arrow pointing from the worker to a satellite in space. Another orange arrow points from the satellite to a person at a computer workstation (Headquarters). The background shows an offshore oil rig and a cloudy sky. + +**Figure I.6 – Real-time interactive multimedia service in remote technical support** + +## Bibliography + +- [b-ITU-T G.8081] Recommendation ITU-T G.8081/Y.1353 (2012), *Terms and definitions for automatically switched optical networks*. +- [b-ITU-T H.264] Recommendation ITU-T H.264 (2021), *Advanced video coding for generic audiovisual services*. +- [b-ITU-T H.265] Recommendation ITU-T H.265 (2021), *High efficiency video coding*. +- [b-ITU-T H.323] Recommendation ITU-T H.323 (2022), *Packet-based multimedia communications systems*. +- [b-ITU-T H.360] Recommendation ITU-T H.360 (2004), *An architecture for end-to-end QoS control and signalling*. +- [b-ITU-T Q.1743] Recommendation ITU-T Q.1743 (2016), *IMT-Advanced references to Release 11 of LTE-Advanced evolved packet core network*. +- [b-ITU-T X.518] Recommendation ITU-T X.518 (2019), *Information technology – Open Systems Interconnection – The Directory: Procedures for distributed operation*. +- [b-Martini] Martini, M.G., Iacobelli, L., Bergeron, C., Hewage, C.T., Panza, G., Piri, E., Vehkapera, J., Amon, P., Mazzotti, M., Savino, K., Bokor, L. (2015). Real-time multimedia communications in medical emergency-the CONCERTO project solution. In: *Proc. 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society*, pp. 7324-7. doi: 10.1109/EMBC.2015.7320083 +- [b-Nguyen] Nguyen, T., Yegenoglu, F., Sciuto, A. Subbarayan, R. (2001). Voice over IP service and performance in satellite networks. *IEEE Commun. Mag.* **39**(3), pp. 164-171. doi: 10.1109/35.910604 +- [b-Ni] Ni, Q., Li, T., Turletti, T., Xiao, Y. (2005). Saturation throughput analysis of error-prone 802.11 wireless networks. *Wireless Commun. Mobile Comput.* **5**(8), pp. 945-956. +- [b-Tsou] Tsou, M.H., Jung, C.T., Allen, C., Yang, J.A., Han, S.Y., Spitzberg, B.H., Dozier, J. (2017). Building a real-time geo-targeted event observation (geo) viewer for disaster management and situation awareness. In: Peterson, M. (eds). *Advances in Cartography and GIScience. ICACI 2017. Lecture Notes in Geoinformation and Cartography*. Berlin: Springer Nature. doi: 10.1007/978-3-319-57336-6\_7 +- [b-Wang] Wang, Y., Jing, X., Tham, J.Y., Goh, K.H., Lee, W.S. (2013). Error-resilient packet reordering for compressed video transmission over error-prone networks. In: *Proc. 7th International Conference on Ubiquitous Information Management and Communication (ICUIMC '13)*, Article 28, 1–6. New York, NY: Association for Computing Machinery. doi: 10.1145/2448556.2448584 +- [b-Zhang] Zhang, Y., Gao, W., Lu, Y., Huang, Q., Zhao, D. (2007). Joint source-channel rate-distortion optimization for H. 264 video coding over error-prone networks. *IEEE Trans. Multimedia* **9**(3), pp. 445-454. doi: 10.1109/TMM.2006.887989. + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.746.13-202203-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.746.13-202203-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7fc0ce00090f66b87404c12debf099b47a838ec0 --- /dev/null +++ b/marked/F/T-REC-F.746.13-202203-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0efd3ea366c9c151bbcfa01a99d20b4bfbdb579443e1a9dd88122c3e286dae51 +size 6108 diff --git a/marked/F/T-REC-F.746.13-202203-I_PDF-E/a234352dfaccdc24745c88eef7724cc6_img.jpg b/marked/F/T-REC-F.746.13-202203-I_PDF-E/a234352dfaccdc24745c88eef7724cc6_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..0c86ca04d800215a96e16aa6ee19e9dc777b7ea8 --- /dev/null +++ b/marked/F/T-REC-F.746.13-202203-I_PDF-E/a234352dfaccdc24745c88eef7724cc6_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c9f55c38d51cb9b76ef6a3abc1efcac43df701256bf4b1c44c6ac87a357bf5fe +size 46146 diff --git a/marked/F/T-REC-F.746.13-202203-I_PDF-E/d4af765160d04ecef538e5066006dc77_img.jpg b/marked/F/T-REC-F.746.13-202203-I_PDF-E/d4af765160d04ecef538e5066006dc77_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..78046cfdd13d94e32d6aaaa6d40cec6a751074ad --- /dev/null +++ b/marked/F/T-REC-F.746.13-202203-I_PDF-E/d4af765160d04ecef538e5066006dc77_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0443c8a1a56946b64cc87d8279c8b24f2b7064dc96ece9726b17bf9bb7bff268 +size 35518 diff --git a/marked/F/T-REC-F.746.14-202212-I_PDF-E/367926125450c2bc3f4bdca9d59a62ba_img.jpg b/marked/F/T-REC-F.746.14-202212-I_PDF-E/367926125450c2bc3f4bdca9d59a62ba_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..227c810d0612ed82614efa4a8e2a2f10674c916a --- /dev/null +++ b/marked/F/T-REC-F.746.14-202212-I_PDF-E/367926125450c2bc3f4bdca9d59a62ba_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2affdb262aba8670cc49d538bd771aa932da1f4bb93ef7ff995894853c44e5e2 +size 60772 diff --git a/marked/F/T-REC-F.746.14-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.746.14-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ac295b1b0c7f14b1c5a33ef6682b1b017c812477 --- /dev/null +++ b/marked/F/T-REC-F.746.14-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37aed4fc114a39843438175326e0b7e1d09a9d6075723d82fea32040fd560ee4 +size 7192 diff --git a/marked/F/T-REC-F.746.14-202212-I_PDF-E/8fbdfc3d17fb1dae7b2d8f5a287fa9fc_img.jpg b/marked/F/T-REC-F.746.14-202212-I_PDF-E/8fbdfc3d17fb1dae7b2d8f5a287fa9fc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d94066b989fc17294c0601c2ed807225129f7d83 --- /dev/null +++ b/marked/F/T-REC-F.746.14-202212-I_PDF-E/8fbdfc3d17fb1dae7b2d8f5a287fa9fc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:271c62d9655ab66c46da237b0aaa15019a326a911223ecf4ed73ef7d579ba68b +size 68711 diff --git a/marked/F/T-REC-F.746.14-202212-I_PDF-E/fa859e4e468bfb2710a94527f2c504af_img.jpg b/marked/F/T-REC-F.746.14-202212-I_PDF-E/fa859e4e468bfb2710a94527f2c504af_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3cfd5f107664560eb25e35a6343b1451493338d9 --- /dev/null +++ b/marked/F/T-REC-F.746.14-202212-I_PDF-E/fa859e4e468bfb2710a94527f2c504af_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d373cb042bfe77cd53a32110151dfc604f45b9393fd5d9face19dd3bb3d159f4 +size 67659 diff --git a/marked/F/T-REC-F.746.14-202212-I_PDF-E/ff7ce44f3fdd51bae7b231f34df07c6a_img.jpg b/marked/F/T-REC-F.746.14-202212-I_PDF-E/ff7ce44f3fdd51bae7b231f34df07c6a_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..0c6f99bc66bd5378828af944aa38d7118da56d19 --- /dev/null +++ b/marked/F/T-REC-F.746.14-202212-I_PDF-E/ff7ce44f3fdd51bae7b231f34df07c6a_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:793ca3398b6a027b5e27f8ec31c3445ed463e6cab2ec606d30e27b71607e0cf1 +size 92416 diff --git a/marked/F/T-REC-F.746.15-202212-I_PDF-E/04b5df354878bcb2868795f8dbdf2726_img.jpg b/marked/F/T-REC-F.746.15-202212-I_PDF-E/04b5df354878bcb2868795f8dbdf2726_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7414c91b8f6cb277093dd1813a10e956b3c15d6f --- /dev/null +++ b/marked/F/T-REC-F.746.15-202212-I_PDF-E/04b5df354878bcb2868795f8dbdf2726_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:34bcb4a7be7670e7e639c2a10e8470b51019cdfa3be9b38effaf0cef9e085ee7 +size 38082 diff --git a/marked/F/T-REC-F.746.15-202212-I_PDF-E/16c1175b5f05a4b55e6d396fc51b15b3_img.jpg b/marked/F/T-REC-F.746.15-202212-I_PDF-E/16c1175b5f05a4b55e6d396fc51b15b3_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..309a3e0806c13a8d87ac43a892504178b24d285d --- /dev/null +++ b/marked/F/T-REC-F.746.15-202212-I_PDF-E/16c1175b5f05a4b55e6d396fc51b15b3_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:86ae08432b1fed7881d0484eaa76d042ca01589644a2394ea8bc1e6973b60df9 +size 68083 diff --git a/marked/F/T-REC-F.746.15-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.746.15-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..0f282db7fd0b6dc7ae4ea2192bbbaf302c4affd1 --- /dev/null +++ b/marked/F/T-REC-F.746.15-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:68ed066d3c785d7afb41ef19c55b89df8664472d028ae48a071949b39c0a1142 +size 7317 diff --git a/marked/F/T-REC-F.746.15-202212-I_PDF-E/8fbdfc3d17fb1dae7b2d8f5a287fa9fc_img.jpg b/marked/F/T-REC-F.746.15-202212-I_PDF-E/8fbdfc3d17fb1dae7b2d8f5a287fa9fc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..503a2758beeeffd45b90cd43817956f842811c89 --- /dev/null +++ b/marked/F/T-REC-F.746.15-202212-I_PDF-E/8fbdfc3d17fb1dae7b2d8f5a287fa9fc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5101a43fa0e7c860676721230a03e1078a25d35203406bae2b41e2b94394eef3 +size 81742 diff --git a/marked/F/T-REC-F.746.15-202212-I_PDF-E/a234352dfaccdc24745c88eef7724cc6_img.jpg b/marked/F/T-REC-F.746.15-202212-I_PDF-E/a234352dfaccdc24745c88eef7724cc6_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..570ab5ef34873d2533dff95dbe837f97821b7ebf --- /dev/null +++ b/marked/F/T-REC-F.746.15-202212-I_PDF-E/a234352dfaccdc24745c88eef7724cc6_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a3abf8446ed6cf6c88d29de058a69d8a972fec412fb72c394bb0ed08902b6d4e +size 45743 diff --git a/marked/F/T-REC-F.746.15-202212-I_PDF-E/b03533cc9198ad21d619252486969839_img.jpg b/marked/F/T-REC-F.746.15-202212-I_PDF-E/b03533cc9198ad21d619252486969839_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f7741805732ae592bbe7b4a439dd44f258994c69 --- /dev/null +++ b/marked/F/T-REC-F.746.15-202212-I_PDF-E/b03533cc9198ad21d619252486969839_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1350590641dddc91fdbd768398fc4b4bd00ea9c66da3187deda9ae5da9c68080 +size 50983 diff --git a/marked/F/T-REC-F.746.15-202212-I_PDF-E/cfda9df1319e04207eb28bcefd1dab7b_img.jpg b/marked/F/T-REC-F.746.15-202212-I_PDF-E/cfda9df1319e04207eb28bcefd1dab7b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..52a7df79ee48819821ab12b66024ed598e480e38 --- /dev/null +++ b/marked/F/T-REC-F.746.15-202212-I_PDF-E/cfda9df1319e04207eb28bcefd1dab7b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e6e24423a4305dd9e7ff35cbaec5a40c8d2aa42eb8a5fa4344695613f88d1f05 +size 176845 diff --git a/marked/F/T-REC-F.746.15-202212-I_PDF-E/dfe0567f7f23e65eac31fd8a281ea347_img.jpg b/marked/F/T-REC-F.746.15-202212-I_PDF-E/dfe0567f7f23e65eac31fd8a281ea347_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..886eb1e1806169171e15b1f64542eee2d4a82eaa --- /dev/null +++ b/marked/F/T-REC-F.746.15-202212-I_PDF-E/dfe0567f7f23e65eac31fd8a281ea347_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c768926507d9715e8fdebac852138a4134dc12df704cdc918e16c62412c4d31c +size 40831 diff --git a/marked/F/T-REC-F.746.15-202212-I_PDF-E/fa859e4e468bfb2710a94527f2c504af_img.jpg b/marked/F/T-REC-F.746.15-202212-I_PDF-E/fa859e4e468bfb2710a94527f2c504af_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..dff87c3f8fc3dd83b31616bffeb7173000e76feb --- /dev/null +++ b/marked/F/T-REC-F.746.15-202212-I_PDF-E/fa859e4e468bfb2710a94527f2c504af_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:edd7d5465c3ef151a8f2bf25c1d53bef8f4c8f7a043b20f1f89ee42d33250f62 +size 49173 diff --git a/marked/F/T-REC-F.746.16-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.746.16-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..30d4bc9edd02d463a452fa930b82ff2dfd368c24 --- /dev/null +++ b/marked/F/T-REC-F.746.16-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8bdea22ddbf01b83f16dbc58bc0c9e4b409b0baae46bf6dede68381ee2f7df83 +size 7130 diff --git a/marked/F/T-REC-F.746.16-202212-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.746.16-202212-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c3ccc80031991559ec831585244f2536f3deb7b7 --- /dev/null +++ b/marked/F/T-REC-F.746.16-202212-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e1e36b45417807b681c0cac1f7997469d9071add686f36d3e023269e5a55c7ba +size 66802 diff --git a/marked/F/T-REC-F.746.16-202212-I_PDF-E/raw.md b/marked/F/T-REC-F.746.16-202212-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..0f68c02c825847635d3c94d0eafc123bf5ac4b67 --- /dev/null +++ b/marked/F/T-REC-F.746.16-202212-I_PDF-E/raw.md @@ -0,0 +1,762 @@ + + +# Recommendation **ITU-T F.746.16 (12/2022)** + +SERIES F: Non-telephone telecommunication services +Multimedia services + +--- + +## **Technical requirements and evaluation methods of intelligence levels of intelligent customer service systems** + +![ITU logo](390120de4fe440c42fea8154fcaad334_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +### **TELEGRAPH SERVICE** + +| | | +|-----------------------------------------------------------------|-------------| +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessages service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | + +### **MOBILE SERVICE** + +| | | +|--------------------------------------------------------|-------------| +| Mobile services and multdestination satellite services | F.110–F.159 | +|--------------------------------------------------------|-------------| + +### **TELEMATIC SERVICES** + +| | | +|-------------------------------------------|-------------| +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | + +### **MESSAGE HANDLING SERVICES** + +| | | +|---------------------------|-------------| +| MESSAGE HANDLING SERVICES | F.400–F.499 | +|---------------------------|-------------| + +### **DIRECTORY SERVICES** + +| | | +|--------------------|-------------| +| DIRECTORY SERVICES | F.500–F.549 | +|--------------------|-------------| + +### **DOCUMENT COMMUNICATION** + +| | | +|--------------------------------------|-------------| +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | + +### **DATA TRANSMISSION SERVICES** + +| | | +|----------------------------|-------------| +| DATA TRANSMISSION SERVICES | F.600–F.699 | +|----------------------------|-------------| + +### **MULTIMEDIA SERVICES** + +| | | +|----------------------------|--------------------| +| MULTIMEDIA SERVICES | F.700–F.799 | +|----------------------------|--------------------| + +### **ISDN SERVICES** + +| | | +|---------------|-------------| +| ISDN SERVICES | F.800–F.849 | +|---------------|-------------| + +### **UNIVERSAL PERSONAL TELECOMMUNICATION** + +| | | +|--------------------------------------|-------------| +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +|--------------------------------------|-------------| + +### **ACCESSIBILITY AND HUMAN FACTORS** + +| | | +|---------------------------------|-------------| +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | +|---------------------------------|-------------| + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.746.16 + +# Technical requirements and evaluation methods of intelligence levels of intelligent customer service systems + +## Summary + +The intelligent customer service system (ICSS) can provide more convenient, efficient, and stable services for users through the application of artificial intelligence (AI) technologies such as speech recognition, text to speech and natural language processing. There is value therefore in improving and evaluating the intelligence levels of the intelligent customer service system. Recommendation ITU-T F.746.16 specifies the requirements and methods for evaluating the system intelligence of intelligent customer service systems through four aspects which are; the basic functions, the core technologies of AI, the system maturation and the service experience. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.746.16 | 2022-12-14 | 16 | 11.1002/1000/15190 | + +## Keywords + +Intelligent customer service, intelligence levels. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2023 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|------|-------------------------------------------------------------|----| +| 1 | Scope ..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 1 | +| 4 | Abbreviations and acronyms ..... | 2 | +| 5 | Conventions ..... | 2 | +| 6 | Architecture overview ..... | 2 | +| 7 | Intelligence levels of ICSS ..... | 4 | +| 8 | Basic functions of ICSS..... | 4 | +| 8.1 | Multi-channel management ..... | 5 | +| 8.2 | Multi-media message interaction ..... | 5 | +| 8.3 | Intelligent dialogue..... | 5 | +| 8.4 | Identity recognition ..... | 6 | +| 8.5 | Knowledge base management ..... | 6 | +| 8.6 | Work order management ..... | 6 | +| 8.7 | Human-machine collaboration ..... | 7 | +| 8.8 | Abnormal session management..... | 7 | +| 9 | Artificial intelligence core technical ability of ICSS..... | 7 | +| 9.1 | Speech recognition ..... | 8 | +| 9.2 | Text to speech..... | 9 | +| 9.3 | Single-round dialogue ..... | 9 | +| 9.4 | Multi-round dialogue..... | 10 | +| 9.5 | Knowledge graph..... | 11 | +| 9.6 | Multilingual interaction ..... | 11 | +| 9.7 | Optical character recognition ..... | 11 | +| 9.8 | Picture content recommendation ..... | 11 | +| 10 | ICSS system maturity ..... | 12 | +| 10.1 | Data visibility ..... | 12 | +| 10.2 | Usability ..... | 12 | +| 10.3 | Reliability ..... | 13 | +| 10.4 | Security..... | 13 | +| 10.5 | Maintainability ..... | 13 | +| 10.6 | Auditability..... | 14 | +| 11 | Service experience of enterprises ..... | 14 | +| 11.1 | Data access ability ..... | 15 | +| 11.2 | Human intervention ability..... | 15 | + +| | Page | +|-------------------------------------------|-------------| +| 11.3 Emotional pacification ability ..... | 15 | +| 11.4 Assistant service ability ..... | 15 | +| 11.5 Customer satisfaction ..... | 16 | + +# **Technical requirements and evaluation methods of intelligence levels of intelligent customer service systems** + +# **1 Scope** + +This Recommendation aims to establish an indicator system to evaluate the intelligence levels of intelligent customer service systems (ICSS) for human-machine interactions. The intelligent customer service system mentioned in this Recommendation includes a variety of product forms such as outbound service robots, inbound service robots, and text service robots. + +This Recommendation considers four main aspects: + +- The basic functions of ICSS; +- The artificial intelligence core technical abilities of ICSS; +- The system maturation of ICSS; +- The service experience of enterprises. + +## **2 References** + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +# **3 Definitions** + +### **3.1 Terms defined elsewhere** + +None. + +### **3.2 Terms defined in this Recommendation** + +This Recommendation defines the following terms: + +**3.2.1 intelligent customer service system:** A customer service system based on natural language processing, speech recognition, text to speech and other artificial intelligence technologies. After converging the user's questions through single or multiple rounds of conversation, the intelligent customer service system relies on a maintained background knowledge base and interface components to realize intelligent dialogue. + +**3.2.2 single-round dialogue:** A conversation between a human and an intelligent customer service system. + +**3.2.3 multi-round dialogue:** More than one conversation between a human and an intelligent customer service system in the process for solving complex problems. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|-------------------------------------| +| APP | Application | +| BLEU | Bilingual Evaluation Understudy | +| CRM | Customer Relationship Management | +| ICSS | Intelligent Customer Service System | +| KM | Knowledge Management | +| MOS | Mean Opinion Score | +| NLG | Natural Language Generation | +| NLP | Natural Language Processing | +| NLU | Natural Language Understanding | +| OCR | Optical Character Recognition | +| PII | Personally Identifiable Information | +| SER | Sentence Error Rate | +| TTS | Text to Speech | +| WER | Word Error Rate | + +## 5 Conventions + +In this Recommendation: + +- The keywords "**is required to**" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "**is recommended**" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement needs not be present to claim conformance. +- The keywords "**can optionally**" and "**may**" indicate an optional requirement which is permissible, without implying any sense of being recommended. These terms are not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +# 6 Architecture overview + +Figure 1 shows the functional structure of the intelligent customer service system. + +![Flowchart of the Functions structure of ICSS. The flow starts with 'Customers access through multiple channels' at the top, followed by 'Identification module', 'Central control module', and 'Intelligent dialogue management'. The 'Intelligent dialogue management' module contains sub-components: NLP, NLU, NLG, TTS, and Knowledge base. It connects to 'Manual customer service' and 'Work order management'. 'Work order management' connects to 'CRM system' and 'KM system'. A vertical bar on the right labeled 'System functions analysis' spans from the 'Central control module' down to the 'Work order management' level. The reference code F.746.16(22) is at the bottom right.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +``` + +graph TD + A[Customers access through multiple channels] --> B[Identification module] + B --> C[Central control module] + C --> D[Intelligent dialogue management] + C --> E[Manual customer service] + D --> F[Work order management] + E --> F + F --> G[CRM system] + F --> H[KM system] + D --- D1[NLP] + D --- D2[NLU] + D --- D3[NLG] + D --- D4[TTS] + D --- D5[Knowledge base] + D1 & D2 & D3 & D4 & D5 --> E + D1 & D2 & D3 & D4 & D5 --> F + G <--> F + H <--> F + I[System functions analysis] -.-> C + I -.-> D + I -.-> E + I -.-> F + J[F.746.16(22)] + +``` + +Flowchart of the Functions structure of ICSS. The flow starts with 'Customers access through multiple channels' at the top, followed by 'Identification module', 'Central control module', and 'Intelligent dialogue management'. The 'Intelligent dialogue management' module contains sub-components: NLP, NLU, NLG, TTS, and Knowledge base. It connects to 'Manual customer service' and 'Work order management'. 'Work order management' connects to 'CRM system' and 'KM system'. A vertical bar on the right labeled 'System functions analysis' spans from the 'Central control module' down to the 'Work order management' level. The reference code F.746.16(22) is at the bottom right. + +**Figure 1 – Functions structure of ICSS** + +**Customers access through multiple channels:** This module provides users with web pages, apps, phones and other channels to access the ICSS and obtain relevant services. + +**Identification module:** This module of the ICSS mainly recognizes the user's speech and transforms it into text content for subsequent analysis. + +**Central control module:** This module further process the user's message and converts it into input content that can be recognized and understood by AI, and determines whether the message is answered by the intelligent customer service system or by a manual customer service. + +**Intelligent dialogue management:** This module is to understand the user's intention, match the relevant answers in the corpus, and feed back to the user through text or speech. It mainly involves natural language processing (NLP), natural language understanding (NLU), natural language generation (NLG) and other AI technologies. + +**Manual customer service:** Manual customer service is an important part of the ICSS, which is responsible for dealing with problems that are difficult to solve automatically by the system. + +**Work order management:** When the system encounters problems that are difficult to solve, the content of human-machine conversation will be saved in the form of work orders, and finally the centralized manual processing will be carried out. + +**CRM system and KM systems:** Customer relationship management (CRM) and knowledge management (KM) are auxiliary systems of the ICSS, which can help the ICSS better manage customer relations and business knowledge. CRM system refers to an information system for enterprises to collect, manage, analyse and use customer information by using software, hardware and network technology. KM system is an information system that collects, processes and shares all the knowledge of an organization. + +# 7 Intelligence levels of ICSS + +According to the application maturity of involved intelligent technologies, the intelligence ability of ICSS is divided into five levels. The definition and description of each level are shown in Table 1. + +**Table 1 – Intelligence levels for ICSS** + +| Intelligence levels | Definition | Description | +|---------------------|--------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Level-1 | No intelligence | The system only has simple basic functions and system maturity, the user experience is not satisfactory. | +| Level-2 | Partial intelligence | Based on some basic functions and system maturity, the system has simple AI core technical capabilities, but its performance and service experience are mediocre, and users can barely accept the system. | +| Level-3 | Conditional intelligence | The system has many basic functions, and the system supports a variety of AI core technical capabilities, and its performance is good. The service capability and system maturity of the enterprise are basically acceptable to users. | +| Level-4 | High intelligence | The system has rich basic functions, and the system supports a variety of AI core technical capabilities, and its performance is excellent. The service ability and system maturity of the enterprise are strong, and the user acceptance is strong. | +| Level-5 | Full intelligence | In addition to the basic functions, the system also has other extended functions, the system supports a variety of AI core technical capabilities, and its performance is excellent. The service capability and system maturity are completely acceptable to users. | + +## 8 Basic functions of ICSS + +Basic functions include multi-channel management, multimedia message interaction, intelligent recommendation, identity recognition, knowledge base management, work order management, human-machine collaboration, and session abnormal management. The intelligence level description of this dimension can be seen in Table 2. + +**Table 2 – Intelligence levels for basic function of ICSS** + +| Intelligence levels | Description | +|---------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Level-1 | 1) Supports at least one way to access channel.
2) Supports at least one multimedia message type. | +| Level-2 | New relatives to the previous level:
1) Supports at least two ways to access channel.
2) Supports at least two types of multimedia messages.
3) Supports at least one way to identify the identity. | +| Level-3 | New relatives to the previous level:
1) Supports at least three ways to access channel.
2) Supports at least three types of multimedia messages.
3) Supports the intelligent dialogue function, and predict the user's intention according to the user's behaviour track and historical question information. | + +**Table 2 – Intelligence levels for basic function of ICSS** + +| Intelligence levels | Description | +|---------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| | 4) Supports at least two ways to identify the identity.
5) Supports knowledge base management. | +| Level-4 | New relatives to the previous level:
1) Supports at least four ways to access channel.
2) Supports at least four types of multimedia messages.
3) Supports intelligent dialogue, and can automatically complete, prompt or correct the entered content in the process of user entering questions.
4) Supports at least three ways to identify the identity.
5) Supports common operations such as adding, deleting, modifying, and querying the knowledge base.
6) Supports work order management functions, including rapid creation of work orders, flexible configuration of work orders, one click to work orders, work order prioritization, and work order status tracking.
7) Supports intelligent customer service such as responding quickly and transferring to manual customer service in time. | +| Level-5 | New relatives to the previous level:
1) Supports work order management functions other than the above functions.
2) Supports intelligent customer service with human-computer cooperation function, and provide process records, problem push and other assistance for manual customer service during the interaction process.
3) Supports abnormal session management. | + +### 8.1 Multi-channel management + +ICSS is recommended to support websites, APPs, e-mails and other channels. + +Evaluation methods: + +- Determine whether the customer service product supports the above channel access. +- Determine whether the customer service product supports access through channels other than the above channels. + +### 8.2 Multi-media message interaction + +Determine whether the interactive methods supported by intelligent customer service are abundant. Customers can use speech, text, pictures, videos, expressions, custom messages and other real-time interactions during communication with the customer service to increase the friendliness and fun. + +Evaluation methods: + +- Judge the types of messages supported by the customer service system through material review and technical means, and give corresponding scores. + +### 8.3 Intelligent dialogue + +The ICSS recommends related questions based on the user's behaviour tracking, historical consultation and other information. During the interaction process, the system automatically recommends questions related to user consultation, which usually appear in the feedback answers or interaction interface. + +Evaluation methods: + +- a) When the user enters the conversation interface, judge whether the customer service can predict user questions based on historical question records or popular questions, and provide relevant pushes. +- b) When the users ask questions, the system will automatically analyse the content that has been entered to provide completion or relevant prompts. + +### **8.4 Identity recognition** + +This indicator is to evaluate whether the intelligent customer service supports the function of identification. Considering the personally identifiable information (PII) protection, the ICSS can optionally have the function of identifying the user's identity after obtaining the user's permission. For example, the system can use voiceprint recognition, face recognition, short message service (SMS) verification code, voice call and other technologies to determine the user's identity information. + +Evaluation methods: + +- a) Check the smart customer service product information provided by the manufacturer to determine whether the smart customer service supports the identification function. +- b) According to the product description, the test system can correctly identify the user's identity. + +### **8.5 Knowledge base management** + +This indicator is to evaluate the richness and self-learning function of the knowledge base. The ICSS is recommended to include corpus, question database and other related databases. + +Evaluation methods: + +- a) Check the smart customer service product information provided by the manufacturer to determine whether the smart customer service supports the knowledge base management function. +- b) Check the richness of customer service knowledge base. +- c) Test whether the knowledge base can support self-learning and establish custom questions and answers. + +### **8.6 Work order management** + +This indicator is to evaluate whether the ICSS has the work order management function. When encountered problems cannot be solved, customer service staff is required to create work orders to support functions such as flexible work order configuration, prioritization, and work order status tracking. + +Evaluation methods: + +- a) Check the smart customer service product information provided by the manufacturer to determine whether the smart customer service supports the work order management function. +- b) Check the operating instructions of the work order management system, whether it supports flexible configuration of work order rules, one-click transfer of work orders, provision of hierarchical services, priority sorting, and work order status tracking functions according to the customer's customer service needs. +- c) Other functions supported by the work order management system. + +For example: The enterprise can set the work orders of users such as VIP and special strategy risks as priority work orders. When customer service personnel of a certain business type receive other business, they can carry out internal transfer orders. Both the internal system and users can understand the progress of the corresponding work order processing. + +### 8.7 Human-machine collaboration + +This indicator is to evaluate the ability of the intelligent customer service and the manual customer service to solve customer problems collaboratively. When encountered problems cannot be solved by the customer service system of the machine, it supports transfer to manual service. Then the human service and the customer service system will cooperate to solve the customers' problems. + +Evaluation methods: + +- Determine whether there is support for the transfer to manual services. +- Determine whether the smart customer service supports video calls, voice calls, text chats, etc. + +For example: When the machine customer service is not smooth, such as unrecognizable, multiple fuzzy recognition, user dissatisfaction, etc., observe whether the machine customer service responds as always, and judge whether it is a manual combination of machine response. + +### 8.8 Abnormal session management + +Determine whether the intelligent customer service can detect abnormal conditions such as abnormal session duration and abnormal connection, and implement corresponding countermeasures. + +When abnormalities such as disconnection occur in the session, the ICSS is required to save the previous records, determine the cause of the abnormality, and try to reconnect. When the system contacts the user again, it is recommended to be able to continue the topic before the interruption. + +Evaluation methods: + +- Check the session duration to determine whether it is abnormal. +- When the session is abnormal, check whether there is an attempt to reconnect. +- After the session is reconnected, check whether the system can continue to record the breakpoint. + +## 9 Artificial intelligence core technical ability of ICSS + +This indicator contains three types of algorithms: speech processing, natural language processing (NLP), and image processing. The intelligence level description of this dimension can be seen in Table 3. + +**Table 3 – Intelligence levels for artificial intelligence core technical ability** + +| Intelligence levels | Description | +|---------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Level-1 |
  1. Supports speech recognition, but the performance is poor, the user's experience is poor.
  2. Supports speech synthesis, and users reluctantly accept the naturalness of synthetic speech.
  3. Supports single-round dialogue, but the performance is poor, the user's experience is poor.
| + +**Table 3 – Intelligence levels for artificial intelligence core technical ability** + +| Intelligence levels | Description | +|---------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Level-2 | New relatives to the previous level:
1) Supports speech recognition, the system performance is mediocre, the user's experience is flat.
2) Supports speech synthesis, and users basically accept the naturalness of synthetic speech.
3) Supports single-round dialogue and the system performance is mediocre, the user's experience is flat.
4) Supports multiple-round dialogue, but the performance is poor, the user's experience is poor. | +| Level-3 | New relatives to the previous level:
1) Supports speech recognition and the system performance is good.
2) Supports speech synthesis, and users can accept the naturalness of synthetic speech.
3) Supports single-round dialogue, the system performance is good.
4) Supports multiple-round dialogue and the system performance is mediocre, the user's experience is flat.
5) Supports multilingual interaction. | +| Level-4 | New relatives to the previous level:
1) Supports speech recognition and the performance is perfect.
2) Supports speech synthesis, and users can fully accept the naturalness of synthetic speech.
3) Supports single-round dialogue and the performance is perfect.
4) Supports multiple-round dialogue and the performance is good.
5) Supports optical character recognition (OCR) and the performance is good. | +| Level-5 | New relatives to the previous level:
1) Supports multiple-round dialogue and the performance is perfect.
2) Supports OCR and the performance is perfect.
3) Supports picture content recommendation, and the recommended pictures are highly relevant.
4) Supports the ability to construct, use, and maintain a knowledge graph. | + +### 9.1 Speech recognition + +This indicator is to test whether the intelligent customer service has speech recognition capability. This is the ability to test the accuracy of the intelligent customer service's speech recognition mainly through the technical test method. Through recognition and understanding, the voice signal is converted into corresponding text or commands. + +Evaluation methods: + +- Construct a speech data set, obtain the text of speech recognition, and separately count the word accuracy and sentence accuracy of speech recognition. +- Word error rate (WER): Compare the recognized sequence with the label sequence of characters, insert, replace, and delete the total number of characters, divided by the percentage of the total number of words in the label sequence. Each test data will get a word error rate, and the average value is the word error rate of the test set. + +$$WER = \frac{NW_{misrecognized}}{NW_{total}}$$ + +$NW_{misrecognized}$ : the number of misrecognized words. + +$NW_{total}$ : Total words. + +- c) Sentence error rate (SER): If there is a misrecognition of a word or character in a sentence, the sentence is considered to be misrecognized. The number of sentence recognition errors divided by the total number of sentences is the sentence error rate. + +$$SER = \frac{NS_{misrecognized}}{NS_{total}}$$ + +$NS_{misrecognized}$ : the number of misrecognized sentences. + +$NS_{total}$ : Total sentences. + +### 9.2 Text to speech + +This indicator is to test whether the intelligent customer service has text to speech capability. This is the ability to convert any text into speech with a high degree of naturalness. + +Evaluation methods: + +- a) After the multi-round call task initiated by the outbound call system under test is completed, the text and audio are extracted. +- b) Many testers use the mean opinion score (MOS) indicators (five-point scale) to rate the naturalness of the robot's speech synthesis in audio. The MOS scoring instructions can be seen in Table 4. + +**Table 4 – MOS scoring instructions** + +| Scores | Levels | Description | +|--------|-----------|-----------------------------------------------------------------------------------------------------| +| 5 | Excellent | Fluent and natural, can be completely relaxed, no need for attention. | +| 4 | Good | The distortion is not obvious and requires attention, but does not need to be clearly concentrated. | +| 3 | General | It is clearly distorted and requires a moderate level of attention; | +| 2 | Bad | Severe distortion and need for concentration. | +| 1 | Very bad | The quality of the speech is so poor that it is difficult to understand even if you try to listen. | + +- c) In the case of artificial recording and speech synthesis audio splicing, the overall fluency is required to be determined and the jumping and non-smooth conditions are required to be recorded. +- d) In order to reflect the customizability and diversity of speech synthesis dialogue in the external call process, the situation regarding the richness of timbre and variety is to be recorded. + +### 9.3 Single-round dialogue + +Evaluate whether the intelligent customer service product has the capability of single-round conversation. A single round of conversation refers to a simple question and answer. The question can be described in one sentence and does not depend on the context. + +Evaluation methods: + +- a) A single round of dialogue test data set is used to observe the dialogue and question answering effects of intelligent customer service. +- b) The recognition situation is divided into: accurate recognition, fuzzy recommendation is correct, fuzzy recommendation is wrong, not recognized. +- c) Accurate recognition and fuzzy recommendation are regarded as successful intention recognition. The ratio of successful recognition to total tests is the intention recognition rate. The ratio of accurate recognition to total tests is the accurate recognition rate. + +$$\text{Intent recognition rate} = \frac{NT_{\text{true\&vague}}}{NT_{\text{total}}}$$ + +$NT_{\text{true\&vague}}$ : *Accurately identify or indistinctly recommend the correct quantity* + +$NT_{\text{total}}$ : *Total task* + +$$\text{Accurate recognition rate} = \frac{NT_{\text{true}}}{NT_{\text{total}}}$$ + +$NT_{\text{true}}$ : *The number of accurate identifications.* + +### 9.4 Multi-round dialogue + +This indicator is to evaluate the ability of intelligent customer service products to support multiple rounds of dialogue. In the customer service scenario, many problems cannot be solved by simple question and answer in a single round of conversation. When the user's needs are more complex and may need to be presented in multiple rounds of conversation, the machine can help the user find satisfactory results by asking, clarifying, or confirming. Such a round-trip interactive session is called the multi-round dialogue. + +Evaluation methods: + +- a) Use multiple rounds of dialogue to test the data sets. Before the intelligent customer service product gives the user accurate results, the system will ask diverging questions, add details and repeatedly ask questions based on the results to see whether it can be accurately identified. +- b) Multiple rounds of test data sets covering 2-4 rounds. +- c) All rounds of multiple dialogues were successfully identified as task completion, and the ratio of test data volume to task completion rate is given by: + +$$\text{Task completion rate} = \frac{T_{\text{completed}}}{T_{\text{total}}}$$ + +$T_{\text{completed}}$ : *the number of completed tasks.* + +$T_{\text{total}}$ : *the total number of tasks.* + +- d) The ratio of the number of successful rounds identified in the multi-round test to the total number of test rounds is the recognition rate: + +$$\text{Recognition rate} = \frac{T_{\text{co-i}}}{T_{\text{total}}}$$ + +$T_{\text{co-i}}$ : *The number of correctly identified multi – round dialogues.* + +### 9.5 Knowledge graph + +The purpose of this indicator is to evaluate whether the ICSS has the ability to construct, use, and maintain a knowledge graph. The ICSS can optionally have the ability to construct a knowledge graph, and can improve the functions lacking in the traditional knowledge base. + +Evaluation methods: + +- Determine whether the ICSS has the ability to build a knowledge graph through material reviews. +- Test the ability of ICSS to build, generate and maintain knowledge graph through technical means. +- Comprehensive evaluation of the construction, generation and maintenance of knowledge graphs and other indicators to determine whether customer service products have integrated knowledge graph technology. + +### 9.6 Multilingual interaction + +This indicator is to evaluate whether the ICSS has the ability to interact in multiple languages. The intelligent customer service can optionally have the ability of multilingual interaction, that is, the function of bilingual conversion can be automatically performed by machine translation. + +Evaluation methods: + +- Judge whether the ICSS is capable of multi-language interaction through material reviews. +- Test the intelligibility of multilingual interaction of the ICSS and bilingual evaluation understudy (BLEU) score through test corpus. + +### 9.7 Optical character recognition + +The purpose of this indicator is to evaluate whether the ICSS has the capability of text image recognition. The ICSS can optionally have intelligent visual capabilities such as OCR to assist the ICSS in receiving and processing multi-modal user input. + +Evaluation methods: + +- Judge whether the ICSS has the capability of image and text recognition through material review. +- Through the building of text image data to test whether the ICSS has text image recognition capability. + +$$Accuracy\ rate = \frac{Ima_{correctly}}{T_{total}}$$ + +*Imacorrectly: The number of correctly identified images.* + +### 9.8 Picture content recommendation + +This indicator is to evaluate whether the ICSS has the ability to recommend related content of pictures. The ICSS can optionally identify pictures and recommend the same or similar items in specific fields such as e-commerce. + +Evaluation methods: + +- Judge whether the ICSS has the ability to recommend related pictures through material review. +- Test whether the ICSS has the ability to recommend related contents of pictures by building image data, and score the relevancy of pictures on a scale of 0-5. + +## 10 ICSS system maturity + +Indicators of maturation of the ICSS system include data visibility, usability, reliability, security, maintainability, and auditability. The intelligence level description of this dimension can be seen in Table 5. + +**Table 5 – Intelligence levels for ICSS system maturation** + +| Intelligence levels | Description | +|---------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Level-1 | 1) The usability performance is poor, the user's experience is poor.
2) The security performance is poor, the user's experience is poor. | +| Level-2 | New relatives to the previous level:
1) The usability performance is general, and the user experience is basically acceptable.
2) The reliability performance is general.
3) The security performance is good, and the user experience is basically acceptable.
4) Supports auditability. | +| Level-3 | New relatives to the previous level:
1) The usability performance is good.
2) The reliability performance is good.
3) The security performance is perfect.
4) Supports data visibility function, which can generate visual reports for business data in real time. | +| Level-4 | New relatives to the previous level:
1) The usability performance is excellent.
2) The reliability performance is excellent, such as weather and news inquiries.
3) Supports auditability.
4) Supports automatic operation and maintenance.
5) Supports data visibility function, and the system can automatically generate analysis results and documents and export them. | +| Level-5 | New relatives to the previous level:
1) Supports operation monitoring, and the system can quickly locate system operation problems. | + +## 10.1 Data visibility + +Enterprises need to judge the effect of intelligent customer service through visual analysis. The ICSS is required to have a report analysis function to help enterprises improve their operating efficiency. + +Evaluation methods: + +- a) Generate visual reports. +- b) Analyse according to the report. +- c) Export reports and data. + +### 10.2 Usability + +Evaluate whether the ICSS is easy to use. The ICSS is required to obey the principles of user friendliness and ease of use. + +Evaluation methods: + +- a) The reply message is required to be completed within a certain time, and there is a prompt such as "reply in customer service". +- b) User-centred design. Used smoothly in a set of business processes in a guided manner. +- c) Interface design can optionally fully consider the user experience, friendly and clear, simple operation, obvious function points. The main functions are required to be in a prominent position, and the commonly used functions are required to be in a position easy to operate. Remove the redundant steps and get to the result. + +## **10.3 Reliability** + +Evaluate whether the ICSS is mature and reliable. A mature and stable system is very important for the continuous operation of the ICSS. + +Evaluation methods: + +- a) Stable and reliable mature technology is required to be adopted to ensure the long-term safe operation of the system. +- b) The hardware, software and information resources in the system can optionally meet the requirements of reliability design. +- c) The system has fault-tolerant ability, which makes management and maintenance more convenient. +- d) Unified planning and analysis of network design, selection, installation, debugging and other links to ensure reliable operation of the system. +- e) The system provides multi-level security control means, establishes a perfect security management system, guarantees data security and PII protection and integrity, and simultaneously has reliable anti-virus measures. +- f) System failure recovery capability. +- g) Log management function. +- h) System disaster recovery backup. + +### **10.4 Security** + +Evaluate the security of the ICSS. While providing intelligent services, intelligent customer service products must also ensure data security and legality of dialogues. + +Evaluation methods: + +- a) User program security and PII protection, including user permissions, password management, and so on. +- b) System network security, including attack protection, vulnerability detection, Trojan killing, and so on. +- c) Database security, including backup, data encryption, and so on. +- d) Transmission security, including data encryption transmission, and so on. +- e) Content security, including disabled word filtering, image filtering, and so on. +- f) Deployment of security, including the separation of the two networks, export protection, wall protection, and so on. + +### **10.5 Maintainability** + +Evaluate the maintainability of the intelligent customer service system. The ICSS is required to have complete maintenance tools. + +Evaluation methods: + +- a) Whether to have automatic operation and maintenance technology. +- b) Whether there is operation monitoring. + +### 10.6 Auditability + +Evaluate whether the ICSS is auditable. The operation data of the ICSS is required to be able to be checked, reviewed and traced. + +Evaluation methods: + +- a) Supports personnel authority management. +- b) Supports operation audit. + +## 11 Service experience of enterprises + +Service experience of enterprises evaluation includes data access ability, human intervention ability, emotional pacification ability, assistant service ability, customer satisfaction. The intelligence levels description of this dimension can be seen in Table 6. + +**Table 6 – Intelligent levels for service experience of enterprises** + +| Intelligence levels | Description | +|---------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Level-1 |
  • 1) Supports access to user information, and provides targeted services for users.
  • 2) Supports manual intervention to deal with emergencies.
  • 3) The user is not satisfied, and the total user experience score is [0, 60).
| +| Level-2 |

New relatives to the previous level:

  • 1) Supports the manual intervention of the single-round dialogue, and the manual intervention rate is high.
  • 2) The user is basically satisfied, and the total user experience score is [60, 70).
| +| Level-3 |

New relatives to the previous level:

  • 1) Supports data access ability, and the data access rate is high.
  • 2) Supports the manual intervention of the single-round dialogue, and the manual intervention rate is acceptable.
  • 3) Supports the manual intervention of the multi-round dialogue, with a high rate of manual intervention.
  • 4) The user is relatively satisfied, and the total user experience score is [70, 80).
| +| Level-4 |

New relatives to the previous level:

  • 1) Supports the manual intervention of multiple rounds of dialogue, and the manual intervention rate is acceptable.
  • 2) Supports assistant service ability.
  • 3) The user is very satisfied, and the total user experience score is [80, 90).
| +| Level-5 |

New relatives to the previous level:

  • 1) Supports emotional pacification ability, and switches to manual customer service for pacification when necessary.
  • 2) The user is completely satisfied, and the total user experience score is [90, 100].
| + +### 11.1 Data access ability + +The purpose of this indicator is to evaluate whether smart customer service products can access user information and access capabilities. The intelligent customer service has data access capabilities, which can provide users with a more convenient and safe service experience. + +Evaluation methods: + +- a) Businesses that need access to data build a test suite to observe data access. +- b) The data access rate is the number of successful implementations divided by the number of tests. + +$$\text{Data access rate} = \frac{T_{\text{successful}}}{T_{\text{total}}}$$ + +$T_{\text{successful}}$ : successful data access. + +### 11.2 Human intervention ability + +This indicator is to evaluate the manual intervention of the ICSS and in order to judge the autonomous business capabilities of the intelligent customer service. In the process of providing services to users, the ICSS needs to rely on the intervention of manual customer service for specific items such as puzzles, serious questions, and user's mood swings. + +Evaluation methods: + +- a) Count the number of manual interventions during the single-round dialogue and multi-round dialogue testing. + +$$\text{Manual intervention rate} = \frac{T_{\text{manual}}}{T_{\text{total}}}$$ + +$T_{\text{manual}}$ : the number of human interventions. + +### 11.3 Emotional pacification ability + +This indicator is to evaluate whether the ICSS can identify the user's mood swings and provide timely and effective pacification. Emotional pacification is a high-level demand for an ICSS, which requires the system to correctly determine the user's emotions through a series of elements such as speech, tone, wording, punctuation and tone words during the interaction with the user. When recognizing the user's emotions such as expressions of anger, it can respond and comfort in time, and if necessary, it can switch to manual customer service for pacification. + +Evaluation methods: + +- a) A corpus of anger and requests for artificial service was designed to evaluate the capability of providing emotional consolation. + +### 11.4 Assistant service ability + +This indicator is to assess whether the ICSS provides assistant services such as weather inquiries and news inquiries. The intelligent customer service needs to provide related services such as weather inquiries and news inquiries while providing business-related services. + +Evaluation methods: + +- a) Determine whether the ICSS has the related business ability to provide assistant services through material review. +- b) Build the command corpus related to weather inquiry, news inquiry, etc., to test whether the ICSS has relevant abilities. The service ability of assistants is rated according to 0-5 score. + +## 11.5 Customer satisfaction + +This indicator is to test the effect of the user experience of the ICSS. The quality of the user experience directly determines whether the intelligent customer service can play the role of respondent on a certain product or on certain types of products. User experience is a purely subjective feeling established by users during the use of products. For any intelligent customer service product, the user experience comprehensively reflects the subjective experience of the system in terms of channel management, technical effects, and interface design. In the context of user-centred and people-oriented product design concepts, user experience has become a key indicator for evaluating the merits of human-computer interaction systems. + +Evaluation methods: + +- a) Through the investigation and trial of the ICSS, the user experience effect is assessed. The scoring indicators are as follows: +- b) Useful: The ICSS is required to be useful and actually solve the user's problems and confusion. +- c) Findable: Good navigation and positioning of elements is required to be provided so that users can quickly find the interactive entry point of intelligent customer service. +- d) Accessible: The basic and common functions of the intelligent customer service are required to be accessible to all users, including members and non-members. +- e) Desirable: Intelligent customer service is required to satisfy users' various emotional experiences, including emotional comfort, etc. +- f) Credible: Intelligent customer service replies can let the user trust the system, no false forged information. +- g) Each of the five specific indicators above gets 20 points, and the sum of the five items $S$ is the total score of user experience. +- h) The total user experience score is 4 points, $S \in [0, 60)$ no score, $S \in [60, 70)$ 1 score, $S \in [70, 80)$ 2 score, $S \in [80, 90)$ 3 score, $S \in [90, 100]$ 4 score. +- i) Researchers questioned: The rating group shall be composed of a variety of social roles and professionals and shall exceed 100 persons; Each person will score the ICSS independently, and the average score will be the final score of each system. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.746.17-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.746.17-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..abda52cbb1047d6649f89e40c898a1c33bed7267 --- /dev/null +++ b/marked/F/T-REC-F.746.17-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6f3b7e07231c0f3e116c99c493bdda98ecd52313b6735faa6cea23815cf3162b +size 7190 diff --git a/marked/F/T-REC-F.746.17-202212-I_PDF-E/dbe553cf16dd14073b89a8263a428664_img.jpg b/marked/F/T-REC-F.746.17-202212-I_PDF-E/dbe553cf16dd14073b89a8263a428664_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..fa46afa31654962ca441fceb30348a4d903009e9 --- /dev/null +++ b/marked/F/T-REC-F.746.17-202212-I_PDF-E/dbe553cf16dd14073b89a8263a428664_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d675fbc828b0a9ec86d6c39cc81f3bf82df5de77681c37f3a82ad432cf086bfe +size 50612 diff --git a/marked/F/T-REC-F.746.17-202212-I_PDF-E/f4fdd410cdb84df81274da55721e56fb_img.jpg b/marked/F/T-REC-F.746.17-202212-I_PDF-E/f4fdd410cdb84df81274da55721e56fb_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..bc17af4f172645b25cb45f5a573a7fad62cfd646 --- /dev/null +++ b/marked/F/T-REC-F.746.17-202212-I_PDF-E/f4fdd410cdb84df81274da55721e56fb_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5b9d1c5811602ff23b43175e020a6a56eb484958edf208364f18a1816ced45a7 +size 111369 diff --git a/marked/F/T-REC-F.746.17-202212-I_PDF-E/raw.md b/marked/F/T-REC-F.746.17-202212-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..4abb04f27d032724190a1c97531ea88e8b84dd88 --- /dev/null +++ b/marked/F/T-REC-F.746.17-202212-I_PDF-E/raw.md @@ -0,0 +1,465 @@ + + +# Recommendation **ITU-T F.746.17 (12/2022)** + +SERIES F: Non-telephone telecommunication services +Multimedia services + +## --- **Requirements for media processing services** + +![ITU logo](390120de4fe440c42fea8154fcaad334_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with white lines representing latitude and longitude, and the letters 'ITU' in a bold, blue, sans-serif font superimposed on the globe. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +### **TELEGRAPH SERVICE** + +| | | +|-----------------------------------------------------------------|-------------| +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessages service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | + +### **MOBILE SERVICE** + +| | | +|--------------------------------------------------------|-------------| +| Mobile services and multdestination satellite services | F.110–F.159 | +|--------------------------------------------------------|-------------| + +### **TELEMATIC SERVICES** + +| | | +|-------------------------------------------|-------------| +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | + +### **MESSAGE HANDLING SERVICES** + +F.400–F.499 + +### **DIRECTORY SERVICES** + +F.500–F.549 + +### **DOCUMENT COMMUNICATION** + +| | | +|--------------------------------------|-------------| +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | + +### **DATA TRANSMISSION SERVICES** + +F.600–F.699 + +### **MULTIMEDIA SERVICES** + +**F.700–F.799** + +### **ISDN SERVICES** + +F.800–F.849 + +## **UNIVERSAL PERSONAL TELECOMMUNICATION** + +F.850–F.899 + +## **ACCESSIBILITY AND HUMAN FACTORS** + +F.900–F.999 + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.746.17 + +## Requirements for media processing services + +## Summary + +Recommendation ITU-T F.746.17 identifies the functional requirements for the media processing services. In particular, the scope of this Recommendation includes functional requirements and application scenarios. Media processing services utilize a set of techniques including cloud computing, computing resource virtualization, and job queue processing to dynamically control and manage computing resources which improves scalability, flexibility and availability. This Recommendation specifies the functional requirements of general requirements, service provision requirements, service management requirements, security considerations, etc. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.746.17 | 2022-12-14 | 16 | 11.1002/1000/15191 | + +### Keywords + +Artificial intelligence (AI), audio processing, content delivery network (CDN), media processing service (MPS), video services. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2023 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|------|-------------------------------------------------|----| +| 1 | Scope ..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 1 | +| 4 | Abbreviations and acronyms ..... | 1 | +| 5 | Conventions ..... | 2 | +| 6 | Introduction ..... | 3 | +| 7 | Overview of media processing services ..... | 3 | +| 8 | Requirements ..... | 5 | +| 8.1 | General requirements..... | 5 | +| 8.2 | Service provision requirements ..... | 5 | +| 8.3 | Availability and scalability requirements ..... | 6 | +| 8.4 | Service management requirements..... | 6 | +| 9 | Security considerations..... | 7 | +| | Appendix I – Application scenarios..... | 8 | +| I.1 | Face filtering..... | 8 | +| I.2 | Audio processing..... | 8 | +| I.3 | Watermarks..... | 8 | +| I.4 | Specific content detection..... | 8 | +| I.5 | Information insertion ..... | 8 | +| I.6 | Video tagging ..... | 8 | +| I.7 | Speech recognition ..... | 8 | +| I.8 | SD to HD conversion..... | 8 | +| I.9 | Video smart cover..... | 8 | +| I.10 | Video format converting..... | 8 | +| I.11 | Resolution enhancement..... | 9 | +| I.12 | Video clipping and stitching..... | 9 | +| I.13 | Picture in picture (PIP) ..... | 9 | +| I.14 | Encrypting video ..... | 9 | +| | Bibliography..... | 10 | + + + +# Requirements for media processing services + +# 1 Scope + +This Recommendation identifies the requirements for media processing services. In particular, the scope of this Recommendation includes functional requirements and application scenarios. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +# 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 artificial intelligence (AI)** [b-ITU-T F.749.13]: An interdisciplinary field, usually regarded as a branch of computer science, dealing with models and systems for the performance of functions generally associated with human intelligence, such as reasoning and learning. + +**3.1.2 cloud computing** [b-ITU-T Y.3500]: Paradigm for enabling network access to a scalable and elastic pool of shareable physical or virtual resources with self-service provisioning and administration on-demand. + +**3.1.3 content delivery network (CDN)** [b-ITU-T F.750]: A network optimized for delivering digital content. + +**3.1.4 platform as a service (PaaS)** [b-ITU-T Y.3500]: Cloud service category (see clause 3.2.10) in which the cloud capabilities type (see clause 3.2.4 in [b-ITU-T Y.3500]) provided to the cloud service customer (see clause 3.2.11 in [b-ITU-T Y.3500]) is a platform capabilities type (see clause 3.2.31 in [b-ITU-T Y.3500]). + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following term: + +**3.2.1 media processing service (MPS)**: A cloud service for transcoding multimedia data, which can provide a cost-effective, easy-to-use, elastic, and highly scalable method to convert audio and video into formats suitable for playing on PCs, TVs and mobile interfaces, and can also conduct deep learning based technology on massive data to analyse the content including text, voice and other scenarios of the media in multiple modes, and to achieve intelligent review, content understanding, intelligent editing, etc. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|----------|--------------------------------------------------------------------------| +| 3GP | 3rd Generation Partnership Project (3GPP) file format | +| AI | Artificial Intelligence | +| AR | Augmented Reality | +| ASIC | Application Specific Integrated Circuit | +| AVI | Audio Video Interleaved | +| CDN | Content Delivery Network | +| CPU | Central Processing Unit | +| DAG | Direct Acyclic Graph | +| DRM | Digital Rights Management | +| FLV | Flash Video | +| FPGA | Field Programmable Gate Array | +| GPU | Graphics Processing Unit | +| HD | High-Definition | +| IO | Input/Output | +| M3U8 | Moving Picture Experts Group Audio Layer 3 Uniform Resource Locator-UTF8 | +| MP4 | Moving Picture Experts Group 4 | +| MPG/MPEG | Moving Picture Experts Group | +| MPS | Media Processing Service | +| PaaS | Platform as a Service | +| PGC | Professionally Generated Content | +| PIP | Picture in Picture | +| SD | Standard-Definition | +| UGC | User Generated Content | +| URL | Uniform Resource Locator | +| UTF-8 | Unicode Transformation Format | +| VR | Virtual Reality | + +# 5 Conventions + +In this Recommendation: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement needs not be present to claim conformance. +- The keywords "can optionally" and "may" indicate an optional requirement which is permissible, without implying any sense of being recommended. These terms are not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +# 6 Introduction + +With the development of Internet technology, the network traffic brought by multimedia content, especially audio and video services has had an explosive growth. Meanwhile, video services have been widely used in entertainment, finance, education, medical and other fields. 4K/8K and virtual reality (VR) / augmented reality (AR) video services will be popular. The rise of live broadcasting and short video services has driven the development of professionally generated content (PGC) and user generated content (UGC). With these massive media contents, how to provide superior service to ensure that media content can be distributed quickly, reliably, and safely is the most significant issue to be solved currently. + +Traditionally, content delivery network (CDN) is used to deliver content from the head-end of content sources to the terminal devices. CDN only transmits media content without any processing. With these requirements of new video services, media contents are required to be processed before transmission through the CDN. For example, media contents are required to be transcoded adaptively according to the quality of the network. UGC contents should not contain specific features. Some image and text contents are required to be detected, located, and identified to provide personalized service for consumers. + +Regarding the huge amount of video data and the variety of media processing, the way of processing in the local computing system cannot meet the current computing resource requirements of video services. At present, computing technology is widely applied. Media processing services are based on the concept of cloud computing technology. Media processing services provide a cost-effective, easy-to-use, elastic, on-demand, and highly scalable method to process audio and video in various video services. + +In a typical media processing service flow from end to end, video contents are captured by the content provider. After the video is uploaded to the cloud server, media processing services begin to work. In order to adapt to different network environments and terminals, the video will be usually transcoded in the cloud. Some personalized value-added services are provided, which include data statistics service, content review service, real-time watermarking service and so on. After media processing, the video contents are distributed and speeded up to the terminal devices by the CDN. Video contents are decoded and played on the terminal devices. + +Figure 6-1 shows the reference model for media processing services (MPSs). + +![Figure 6-1 – Reference model for media processing services. The diagram illustrates a sequential flow from left to right: Video capture (by Content provider) -> Video coding -> Video processing (containing Watermarks, Face filters, Video tagging, Specific content detection, Information insertion, Speech recognition, and ellipsis) -> Content distributing (by CDN) -> Video decoding -> Video playing (by Content consumer). Below the flow, three large chevrons group these steps into 'Video provider' (capture and coding), 'Video service' (processing), and 'Video display' (distributing, decoding, and playing). A small code 'F.746.17(22)' is in the bottom right corner.](dbe553cf16dd14073b89a8263a428664_img.jpg) + +F.746.17(22) + +Figure 6-1 – Reference model for media processing services. The diagram illustrates a sequential flow from left to right: Video capture (by Content provider) -> Video coding -> Video processing (containing Watermarks, Face filters, Video tagging, Specific content detection, Information insertion, Speech recognition, and ellipsis) -> Content distributing (by CDN) -> Video decoding -> Video playing (by Content consumer). Below the flow, three large chevrons group these steps into 'Video provider' (capture and coding), 'Video service' (processing), and 'Video display' (distributing, decoding, and playing). A small code 'F.746.17(22)' is in the bottom right corner. + +**Figure 6-1 – Reference model for media processing services** + +# 7 Overview of media processing services + +Media processing services utilize a set of techniques including cloud computing, computing resource virtualization, and job queue processing to dynamically control and manage various computing resources, which improves scalability, flexibility and availability. + +Figure 7-1 illustrates the typical framework of media processing services. + +![Figure 7-1: Typical framework of media processing services. The diagram shows a hierarchical structure of management and resource layers. At the top, Service management is connected to a grid of media processing services including Face filters, Audio processing, Watermarks, Specific content detection, Information insertion, Video tagging, Speech recognition, and SD to HD conversion. Below this, Job management is connected to a queueing system with Job1, Job2, and Job3 entering from an 'Enter queue' and exiting through an 'Exit queue'. These jobs are then processed through three layers: Logical partition (represented by colored ovals), Virtual resource (represented by server icons), and Physical resource (represented by server icons). Resource management is shown at the bottom, connected to the Virtual and Physical resource layers and also connected to Job management.](f4fdd410cdb84df81274da55721e56fb_img.jpg) + +The diagram illustrates the typical framework of media processing services, organized into three main layers of management and resource allocation: + +- Service management:** Connected to a grid of media processing services including Face filters, Audio processing, Watermarks, Specific content detection, Information insertion, Video tagging, Speech recognition, SD to HD conversion, and others (...). +- Job management:** Connected to a queueing system where jobs (Job1, Job2, Job3) enter from an "Enter queue" and exit through an "Exit queue". +- Resource management:** Connected to the "Logical partition", "Virtual resource", and "Physical resource" layers. It also has a bidirectional connection to "Job management". + +The flow of data and jobs is as follows: Service management provides services to the job queue. Jobs enter the queue, are assigned to a logical partition, then to virtual resources, and finally to physical resources for processing. Resource management oversees the allocation of virtual and physical resources to the logical partitions. + +F.746.17(22) + +Figure 7-1: Typical framework of media processing services. The diagram shows a hierarchical structure of management and resource layers. At the top, Service management is connected to a grid of media processing services including Face filters, Audio processing, Watermarks, Specific content detection, Information insertion, Video tagging, Speech recognition, and SD to HD conversion. Below this, Job management is connected to a queueing system with Job1, Job2, and Job3 entering from an 'Enter queue' and exiting through an 'Exit queue'. These jobs are then processed through three layers: Logical partition (represented by colored ovals), Virtual resource (represented by server icons), and Physical resource (represented by server icons). Resource management is shown at the bottom, connected to the Virtual and Physical resource layers and also connected to Job management. + +**Figure 7-1 – Typical framework of media processing services** + +A typical media processing service generally consists of many physical resource nodes which are located in a resource pool supported by the cloud computing technology. The nodes are deployed as physical resources such as servers and storage which include many types of computing units, such as the central processing unit (CPU), graphics processing unit (GPU), field programmable gate array (FPGA), and the application specific integrated circuit (ASIC). These computing units are the main hardware to execute various media processing services. The physical resources are abstracted to the virtual resources such as virtual machines and virtual storage, which form a virtual resources pool utilizing virtualization technology. Computing units are virtualized into many virtual computing units. In order to provide service to multiple vendors, virtual resources and virtual computing units are aggregated and isolated according to the vendors' service requirements. Every logical partition is allocated to a specific job. Resource management capability is responsible to manage physical resources and virtualized resources. + +A job is a media processing job according to the vendor's requirements. A job consists of three main features which are input resources, media processing service type and parameters, and output video or data. Input resources are original media such as videos, files, or streams that vendors want to process. Media processing service types are mentioned in scenarios while parameters are set by the vendors according to their requirements. Output videos and data are products of media processing, which are various types or formats according to the requirements. + +Job management capability is responsible for managing the life span of a job. It is also responsible for job scheduling management, selecting computing resources, and logical partition to execute jobs based on workload and resource loads. Job scheduling can dispatch jobs to different queues for execution according to information such as media processing service type, job priority, submission time, etc. + +Service management capability is responsible for managing all kinds of media processing services. Service management provides a portal for vendors to select their services and set up parameters. Service management submits this information to the job management, job management then configures the related logical partitions and physical nodes, and executes the jobs by scheduling. + +The logical partition is managed by job management which controls and configures the logical partition in terms of the vendors' requirements. When the requirements of a logical partition are changing, job management requires resource management to perform functions such as adding or diminishing the virtual resources to the logical partition. + +# **8 Requirements** + +This clause addresses the requirements for media processing services. + +## **8.1 General requirements** + +Media processing service is a platform as a service (PaaS) service. It provides various types of services by concatenating or assembling various media related processing function modules. These related modules involve the mux layer, codec-stream layer, and transport protocol layer. It includes various processing conversions such as video image processing, audio processing, etc. All of these functions can be combined. Since each module requires different computing resources, network input/output (IO), storage, etc., thus different hardware environments required for each module need to be scheduled together for parallel or serial processing to meet the various functional requirements of users. Due to the inconsistent processing time of each function module, it may cause them to be influenced by each other. This could lead the entire processing link to be too long in time. Since some types of media processing services need more processing capabilities to be consumed than that of the common services, they can also meet the same efficiency requirements of the media processing. The system is recommended or required to provide the following capabilities: + +GR-01: The system is required to support the orchestration capability of the function modules based on the direct acyclic graph (DAG). By freely combining various processors, the functions could get integrity and variety. + +GR-02: The system is recommended to support the setting of the efficiency requirements of a certain function. For example, a 30-minute video can be processed within 5 minutes or 30 minutes. + +GR-03: The system is required to support heterogeneous capabilities to coordinate different types of devices to complete the same job, and to achieve a balance between performance and cost. Algorithms of media processing may be based on different computing units and hardware (such as CPU, GPU, FPGA, ASIC, etc.) It is necessary to complete specific processing sub-job by the corresponding computing units and hardware, and finally serially or parallel coordinate these processing results to complete a job. + +In general, the system is a distributed system, which is required to support streaming and parallel processing. + +## **8.2 Service provision requirements** + +SPR-01: The system is required to provide load-balance capability, which supports the concurrent processing of multiple jobs and avoid excessive load on a single hardware node, which affects the processing efficiency. + +SPR-02: The system is recommended to support elastic expansion and contraction of computing resources on time or on demand, thereby improving processing efficiency and reducing costs. + +SPR-03: The system is recommended to support grey releasing or upgrading among different versions. The version iteration cycle of cloud services is not fixed, and there are jobs running during release. In order to avoid the risk of potential defects introduced by the released version, which may cause the system to be fully affected, the grey releasing or upgrading can be performed in different arrangements such as groups, intervals, etc., and ways such as splitting dimensions on a user's importance/priorities, job function module, regions of service supporting/infrastructure, etc. + +SPR-04: The system is recommended to support the online job duplication ability for verification of functions. In order to ensure the stability of the version release and reduce the introduction of potential defects, A/B testing can be performed for online jobs and the test results can be compared. Some jobs may be suitable for running A/B testing both online, for an original job and offline for a duplicated job. + +SPR-05: The system is recommended to support cloud-native deployment capabilities with the container. It is convenient for operation, maintenance, management and control. + +SPR-06: The system is required to provide basic capabilities such as logging, end-to-end link inspection, capability usage / occupancy ratio monitoring and alarm, cost and billing, etc. + +## **8.3 Availability and scalability requirements** + +ASR-01: The system is required to provide unitized (in logical dimension) and rationalized (in physical dimension) deployment capabilities, and can provide high availability at the region-level (divided following service supporting / infrastructure, etc.) When an input file (such as a video segment) comes from region A with lower availability than region B where an MPS system is located, the availability of service after the MPS is processed may not be lower than that in region A. When the input file (such as a video segment) comes from region A with higher availability than region B where the MPS system is located, the availability will be the same as that in region B. + +ASR-02: The system is required to support extendibility and can dynamically expand or contract resources to quickly respond to service requirements when service scale quickly increases or decreases. + +ASR-03: The system is required to provide the alarm capability with pre-set values for alarm conditions or thresholds. + +ASR-04: The system is recommended to provide the back-off and degradation mechanism to ensure the stability of the basic functions for the services as far as possible under abnormal conditions. + +ASR-05: The system is recommended to support automatic or manual intervention for failed jobs, then get compensation via quick rerunning. It should prioritize the requirement to restore jobs from users as normal under a flexible strategy. + +## **8.4 Service management requirements** + +The system is required to provide capabilities such as operable, maintainable, management, control, configurable, traceable and troubleshooting. + +SMR-01: Operability, maintainability, manageability and controllability: It is required to include batch submission of jobs, job queue management (with job processing priority) and job cancellation, etc. + +SMR-02: Configurability: It is required to be changed only through configuration without modifying the hardware / software. + +SMR-03: Trackability: It is required to include the status of the overall capability usage / occupancy ratio and the status of specific jobs. + +SMR-04: Troubleshooting capability: According to the end-to-end link log, it is required to support the problem that could be located at the system-level and job-level. + +# **9 Security considerations** + +The system is required to provide security for accessing data. The address of the input video or output video processed by the system should be the address of the stored object bucket, rather than the uniform resource locator (URL) which is accessible on the public network, to ensure the security of the reading / writing data of the system. + +SC-01: The system is required to provide security for the storage service of data. The object storage service selected by the system should have permission configuration management and security control. + +SC-02: The system is required to provide security for the operation of the system. As a public cloud service, the system should have a unified account management system consistent with other cloud products and services so that it can build mechanisms such as account permission management, service subscribing or unsubscribing, and anti-attack. + +SC-03: The system is required to provide compliance for the user's service data. The system must not store the user's video content, and the logs related to the user's service data shall be accessed only within the specific area/region scope and must not be accessed across borders or areas/regions. The related log data should be cleared periodically. The alarm information related to the logs should also be desensitized. + +SC-04: The system is required to provide security for the operation and maintenance of the system itself. The core management and control application service of the system is the operation and maintenance system of the internal platform, which is separate from the public cloud management and control system. The internal operation and maintenance system should be operated under multi-role with different permission levels. + +SC-05: The system is recommended to provide encryption of the user's data. The system needs to provide an encryption (i.e., digital rights management (DRM)) function for video content, which can be used as content protection for distribution and playback. + +## Appendix I + +## Application scenarios + +(This appendix does not form an integral part of this Recommendation.) + +The typical service application scenarios for media processing services are the following. + +### **I.1 Face filtering** + +Face filtering service offers various filters and methods such as beautifying the facial features for anchors in the Internet broadcast services. + +### **I.2 Audio processing** + +Audio processing service offers audio mixing, in-ear monitoring, and other audio processing technologies for anchors in the Internet broadcast services. + +### **I.3 Watermarks** + +Watermarks service helps to add an identifying image or text as watermarks in the output video to safeguard video copyright for content providers. + +### **I.4 Specific content detection** + +Specific content detection service offers intelligent identification of specific content to reduce the risk of violations for content providers and users. + +### **I.5 Information insertion** + +Information insertion services can insert supplemental information in live streaming, which is suitable for live streaming quizzes, presentations, and other scenarios. + +### **I.6 Video tagging** + +Video tagging service offers the ability to provide video classification and multi-dimension content tags, which is suitable for personalized recommendation and searching services for content providers. + +### **I.7 Speech recognition** + +Speech recognition service offers multiple speech recognition solutions to transfer audio information into texts in various scenarios, such as meetings, courts and interviews. + +### **I.8 SD to HD conversion** + +Standard-definition (SD) to high-definition (HD) conversion service is able to convert standard-definition videos to high-definition videos by removing the compression noise of transcoding technology. + +### **I.9 Video smart cover** + +Video smart cover service offers an automatic capture of wonderful pictures such as the cover of the video content by artificial intelligence (AI) analysis, which can be applied to the automatic cover selection of new and old videos and improve the click rate and user experience of the video. + +### **I.10 Video format converting** + +In order to be compatible with different demands for video storage, transferring and playing, a video format converting service is needed. Video format includes a 3rd generation partnership project + +(3GPP) file format (3GP), audio video interleaved (AVI), flash video (FLV), moving picture experts group 4 (MP4), moving picture experts group audio layer 3 uniform resource locator – unicode transformation format (UTF-8) (M3U8), moving picture experts group (MPG/MPEG), etc. + +### **I.11 Resolution enhancement** + +The resolution of old historical videos is low. It is necessary to convert the low resolution of the original video into high resolution, which is called resolution enhancement. + +### **I.12 Video clipping and stitching** + +Video clipping and stitching service offers multiple videos to be edited, spliced, cut, and processed to make a new video programme. + +### **I.13 Picture in picture (PIP)** + +PIP service offers the service to embed one video as part of the frame into another video. + +### **I.14 Encrypting video** + +Encrypting video service offers an encrypted content to avoid specific downloading and disseminating. Services such as online education, paid video, and other paid contents use this service. + +## Bibliography + +- [b-ITU-T F.749.13] Recommendation ITU-T F.749.13 (2021), *Framework and requirements for civilian unmanned aerial vehicle flight control using artificial intelligence.* +- [b-ITU-T F.750] Recommendation ITU-T F.750 (2005), *Metadata framework.* +- [b-ITU-T Y.3500] Recommendation ITU-T Y.3500 (2014), *Information technology – Cloud computing – Overview and vocabulary.* + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.746.18-202406-I_PDF-E/raw.md b/marked/F/T-REC-F.746.18-202406-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..04576207b374e3b42bf5999b99a2fe551b68e4a0 --- /dev/null +++ b/marked/F/T-REC-F.746.18-202406-I_PDF-E/raw.md @@ -0,0 +1,415 @@ + + +# Recommendation **ITU-T F.746.18 (06/2024)** + +SERIES F: Non-telephone telecommunication services +Multimedia services + +--- + +**Requirements for interactive low-latency +multimedia transmission system over the +Internet** + +![ITU logo](390120de4fe440c42fea8154fcaad334_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with white lines representing latitude and longitude, and the letters 'ITU' in a bold, blue, sans-serif font superimposed on the globe. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.746.18 + +# Requirements for interactive low-latency multimedia transmission system over the Internet + +## Summary + +Recommendation ITU-T F.746.18 specifies the functional requirements of interactive low-latency multimedia transmission system (ILMTS) over the Internet. It identifies the key functional requirements from the different aspects, such as service level, application level and management level. According to those requirements, this Recommendation also introduces a reference framework of ILMTS. Some use cases are described in the appendix. + +Service providers who provide the interactive live content delivery service with low-latency media delivery can design a standardized real-time multimedia service system by following the requirements defined in this Recommendation. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.746.18 | 2024-06-13 | 16 | 11.1002/1000/15907 | + +## Keywords + +Interactive, low latency, multimedia transmission. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2024 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|-----|----------------------------------------------------------------------------------------------------------|----| +| 1 | Scope ..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 2 | +| 4 | Abbreviations and acronyms ..... | 2 | +| 5 | Conventions ..... | 2 | +| 6 | Introduction ..... | 3 | +| 7 | Requirements for interactive low-latency multimedia transmission system over the Internet..... | 4 | +| 7.1 | Service requirements ..... | 4 | +| 7.2 | Management requirements ..... | 4 | +| 7.3 | Accounting requirements..... | 5 | +| 7.4 | Networking requirements ..... | 5 | +| 7.5 | Quality of Service (QoS) and performance requirements ..... | 5 | +| 7.6 | Scalability requirements ..... | 5 | +| 8 | Reference framework..... | 6 | +| | Appendix I – Use cases for interactive low-latency multimedia transmission system over the Internet..... | 8 | +| I.1 | E-commerce live broadcasting ..... | 8 | +| I.2 | Multiparty visual meeting..... | 8 | +| I.3 | Live distance learning..... | 8 | +| I.4 | Video chat..... | 8 | +| I.5 | Cloud games ..... | 8 | +| I.6 | Cloud computer ..... | 9 | +| I.7 | Cloud rendering..... | 9 | +| | Bibliography..... | 10 | + + + +# Requirements for interactive low-latency multimedia transmission system over the Internet + +## 1 Scope + +This Recommendation defines the requirements for interactive low-latency multimedia transmission system (ILMTS) over the Internet, including service requirements, management requirements, accounting requirements, networking requirements, quality of service (QoS) and performance requirements, and scalability requirements, and specifies the use cases for the ILMTS. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions, which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +# 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 cloud-edge collaboration** [b-ITU-T F.747.13]: A process (or method) that coordinates cloud computing and edge computing, dynamically allocates required computing, algorithm models, data, or other resources, and jointly completes the same tasks (or objectives) agreed in advance. + +**3.1.2 container** [b-ITU-T Y.3535]: A set of software to provide isolation, resource control and portability for virtualization processing of an application. + +NOTE 1 – A container runs on the kernel in a bare-metal machine or virtual machine. + +NOTE 2 – "Application" implies business logic including a required library or binary to run in a container. + +**3.1.3 control plane** [b-ITU-T Y.2011]: The set of functions that controls the operation of entities in the stratum or layer under consideration, plus the functions required to support this control (see clause 8.1.1 of ITU-T Y.2011 for some details). + +**3.1.4 data plane** [b-ITU-T Y.2011]: The set of functions used to transfer data in the stratum or layer under consideration. + +**3.1.5 microservice** [b-ITU-T J.1301]: An architectural and organizational approach to software development where software is composed of small independent microservices that communicate over well-defined application programming interfaces. Microservice architecture makes applications easier to scale and faster to develop, enabling innovation and accelerating time-to-market for new features. + +**3.1.6 microservice architecture** [b-ISO/IEC TS 23167]: Design approach that divides an application into a set of microservices. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 access scheduling:** A mechanism used to select the most appropriate network access point for a client according to its access request and network status. + +**3.2.2 cloud-edge-client collaboration:** An extended cloud-edge collaboration that involves clients or terminal devices as a participation part of a joint task processing. + +**3.2.3 container cloud:** A cloud computing platform based on container technology. + +**3.2.4 routing scheduling:** A mechanism used to select the optimal relay node and data forwarding path based on the network topology and network traffic status. + +**3.2.5 thin terminal:** A lightweight terminal device with limited local functions and capabilities for computing, whose applications and storage functions are mostly deployed remotely. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-------|--------------------------------------------------------| +| CDN | Content Delivery Network | +| CPU | Central Processing Unit | +| HLS | HTTP Live Streaming | +| HTTP | Hypertext Transfer Protocol | +| ILMTS | Interactive Low-Latency Multimedia Transmission System | +| P2P | Peer-to-Peer | +| PaaS | Platform as a Service | +| QoS | Quality of Service | +| RTMP | Real Time Messaging Protocol | +| SLA | Service Level Agreement | +| TCP | Transmission Control Protocol | +| VM | Virtual Machine | +| VoIP | Voice over Internet Protocol | +| XR | Extended Reality | + +## 5 Conventions + +In this Recommendation: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. +- The keywords "can optionally" and "may" indicate an optional requirement which is permissible, without implying any sense of being recommended. These terms are not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +# 6 Introduction + +As the mobile Internet rushes in, a variety of video live streaming applications begin to emerge. In addition, online live shopping, online live education and other real-time communication multimedia applications, which are known as interactive live streaming applications, have become popular. The "interactivity" brings another practical requirement of live streaming business. Different from the one-way transmission of traditional media streams, interactive live streaming applications require new two-way channels to meet interactive requirements. Traditional live video systems over the Internet usually utilize content delivery network (CDN) for video distribution. The mainstream protocols adopted include real time messaging protocol (RTMP), hypertext transfer protocol (HTTP)/FLV, HTTP live streaming (HLS), etc., and there have been numerous diversified optimizations for traditional live broadcasting application scenarios. However, CDN adopts a tree-like hierarchical structure (which can be referenced from [b-ITU-T H.644.3]) and uses transmission control protocol (TCP) protocol for video transmission. Consequently, it is desirable to significantly reduce the end-to-end transmission delay. + +A two-way low-latency multimedia delivery system over the Internet is required to satisfy the above requirements. In this Recommendation, an interactive low-latency multimedia transmission system (ILMTS) over the Internet is introduced. The ILMTS has many interactive network audio/video-based application scenarios, such as e-commerce live broadcasting, online multi-party meeting, live distance learning, video chat and cloud games. + +Figure 6-1 presents an overview of ILMTS deployment. + +![Figure 6-1: Example of ILMTS over Internet. The diagram illustrates the ILMTS architecture. At the top, a 'Control centre' (represented by three server icons) is connected to a central 'Bi-direction delivery network' (a cloud containing five 'Relay' nodes). Below the control centre, two zones, 'Zone A' and 'Zone B', are shown. Each zone contains an 'Access point' (represented by a tower icon) which is connected to multiple 'Client' devices (represented by icons of a laptop, monitor, and smartphone). The 'Access point' in Zone A is connected to the 'Bi-direction delivery network' via a 'Relay' node, and similarly for Zone B. The 'Bi-direction delivery network' is connected to the 'Internet' (represented by a cloud containing several nodes) via an 'Overlay' connection. The entire ILMTS structure is enclosed in a dashed box labeled 'ILMTS'.](7a0db9703b68b3d06cdaeefc084c0006_img.jpg) + +The diagram shows the ILMTS architecture. At the top, a 'Control centre' (three server icons) is connected to a central 'Bi-direction delivery network' (cloud with five 'Relay' nodes). Below the control centre, two zones, 'Zone A' and 'Zone B', are shown. Each zone contains an 'Access point' (tower icon) connected to multiple 'Client' devices (laptop, monitor, smartphone icons). The 'Access point' in Zone A is connected to the 'Bi-direction delivery network' via a 'Relay' node, and similarly for Zone B. The 'Bi-direction delivery network' is connected to the 'Internet' (cloud with nodes) via an 'Overlay' connection. The entire ILMTS structure is enclosed in a dashed box labeled 'ILMTS'. + +Figure 6-1: Example of ILMTS over Internet. The diagram illustrates the ILMTS architecture. At the top, a 'Control centre' (represented by three server icons) is connected to a central 'Bi-direction delivery network' (a cloud containing five 'Relay' nodes). Below the control centre, two zones, 'Zone A' and 'Zone B', are shown. Each zone contains an 'Access point' (represented by a tower icon) which is connected to multiple 'Client' devices (represented by icons of a laptop, monitor, and smartphone). The 'Access point' in Zone A is connected to the 'Bi-direction delivery network' via a 'Relay' node, and similarly for Zone B. The 'Bi-direction delivery network' is connected to the 'Internet' (represented by a cloud containing several nodes) via an 'Overlay' connection. The entire ILMTS structure is enclosed in a dashed box labeled 'ILMTS'. + +F.746.18(24) + +**Figure 6-1 – Example of ILMTS over Internet** + +ILMTS adopts a hybrid network structure that combines the hierarchy and peer-to-peer (P2P) networking modes. Meanwhile, the multiple efficient video transmission technologies can be used for optimizing the end-to-end performance of video delivery, including dynamic P2P routing planning, low-latency content transport protocol and distributed edge content processing. Through + +elaborate design and fine-tuning of the delivery mechanism, ILMTS can achieve an end-to-end transmission delay of hundreds of milliseconds. + +# **7 Requirements for interactive low-latency multimedia transmission system over the Internet** + +### **7.1 Service requirements** + +[SR-01]: ILMTS is recommended to support the convergence of cloud and network technologies, which means that ILMTS can be composed of overlay network (cloud) and infrastructure network (network). + +[SR-02]: The server-side functions of ILMTS, specifically the control, relay and access functions outlined in this Recommendation, are recommended to be deployed in an overlay network utilizing a cloud-based approach. + +[SR-03]: ILMTS is recommended to support the collaboration of cloud-edge-client, that is, the functional components of ILMTS can be deployed in the different areas of network and are capable of working collaboratively by interaction of control signalling and interaction of media transmission. + +[SR-04]: ILMTS is recommended to support the layered architecture by the separation of data plane and control plane. + +## **7.2 Management requirements** + +[MGR-01]: The control plane of ILMTS is recommended to be deployed in the central cloud, as a control centre. + +[MGR-02]: The control plane of ILMTS is recommended to support the management and programming of access point and relay node, which includes access scheduling and routing scheduling. + +[MGR-03]: Access scheduling is recommended to support the following aspects, but not limited to: + +- Support data aggregation: to be capable of collecting the information of access point, the network information detected between client and access point. +- Support maintenance of static/dynamic IP address database: to be capable of filtering the aggregated data and categorizing the data according to the position of the access point. +- Support intelligent allocation of access point: to be capable of selecting the optimal access point according to the parameter of service request and pre-processed data. +- Support query of candidate access point: to be capable of providing the most appropriate access point to the client. + +[MGR-04]: The routing scheduling is recommended to support the following aspects, but not limited to: + +- Support data aggregation: to be capable of collecting the information of the relay node, the network information detected within relay nodes. +- Support data pre-process: to be capable of filtering the aggregated data and categorizing the data according to the different conditions. +- Support intelligent routing among the relay nodes: to be capable of computing optimal network routing and dynamically generating of quantified network topology, according to the pre-processed data. +- Support query of routing information: to be capable of providing the customized query of routing between relay nodes. + +### 7.3 Accounting requirements + +[AR-01]: ILMTS is recommended to support the diversified billing models based on the accounting factors such as duration, traffic and capacity types. + +[AR-02]: The billing model can optionally be divided into different levels and package rates. For example, it can be billed according to monthly/annual subscription, license, service level agreement (SLA), etc. + +### 7.4 Networking requirements + +[NTR-01]: ILMTS is recommended to deploy multiple relay nodes and access points for a high-quality and high-reliability media transmission network. + +[NTR-02]: The networking of ILMTS relay nodes and access points is recommended to be harmonized with the network architecture of infrastructure transport network. + +[NTR-03]: The access point of ILMTS is recommended to be deployed in the edge network, as the geographic location closest to the end-user. It is recommended to support the following aspects, but not limited to: + +- It composes of multiple light-weight nodes and covers all areas within ILMTS. +- It is capable of transmitting the detection information of client accessing and the information of access point itself to the control centre. +- It is capable of high-reliability and high-stability accessing. + +[NTR-04]: The Relay node of ILMTS is recommended to be deployed in the backbone network, as the data relaying between different zones. It is recommended to support the following aspects, but not limited to: + +- It is a large-scaled node composed of high-performance servers and a large quantity of storage. +- It supports multipoint-to-multipoint network connection. +- It is capable of transmitting the network detection information among the relay nodes and the information of relay node itself to the control centre. +- It is capable of high-speed and high-quality data forwarding. + +### 7.5 Quality of Service (QoS) and performance requirements + +[QPR-01]: ILMTS can optionally support the QoS guaranteed by collaboration of overlay network layer (cloud) and infrastructure network layer (network). + +### 7.6 Scalability requirements + +[SCR-01]: ILMTS is recommended to run its system resource on container cloud and microservice architecture, which is based on a cloud-based infrastructure. + +[SCR-02]: The server hardware resource of ILMTS is recommended to be cloudified so that it is able to construct a cloud-based infrastructure like platform as a service (PaaS). + +[SCR-03]: The system resource of ILMTS is recommended to support the flexible resource scaling. The scaling strategy can be: + +- Adjust the quantity of running instance dynamically based on the scale of service capability. +- Configure the strategy of scaling and threshold for each specific system element, based on the monitoring of the related element resource occupancy status, such as the central processing unit (CPU)/memory occupancy rate. +- Periodical resource scalability. + +[SCR-04]: ILMTS can optionally adopt the docker model to reduce the wastage of conventional virtual machine (VM) resources. + +# 8 Reference framework + +The entire ILMTS consists of four parts: Control centre, relay node, access points (also called edge node) and client. To provide a low-latency interactive media delivery, a dynamic media delivery path optimization mechanism is required to be applied and it relies on the collaboration of all components of ILMTS. + +Figure 8-1 presents the high-level framework of ILMTS. + +![Figure 8-1 – Reference framework diagram showing the four layers of ILMTS: Control centre, Relay node, Access point (Edge node), and End-user (Client). The Control centre is part of the Control plane, while the other three are part of the Data plane.](1439cb942d9e363bbb3161b5540dd8c6_img.jpg) + +The diagram illustrates the high-level framework of ILMTS, organized into four stacked layers. The top layer, 'Control centre', is grouped under the label 'Control plane' and contains three modules: 'Network awareness', 'Path planning', and 'Scheduling management'. The subsequent three layers are grouped under the label 'Data plane'. These are: 'Relay node' (containing 'Data forwarding', 'Data receiving', 'Path management', and 'Forwarding detection'), 'Access point (Edge node)' (containing 'Data forwarding', 'Data receiving', 'Path management', and 'Access detection'), and 'End-user (Client)' (containing 'Data generation', 'Content processing', 'Path management', and 'Client detection'). A small code 'F.746.18(24)' is visible at the bottom right of the diagram. + +Figure 8-1 – Reference framework diagram showing the four layers of ILMTS: Control centre, Relay node, Access point (Edge node), and End-user (Client). The Control centre is part of the Control plane, while the other three are part of the Data plane. + +**Figure 8-1 – Reference framework** + +### – Control centre + +The control centre is responsible for the management of diverse data. It controls the relay node and access point for data forwarding and routing path planning. The control centre is capable of network performance awareness and performs global routing path management and scheduling. Typically, network status and the running status of the relay node, access point and client should be processed in the control centre, as the basis of routing path planning. + +### – Relay node + +The relay node can be regarded as the application-layer data forwarding router, capable of forwarding data with high-speed and high-quality among relay nodes. It is further composed of modules for forwarding detection, data forwarding, data receiving and path management. + +### – Access point + +Access point, also called edge node, is responsible for connecting the client to the ILMTS with high-reliability and high-stability. The access point can be further composed of access detection, path management, data receiving and data forwarding modules. + +### – Client (end-user/terminal device) + +The client consists of a variety of physical devices or software applications that support the generation, transmission and display of media data. For example, in video conferencing services, the client may include PC clients, mobile phone clients, and conference terminals in conference rooms. In cloud computing services, it could encompass thin terminals, PC, back-end servers, etc. For extended reality (XR) services, it may include glasses terminals, head-mounted terminals and back-end servers. The client can be further subdivided into modules of client detection, path management, content processing, data forwarding. + +## Appendix I + +### Use cases for interactive low-latency multimedia transmission system over the Internet + +(This appendix does not form an integral part of this Recommendation.) + +### I.1 E-commerce live broadcasting + +E-commerce live broadcast refers to an enhanced application of live video broadcast. It adds interactive functions in live video broadcast, including large concurrent live broadcast oriented to users and two-way video interaction with viewers. Interactive enabled live broadcast can establish a low-latency video interaction between the main broadcast and viewers at any time. It has made essential changes in viewing experience and presentation effects and represents the development trend of live video broadcast applications. E-commerce live broadcast, online interactive education, and other popular forms on the network are typical application scenarios of interactive live broadcast. + +### I.2 Multiparty visual meeting + +A multiparty visual meeting refers to a simulated face-to-face conference between two or more individuals or groups in different places through a low-latency video/audio transport system and multimedia devices. Video conferencing is a kind of real-time telepresence, where in addition to seeing the caller and verbal communication, the person's expression and actions can be seen, making people in different places communicate as if they were in the same room. + +### I.3 Live distance learning + +Live distance learning refers to an education mode provided without the teacher and students being physically present in the same place. The traditional education mode usually offers correspondence courses materials sent to a student by paper or mail, and assignments were completed online or returned to the teacher by mail. The live distance learning programs can provide the opportunities for the students to obtain the online course materials while they are involving in a real-time online class. They can also use interactive tools, such as a virtual white board, to discuss with other classmates and complete the online assignments within a limited time. + +### I.4 Video chat + +A video chat is a phone call that uses an Internet connection, sometimes called voice over internet protocol (VoIP), which utilizes video to transmit a live picture of the person making the call. Video chats are made using a computer's webcam or other electronic hardware devices with a video-capable camera, like a smartphone, tablet or video-capable phone system. In some cases, video calling can also provide interactive elements such as live white board (similar to the video conference) while two or more users are still in a video call. + +### I.5 Cloud games + +A cloud game (cloud gaming), which may also be referred to as game on demand (gaming on demand), is an online game technology based on a cloud computing technology. Cloud gaming technology enables light-end devices (thin client) with relatively limited graphics and data computing capabilities to a run high-quality game. In a cloud gaming scenario, the game is not performed on a player's game terminal, but runs on a cloud server. The cloud server renders the game scenario into a video and audio stream and transmits the video and audio stream to the player's game terminal over an ultra-low latency network. Gamers do not need to have powerful graphic computing and data processing capabilities, but only need to have basic streaming media playback capabilities and the capability of getting instructions from gamers and sending them to the cloud server. + +### **I.6 Cloud computer** + +A cloud computer, also called a cloud desktop, is a new model that replaces traditional computers. Compared with a traditional computer, a cloud computer does not have hardware such as a CPU, memory and a hard disk. All the hardware is integrated in a data centre in the cloud. Users can access their personal desktop, data and applications from a single, compact client device that can be connected anywhere where there is a network, connecting to a keyboard, mouse and monitor the same way as with an ordinary personal computer. The user's operation and the operation results should be reflected in a very short time between user and the data centre in the cloud. Therefore, the control signal and the media steam should be transmitted relied on an interactive low-latency media transport system. + +### **I.7 Cloud rendering** + +Cloud rendering means that a 3D program is rendered on a remote server. A user client uses web software or a local 3D program to click a "cloud rendering" button and access resources through an interactive low-latency Internet access. An instruction is sent from the user client, and the server executes the corresponding rendering task according to the instruction. A rendering result image is sent back to the user client for display. + +## Bibliography + +- [b-ITU-T F.747.13] Recommendation ITU-T F.747.13 (2023), *Requirements and reference framework of cloud-edge collaboration in industrial machine vision systems.* +- [b-ITU-T H.644.3] Recommendation ITU-T H.644.3 (2020), *Functional architecture of multimedia content delivery networks.* +- [b-ITU-T J.1301] Recommendation ITU-T J.1301 (2021), *Specification of cloud-based converged media service to support Internet protocol and broadcast cable television – Requirements.* +- [b-ITU-T J.1305] Recommendation J.1305 (2023), *Requirements for microservices architecture for audiovisual media in the converged media cloud.* +- [b-ITU-T Y.2011] Recommendation ITU-T Y.2011 (2004), *General principles and general reference model for Next Generation Networks.* +- [b-ITU-T Y.3535] Recommendation ITU-T Y.3535 (2022), *Cloud computing – Functional requirements for a container.* +- [b-ISO/IEC TS 23167] Technical Specification ISO/IEC TS 23167:2020, *Information technology – Cloud computing – Common technologies and techniques.* + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.746.19-202512-I_PDF-E/724c7777b608e53be38b12b6fb3c43bc_img.jpg b/marked/F/T-REC-F.746.19-202512-I_PDF-E/724c7777b608e53be38b12b6fb3c43bc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..302496167df9c5b6212a7079578f9075e160d12a --- /dev/null +++ b/marked/F/T-REC-F.746.19-202512-I_PDF-E/724c7777b608e53be38b12b6fb3c43bc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2f6151d6e2e77de45a13f336c79f4206ef6c3dff1c7ced2c3da5103c5014f54e +size 56422 diff --git a/marked/F/T-REC-F.746.19-202512-I_PDF-E/731f533b0599c8e42a063f06e4332045_img.jpg b/marked/F/T-REC-F.746.19-202512-I_PDF-E/731f533b0599c8e42a063f06e4332045_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d31403661cc65fa06f1b2c1743d67015b5876619 --- /dev/null +++ b/marked/F/T-REC-F.746.19-202512-I_PDF-E/731f533b0599c8e42a063f06e4332045_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fca75774424a7b486aef2ecebb9dc438910180c9941106fa1394d3afa10c535f +size 28844 diff --git a/marked/F/T-REC-F.746.19-202512-I_PDF-E/7a0db9703b68b3d06cdaeefc084c0006_img.jpg b/marked/F/T-REC-F.746.19-202512-I_PDF-E/7a0db9703b68b3d06cdaeefc084c0006_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..76620e543b38dda382a5ddc8ba51dfe3817ad456 --- /dev/null +++ b/marked/F/T-REC-F.746.19-202512-I_PDF-E/7a0db9703b68b3d06cdaeefc084c0006_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b6f38408fcc9c28f442a96213a6c64469f5be874a234197447661e671c7ce007 +size 25251 diff --git a/marked/F/T-REC-F.746.19-202512-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.746.19-202512-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ac295b1b0c7f14b1c5a33ef6682b1b017c812477 --- /dev/null +++ b/marked/F/T-REC-F.746.19-202512-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37aed4fc114a39843438175326e0b7e1d09a9d6075723d82fea32040fd560ee4 +size 7192 diff --git a/marked/F/T-REC-F.746.19-202512-I_PDF-E/967c30813761a8952ecc5e16bf42ea45_img.jpg b/marked/F/T-REC-F.746.19-202512-I_PDF-E/967c30813761a8952ecc5e16bf42ea45_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..54a22a209dd9219e620e85a462f5249c6bf78771 --- /dev/null +++ b/marked/F/T-REC-F.746.19-202512-I_PDF-E/967c30813761a8952ecc5e16bf42ea45_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e8bbfdc03ddd26dc2c8259ebcf90bfb39068ac12643751ea59bfed5a2c51ddfd +size 36934 diff --git a/marked/F/T-REC-F.746.19-202512-I_PDF-E/c036e2540a94b31357ceb0002f0cacab_img.jpg b/marked/F/T-REC-F.746.19-202512-I_PDF-E/c036e2540a94b31357ceb0002f0cacab_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..e8cb7017dbd6981b9c85a9ae5f1632ba14ec43b6 --- /dev/null +++ b/marked/F/T-REC-F.746.19-202512-I_PDF-E/c036e2540a94b31357ceb0002f0cacab_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3f2b85fbfa7a6c93b6243f197464a25ce9d6c77354442046bfdb7304b519d9d7 +size 36087 diff --git a/marked/F/T-REC-F.746.2-201402-I_PDF-E/raw.md b/marked/F/T-REC-F.746.2-201402-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..a368ecb905cd10cfbb3c5186345d4861dc49fdae --- /dev/null +++ b/marked/F/T-REC-F.746.2-201402-I_PDF-E/raw.md @@ -0,0 +1,570 @@ + + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.746.2** + +(02/2014) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Audiovisual services + +--- + +**Service description for user data exchange +between functional components in network +entities and terminals** + +Recommendation ITU-T F.746.2 + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| AUDIOVISUAL SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.746.2 + +# Service description for user data exchange between functional components in network entities and terminals + +## Summary + +Recommendation ITU-T F.746.2 describes user data exchange scenarios and a basic approach of exchanging user data between terminals or between terminals and network entities by using data packages encapsulating formatted user data. The functionality of the components which implement the data exchange and general guidance of generating and parsing such data package are given as well. This Recommendation also describes the preferred extensible markup language (XML) representations used when encapsulating corresponding user data, including, but not limited to, contacts, messages and other user data categories. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.746.2 | 2014-02-13 | 16 | 11.1002/1000/12050 | + +## Keywords + +Data exchanging agent (DXA), user data exchange, XML. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2014 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|--------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 2 | +| 3.1 Terms defined elsewhere ..... | 2 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 3 | +| 6 General..... | 3 | +| 7 Scenarios of user data exchange ..... | 3 | +| 7.1 User data grouping ..... | 3 | +| 7.2 Scenario description ..... | 4 | +| 8 User data exchange method ..... | 4 | +| 8.1 General procedure ..... | 4 | +| 8.2 Data package ..... | 5 | +| 9 Functionality of data exchange functional components ..... | 6 | +| 9.1 Basic functionality ..... | 6 | +| 9.2 Security consideration ..... | 6 | +| 10 Preferred XML user data syntax..... | 7 | +| 10.1 General ..... | 7 | +| 10.2 Contacts data syntax ..... | 7 | +| 10.3 Calendar data syntax..... | 7 | +| 10.4 Messaging data syntax..... | 7 | +| 10.5 Other ..... | 10 | + + + +# Service description for user data exchange between functional components in network entities and terminals + +# 1 Scope + +The general purpose of this Recommendation is to describe user data exchange scenarios and a basic approach to exchange user data between terminals or between terminals and network entities by using data packages encapsulating formatted user data. The functionality of the components which implement the data exchange and general guidance of generating and parsing such data packages are given as well. This Recommendation also describes the preferred extensible markup language (XML) representations used when encapsulating corresponding user data, including, but not limited, to contacts, messages and other user data categories. + +This Recommendation applies to terminals and network entities which choose to implement user data exchanging functions as described in this Recommendation. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +- [ETSI TS 123 140] ETSI TS 123 140 V6.16.0 (2009), *Digital cellular telecommunications system (Phase 2+); Universal Mobile Telecommunications System (UMTS); Multimedia Messaging Service (MMS); Functional description; Stage 2* (3GPP TS 23.140 version 6.16.0 Release 6). +- [ETSI TS 126 140] ETSI TS 126 140 V7.1.0 (2007), *Digital cellular telecommunications system (Phase 2+); Universal Mobile Telecommunications System (UMTS); Multimedia Messaging Service (MMS); Media formats and codes* (3GPP TS 26.140 version 7.1.0 Release 7). +- [ETSI TS 127 005] ETSI TS 127 005 V3.2.0 (2002), *Digital cellular telecommunications system (Phase 2+); Universal Mobile Telecommunications System (UMTS); Use of Data Terminal Equipment – Data Circuit terminating Equipment (DTE – DCE) interface for Short Message Service (SMS) and Cell Broadcast Service (CBS)* (3GPP TS 27.005 version 3.2.0 Release 1999). +- [IETF RFC 5545] IETF RFC 5545 (2009), *Internet Calendaring and Scheduling Core Object Specification (iCalendar)*. +- [IETF RFC 6321] IETF RFC 6321 (2011), *xCal: The XML Format for iCalendar*. +- [IETF RFC 6350] IETF RFC 6350 (2011), *vCard Format Specification*. +- [IETF RFC 6351] IETF RFC 6351 (2011), *xCard: vCard XML Representation*. +- [W3C XML] W3C XML 1.0 (2008), *Extensible Markup Language (XML) 1.0 (Fifth Edition)*. +- [W3C XMLEnc] W3C XML Encryption (2002), *XML Encryption Syntax and Processing*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +None. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 data exchanging agent:** Logic in terminals or entities where the main function is to put data into or extract data from the data package described in this Recommendation. Furthermore, it monitors the data exchange process and takes actions when specific user data is encountered after the data exchange process. A typical form of the functional component is an extensible markup language (XML) parser combined with application-dependent logic. + +**3.2.2 depository:** A depository is a location where the user data could be stored or kept so that it could be accessed later. It can take various forms such as internal or external memory of terminals, network storage servers and other storage devices. + +**3.2.3 user data:** "User data" is the data that is created or stored in terminals or network entities by the users, and is related to the private information, historical operations performed by the user, personal tendencies or favourite user settings. It is independent of whether this data is transported by terminal or network entities, and has no correlation with the counterpart of "Control data". + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|-------------------------------| +| DRM | Digital Rights Management | +| DXA | Data exchanging Agent | +| FTP | File Transfer Protocol | +| HTTP | Hypertext Transfer Protocol | +| IM | Instant Message | +| MMS | Multimedia Message Service | +| RAM | Random Access Memory | +| ROM | Read Only Memory | +| SD | Secure Digital | +| SIM | Subscriber Identity Module | +| SMS | Short Message Service | +| URL | Uniform Resource Locator | +| USB | Universal Serial Bus | +| USIM | User Services Identity Module | +| W3C | World Wide Web Consortium | +| XML | Extensible Markup Language | + +# 5 Conventions + +None. + +# 6 General + +Under many circumstances there is a need to exchange user data between terminals or network entities and another data depository. For example, a user may want to replace a terminal with a new terminal without losing personal data, such as contacts and received messages, stored in the original one. The data exchange process might be initiated by the user manually or by a network application automatically, and multiple categories of data might be involved in an exchange process. However, the terminals or entities acting as the source and destination in the exchange process may use different operating systems or different internal data expressions for the same kind of data. Without a consistent approach, accurate exchange of the user data is difficult to realize. + +To facilitate the exchange process and make it as simple as possible for the users, this Recommendation describes an exchange approach by using a data package as a container to encapsulate user data and specifies the recommended data syntax for parsing or generating the data package. + +# 7 Scenarios of user data exchange + +## 7.1 User data grouping + +The user data includes the user's personal data, device configuration parameters and personalized application data, and can be stored in terminals or other entities, no matter the device type. User data may take various forms such as binary, text or multimedia files depending on the corresponding data category. It could be stored in different locations such as internal memory (RAM, ROM, etc.) and external memory (SIM/USIM, SD card, hard disk, etc.). + +### 7.1.1 Private user data + +Private user data is created by the user for a special purpose or generated by the users operations and contains important information about the user. Examples of private user data include: + +- Contact information (including phonebook) +- Messages (including SMS, MMS, IM and e-mail) +- Calendar information +- Favourite uniform resource locator (URL) lists and bookmarks +- Call and browsing history + +### 7.1.2 Device configuration data + +Device configuration data is the data that a terminal allows users to access and modify, so as to personalize what the terminal displays and the way it behaves. Examples of device configuration data include: + +- Terminal theme, fonts, wallpaper and screensaver +- Language settings +- Date/Time and location settings +- Energy saving settings +- Accessibility settings + +### 7.1.3 Personalized application data + +Personalized application data is the data that applications allow users to access and modify, so as to personalize the way an application operates or behaves, including: + +- Service dependent settings or parameters +- Application settings or application files + +## 7.2 Scenario description + +A basic user data exchange scenario is that a terminal or an entity wants to receive discrete sets of user data from another terminal or to send data to another terminal. For example, a mobile phone desires to retrieve all the contacts and calendar data stored in another terminal. Another example would be that an entity in the network needs to receive a status report or needs to retrieve the browsing history of the streaming media player on the terminal to use for later statistical purposes. + +In those cases, typically user data is of small size and may be stored in different formats in the two terminals because of different operating platforms or different data management applications. To exchange data, normally the data providing party will generate a data package which contains multiple categories of data, and the other party receives the data package and performs data analysis, followed by certain operations according to the original purpose of this data acquisition process. + +# 8 User data exchange method + +## 8.1 General procedure + +For a typical user data exchange scenario, the data exchange process can be summarized into several steps: Export, transfer and analyse. + +- Export: Export selected user data from party A as an accessible data package. +- Transfer: Party A transfers the selected package to party B by any approach. This step does not need to be in real time. +- Analyse: Party B opens the package and performs the analysis. It retrieves all the needed data or supported data according to the pre-defined data format or expression. This is followed by certain operations based on the retrieved data. + +The data package which contains all the user data forms the basis of data exchange. To ensure compatibility, the data package is defined in an extensible markup language (XML) format as specified in [W3C XML] so that it may be supported and applied in various applications across different operating systems and terminals. Furthermore, the data package is able to encapsulate multiple categories of data and is also extendable to new data categories. This Recommendation uses XML for the user data in the data package as it is designed to describe structured data, is naturally extendable and widely supported by many applications and operating systems. + +The transfer step encompasses transferring a data package between a terminal and another entity. There are various techniques for transferring files (such as the XML data package files), including file transfer protocol (FTP), hypertext transfer protocol (HTTP), Bluetooth and infrared via short ranged wireless communication and universal serial bus (USB) storage via wired connections. The transfer may be real-time or may be a store and forward arrangement. There may be extra processing during the data package transfer. The data package may also be moved to another depository and accessed at a later time as necessary. + +The export and analyse steps could be implemented by functional components. These components are deployed in the terminals or the entities acting as data exchange agents to manage the data exchange. The data exchange process is illustrated in Figure 1. + +![Figure 1 – Data exchange procedure diagram. The diagram shows two 'Terminal or entities' boxes, each enclosed in a dashed rectangle. The left box contains 'Required data' pointing to a 'DXA (Export)' box, which points to a 'Data package' (represented by a document icon). A dashed arrow labeled 'Transferring or processing' connects the first 'Data package' to a second 'Data package'. The second 'Data package' points to a 'DXA (Analyse)' box, which points to 'Retrieved data' inside the right 'Terminal or entities' box. The text 'F.746.2(14)_F01' is at the bottom right.](cfef993dcc8fb513de79eb1f93cf26ae_img.jpg) + +Figure 1 – Data exchange procedure diagram. The diagram shows two 'Terminal or entities' boxes, each enclosed in a dashed rectangle. The left box contains 'Required data' pointing to a 'DXA (Export)' box, which points to a 'Data package' (represented by a document icon). A dashed arrow labeled 'Transferring or processing' connects the first 'Data package' to a second 'Data package'. The second 'Data package' points to a 'DXA (Analyse)' box, which points to 'Retrieved data' inside the right 'Terminal or entities' box. The text 'F.746.2(14)\_F01' is at the bottom right. + +**Figure 1 – Data exchange procedure** + +## 8.2 Data package + +### 8.2.1 Example + +The data package is an XML encoded structure. A simple example of a data package containing the contacts information, calendar records and instant message history, which is used for exchanging private user data between two terminals, is illustrated as follows: + +``` + + + 1.0 + 0 + + 2010-08-10T14:30:00Z + MobilePhone_Vendor_X + TypeNumber_2321C + Data_export_Manager + + + + + + + + + + + + + +``` + +### 8.2.2 element + +The root element in this XML structure is which is also the beginning of the XML structure. When packing data into the element, the format should follow the XML schema 1.1 description of the element, which follows [W3C XSD], as listed below: + +``` + + + + + + + + + + + + +``` + +``` + + + + + + + + + + + + + + + + +``` + +In the element, there are several child elements: + +- : this element indicates the version of this XML file. +- : this element indicates the encryption status of this XML file. +- : this element contains the following descriptive information about the data package: + - : Contains a string indicating the creation time of this package according to the format + - : Contains a string indicating information about the terminal vendor. + - : Contains a string indicating information about the terminal itself. + - : Contains a string indicating information about the application generating the data package. +- : this element contains the main body of the data to be exchanged. + +### 8.2.3 element + +The is the main element which contains the data to be exchanged. Any XML formulated user data that follows the corresponding XML representations defined in the existing specifications or defined by applications itself may be stored in this element. + +# 9 Functionality of data exchange functional components + +## 9.1 Basic functionality + +The data exchanging agent (DXA) should be able to generate and parse the XML data package. When parsing the XML data package, the DXA should ignore unsupported data in the data package without interrupting the data exchange process or generating an unexpected exception error. + +The DXA should be capable of supporting the updating of the data set definition. This functionality may be realized by updating the corresponding XML definition of the supported data. + +The DXA should, when appropriate, avoid data collision when storing the retrieved data from the exchange process, possibly by noting the existing user data records. When data collision might occur, actions such as prompts to user or user data merging according to predefined rules should be taken. + +## 9.2 Security consideration + +The decision to take action to ensure the security of user data exchange depends on the application implementation and the data sensitivity. For some real-time implementations of user data exchange, security issues could be addressed in the transport layer by secure file transfer. For other non-real- + +time implementations, for example where the generated data package would be temporarily stored in public websites, simple password protection or XML encryption according to [W3C XMLEnc] may be applied on the data package to prevent unauthorized access of the user data. + +When XML encryption is used to create the data package, the DXA should be capable of encrypting the user data and decrypting the XML encrypted data according to the [W3C XMLEnc]. + +# 10 Preferred XML user data syntax + +## 10.1 General + +Whether the data retrieved from the data package can be successfully understood will depend on the XML user data syntax. If the XML user data syntax in the data package perfectly matches the internal storage format used by the terminal or the entities which receive the data package, then the user data exchange can be performed without any compatibility problem, otherwise the exchange process may suffer data loss. + +User data typically has its own distinctive format, which may be different between multiple terminals and/or operating systems. Moreover some kinds of user data do not use popular or globally-accepted formats. This Recommendation defines XML syntax for several common forms of user data in order to improve interoperability between the entities. + +## 10.2 Contacts data syntax + +For representing and exchanging contact and relevant information about individuals and other entities, IETF has specified vCard data format in [IETF RFC 6350], which is a text-based format. + +The XML representation for vCard is defined by [IETF RFC 6351]. The underlying data structure is exactly the same as the original vCard text format, which supplies the one-to-one mapping between the original vCard format and the XML representation. + +The contacts XML representation should follow the vCard XML format, which is specified in [IETF RFC 6351]. + +## 10.3 Calendar data syntax + +For the calendaring and scheduling data, IETF has defined the iCalendar data format in [IETF RFC 5545]. The iCalendar data format is a widely deployed interchange format. While many applications and services on the Internet support and generate calendar data, similar to the vCard format, an iCalendar XML representation has been defined. This supplies a one-to-one mapping between the original iCalendar format and the XML representation. + +The calendaring XML representation should follow the iCalendar XML formats, which is specified in [IETF RFC 6321]. + +## 10.4 Messaging data syntax + +### 10.4.1 SMS syntax + +For exchanging the SMS data stored in mobile terminals or other storages, the recommended XML schema representation following XML schema 1.1 is shown as below: + +``` + + + + + + + + + +``` + +``` + + + + + + + + + + + + + + + + + + + + + + +``` + +When converting SMS data to XML representation, the SMS information data fields should be mapped to XML elements correctly. The specific data format and meaning for each element is described as below: + + is the root element for storing SMS. It may contain multiple elements. + + is the element containing data for one SMS record, it has several fields: + +- : Contains a string with the address information of the terminals or the entities sending the SMS, the specific format shall follow the address definition of SMS in [ETSI TS 127 005]. +- : Contains a string with the address information of the recipient of the SMS, the specific format shall follow the address definition of SMS in [ETSI TS 127 005]. +- : Contains a string with the text content of the SMS, the format of the string shall follow the short message data format definition in [ETSI TS 127 005]. +- : Contains a string indicating the creating time of this SMS record when it is received, sent out or generated within the terminal or the entities, the format of the time shall follow the time string format defined in [ETSI TS 127 005]. +- : this is the elements indicating the status of the SMS: + - : If the SMS is sent to the terminals or the entities by other terminals or entities, this field shall be set to 1; otherwise it shall be set to 0. + - : If the SMS is sent out by the terminals or the entities to other terminals or entities, this field shall be set to 1; otherwise it shall be set to 0. + - : If the SMS is generated within the terminals or the entities and hasn't been sent out, this field shall be set to 1; otherwise it shall be set to 0. + - : If the SMS is a delivery report from the SMS service centre located in the network, this field shall be set to 1; otherwise it shall be set to 0. +- : Contains a string about any extra descriptive information of the SMS if supported by the terminals or the entities. + +### 10.4.2 MMS Syntax + +For exchanging the multimedia message service (MMS) data stored in the mobile terminals or other storages, the recommended XML schema representation following XML schema 1.1 is shown as below: + +``` + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +``` + +When converting MMS data to XML representation, the MMS information data fields should be mapped to the XML elements correctly, the specific data format and the meaning for each MMS information field is described below: + + is the root element for storing MMS. It may contain multiple elements. + + is the element containing data for one MMS record, it has several fields: + +- : Contains several fields holding the content of the MMS. + +- : Contains a string indicating the character encoding for text as specified in [ETSI TS 126 140]. +- : Contains a string indicating the content encoding type as specified in [ETSI TS 123 140]. +- : Contains a string with the reference location of MMS content as specified in [ETSI TS 123 140]. +- : Contains a string with the encoded data of the MMS. +- : Contains a string with the indicator of content type as defined in [ETSI TS 123 140]. +- : Contains several fields of more descriptive information. + - : Contains a string with the descriptive information of the nature of the content, such as indications of the classified contents of the MMS and the digital rights management (DRM) status, as defined in [ETSI TS 123 140]. + - : Contains a string with the media type and formats of the classified content as defined in [ETSI TS 123 140]. + - : Contains a string about the index of the MMS as defined in [ETSI TS 123 140]. +- : Contains a string with the message type of MMS, as defined in [ETSI TS 123 140]. +- : Contains a string with the current version of the MMS. +- : Contains a string with the message identification of the MMS as defined in [ETSI TS 123 140]. +- : Contains a string with the address information of the terminals or the entities sending the MMS, the specific format shall follow the address definition of MMS in [ETSI TS 123 140]. +- : Contains a string with the address information of the terminals or the entities receiving the MMS, the specific format shall follow the address definition of MMS in [ETSI TS 123 140]. +- : Contains a string with the class of the MMS, e.g., personal, advertisement, informational or auto as defined in [ETSI TS 123 140]. +- : Contains a string with the date and time as the MMS is handled. +- : Contains a string with the priority of the MMS when handled, which might be "low", "normal" or "high". +- : Contains a string with the subject of the MMS. +- : Contains a string with the current status of the MMS, which might be "Draft", "Sent", "New", "Retrieved", "Forwarded" as defined in [ETSI TS 123 140]. +- : Contains a string with any extra descriptive information of the MMS if supported by the terminals or the entities. + +## 10.5 Other + +XML syntax for additional user data is for further study. + + + +# SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|---------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | General tariff principles | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Terminals and subjective and objective assessment methods | +| Series Q | Switching and signalling | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects and next-generation networks | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.746.20-202512-I_PDF-E/raw.md b/marked/F/T-REC-F.746.20-202512-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..1cb0ef235ea61e896f50002e19b5c1ba0adf9889 --- /dev/null +++ b/marked/F/T-REC-F.746.20-202512-I_PDF-E/raw.md @@ -0,0 +1,466 @@ + + +# Recommendation **ITU-T F.746.20 (12/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +## --- **Framework for interactive virtual performing arts services** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.746.20 + +# Framework for interactive virtual performing arts services + +## Summary + +Performing arts is a generic term for artistic acts presented in real time on stage for an audience. These performances are rapidly shifting from offline to online virtual environments due to the spread of non-face-to-face culture, influenced by external factors such as the COVID-19 pandemic and the development of digital technologies. + +The interactive virtual performing arts services framework is an ICT-based framework that enables stage performances in a virtual environment, allowing performers and audiences to interact through their avatars. + +Recommendation ITU-T F.746.20 specifies a framework for interactive virtual performing arts services. The framework establishes common requirements for avatar-based representation, real-time interaction, and coordinated management of the stage, auditorium, assets, and audiences, supported by seamless connectivity and stakeholders. It defines components, data and control architecture, core functional components and an operation model to enable interoperability and comparability of interactive virtual performing arts services across different platforms and providers. + +## History\* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.746.20 | 2025-12-14 | 21 | 11.1002/1000/16580 | + +## Keywords + +ICT-based avatar interaction, virtual performing arts. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2026 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|---------------------------------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Background..... | 2 | +| 7 Interactive virtual performing arts services framework..... | 2 | +| 7.1 System requirements ..... | 3 | +| 7.2 Stakeholders ..... | 3 | +| 7.3 Components and roles of the interactive virtual performing arts services framework..... | 5 | +| 7.4 Data and control architecture..... | 6 | +| 7.5 Operational model ..... | 9 | +| Bibliography..... | 12 | + + + +# Recommendation ITU-T F.746.20 + +## Framework for interactive virtual performing arts services + +## 1 Scope + +This Recommendation establishes requirements to enable interoperability and comparability of interactive virtual performing arts systems across different platforms and providers. + +This Recommendation specifies the interactive virtual performing arts services framework, including requirements, stakeholders, components, data and control architecture, and operational models. + +The scope of this Recommendation includes: + +- Requirements for ICT-based interactive virtual performing arts services framework. +- Definition of stakeholders and framework components. +- Specification of the control architecture and the operational models for interoperability and consistent interaction. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[[ITU-T H.430.1](#)] Recommendation ITU-T H.430.1 (2018), *Requirements for immersive live experience (ILE) services*. + +[[ITU-T H.430.2](#)] Recommendation ITU-T H.430.2 (2018), *Architectural framework for immersive live experience (ILE) services*. + +[[ITU-T H.430.3](#)] Recommendation ITU-T H.430.3 (V2) (2023), *Service scenario for immersive live experience (ILE)*. + +[[ITU-T H.430.4](#)] Recommendation ITU-T H.430.4 (2019), *Service configuration, media transport protocols, signalling information of MPEG media transport for immersive live experience (ILE) systems*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 metaverse** [b-FGMV-20] and [b-ITU FGMV-33]: A converged networked environment integrating digital and physical realities with persistent, shared, and interoperable experiences. + +**3.1.2 avatar** [b-FGMV-20] and [b-ITU FGMV-33]: A user's digital representation in a metaverse environment, facilitating interaction and expression. + +**3.1.3 service** [b-ITU-T Y.2091]: A set of functions and facilities offered to a user by a provider. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 virtual stage:** A virtual environment that provides a space for presenting performances. + +NOTE – The performances of a virtual stage include avatars of performers and stage effects. + +**3.2.2 virtual auditorium:** A virtual environment that provides a space for audiences to experience performances through their avatars and interactions. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviation and acronym: + +ICT Information and Communication Technology + +UI User Interface + +## 5 Conventions + +In this Recommendation: + +- The keywords "**is required to**" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. + +# 6 Background + +The performing arts traditionally refer to artistic acts presented in real time on a stage for an audience. With the spread of non-face-to-face culture, driven by external factors such as the COVID-19 pandemic and the development of digital technologies, these performances have increasingly shifted from physical venues to virtual environments. + +To reproduce the sense of presence and immersion that is experienced in physical performances, virtual performing arts services require realistic stage composition and interactive accessibility. This includes configuring virtual stages and auditoriums analogous to physical venues, and enabling real-time interaction between performers and audiences through avatars connected over communication networks. + +This Recommendation specifies a standardized framework that, with respect to human factors such as avatar-based representation, real-time interaction, and user experience, defines functional components and procedures for interactive virtual performing arts services, including virtual stage composition, stage operation, virtual auditorium composition, auditorium operation, asset management, and audience management. + +The metaverse concept addresses broader converged environments integrating digital and physical realities. In contrast, this Recommendation specifies a dedicated framework for interactive virtual performing arts services enabling artistic direction and stage control interfaces, real-time broadcast streaming of the virtual stage, massive audience interaction and virtual audience management. It establishes normative requirements and framework components to support interoperable deployment. + +## 7 Interactive virtual performing arts services framework + +The framework is an information and communication technology (ICT)-based system that supports stage performances in a virtual environment, allowing performers and audiences to interact through avatars. By defining common requirements and components, this Recommendation establishes a harmonized basis for interoperability, comparability of results, and consistency in the delivery of interactive virtual performing arts services. + +This clause specifies system requirements, stakeholders, overall components, the data and control architecture, and the operation model of the interactive virtual performing arts services framework, which enables interoperability and comparability across different platforms and service providers. + +## 7.1 System requirements + +This clause specifies the system requirements that the interactive virtual performing arts services framework is required to fulfil in order to support an interface that enables an artistic direction and stage control, real-time broadcast streaming of the virtual stage, large-scale audience interaction and virtual audience management. It also specifies requirements for synchronized media exchange using the ITU-T H.430 series Recommendations ([ITU-T H.430.1] [ITU-T H.430.2] [ITU-T H.430.3] [ITU-T H.430.4]), and avatar representation compatibility using the ITU-T FG-MV deliverables: + +- The system is required to ensure interoperability and consistent real-time exchange of audio, video, and motion data among performer and audience environments, utilizing immersive media transport and synchronization mechanisms ([ITU-T H.430.1] [ITU-T H.430.2] [ITU-T H.430.3] [ITU-T H.430.4]). + +NOTE 1 – This requirement refers to clause 6 in [ITU-T H.430.4] for synchronization. Additional mechanisms supporting immersive media transport and overall architecture refers to the ITU-T H.430 series Recommendations ([ITU-T H.430.1] [ITU-T H.430.2] [ITU-T H.430.3] [ITU-T H.430.4]). + +- The system is required to ensure compatibility with avatar representation frameworks in order to enable integration and consistent behaviour of performer and audience avatars within the virtual performance environment. + +NOTE 2 – This requirement is related to the specification of metaverse interoperability [b-FGMV-19] [b-FGMV-42] [b-FGMV-43]. + +- The system is required to support an interface that enables an artistic director to manage and control performance flow, stage effects, and synchronization among performers, audiences, and digital assets in real time. +- The system is required to support stage management interface for real-time broadcast streaming of the virtual stage for synchronized delivery of performance scenes to multiple audiences. +- The system is required to support virtual audience management interface, including registration, authentication, admission control, and participation data management. + +While building upon existing immersive media transport and avatar representation frameworks, this Recommendation defines additional requirements for an interface that enables an artistic direction and stage control, real-time broadcast streaming of the virtual stage, large-scale audience interaction, and virtual audience management, which are not specified in [ITU-T H.430.1] [ITU-T H.430.2] [ITU-T H.430.3] [ITU-T H.430.4]. + +## 7.2 Stakeholders + +This clause identifies stakeholders and summarizes their functions. Table 1 provides role-level descriptions: + +- Performer: generates artistic content represented as avatars on the virtual stage. +- Audience: consumes and interacts with performance content (passive or interactive) and may broadcast the performance. +- Designer/maker: provides creative assets (stage design, props, avatar models, lighting, sound, special effects). +- Director/manager: coordinates artistic and technical processes (direction, stage management, performance direction, lighting/sound operation). + +- ICT-based service provider: operates the platform and multimedia communication (e.g., platform operator, streaming operator, platform administrator, platform maintenance provider). + +**Table 1 – Stakeholders in interactive virtual performing arts services framework** + +| Stakeholders | Subclass | Description | +|----------------------------------|-------------------------------|-------------------------------------------------------------------------------------------------------------------| +| Performer | Actor/actress | Person who plays roles for virtual performing arts | +| | Dancer | Person who expresses body movements in virtual performing arts | +| | Musician | Person who plays music in virtual performing arts | +| | Singer | Person who sings in virtual performing arts | +| Audience | One-way audience | Audience who participates in a one-way manner (no real-time interaction) | +| | Interactive audience | Audience who interactively communicates | +| | Streamer | Audience who broadcasts virtual performing arts | +| Designer/
maker | Set designer | Person who designs equipment for virtual performance stages | +| | Property maker | Person who makes props for virtual performance stages | +| | Avatar designer | Person who designs and creates avatars for virtual performance stages | +| | Special effects designer | Person who creates special effects for virtual performance stages | +| | Lighting designer | Person who designs concept of light and creates an action plan in the production of performance | +| | Sound designer | Person who designs concept of sound and creates an action plan in the production of performance | +| Director/
manager | Main director | Person responsible for integrating and coordinating the entire artistic field of performance production | +| | Assistant director | Person who assists in artistic process of directing | +| | Stage manager | Person responsible for coordinating lighting, sound, stage equipment, stage transitions, and performance progress | +| | Performance director | Person who directs live actors for motion capture and manages avatar movements | +| | Special effects operator | Person who operates special effects equipment for virtual performing arts | +| | Lighting director | Person who designs the lighting concept and creates an action plan for the performance production | +| | Sound director | Person who designs concept of sound and creates an action plan for the performance production | +| ICT-based
service
provider | Platform operator | Person who operates platform and multimedia communication | +| | Streaming operator | Person who runs streaming and ensures synchronization | +| | Platform administrator | Person who manages authorization and administration | +| | Platform maintenance provider | Person who provides maintenance for the platform and the multimedia communication | + +## 7.3 Components and roles of the interactive virtual performing arts services framework + +Components and roles of the framework are provided in Figure 1. Clauses 7.4 and 7.5 specify the supporting architecture and procedures. For the purpose of description, these eight components, namely direction orchestration, virtual stage composition, stage operation, virtual auditorium, auditorium operation, audience management, asset management and multimedia communication, are grouped into an application subsystem and a communication subsystem as shown in Figure 1, which correspond to the operational entities specified in clause 7.5. In Figure 1, solid rectangles represent components, while dashed rectangles represent roles within each component. + +![Figure 1: Components and roles of interactive virtual performing arts services framework. The diagram shows a hierarchical structure of components and roles. At the top is 'UI direction orchestration' (solid rectangle) containing 'Stage direction', 'Scene synchronization', and 'Audience coordination' (dashed rectangles). Below this are four main components: 'Asset management' (solid), 'UI stage operation' (solid), 'UI auditorium operation' (solid), and 'Audience management' (solid). 'Asset management' contains 'Avatar asset', 'Stage asset', 'Effects asset', and 'Interaction asset' (dashed). 'UI stage operation' contains 'Virtual stage composition' (solid), which includes 'Stage effects' (solid) with 'Lighting effects', 'Sound effects', 'Special effects', and 'Video walls' (dashed), and 'Performer avatars' (solid) with 'Motion', 'Voice', 'Appearance', and 'Facial expression' (dashed). 'UI auditorium operation' contains 'Virtual auditorium composition' (solid), which includes 'Auditorium interface' (solid) with 'Props interaction', 'Viewpoint switch', 'Billboard interaction', and 'Spatial movement' (dashed), and 'Audience avatars' (solid) with 'Motion', 'Voice', 'Appearance', and 'Facial expression' (dashed). 'Audience management' contains 'User management', 'Payment management', and 'Admission management' (dashed). At the bottom is 'Multimedia communication' (solid) containing 'Interactive streaming' and 'One-way streaming' (dashed). The top section is labeled 'Application subsystem' and the bottom 'Communication subsystem'. Arrows indicate flow from top to middle and bidirectional flow between middle and bottom. A small code 'F.746.20(25)' is in the bottom right corner of the diagram area.](e9314c83043183351ed74908e9bf2f90_img.jpg) + +Figure 1: Components and roles of interactive virtual performing arts services framework. The diagram shows a hierarchical structure of components and roles. At the top is 'UI direction orchestration' (solid rectangle) containing 'Stage direction', 'Scene synchronization', and 'Audience coordination' (dashed rectangles). Below this are four main components: 'Asset management' (solid), 'UI stage operation' (solid), 'UI auditorium operation' (solid), and 'Audience management' (solid). 'Asset management' contains 'Avatar asset', 'Stage asset', 'Effects asset', and 'Interaction asset' (dashed). 'UI stage operation' contains 'Virtual stage composition' (solid), which includes 'Stage effects' (solid) with 'Lighting effects', 'Sound effects', 'Special effects', and 'Video walls' (dashed), and 'Performer avatars' (solid) with 'Motion', 'Voice', 'Appearance', and 'Facial expression' (dashed). 'UI auditorium operation' contains 'Virtual auditorium composition' (solid), which includes 'Auditorium interface' (solid) with 'Props interaction', 'Viewpoint switch', 'Billboard interaction', and 'Spatial movement' (dashed), and 'Audience avatars' (solid) with 'Motion', 'Voice', 'Appearance', and 'Facial expression' (dashed). 'Audience management' contains 'User management', 'Payment management', and 'Admission management' (dashed). At the bottom is 'Multimedia communication' (solid) containing 'Interactive streaming' and 'One-way streaming' (dashed). The top section is labeled 'Application subsystem' and the bottom 'Communication subsystem'. Arrows indicate flow from top to middle and bidirectional flow between middle and bottom. A small code 'F.746.20(25)' is in the bottom right corner of the diagram area. + +Figure 1 – Components and roles of interactive virtual performing arts services framework + +### 7.3.1 Virtual stage composition and UI stage operation + +The virtual stage composition and user interface (UI) stage operation process performers' motion and audio data and applies them to avatar models. They synthesize and render stage scenes with avatar motion, voice, appearance, and facial expressions, and support stage effects including lighting, sound, special effects, and video walls. + +### 7.3.2 Virtual auditorium composition and UI auditorium operation + +The virtual auditorium composition and UI auditorium operation process audience motion, voice, appearance, and facial expressions, apply them to audience avatars, and render the auditorium scene by combining these data with environmental and interaction elements. They support interaction functions such as props interaction, viewpoint switching, billboard interaction, and spatial movement. + +#### 7.3.3 Audience management + +The audience management component performs audience-related administrative functions. It provides user registration, authentication, and admission control, and processes payment and profile + +data. This information is provided to the virtual auditorium composition via the auditorium operation to ensure consistent audience representation and interaction. + +#### **7.3.4 Asset management** + +The asset management also supplies the required assets to other components in a synchronized manner. asset management manages digital assets used in the virtual stage composition and virtual auditorium composition. It stores and updates stage, prop, and avatar models, and controls lighting, sound, special effects, and video walls. It also supplies assets to other components in a synchronized manner. + +#### **7.3.5 UI direction orchestration** + +The direction orchestration coordinates the overall operation of the framework. It manages stage direction, performance flow, effect triggering, and synchronization between the virtual stage composition and virtual auditorium composition. It also coordinates audience interaction to align responses with stage events and ensures coherent operation across all components, including multimedia communication and management functions. + +#### **7.3.6 Multimedia communication** + +The multimedia communication integrates and delivers data streams among components. It supports interactive communication and one-way delivery and provides data mixing and visualization to ensure consistent representation across the virtual stage composition and the virtual auditorium composition. + +## **7.4 Data and control architecture** + +This clause specifies the data and control architecture of the interactive virtual performing arts services framework. Figures 2 and 3 illustrate the architecture from two complementary perspectives. Figure 2 shows the application subsystem flows among functional entities such as virtual stage composition, stage operation, virtual auditorium composition, auditorium operation, audience management, asset management, and direction orchestration. Figure 3 shows the communication subsystem flows, which are provided by multimedia communication through storage, service, and session functions that enable the transfer and synchronization of data used by the application subsystem entities. Together, Figures 2 and 3 specify the overall data and control architecture of the framework, in which the communication subsystem provides storage, session and delivery functions that transport, synchronize and manage data and control flows between application subsystem entities, enabling consistent and reliable operation. In these figures, solid arrows denote control flows, dashed arrows denote data flows, rectangles denote data entities, rounded rectangles denote functional entities, and human figures denote actors representing sources or recipients of data and control flows. + +#### **7.4.1 Application subsystem flows** + +This clause specifies the data and control flows among functional entities of the interactive virtual performing arts services framework. These flows enable consistent representation and synchronized interaction between virtual stage composition, UI stage operation, virtual auditorium composition, UI auditorium operation, audience management, asset management, and UI direction orchestration. + +![Figure 2 – Data and control architecture of application subsystem. This block diagram illustrates the flow of data and control signals between various components. On the left, a 'Performer' provides 'Performer motion data' and 'Performer audio data' to the 'UI Stage Operation'. A 'Designer/ Maker' provides 'Stage/ Effects asset data', 'Avatar asset data', and 'Interaction asset data' to 'Asset Management'. On the right, an 'Audience' provides 'Audience motion data', 'Audience message data', 'Audience position data', and 'User payment/ Admission data' to 'UI Auditorium Operation'. The central 'UI Direction Orchestration' (containing a 'Director/ Manager' icon) receives data from all four operations: 'UI Stage Operation', 'Virtual Stage Composition', 'Virtual Auditorium Composition', and 'UI Auditorium Operation'. It also receives data from 'Asset Management' and 'Audience Management'. The 'UI Stage Operation' and 'UI Auditorium Operation' both have bidirectional connections to the 'Virtual Stage Composition' and 'Virtual Auditorium Composition'. The 'UI Auditorium Operation' also has a bidirectional connection to 'Audience Management'. The 'UI Direction Orchestration' has bidirectional connections to the 'UI Stage Operation', 'Virtual Stage Composition', 'Virtual Auditorium Composition', and 'UI Auditorium Operation'. The 'UI Direction Orchestration' also has bidirectional connections to 'Asset Management' and 'Audience Management'. The 'UI Direction Orchestration' is connected to the 'Director/ Manager' icon. The 'UI Direction Orchestration' is connected to the 'Director/ Manager' icon. The 'UI Direction Orchestration' is connected to the 'Director/ Manager' icon.](af7916c89a458fdab6c3f443217388ae_img.jpg) + +Figure 2 – Data and control architecture of application subsystem. This block diagram illustrates the flow of data and control signals between various components. On the left, a 'Performer' provides 'Performer motion data' and 'Performer audio data' to the 'UI Stage Operation'. A 'Designer/ Maker' provides 'Stage/ Effects asset data', 'Avatar asset data', and 'Interaction asset data' to 'Asset Management'. On the right, an 'Audience' provides 'Audience motion data', 'Audience message data', 'Audience position data', and 'User payment/ Admission data' to 'UI Auditorium Operation'. The central 'UI Direction Orchestration' (containing a 'Director/ Manager' icon) receives data from all four operations: 'UI Stage Operation', 'Virtual Stage Composition', 'Virtual Auditorium Composition', and 'UI Auditorium Operation'. It also receives data from 'Asset Management' and 'Audience Management'. The 'UI Stage Operation' and 'UI Auditorium Operation' both have bidirectional connections to the 'Virtual Stage Composition' and 'Virtual Auditorium Composition'. The 'UI Auditorium Operation' also has a bidirectional connection to 'Audience Management'. The 'UI Direction Orchestration' has bidirectional connections to the 'UI Stage Operation', 'Virtual Stage Composition', 'Virtual Auditorium Composition', and 'UI Auditorium Operation'. The 'UI Direction Orchestration' also has bidirectional connections to 'Asset Management' and 'Audience Management'. The 'UI Direction Orchestration' is connected to the 'Director/ Manager' icon. The 'UI Direction Orchestration' is connected to the 'Director/ Manager' icon. The 'UI Direction Orchestration' is connected to the 'Director/ Manager' icon. + +**Figure 2 – Data and control architecture of application subsystem** + +##### 7.4.1.1 Data transmission procedure + +The data transmission procedure specifies how data are transferred among functional entities: + +- Motion-capture and audio data generated by the performer are transferred to the Virtual stage composition via the UI stage operation. +- Stage, props and avatar model data generated by designers and makers are managed by Asset Management and are delivered to the virtual stage composition and the virtual auditorium composition via the UI stage operation as required. +- Audience-reaction data, including gesture, message and position are transferred to the virtual auditorium composition via the UI auditorium operation for representation and interaction control. +- Admission, payment and profile data are processed by audience management and are provided to the virtual auditorium composition via the UI auditorium operation when necessary. +- Streaming data generated by the virtual stage composition and the virtual auditorium composition are delivered to performers and audiences. +- Control-related data associated with each operation of the UI direction orchestration are received from relevant functional entities to maintain synchronized coordination and system feedback. + +##### 7.4.1.2 Control procedure + +The control procedure specifies how control functions are coordinated among functional entities: + +- Virtual stage composition applies performer data to avatars for reproduction. +- Asset management provides stage, props and avatar models to the virtual stage composition and the virtual auditorium composition via the stage Operation. +- Virtual auditorium composition processes audience data and manages the auditorium state. +- Synchronization, scene control and effect triggering is exchanged among the virtual stage composition, the virtual auditorium composition and the asset management. +- The Direction Orchestration coordinates these control flows by providing stage direction operation, scene synchronization and audience interaction coordination. + +##### 7.4.1.3 Functional entities + +The functional entities perform the following tasks in supporting subsystem operation: + +- Virtual stage composition and stage operation process performer motion and audio data, applies them to avatar models and render stage scenes with stage effects. +- Virtual auditorium composition and auditorium operation represent audience avatars and supports props interaction, viewpoint switching, billboard interaction and spatial movement. +- Audience management handles user registration, authentication, admission and payment, and provides user data to the virtual auditorium composition. +- Asset management maintains digital assets such as stage, props and avatar models and controls lighting, sound, special effects and video walls. +- Direction orchestration coordinates the overall framework by managing stage direction, synchronizing scenes between stage and auditorium, and aligning audience interaction with stage events. + +#### 7.4.2 Communication subsystem flows + +This clause specifies the communication subsystem data and control flows that support the functional entities of the interactive virtual performing arts services framework. These flows are provided by the multimedia communication, which consists of storage repository, service delivery and session coordination, supporting virtual stage delivery, interactive communication and virtual auditorium delivery. The communication subsystem flows enable motion, audio, asset and interaction data to be consistently stored, processed and delivered, and channels to be established and maintained for synchronized operation across virtual stage composition, virtual auditorium composition, audience management, asset management and direction orchestration. + +![Figure 3: Data and control architecture of communication subsystem. The diagram shows three main components: Storage repository, Service delivery, and Session coordination. The Storage repository contains four data types: User payment/admission data, Audience motion/message/position data, Performer motion/audio data, and Stage/Avatar/Interaction asset data. The Service delivery contains three functions: Virtual auditorium delivery, Interactive communication, and Virtual stage delivery. The Session coordination contains four channels: Audience channel, Interaction channel, Performer channel, and Asset channel. Arrows indicate data flow from the Storage repository to the Service delivery functions and from the Service delivery functions to the Session coordination channels. A legend at the bottom shows an icon for 'ICT-based service provider' and the code 'F.746.20(25)'.](a33da0f14e456f92539ce3e9b7d81f9a_img.jpg) + +The diagram illustrates the data and control architecture of the communication subsystem. It is organized into three main vertical columns: + + +- Storage repository:** Contains four stacked boxes representing data types: 'User payment/admission data', 'Audience motion/message/position data', 'Performer motion/audio data', and 'Stage/Avatar/Interaction asset data'. +- Service delivery:** Contains three stacked boxes representing functions: 'Virtual auditorium delivery', 'Interactive communication', and 'Virtual stage delivery'. +- Session coordination:** Contains four stacked boxes representing channels: 'Audience channel', 'Interaction channel', 'Performer channel', and 'Asset channel'. + + Data flow is indicated by arrows: + + +- Dashed arrows point from each of the four data types in the 'Storage repository' to the 'Virtual auditorium delivery', 'Interactive communication', and 'Virtual stage delivery' functions. +- Solid arrows point from each of the three 'Service delivery' functions to each of the four 'Session coordination' channels. + + At the bottom left, there is an icon of a person labeled 'ICT-based service provider'. At the bottom right, the text 'F.746.20(25)' is present. + +Figure 3: Data and control architecture of communication subsystem. The diagram shows three main components: Storage repository, Service delivery, and Session coordination. The Storage repository contains four data types: User payment/admission data, Audience motion/message/position data, Performer motion/audio data, and Stage/Avatar/Interaction asset data. The Service delivery contains three functions: Virtual auditorium delivery, Interactive communication, and Virtual stage delivery. The Session coordination contains four channels: Audience channel, Interaction channel, Performer channel, and Asset channel. Arrows indicate data flow from the Storage repository to the Service delivery functions and from the Service delivery functions to the Session coordination channels. A legend at the bottom shows an icon for 'ICT-based service provider' and the code 'F.746.20(25)'. + +**Figure 3 – Data and control architecture of communication subsystem** + +##### 7.4.2.1 Data transmission procedure + +The data transmission procedure specifies how data types are transmitted within the multimedia communication: + +- Performer motion and audio data are transmitted to service delivery hosting the virtual stage delivery function, where they are combined with stage effects. +- Audience-reaction data, including gestures, messages and positions, are transmitted to service delivery hosting the interactive communication function, where they are mixed with performer data. + +- User payment and admission data, as provided by audience management, are transported via the multimedia communication to service delivery hosting the virtual auditorium delivery function. +- Asset data, including stage, effects, avatars and interaction assets, are transmitted to the storage repository of the multimedia communication and are delivered to service delivery as required. +- Control-related data exchanged between service delivery functions and session coordination functions, as well as storage repository data, are subsequently delivered to the session coordination functions through the service delivery functions. + +##### 7.4.2.2 Control procedure + +The control procedure specifies how session coordination establishes and manages channels in coordination with service delivery within the multimedia communication, as shown in Figure 3: + +- Session coordination establishes and maintains the channels supporting the three service delivery functions: virtual stage delivery, interactive communication, and virtual auditorium delivery. +- Coordination actions include channel establishment, synchronization of data streams and effect triggering to ensure consistent rendering of stage and auditorium environments. +- Each service delivery function operates its associated channel through session coordination: virtual stage delivery for the performer channel, interactive communication for the interaction channel, and virtual auditorium delivery for the audience channel. + +##### 7.4.2.3 Functional entities + +The functional entities specify the tasks that are performed within the multimedia communication: + +- Storage repository provides repositories for performer motion and audio data, audience-reaction data, user admission data and performance asset data. +- Service delivery supports virtual stage delivery, interactive communication and virtual auditorium delivery, and processes and forwards the corresponding data streams. +- Session coordination establishes and maintains the required channels for these services and ensures that data and control flows are delivered reliably and in synchronization across the framework. + +## 7.5 Operational model + +This clause specifies the operational model of the interactive virtual performing arts services framework. Figure 4 illustrates the operational interactions among the functional entities, which are stage operation, direction orchestration, virtual stage and virtual auditorium composition, and auditorium operation. The operational model specifies how performance data, model data, stage scenes and audience reaction data are exchanged among these entities, and how coordination is achieved through functions such as system start-up and control, stage mixing, auditorium mixing, visualization of stage scenes and visualization of updated auditorium scenes. In Figure 4, solid arrows denote control flows and dashed arrows denote data flows, while rectangles denote functional entities and data labels identify the information exchanged between entities. + +![Sequence diagram illustrating the operational model of interactive virtual performing arts services framework. The diagram shows four lifelines: UI Stage Operation, UI Direction Orchestration, Virtual Stage and Virtual Auditorium Compositions, and UI Auditorium Operation. The sequence of interactions is as follows: 1. UI Direction Orchestration sends 'Request initiation' to Virtual Stage and Virtual Auditorium Compositions. 2. Virtual Stage and Virtual Auditorium Compositions performs 'Configure asset' and 'Initiate connection'. 3. Virtual Stage and Virtual Auditorium Compositions sends 'Confirm completion' to UI Direction Orchestration. 4. UI Direction Orchestration sends 'Request performance' to UI Stage Operation. 5. UI Stage Operation sends 'Send performance data' to Virtual Stage and Virtual Auditorium Compositions. 6. Virtual Stage and Virtual Auditorium Compositions performs 'Mix stage'. 7. Virtual Stage and Virtual Auditorium Compositions sends 'Send stage data' to UI Auditorium Operation. 8. UI Auditorium Operation performs 'Visualize stage'. 9. UI Auditorium Operation sends 'Send reaction data' to Virtual Stage and Virtual Auditorium Compositions. 10. Virtual Stage and Virtual Auditorium Compositions performs 'Mix auditorium'. 11. Virtual Stage and Virtual Auditorium Compositions sends 'Send auditorium data' to UI Auditorium Operation. 12. UI Auditorium Operation performs 'Visualize auditorium'. 13. Virtual Stage and Virtual Auditorium Compositions sends 'Send auditorium data' to UI Stage Operation. 14. UI Stage Operation performs 'Visualize auditorium'. A label F.746.20(25) is present at the bottom right.](fa859e4e468bfb2710a94527f2c504af_img.jpg) + +Sequence diagram illustrating the operational model of interactive virtual performing arts services framework. The diagram shows four lifelines: UI Stage Operation, UI Direction Orchestration, Virtual Stage and Virtual Auditorium Compositions, and UI Auditorium Operation. The sequence of interactions is as follows: 1. UI Direction Orchestration sends 'Request initiation' to Virtual Stage and Virtual Auditorium Compositions. 2. Virtual Stage and Virtual Auditorium Compositions performs 'Configure asset' and 'Initiate connection'. 3. Virtual Stage and Virtual Auditorium Compositions sends 'Confirm completion' to UI Direction Orchestration. 4. UI Direction Orchestration sends 'Request performance' to UI Stage Operation. 5. UI Stage Operation sends 'Send performance data' to Virtual Stage and Virtual Auditorium Compositions. 6. Virtual Stage and Virtual Auditorium Compositions performs 'Mix stage'. 7. Virtual Stage and Virtual Auditorium Compositions sends 'Send stage data' to UI Auditorium Operation. 8. UI Auditorium Operation performs 'Visualize stage'. 9. UI Auditorium Operation sends 'Send reaction data' to Virtual Stage and Virtual Auditorium Compositions. 10. Virtual Stage and Virtual Auditorium Compositions performs 'Mix auditorium'. 11. Virtual Stage and Virtual Auditorium Compositions sends 'Send auditorium data' to UI Auditorium Operation. 12. UI Auditorium Operation performs 'Visualize auditorium'. 13. Virtual Stage and Virtual Auditorium Compositions sends 'Send auditorium data' to UI Stage Operation. 14. UI Stage Operation performs 'Visualize auditorium'. A label F.746.20(25) is present at the bottom right. + +**Figure 4 – Operational model of interactive virtual performing arts services framework** + +#### 7.5.1 Functional entities + +The functional entities shown in Figure 4 exchange data and control flows to realize the operational model of the framework: + +- UI stage operation processes performance data and applies it to avatar models and stage effects for scene generation. +- UI direction orchestration requests the initiation of asset and connection as pre-operational steps, and issues stage control, manages synchronization of data and scenes, and coordinate stage and auditorium mixing. +- Virtual stage and virtual auditorium compositions perform asset configuration and connection initiation upon request, and synthesize and update scenes by combining processed performance data, model data and audience reaction data through stage and auditorium mixing. +- UI auditorium operation renders stage and auditorium scenes and collects audience reaction data for feedback. + +#### 7.5.2 Behaviour + +The behaviour defines the sequence of actions performed by the functional entities in exchanging data and control flows: + +- UI direction orchestration requests the initiation of asset and connection as pre-operational steps. +- Virtual stage and virtual auditorium compositions perform asset and connection initiation upon request and confirm completion to UI direction orchestration. +- After confirmation, UI direction orchestration sends control to UI stage operation. + +- UI stage operation processes performance and model data and delivers the processed data to virtual stage and virtual auditorium compositions. +- Virtual stage and virtual auditorium compositions perform stage mixing to generate a stage scene. +- The generated stage scene is delivered to UI auditorium operation, which renders the stage scene to the audience. +- UI auditorium operation collects audience reaction data and provides it to virtual stage and virtual auditorium compositions. +- Virtual stage and virtual auditorium compositions perform auditorium mixing by combining the stage scene with audience reaction data to generate an updated auditorium scene. +- UI auditorium operation renders the updated auditorium scene to the audience. + +## Bibliography + +- [b-ITU-T Y.2091] Recommendation ITU-T Y.2091 (2011), *Terms and definitions for next generation networks*. +- [b-FGMV-19] ITU-T Focus Group on metaverse deliverable number 19 (Technical Specification) (2023), *Service scenarios and high-level requirements for metaverse cross-platform interoperability*. +- [b-FGMV-20] ITU-T Focus Group on Metaverse deliverable number 20 (Technical Specification) (2023), *Definition of metaverse*. +- [b-FGMV-33] ITU-T Focus Group on Metaverse deliverable number 33 (Technical Specification) (2024), *Glossary for metaverse*. +- [b-FGMV-42] ITU-T Focus Group on metaverse deliverable number 42 (Technical Report) (2024), *Interoperability of identity of things across metaverse platforms*. +- [b-FGMV-43] ITU-T Focus Group on metaverse deliverable number 43 (Technical Specification) (2024), *High-level interoperability architecture for cross-platform metaverse*. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.746.21-202512-I_PDF-E/007b053fe94a8348f75128a584503fd0_img.jpg b/marked/F/T-REC-F.746.21-202512-I_PDF-E/007b053fe94a8348f75128a584503fd0_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f77998201e8df463887001eb365736955943b695 --- /dev/null +++ b/marked/F/T-REC-F.746.21-202512-I_PDF-E/007b053fe94a8348f75128a584503fd0_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b29d12bd01a8d30c524ef133e7b53f62998f983d51f19b1a623e6c5cccd58a51 +size 59803 diff --git a/marked/F/T-REC-F.746.21-202512-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg b/marked/F/T-REC-F.746.21-202512-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d4d8969f805414689ab79cfd08c333d454d482b3 --- /dev/null +++ b/marked/F/T-REC-F.746.21-202512-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:636c3839edff955a5b6208530b05fa1c983ab153d90f00216517a7a99478b988 +size 7074 diff --git a/marked/F/T-REC-F.746.21-202512-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg b/marked/F/T-REC-F.746.21-202512-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ba511a90cab93c817b1049afda6e3325ccaa1344 --- /dev/null +++ b/marked/F/T-REC-F.746.21-202512-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:68ad829d89c5579fafe9926ae507cbf7fd7802488ccf3b8209eb5383519dbce6 +size 55029 diff --git a/marked/F/T-REC-F.746.21-202512-I_PDF-E/bd0b93e7a46ede276d0a3b79ac487bd9_img.jpg b/marked/F/T-REC-F.746.21-202512-I_PDF-E/bd0b93e7a46ede276d0a3b79ac487bd9_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c2bb657e9c21a23a7d504f0a957c6ba5ebf78016 --- /dev/null +++ b/marked/F/T-REC-F.746.21-202512-I_PDF-E/bd0b93e7a46ede276d0a3b79ac487bd9_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6c1f8d09f35882b68255045f64c72ea1faac3b118a9fb3208792cf22fd630b9b +size 76863 diff --git a/marked/F/T-REC-F.746.21-202512-I_PDF-E/raw.md b/marked/F/T-REC-F.746.21-202512-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..a9ed86292c7f35ecdc14cf41296362d4407ef73e --- /dev/null +++ b/marked/F/T-REC-F.746.21-202512-I_PDF-E/raw.md @@ -0,0 +1,682 @@ + + +# Recommendation **ITU-T F.746.21 (12/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +## **Requirements for intelligent user interface services for persons with age-related disabilities** + +![ITU logo](6ed175c791b5e156d9c98a8dbcc3318c_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.746.21 + +# Requirements for intelligent user interface services for persons with age-related disabilities + +## Summary + +Owing to the recent development of artificial intelligence (AI), especially deep learning technology, multimodal user interfaces (UIs) are increasingly used as intelligent tools for systems requiring improved usability. Human-machine interaction is also critical in the domains of services for persons with age-related disabilities, including various devices. Therefore, UIs should be designed for various users and their tasks, based on the characteristics of users, tasks and devices. The AI technologies will help to devise trustworthy and convenient UIs in the corresponding care services for persons with age-related disabilities. + +Recommendation ITU-T F.746.21 introduces intelligent user interface (IUI) services for persons with age-related disabilities. In order to support these various IUI services, it reviews use cases and proposes requirements, and then defines common components and characteristics of IUI services. + +## History\* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.746.21 | 2025-12-14 | 21 | 11.1002/1000/16581 | + +## Keywords + +Accessibility, intelligent user interface (IUI), multimodal interaction, services for persons with age-related disabilities. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2026 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|----------------------------------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation ..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Background..... | 2 | +| 7 Characteristics of IUI services for persons with age-related disabilities..... | 3 | +| 7.1 Consideration of age-related characteristics..... | 3 | +| 7.2 Intelligent user interface features for persons with age-related disabilities ... | 4 | +| 7.3 Service models for persons with age-related disabilities ..... | 5 | +| 8 Conceptual model of IUI for persons with age-related disabilities care systems..... | 9 | +| 8.1 Components of the conceptual model ..... | 9 | +| 8.2 Input and output module..... | 10 | +| 8.3 Data and interaction flow ..... | 11 | +| 8.4 Conceptual model diagram..... | 11 | +| 9 Requirements for common components and characteristics of the intelligent UI service ..... | 12 | +| 9.1 Requirements for common components ..... | 12 | +| 9.2 Requirements for common characteristics ..... | 13 | +| 10 Intelligent UI service requirements from a hardware perspective..... | 14 | +| 11 Intelligent UI service requirements from a software perspective ..... | 14 | +| Bibliography..... | 16 | + + + +## Requirements for intelligent user interface services for persons with age-related disabilities + +## 1 Scope + +This Recommendation specifies intelligent user interface (IUI) services for persons with age-related disabilities, reviews use cases and proposes requirements, and then defines the components and characteristics. + +The scope of this Recommendation includes: + +- Specification of IUI services for persons with age-related disabilities. +- Use cases and requirements of IUI services for persons with age-related disabilities. +- Service components and characteristics of IUI services for persons with age-related disabilities. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 user interface (UI)** [b-ITU-T F.902]: Software and hardware components through which a user can interact with a system. + +**3.1.2 artificial emotional intelligence (AEI)** [b-ITU-T H.862.6]: A multimodal user interface (UI) based on emotion analysis with some properties and illustrations and data. + +**3.1.3 machine learning (ML)** [b-ITU-T Y.3172]: Processes that enable computational systems to understand data and gain knowledge from it without necessarily being explicitly programmed. + +NOTE 1 – Definition adapted from [b-ETSI GR ENI 004]. + +NOTE 2 – Supervised machine learning and unsupervised machine learning are two examples of machine learning types. + +**3.1.4 person with age-related disabilities** [b-ITU-T F.791]: A person with cognitive or physical disabilities caused by the aging process. Examples are impaired eyesight, deafness in varying degrees, reduced mobility or cognitive abilities. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 intelligence user interface:** A user interface that uses artificial intelligence (AI) to adapt to the user's needs and context. + +**3.2.2 multimodal interaction:** An interaction method using more than one input or output mode. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +AEI Artificial Emotional Intelligence + +AI Artificial Intelligence + +IUI Intelligent User Interface + +ML Machine Learning + +UI User Interface + +UX User Experience + +VUI Voice User Interface + +## 5 Conventions + +In this Recommendation: + +The keywords "**is required to**" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. + +The keywords "**is recommended**" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. + +## 6 Background + +User interface (UI) services for persons with age-related disabilities encompass a comprehensive set of functions designed to address their unique needs and challenges. These services leverage AI technologies, particularly deep learning, to provide intuitive, accessible, and user-friendly interfaces that enhance the user experience (UX) and quality of care for persons with age-related disabilities individuals. + +Owing to the recent development of AI, especially deep learning technology, multimodal UIs tend to be used as an intelligent tool for different systems that require usability of human factors. Human-machine interaction is also critical in the domains of services for persons with age-related disabilities, including different devices in terms of usability. Therefore, UI should be designed for a wide range of different users and their tasks in the systems based on the characteristics of users, tasks and devices. The IUI should reflect the characteristics of different users and tasks so that the information can be treated and processed appropriately, in a trustworthy manner, and with ease of use in the services for persons with age-related disabilities. AI technology will be of help to devise such convenient and trustworthy UIs in the corresponding elder care systems. + +UI services for persons with age-related disabilities are continually evolving, driven by ongoing research and development in AI and human-computer interaction. The focus is on creating interfaces that are adaptive, responsive, adaptable, and empowering, catering to the specific needs and preferences of persons with age-related disabilities. The ultimate goal is to enhance their quality of life, promote independence, and provide comprehensive support for their physical, cognitive, and emotional well-being. + +## **7 Characteristics of IUI services for persons with age-related disabilities** + +IUI services are recommended to be designed with consideration of the diverse physical, cognitive, sensory, and psychological/emotional characteristics of persons with age-related disabilities. These characteristics significantly affect UX, including accessibility, ease of use, and emotional responsiveness. + +Therefore, this clause explains the main considerations and characteristics of IUI services and presents representative service models based on them. + +### **7.1 Consideration of age-related characteristics** + +It is recommended that the design of UIs for persons with age-related disabilities consider five key characteristics: physical, cognitive, sensory, psychological/emotional, and environmental. Each characteristic reflects different age-related changes and requires tailored design approaches. + +#### **1) Physical characteristics** + +Physical characteristics include declines in muscular strength, motor skills, reaction speed, and joint flexibility. These changes may cause difficulties in precise manipulation, reduced mobility, and slower responses. + +Therefore, it is recommended that interfaces consider the following: + +- Use large text and buttons for ease of interaction. +- Support contactless control methods such as voice and gesture inputs. +- Minimize requirements for rapid or complex physical actions. + +#### **2) Cognitive characteristics** + +Cognitive characteristics encompass impairments in memory, attention, decision-making, and information processing speed. Persons with age-related disabilities often face challenges with complex or multi-step procedures. + +Therefore, it is recommended that interfaces consider the following: + +- Provide step-by-step guidance throughout interactions. +- Implement predictable and simple menu structures. +- Use visual emphasis to reinforce learning and reduce cognitive load. + +#### **3) Sensory characteristics** + +Sensory characteristics refer to deterioration in vision, hearing, and tactile perception. Visual impairments may hinder recognition of small or low-contrast text; hearing loss can reduce comprehension of audio cues; and diminished tactile sensitivity can weaken haptic feedback. + +Therefore, it is recommended that interfaces consider the following: + +- Employ high-contrast colours and large fonts. +- Offer multimodal feedback combining auditory, visual, and tactile signals. +- Support alternative sensory channels to compensate for deficits. + +#### **4) Psychological and emotional characteristics** + +Psychological and emotional characteristics include experiences of loneliness, anxiety, social isolation, and the need for emotional support. Emotional well-being and social connection greatly affect user engagement. + +Therefore, it is recommended that interfaces consider the following: + +- Incorporate empathetic interaction designs. +- Provide positive reinforcement and encouragement. + +- Facilitate friendly and natural conversational elements to build emotional rapport. + +#### 5) Environmental characteristics + +Environmental characteristics relate to the user's physical, cultural, and linguistic context. Factors such as lighting, ambient noise, temperature, and device placement affect accessibility. Additionally, language comprehension and cultural symbolism influence user acceptance. + +Therefore, it is recommended that interfaces consider the following: + +- Adapt screen brightness according to ambient lighting conditions. +- Employ robust voice recognition capable of functioning in noisy environments. +- Support ergonomically optimized input methods. +- Use clear, culturally appropriate language and provide multilingual or dialect options. + +### 7.2 Intelligent user interface features for persons with age-related disabilities + +It is recommended intelligent user interfaces (IUIs) be designed for persons with age-related disabilities to reflect age-related characteristics and diverse user needs. + +The key features are as follows: + +- Enhanced accessibility: + - To accommodate physical and sensory limitations, IUIs provide voice commands, gesture recognition, large displays, and high-contrast user interfaces. Contactless control methods improve ease of use. +- Cognitive assistance: + - To reduce memory decline and cognitive load, IUIs offer step-by-step guidance, intuitive menu structures, clear feedback, and notifications. They also support personalized information delivery and repetitive learning. +- Multimodal interaction: + - IUIs support interaction through multiple sensory channels such as visual, auditory, and tactile pathways, enabling communication tailored to individual sensory capabilities. +- Emotional empathy and support: + - To promote emotional stability and social connectedness, IUIs incorporate friendly conversation, positive reinforcement, emotion recognition, and appropriate responses, helping to alleviate loneliness and anxiety. +- Context and environmental adaptability: + - IUIs actively adapt to environmental changes by adjusting screen brightness, improving voice recognition in noisy settings, and providing personalized services based on user context. +- Personalization and customization: + - IUIs automatically adjust interfaces and services based on users' health status, preferences, and usage patterns, delivering personalized experiences. +- Continuous learning and feedback: + - AI-based systems analyse and learn from user behaviour to continuously improve the interface, providing useful feedback and support. + +### 7.3 Service models for persons with age-related disabilities + +This clause presents four representative service models based on IUI technologies that reflect the characteristics and needs of persons with age-related disabilities. + +A smart healthcare service using AI and a speech recognition interface is illustrated in Figure 1. + +- Service model: AI-based care system for persons with age-related disabilities. +- Scope: Real-time monitoring of cognitive and health status and the provision of personalized care services through AI. +- Actors: + - User: Persons with age-related disabilities, family members and caregivers. + - System: AI care system. +- Preconditions: Persons with age-related disabilities wear sensors and wearable devices, and the AI system is operating normally. +- Basic Flow: + - Sensors and input devices collect biometric signals, sleep conditions, health status, medication adherence, and overall well-being data from persons with age-related disabilities. + - The AI and speech recognition interface analyse the collected voice and biometric data to provide appropriate responses based on the user's condition and detect any unusual circumstances compared to normal daily patterns. + - Upon detecting anomalies, the AI sends real-time automated messages to caregivers to enable prompt intervention. + - Personalized health management, notifications, and care actions are provided, with status alerts and feedback communicated to caregivers and family members. +- Postconditions: Maintenance of health status and early risk response for persons with age-related disabilities. +- Special Requirements: It is recommended that AI adjust the intensity and frequency of notifications considering the cognitive decline characteristics of persons with age-related disabilities. + +![Diagram of the Elder care system for persons with age-related disabilities using AI.](d26959f4514c26ca19c3d6f00da85956_img.jpg) + +The diagram illustrates the flow of data and interaction within an 'Elder care system'. On the left, 'Persons with age related disabilities' provide 'Input (User voice)' via a 'Phone call ....'. This input is processed by the 'Elder care system', which contains 'Speech-to-text file' and 'Data storage' components. The data is then sent to 'Analysing data', which in turn connects to an 'AI-based online platform'. From this platform, there are two outputs: 'Output (Text-to-speech)' represented by a speaker icon, and 'Output (Report)' which is sent to a 'Caregiver' represented by three person icons. The reference code 'F.746.21(25)' is shown at the bottom right. + +Diagram of the Elder care system for persons with age-related disabilities using AI. + +**Figure 1 – Care system for persons with age-related disabilities using AI** + +A counselling service model based on artificial emotional intelligence (AEI) technologies is illustrated in Figure 2. + +- Service model: AI-based counselling service. +- Scope: AI counselling service for emotional stability and psychological support for persons with age-related disabilities. + +- Actors: + - User: Persons with age-related disabilities, family member and, psychological experts. + - System: AI counselling system. +- Preconditions: Persons with age-related disabilities have access to the AI counselling service. +- Basic Flow: + - The AI counselling service integrates AEI technologies, including emotional recognition, generation, and augmentation, to analyse the emotional state of persons with age-related disabilities. + - By learning through AI algorithms, it provides personalized emotional support solutions based on comprehensive analysis of human faces, voices, and gestures. + - The service offers advice on appropriate behaviour in various situations such as interviews and supports communication and relationship improvement by identifying emotional relationship characteristics in interactions with others. + - It enhances emotional and psychological stability by understanding emotional states through biometric information, facial expressions, images, or voice. +- Postconditions: Enhanced emotional stability and psychological support for persons with age-related disabilities. +- Special requirements: It is recommended that the AI counsellor accurately recognize persons with age-related disabilities' language patterns and emotional changes. +- Notes: AI counselling services play a vital role in improving emotional health and strengthening social bonds among persons with age-related disabilities. + +![Conceptual diagram of a counselling service using AEI. The diagram shows five input modalities on the left: Facial expression (represented by an eye icon), Acoustic features of voice (represented by a waveform), Gesture (represented by four human figures), Dialogue (represented by a speech bubble with 't e x t'), and an unlabeled icon of a person with a question mark. Arrows from these inputs point to a central 'Machine learning' node, which is depicted as a neural network diagram. From this node, an arrow points to a vertical stack of seven emotion icons: Happy, Sad, Angry, Shocked, Neutral, Excited, and Relaxed. Below this stack is the label 'Emotion' and the code 'F.746.21(25)'. An arrow from the emotion icons points to the text 'Counselling service based on emotional intent'.](007b053fe94a8348f75128a584503fd0_img.jpg) + +Conceptual diagram of a counselling service using AEI. The diagram shows five input modalities on the left: Facial expression (represented by an eye icon), Acoustic features of voice (represented by a waveform), Gesture (represented by four human figures), Dialogue (represented by a speech bubble with 't e x t'), and an unlabeled icon of a person with a question mark. Arrows from these inputs point to a central 'Machine learning' node, which is depicted as a neural network diagram. From this node, an arrow points to a vertical stack of seven emotion icons: Happy, Sad, Angry, Shocked, Neutral, Excited, and Relaxed. Below this stack is the label 'Emotion' and the code 'F.746.21(25)'. An arrow from the emotion icons points to the text 'Counselling service based on emotional intent'. + +**Figure 2 – Conceptual diagram of a counselling service using AEI [b-ITU-T H.862.6]** + +An AI speaker service employing a voice user interface (VUI) to respond to user commands is illustrated in Figure 3. + +- Service model: Proactive AI speaker-based interface for persons with age-related disabilities. +- Scope: + - An AI speaker with IUI services recognizes commands from persons with age-related disabilities and delivers contextually appropriate information with emotionally adaptive responses. + +- Through natural language processing and speech recognition, the system provides personalized information and enables interactive communication tailored to the needs of users. +- Actors: + - User: Persons with age-related disabilities. + - System: AI speaker integrated with IUI services. +- Preconditions: Persons with age-related disabilities are able to interact with the AI speaker through voice. +- Basic Flow: + - The user issues a command using natural language without requiring prior training (intuitiveness and accessibility). + - The AI speaker converts the spoken command into text and matches it with its internal command database (AI and machine learning integration). + - If the command is unclear or ambiguous, the system requests additional clarification from the user (continuous support and feedback). + - Based on the user's response, the system provides appropriate information or performs the requested action (personalization and customization). + - Information is delivered through voice along with visual feedback such as holograms (multimodal interaction). + - The system expresses emotions through voice tone or manner of speaking, enabling a human-like interaction experience (user-friendliness, emotional expression). + - The system learns from interactions to continuously improve the relevance and accuracy of future communications (AI and machine learning integration, continuous support). +- Postconditions: Enhanced convenience and quality of daily life for persons with age-related disabilities. +- Special Requirements: It is recommended that the AI incorporate the pronunciation characteristics of persons with age-related disabilities to improve speech recognition accuracy. Additionally, the user interface (UI) response time should be adjusted to accommodate the interaction pace typical of users. + +![Figure 3: AI speaker of active interaction tailored to persons with age-related disabilities. The diagram shows a flow from 'Input (User voice)' to a 'Recording voice files' block, which is part of a larger 'Voice files and network communication control sensor and output control' unit. This unit connects to a 'Speech API (Speech-to-text)' and a 'Speech synthesis API', both of which are connected to a 'Conversational user interface server'. The 'Speech synthesis API' also connects to a 'Speech-to-text file' and an 'Output control' block within the central unit. The 'Output control' block connects to an 'Output (Text-to-speech)' block, which in turn connects to a 'Hologram LED control' unit containing a 'Motor control board' and a 'Motor'. The 'Voice files and network communication control sensor and output control' unit also connects to a 'Temperature, humidity, illumination Sensor', which connects to a 'Temperature, humidity, illumination Database', which then connects to a 'Database value charting Server', which finally connects to a 'User movement monitoring' block. The reference code F.746.21(25) is shown at the bottom right.](4801720824e4b5e2361a5564f91cfb70_img.jpg) + +``` + +graph TD + Input[Input User voice] --> RVF[Recording voice files] + subgraph CentralUnit [Voice files and network communication control sensor and output control] + RVF + STTF[Speech-to-text file] + OC[Output control] + end + RVF --> SAT[Speech API Speech-to-text] + SAT --> CUIS[Conversational user interface server] + CUIS --> SSA[Speech synthesis API] + SSA --> STTF + SSA --> OC + OC --> OTS[Output Text-to-speech] + OTS --> HLC[Hologram LED control] + subgraph HLC [Hologram LED control] + MCB[Motor control board] + Motor[Motor] + end + CentralUnit --> THIS[Temperature, humidity, illumination Sensor] + THIS --> THD[Temperature, humidity, illumination Database] + THD --> DVS[Database value charting Server] + DVS --> UMM[User movement monitoring] + F7462125[F.746.21(25)] + +``` + +Figure 3: AI speaker of active interaction tailored to persons with age-related disabilities. The diagram shows a flow from 'Input (User voice)' to a 'Recording voice files' block, which is part of a larger 'Voice files and network communication control sensor and output control' unit. This unit connects to a 'Speech API (Speech-to-text)' and a 'Speech synthesis API', both of which are connected to a 'Conversational user interface server'. The 'Speech synthesis API' also connects to a 'Speech-to-text file' and an 'Output control' block within the central unit. The 'Output control' block connects to an 'Output (Text-to-speech)' block, which in turn connects to a 'Hologram LED control' unit containing a 'Motor control board' and a 'Motor'. The 'Voice files and network communication control sensor and output control' unit also connects to a 'Temperature, humidity, illumination Sensor', which connects to a 'Temperature, humidity, illumination Database', which then connects to a 'Database value charting Server', which finally connects to a 'User movement monitoring' block. The reference code F.746.21(25) is shown at the bottom right. + +**Figure 3 – AI speaker of active interaction tailored to persons with age-related disabilities** + +A smart health services using speech recognition interface illustrated in Figure 4. + +- Service model: Smart health service based on a speech recognition interface. +- Scope: This service model defines a smart health service that leverages a speech recognition-based user interface (SUI) to detect the voice and emergency situations of persons with age-related disabilities, initiate interaction, and automatically request assistance when necessary. +- Actors: + - User: Persons with age-related disabilities. + - System: Smart health system based on a speech recognition interface (including a smart speaker). +- Preconditions: The user is within range of a device equipped with a microphone; the system has active speech recognition and a registered emergency contact. +- Basic Flow: + - The system receives and analyses voice input through embedded sensors (intuitiveness and accessibility). + - Speech recognition algorithms detect abnormal patterns or emergency indicators (AI and machine learning integration). + - If an emergency is suspected, the system attempts to initiate a conversation with the user (continuous support and feedback). + - If no response is received, the system sends an alert to the emergency contact system (safety, compatibility). + - The user's health status and emergency information are forwarded to the relevant medical institution (multimodal interaction, compatibility). + - The service is deployed through the smart speaker interface and can be easily implemented in existing devices (user-friendliness, compatibility). +- Postconditions: Enhancement of healthcare accessibility and timely response to emergencies. + +- Special Requirements: + - It is recommended that customized recognition models be trained on the speech and emergency vocal cues of persons with age-related disabilities. + - Local processing or encryption technologies should be applied to ensure privacy protection. + - A highly responsive interface is required to avoid misdetection and ensure seamless interaction. + +![Figure 4: Care services for persons with age-related disabilities based on speech recognition interface. The diagram illustrates a flow from sensor-based detection to emergency response. It shows three potential events (Heart attack, Fall, Stroke) leading to a state where conversation is not possible. This triggers a conversation system activation (health status feedback, emergency connections) which then leads to an emergency connection (represented by a red cross and a star icon). A red line graph labeled 'Event occurred' is shown below the events.](bd0b93e7a46ede276d0a3b79ac487bd9_img.jpg) + +The diagram illustrates the care services for persons with age-related disabilities based on a speech recognition interface. It consists of several interconnected components: + +- Sensor based emergency detection** leads to **Active speech recognition to identify situation**, which leads to **Real-time speech recognition and response**, which leads to **Emergency rescue contact system**. +- Below the flow, three potential events are shown: **Heart attack** (represented by a heart with a red cross), **Fall** (represented by a person falling), and **Stroke** (represented by a brain with a yellow starburst). A red line graph labeled **Event occurred** is shown below these events. +- An arrow from the **Real-time speech recognition and response** box points to a box containing an icon of a person with a cane and the text **Determine location**. This arrow is labeled **Conversation is not possible**. +- An arrow from the **Determine location** box points to a box containing two icons: a red cross and a star. This arrow is labeled **Conversation system activated (Health status feedback, emergency connections)**. +- The final box is labeled **Emergency connection**. +- The text **F.746.21(25)** is shown in the bottom right corner. + +Figure 4: Care services for persons with age-related disabilities based on speech recognition interface. The diagram illustrates a flow from sensor-based detection to emergency response. It shows three potential events (Heart attack, Fall, Stroke) leading to a state where conversation is not possible. This triggers a conversation system activation (health status feedback, emergency connections) which then leads to an emergency connection (represented by a red cross and a star icon). A red line graph labeled 'Event occurred' is shown below the events. + +**Figure 4 – Care services for persons with age-related disabilities based on speech recognition interface** + +## 8 Conceptual model of IUI for persons with age-related disabilities care systems + +This clause describes the conceptual model of IUI in care systems for persons with age-related disabilities. The model aims to define the data flows and interaction flows among hardware components, software services, and intelligent systems that facilitate personalized, adaptive, and safe user experiences. + +The input and output data are multimodal and incorporate various modalities such as speech, text, video, facial expressions, and gestures for both user interaction and system response. In some intelligent care systems for persons with age-related disabilities, the emotional content of the input data serves as a critical parameter for system functions and personalized service delivery. + +### 8.1 Components of the conceptual model + +The IUI system consists of the following main components: + +- User: Persons with age-related disabilities. +- HW: + - Input devices: Microphone (speech), camera or depth sensor (video, facial expression, gesture). + +- Output devices: Speaker (for speech), display/projector (for facial animation or visual cues). +- Embedded systems: Edge computing processors for low-latency processing. +- SW: + - Software that processes multimodal inputs to extract meaning and emotion. +- AI/ML: + - Algorithms that analyse user data, detect changes in cognitive or physical states, and adapt the UI accordingly for personalization and continuous support. +- External service interfaces. + - Integration points for healthcare, emergency response, and other support services to enhance safety and provide timely assistance. + +### 8.2 Input and output module + +This module is responsible for input and output processing functions within the IUI system, focusing on the integrated analysis of multimodal signals and the provision of personalized feedback. + +- **Input:** Multimodal signals (e.g., voice, gesture, biosignals) are received through sensors or user input devices. +- **Processing:** AI models analyse intent, context, and emotional state, referencing personalization settings. +- **Output:** The system provides appropriate feedback (e.g., voice, visual cues, ambient signals) tailored to the user's needs. +- **Feedback:** User responses and system logs are used for further tuning and real-time error handling. + +Table 1 shows the input and output elements of the IUI service. + +**Table 1 – Input and output elements of IUI** + +| Category | Element | Description | +|-------------------|---------------------|----------------------------------------------------------------------------------------------------| +| Input | Voice | Voice signals captured via microphone | +| | Gesture | Hand and body movements detected by cameras/sensors | +| | Biosignals | Physiological signals such as heart rate, electroencephalogram (EEG), galvanic skin response (GSR) | +| Processing | Intent analysis | AI models for natural language and behaviour recognition | +| | Context analysis | Interpretation of situational and environmental data | +| | Emotion recognition | AI algorithms for detecting emotional states | +| | Personalization | Incorporation of user profile and preferences | +| Output | Voice feedback | Speech synthesis and voice output | +| | Visual cues | Visual information displayed on screens or devices | +| | Ambient signals | Environmental signals such as lighting and vibration | +| Feedback | User response | Collection of user reactions | +| | System logs | Logs of errors and usage history | +| | Real-time tuning | Real-time adjustment of AI models based on feedback | + +### 8.3 Data and interaction flow + +The interaction process within the conceptual model consists of the following steps: + +- **Data collection:** Sensors and input devices collect user activity, physiological signals, and environmental context. +- **Analysis:** AI/ML algorithms process the collected data to detect cognitive or physical changes, usage patterns, and potential risks. +- **Adaptation:** Based on the analysis results, the UI dynamically adapts to simplify navigation, adjust input methods, and customize feedback. +- **Feedback:** Continuous support is provided through alerts, assistance, or adjustments, enhancing user confidence and safety. +- **User interaction:** The user interacts through multimodal inputs (voice, touch, gesture), with the system responding in an intuitive and accessible manner. + +Table 2 shows data flow and interaction processes in the IUI system. + +**Table 2 – Data flow and interaction processes in the IUI system** + +| From → To | Description | +|---------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| User → Input devices | Users provide input via voice, touch, motion and other modalities. | +| Input devices → AI/ML | Collected data are analysed by AI. | +| AI/ML → Software | The UI is customized for the user. | +| Software → User | Adaptive UI feedback is provided (e.g., screen, voice, feedback). | +| AI/ML ↔ External services | Integration with external services such as health information and emergency response.
External service data are exchanged with AI/ML for enhanced personalization. | + +### 8.4 Conceptual model diagram + +The conceptual model includes a bidirectional interaction flow between persons with age-related disabilities and the IUI system. + +Input modalities such as voice, touch, and environmental sensors provide real-time data to the processing unit of the system. These data are interpreted by AI/ML models and delivered through multimodal output formats such as visual cues, voice feedback, and tactile responses. These input and output layers are essential for ensuring the intuitiveness, accessibility, and real-time responsiveness of IUI services. + +Figure 5 illustrates the conceptual model diagram of the IUI system. + +![Flowchart of the conceptual model of IUI for care systems for persons with age-related disabilities. The flow consists of five rectangular boxes connected by downward-pointing arrows. The first box is 'Person with age-related disabilities (Age-related needs)'. The second box is 'Input collection (Sensor, camera, mic, UI devices)' with the text 'Input: Voice, gesture, biosignals' above it. The third box is 'AI/ML processing unit' with sub-points: '- Emotion recognition', '- Intent prediction', and '- Personalization logic'. The fourth box is 'Output module' with sub-points: '- Voice feedback', '- Visual display', and '- Haptic or ambient signals'. The fifth box is 'Monitoring and feedback loop' with sub-points: '- User satisfaction', '- Error handling', and '- Learning from interaction'. A small code 'F.746.21(25)' is at the bottom right of the diagram.](76b0cd79baaedd942af4dc42f2e764b8_img.jpg) + +``` + +graph TD + A["Person with age-related disabilities +(Age-related needs)"] -- "Input: Voice, gesture, biosignals" --> B["Input collection +(Sensor, camera, mic, UI devices)"] + B --> C["AI/ML processing unit +- Emotion recognition +- Intent prediction +- Personalization logic"] + C --> D["Output module +- Voice feedback +- Visual display +- Haptic or ambient signals"] + D --> E["Monitoring and feedback loop +- User satisfaction +- Error handling +- Learning from interaction"] + +``` + +F.746.21(25) + +Flowchart of the conceptual model of IUI for care systems for persons with age-related disabilities. The flow consists of five rectangular boxes connected by downward-pointing arrows. The first box is 'Person with age-related disabilities (Age-related needs)'. The second box is 'Input collection (Sensor, camera, mic, UI devices)' with the text 'Input: Voice, gesture, biosignals' above it. The third box is 'AI/ML processing unit' with sub-points: '- Emotion recognition', '- Intent prediction', and '- Personalization logic'. The fourth box is 'Output module' with sub-points: '- Voice feedback', '- Visual display', and '- Haptic or ambient signals'. The fifth box is 'Monitoring and feedback loop' with sub-points: '- User satisfaction', '- Error handling', and '- Learning from interaction'. A small code 'F.746.21(25)' is at the bottom right of the diagram. + +**Figure 5 – Conceptual model of IUI for care systems for persons with age-related disabilities** + +## 9 Requirements for common components and characteristics of the intelligent UI service + +This clause defines the common components and characteristics that form the foundation of IUI services for persons with age-related disabilities. These elements ensure consistency, interoperability, and usability across various service scenarios. + +### 9.1 Requirements for common components + +Common components include: + +- User profile and context manager: + - It is recommended to maintain demographic information, cognitive and sensory abilities, and user preferences of persons with age-related disabilities. + - It is recommended to update the profile dynamically according to usage history and real-time conditions. +- Multimodal input processing units: + - It is recommended to integrate data collected through voice, gestures, and biosignals. + - It is recommended to perform synchronization and fusion of multimodal data. +- AI/ML analysis engine: + - It is recommended to execute intent recognition, context analysis, and emotion detection. + +- It is recommended to adaptively adjust service parameters to provide personalized services. +- Interaction management module: + - It is recommended to control the flow of user interaction to support smooth dialogue and task execution. + - It is recommended to maintain a balance between automation and user control. +- Multimodal output generator: + - It is recommended to provide information through voice, visual, and ambient feedback channels. + - It is recommended to adjust output methods according to the user's sensory characteristics. +- Feedback and learning loop: + - It is recommended to collect system usage logs and user responses. + - It is recommended to support long-term personalization and improve service quality. +- External service interface: + - It is recommended to connect with healthcare systems, caregivers, and emergency response services. + - It is recommended to ensure interoperability with existing information and communication technology (ICT) and Internet of things (IoT) platforms. + +### 9.2 Requirements for common characteristics + +IUI services for persons with age-related disabilities share several common characteristics that support usability, accessibility, and trust. These characteristics describe how the services should behave and respond across different applications. + +- Accessibility and inclusiveness: + - It is recommended to design interfaces to be easily used by persons with age-related disabilities with different physical, cognitive, and sensory abilities. +- Adaptability and personalization: + - It is recommended to allow customization of font size, colour, voice speed, and interaction pace. +- Multimodality: + - It is recommended to support multiple input and output channels such as voice, text, gesture, and vibration. + - It is recommended to provide redundant communication paths (e.g., both audio and visual) to reduce the risk of misunderstanding. +- Empathy and emotional responsiveness: + - It is recommended to include empathetic interaction methods such as friendly voice, positive feedback, and natural conversation. +- Reliability and trustworthiness: + - It is recommended to ensure the system is stable, safe, and consistent in its operation. +- Interoperability: + - It is recommended to support compatibility with different devices, platforms, and external services such as healthcare or emergency systems. + +- Continuous improvement: + - It is recommended to collect feedback and usage data to improve the service over time. + - It is recommended to apply AI/ML learning techniques to adapt the interface as user needs evolve. + +## **10 Intelligent UI service requirements from a hardware perspective** + +IUI hardware requirements are required to consider the physical, cognitive, and sensory characteristics of persons with age-related disabilities. The requirements cover input devices, output devices, ergonomic design, and connectivity. + +- Input devices + - Input devices are required to support various input methods such as touch, voice, gesture, and physical buttons. + - Input devices are recommended to be designed with a large size and wide spacing to minimize erroneous operations. + - Input devices are recommended to provide immediate feedback (e.g., vibration, audio, visual indication) to user actions. +- Output devices + - Displays are required to support large fonts, high contrast, and wide viewing angles. + - Output devices are recommended to provide adjustable sound output and caption functions for users with hearing impairments. + - Tactile feedback (e.g., vibration, Braille) is recommended to be provided as a complementary channel. + - Emergency alerts are required to provide simultaneous visual, auditory, and tactile signals. +- Ergonomic design + - Devices are recommended to be lightweight and designed to minimize burden for persons with age-related disabilities during long-term use. + - Devices are recommended to be designed with consideration of the hand size, muscular strength, and reaction speed of persons with age-related disabilities. +- Connectivity and interoperability + - Devices are recommended to support wireless communication such as Bluetooth, Wi-Fi, and NFC. + - Devices are recommended to ensure interoperability with smart home and healthcare devices. + +## **11 Intelligent UI service requirements from a software perspective** + +- User interface: + - UI is recommended to minimize the number of steps required to access core functions. + - UI is recommended to be designed so that users can adjust font size, colour contrast, and language expressions. + - UI is recommended to maintain consistent menu structures and icon design. +- Accessibility: + - Software is recommended to support voice interfaces optimized for the speech characteristics of persons with age-related disabilities. + +- It is recommended to utilize the multimodal feedback functions of hardware output devices to provide combined interaction through visual, auditory, and tactile channels. + +- User experience: + - Software is recommended to provide tutorials, step-by-step guidance, and voice assistance to improve learnability. + - It is recommended that software encrypt sensitive information (e.g., health, location, financial data) for storage and transmission. + +Error messages are recommended to be expressed in simple and intuitive language. + +- Security and privacy: + - Software is recommended to encrypt sensitive information (e.g., health, location, financial data). + - Software is recommended to provide functions that allow caregivers and family members to remotely manage the system with limited authority. +- Intelligent features: + - Software is recommended to provide multimodal notifications (visual, auditory, tactile) for medication, schedules, and emergency situations. + - Software is recommended to learn user habits and provide personalized function recommendations. + +## Bibliography + +- [b-ITU-T F.791] Recommendation ITU-T F.791 (2018), *Accessibility terms and definitions*. +- [b-ITU-T F.902] Recommendation ITU-T F.902 (1995), *Interactive services design guidelines*. +- [b-ITU-T H.862.6] Recommendation ITU-T H.862.6 (2023), *Functional requirements for counselling services based on artificial emotional intelligence*. +- [b-ITU-T Y.3172] Recommendation ITU-T Y.3172 (2019), *Architectural framework for machine learning in future networks including IMT-2020*. +- [b-ETSI GR ENI 004] ETSI GR ENI 004 V1.1.1 (2018), *Experiential Networked Intelligence (ENI); Terminology for Main Concepts in ENI*. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.746.3-201511-I_PDF-E/raw.md b/marked/F/T-REC-F.746.3-201511-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..2573c8647fdce80ec5cf717366cec1eddcdf0a60 --- /dev/null +++ b/marked/F/T-REC-F.746.3-201511-I_PDF-E/raw.md @@ -0,0 +1,523 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.746.3** + +(11/2015) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Audiovisual services + +--- + +**Intelligent question answering service +framework** + +Recommendation ITU-T F.746.3 + +![ITU logo: A globe with a red lightning bolt and the text 'ITU International Telecommunication Union'.](84a1d09fb489061482111515543b60dc_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a stylized globe with a red lightning bolt striking across it. To the right of the globe, the text "ITU" is written in a bold, blue font, and below it, "International Telecommunication Union" is written in a smaller, blue font. + +ITU logo: A globe with a red lightning bolt and the text 'ITU International Telecommunication Union'. + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| AUDIOVISUAL SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.746.3 + +# Intelligent question answering service framework + +## Summary + +Recommendation ITU-T F.746.3 defines the intelligent question answering service (IQAS) framework as well as the features, requirements and the functionality to support the intelligent question answering (QA) service. The intelligent QA system is an advanced function to generate answers to a user's questions in a natural language. More and more systems in the future are expected to be equipped with QA functions for an advanced user experience. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.746.3 | 2015-11-29 | 16 | 11.1002/1000/12619 | + +## Keywords + +Intelligent system, natural language processing, QA, question answering. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2016 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|--------------------------------------------------------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 3 | +| 6 Introduction..... | 3 | +| 7 Requirements ..... | 4 | +| 7.1 General requirements for intelligent question answering service and the system to support the QA service ..... | 4 | +| 7.2 Natural language processing requirements..... | 5 | +| 7.3 Question analysis requirements ..... | 5 | +| 7.4 Candidate answer generation requirements..... | 6 | +| 7.5 Answer inference and generation requirements ..... | 6 | +| 8 Functional components and interfaces for intelligent question answering service ..... | 7 | +| 8.1 Natural language processing..... | 7 | +| 8.2 Question analysis..... | 9 | +| 8.3 Candidate answer generation..... | 11 | +| 8.4 Answer inference and generation ..... | 12 | +| Bibliography..... | 13 | + + + +# Recommendation ITU-T F.746.3 + +## Intelligent question answering service framework + +# 1 Scope + +This Recommendation provides an overview of the intelligent question answering service (IQAS) framework and describes the features, general requirements and functionality which form a framework to support the intelligent question answering (QA) service. The scope of this Recommendation focuses on QA architecture, devices, servers, interfaces and communication protocols among QA servers and clients in a high-level approach. + +# 2 References + +None. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 application** [b-ITU-T Y.101]: A structured set of capabilities, which provide value-added functionality supported by one or more services. + +**3.1.2 content** [b-ITU-T H.780]: A combination of audio, still image, graphic, video, or data. + +NOTE – Variety of formats is classified as the "data" (e.g., text, encoded values, multimedia description language introduced by [b-ITU-T H.760]). + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 co-reference resolution**: A function that detects the preceding referents of the pronouns which replace the noun phrases of the input sentences. + +**3.2.2 CV\_occupation**: A named entity tag that belongs to the civilization (CV) category that indicates a person's occupation (e.g., scientist). + +**3.2.3 CV\_prize**: A named entity tag that belongs to the civilization (CV) category that indicates a prize (e.g., the Nobel prize). + +**3.2.4 IsA relation**: A hierarchical relation between two words in such a way that the meaning of one word includes that of the other word. + +**3.2.5 knowledge base**: A collection of knowledge resources that consist of structured and unstructured data. The knowledge base is used to provide information to the various applications that are related to information provision such as the QA system and search system. + +**3.2.6 named entity recognition**: A function that recognizes named entities such as PLO entities which are people, locations and organizations from the sentences. The PLO can be decomposed into more specific named entities depending on the applications. + +**3.2.7 natural language processing**: A method that analyses text in natural languages through several processes such as part-of-speech recognition, syntactic analysis and semantic analysis. + +**3.2.8 part-of-speech recognition**: A function that recognizes parts of speech (POS) in the sentences and assigns relevant POS tags considering contextual meaning of the target sentences. + +**3.2.9 question answering:** A system that provides answers in natural language to questions which are in natural language form by analysing the questions and all the knowledge resources that are available to the system. + +**3.2.10 semantic analysis:** A function that recognizes the semantic relations among the words around predicates that exist in the same sentence. The semantic analysis function then generates a semantic predicate-argument structure (PAS). + +**3.2.11 syntactic analysis:** A function that analyses sentence structures and generates dependency relations among words based on dependency grammars. + +# **4 Abbreviations and acronyms** + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|--------|----------------------------------------| +| ARG | Argument | +| CV | Civilization | +| DB | Database | +| DS | Digital Signage | +| ID | Identifier | +| IQAS | Intelligent Question Answering Service | +| KB | Knowledge Base | +| NLP | Natural Language Processing | +| NLU | Natural Language Understanding | +| NNG | Noun General | +| NNP | Proper Noun | +| NP_OBJ | Noun Phrase Objective | +| NP_SBJ | Noun Phrase Subjective | +| PAS | Predicate-Argument Structure | +| PLO | People, Locations and Organizations | +| POS | Parts Of Speech | +| QA | Question Answering | +| SL | Sub Language | +| SPARQL | SPARQL protocol and RDF Query Language | +| SQL | Structured Query Language | +| VP_MOD | Verb Phrase Modal | +| VV | Verb | +| XSNG | Auxiliary Suffix from Noun General | +| XSV | Auxiliary Suffix from Verb | + +# 5 Conventions + +The following conventions are used in this Recommendation: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted, if conformance to this Recommendation is to be claimed. +- The keywords "is prohibited from" indicate a requirement which must be strictly followed and from which no deviation is permitted, if conformance to this Recommendation is to be claimed. +- The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. +- The keywords "is not recommended" indicate a requirement which is not recommended but which is not specifically prohibited. Thus, conformance with this Recommendation can still be claimed even if this requirement is present. +- The keywords "can optionally" indicate an optional requirement which is permissible, without implying any sense of being recommended. This term is not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with this Recommendation. + +# 6 Introduction + +The intelligent question answering (QA) system is an advanced function to generate answers to a user's question in a natural language. More and more systems in the future are expected to be equipped with QA functions for an advanced user experience. Currently, the most widely recognized QA system is an intelligent agent which is provided in smart phones. Although not a completely intelligent agent able to answer all questions that users may ask, a QA service would be useful for a defined environment. A QA service would be useful if a standardized QA service framework existed which could be applied to various domains such as TV-related services, digital signage (DS) and e-health services. A QA service would also be useful in terms of accessibility because natural languages are used to request and provide information conveniently with a speech interface. Figure 1 illustrates the concept of a QA system in the DS domain, where a user requests information about shoe stores from the QA system at a shopping mall. + +- Scenario of an intelligent question answering service (IQAS) system: An IQAS system at a shopping mall. + +Michelle visits a shopping mall in Seattle. There is a DS helper which displays information about the various shops at the mall. It also answers customers' questions about the shops. Michelle wants to buy a pair of shoes and asks the DS helper "Where can I find some nice shoes?" The DS helper answers "Go to the third floor". Michelle goes to the third floor where she can find some shoe stores. The answer can also be shown in text form on the screen or can be provided by speech synthesis in a spoken form according to the needs of the user. + +![Figure 1: Digital signage with a QA function: At the mall. The diagram shows a digital signage screen displaying a mall interior. A person is interacting with the screen, asking 'Go to the third floor.' and 'Where can I find some nice shoes?'. Above the screen, a 'QA server' is connected to a series of server icons. The text 'F.746.3(15)_F01' is visible in the bottom right corner.](a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg) + +Figure 1: Digital signage with a QA function: At the mall. The diagram shows a digital signage screen displaying a mall interior. A person is interacting with the screen, asking 'Go to the third floor.' and 'Where can I find some nice shoes?'. Above the screen, a 'QA server' is connected to a series of server icons. The text 'F.746.3(15)\_F01' is visible in the bottom right corner. + +**Figure 1 – Digital signage with a QA function: At the mall** + +Figure 2 presents an example of functional blocks in QA architecture. The QA system consists of several functional blocks: natural language processing (NLP), question analysis, candidate answer generation, answer inference and answer generation. + +![Figure 2: Example of functional blocks in QA architecture. This flowchart illustrates the QA process: Question -> Question analysis (containing Question understanding) -> Question decomposition -> Candidate answer generation (containing Multiple information search) -> Answer inference (containing Confidence calculation) -> Answer generation (containing Best answer generation) -> Answer. A 'Natural language processing' block feeds into Question analysis and Question decomposition. A globe and network icons represent the information sources. The text 'F.746.3(15)_F02' is visible in the bottom right corner.](e6df2733626a85205c1db682e6259c46_img.jpg) + +Figure 2: Example of functional blocks in QA architecture. This flowchart illustrates the QA process: Question -> Question analysis (containing Question understanding) -> Question decomposition -> Candidate answer generation (containing Multiple information search) -> Answer inference (containing Confidence calculation) -> Answer generation (containing Best answer generation) -> Answer. A 'Natural language processing' block feeds into Question analysis and Question decomposition. A globe and network icons represent the information sources. The text 'F.746.3(15)\_F02' is visible in the bottom right corner. + +**Figure 2 – Example of functional blocks in QA architecture** + +# 7 Requirements + +## 7.1 General requirements for IQAS and the system to support the QA service + +- The IQAS is required to provide answers to the user's various questions considering the user's intention. +- The IQAS is recommended to provide answers in natural language, whether it is in text or in spoken form depending on the user's environment. +- The IQAS is recommended to provide answers in natural language from multiple information sources. +- The IQAS is recommended to provide accessibility for different user needs. + +- The IQAS is recommended to provide answers to the user's questions within a proper time span. +- It is recommended to provide an input editor for the user's question whether it consists of a single sentence or multiple sentences. +- It is recommended that the user's question is input in voice or text form depending on the user's device. +- The IQAS is recommended to provide example questions for the user's convenience. +- The IQAS is recommended to provide a manual on the use of the system for the user. +- The IQAS is recommended to provide a human intervention mechanism where the reliability of the answer is below a predefined threshold value. +- The IQAS is recommended to be able to save and update profile information such as the history of users' questions and answers. +- The IQAS is recommended to be able to provide answers appropriate to the user's level of knowledge. + +## 7.2 Natural language processing requirements + +The following requirements relate to NLP in an IQAS and the system to support the IQAS: + +- The IQAS system is recommended to include a NLP module that analyses natural language sources at a deep level. +- The NLP module is recommended to include a pre-processing block that filters non-essential text to better understand the text and increase the speed of the IQAS. +- The NLP module is recommended to include a part of speech processing block that analyses the natural language text and assigns the correct part of speech tag to the output. +- The NLP module is recommended to include a named entity processing block that analyses the natural language text and assigns the correct named entity tag to the output for language understanding. +- The NLP module is recommended to include a syntactic analysis processing block that analyses the structure of the natural language sentences to help understand the information sources. +- The NLP module is recommended to include a semantic analysis processing block that analyses the meaning and the role of each semantic part of the natural language sentences to help understand the information sources. +- The NLP module is recommended to include a co-reference resolution processing block that relates nouns with the pronouns that refer to the nouns to help understand the information sources. +- The NLP module is recommended to continuously update all the language analysis results and extract time/location information and semantic relation information among words based on the continuous update for the IQAS. + +## 7.3 Question analysis requirements + +The following requirements relate to question analysis in an IQAS and the system to support the IQAS: + +- The question analysis module is recommended to analyse the user's question and understand the intention of the user. +- The question analysis module is recommended to process the user's question, whether it comprises of a single sentence, complex clauses or multiple sentences. + +- The question analysis module is recommended to find relevant information on the type of candidate answers. +- The question analysis module is recommended to decide the best strategy to find evidence for the correct answer and to present it to the user. +- The question analysis module is recommended to include the functions of question classification, answer type recognition, dictionary title recognition, question decomposition and semantic information analysis for the better understanding of the user's question. +- It is recommended that the question decomposition results be presented as an intermediate analysis result. +- It is recommended that the final question analysis results be presented to the user. +- It is recommended that the question paraphrases or question hypotheses which have the same meaning as the user's question be provided with reliability information. +- The question analysis module is recommended to be able to make inferences on the user's level of knowledge. + +## **7.4 Candidate answer generation requirements** + +The following requirements relate to candidate answer generation in an IQAS and the system to support the IQAS: + +- The candidate answer generation module is recommended to provide related language knowledge that is the basis for answer candidates to the user's question. +- The candidate answer generation module is recommended to use the existing knowledge bases for the candidate answer generation. +- The candidate answer generation module is recommended to add new language knowledge increasingly to the existing knowledge bases. +- The candidate answer generation module is recommended to provide the interface to retrieve taxonomy from the knowledge bases. +- It is recommended that the candidate answer generation module provides the ranking of the information that it finds. +- The IQAS system is required to search for the candidate answers from multiple information sources. +- The IQAS system is recommended to have various indexed databases of all different informal documents +- The IQAS system, based on the question analysis results, is recommended to provide the search results, answer rankings and weight information extracted from each of the informal indexed databases. + +## **7.5 Answer inference and generation requirements** + +The following requirements relate to answer inference and generation in an IQAS and the system to support the IQAS: + +- The IQAS system is recommended to calculate the reliability score for each candidate answer based on multiple proofs to support the candidate answers. +- The information sources are recommended to include big data such as web data, Wikipedia and different knowledge bases that are publically available. +- It is recommended to provide the source of the extracted documents, paragraphs and sentences to be able to confirm the reasons for the searched candidate answers. + +- The IQAS system is required to provide the best answers and their ranks to the user from among the candidate answers based on the confidence scores. +- It is recommended to provide the information on the reliability of the best answer. +- It is recommended to provide the best answer from among multiple candidate answers. +- When the answer is presented by real-time answer generation, the documents and paragraphs from which the answer is extracted are recommended to be presented along with the answer. +- When the answer is presented based on the knowledge base (KB), the information on the KB is recommended to be provided accordingly. + +# 8 Functional components and interfaces for intelligent question answering service + +IQAS functional components are basically composed of natural language processing, question analysis, candidate answer generation, answer inference and answer generation functional blocks as follows: + +- **Natural language processing functional block:** supports natural language processing on the QA server; +- **Question analysis functional block:** supports question analysis on the terminal; +- **Candidate answer generation functional block:** supports candidate answer generation by searching information in various databases (DBs); +- **Answer inference/generation functional block:** supports answer inference based on feature normalization and ranking of candidate answers and best answer generation on the terminal. + +Functional blocks of the intelligent question answering service are shown in Figure 3. + +![Figure 3 – Logical functional block diagram of an IQAS. The diagram illustrates the flow of information through four main functional blocks: Question analysis, Candidate answer generation, Answer inference/generation, and Natural language processing. The process starts with a 'Question' input to the 'Question analysis' block, which includes 'Question classification', 'User knowledge inference', 'Question decomposition', 'Question intermediate format generation', and 'Question hypothesis generation'. This block feeds into the 'Candidate answer generation' block, which includes 'Real-time information search', 'Real-time candidate answer generation', 'Multi-source candidate answer generation', 'Text based candidate answer generation', and 'KB-based candidate answer generation'. This block feeds into the 'Answer inference/generation' block, which includes 'Type coercion', 'Answer verification', 'Answer constraints', 'Ranking candidate answers', and 'Answer synthesis'. The output of this block is 'Answer reliability, reason', which leads to 'Human intervention' and then to the final 'Answer'. The 'Natural language processing' block, which includes 'POS recognition', 'Named entity recognition', 'Syntactic analysis', 'Semantic analysis', 'Co-reference resolution', 'Time/location information extraction', 'Semantic relation extraction', 'Continuous linguistic knowledge acquisition/extraction', and 'Learning and extraction of linguistic knowledge', is shown below the other blocks and feeds into the 'Candidate answer generation' block. The entire diagram is enclosed in a dashed box, and the label 'F.746.3(15)_F03' is in the bottom right corner.](410562339ce067fdc6fa41940c118658_img.jpg) + +Figure 3 – Logical functional block diagram of an IQAS. The diagram illustrates the flow of information through four main functional blocks: Question analysis, Candidate answer generation, Answer inference/generation, and Natural language processing. The process starts with a 'Question' input to the 'Question analysis' block, which includes 'Question classification', 'User knowledge inference', 'Question decomposition', 'Question intermediate format generation', and 'Question hypothesis generation'. This block feeds into the 'Candidate answer generation' block, which includes 'Real-time information search', 'Real-time candidate answer generation', 'Multi-source candidate answer generation', 'Text based candidate answer generation', and 'KB-based candidate answer generation'. This block feeds into the 'Answer inference/generation' block, which includes 'Type coercion', 'Answer verification', 'Answer constraints', 'Ranking candidate answers', and 'Answer synthesis'. The output of this block is 'Answer reliability, reason', which leads to 'Human intervention' and then to the final 'Answer'. The 'Natural language processing' block, which includes 'POS recognition', 'Named entity recognition', 'Syntactic analysis', 'Semantic analysis', 'Co-reference resolution', 'Time/location information extraction', 'Semantic relation extraction', 'Continuous linguistic knowledge acquisition/extraction', and 'Learning and extraction of linguistic knowledge', is shown below the other blocks and feeds into the 'Candidate answer generation' block. The entire diagram is enclosed in a dashed box, and the label 'F.746.3(15)\_F03' is in the bottom right corner. + +Figure 3 – Logical functional block diagram of an IQAS + +## 8.1 Natural language processing + +The NLP functional block consists of the natural language understanding function and the linguistic knowledge learning and extraction function. The natural language understanding function analyses the user's questions as well as target documents for knowledge extraction using the technologies of semantic level analysis such as semantic entity recognition and co-reference resolution in addition to the well-known parts of speech (POS) recognition, named entity recognition and syntactic analysis for sentence structures. The linguistic knowledge learning and extraction function shown in Figure 4, extracts semantic relations among words and time/location information in the documents and provides the framework to continuously acquire linguistic knowledge and to extend it to deal with new knowledge. + +- Natural language understanding function consists of: the POS recognition sub-function, named entity recognition sub-function, syntactic analysis sub-function, semantic entity recognition sub-function and the co-reference resolution sub-function. + - Part of speech recognition: + +The POS recognition sub-function analyses sentences by tagging POS tags to the minimum semantic unit which has the meaning. The POS analysis technology recognizes parts of speech in the sentences and documents and assigns relevant POS tags considering contextual meaning of the target sentences. The technology to analyse POS includes rule-based methods and statistical recognition methods. + - Named entity recognition: + +The named entity recognition sub-function recognizes named entities such as people, locations and organizations (PLO) from the sentences. The PLO can be decomposed to more detailed named entities depending on the applications. Both the rule-based method and the statistical method are used to analyse named entities. + - Syntactic analysis: + +The syntactic analysis sub-function analyses sentence structures that are input to the module and generates dependency relations among words based on dependency grammars. It performs the functions of head-dependant recognition function, labelling of sentence structure for the dependency relations recognized and storing the recognized dependency and labels to the predefined data structures. + - Semantic analysis: + +The function of the semantic analysis sub-functional block is to recognize the semantic relations among the words around predicates that exist in the same sentence. It then generates the semantic predicate-argument structure (PAS). For the semantic entity recognition, semantic frames, such as FrameNet, are required that define the semantic roles of the arguments for each predicate. Because it is expensive to manually build all of the different frames one by one, it is required to extend and validate the semantic frames automatically based on a large collection of text. + +NOTE – FrameNet is a free downloadable lexical database of English that is both human- and machine-readable, based on annotating examples of how words are used in actual texts. More details can be found at [b-FrameNet]. + - Co-reference resolution: + +The function of the co-reference resolution sub-functional block is to detect the preceding referents of the pronouns which replace the noun phrases of the input sentences. +- Learning and extraction function of linguistic knowledge + - Semantic relation extraction: + +The function of the semantic relation extraction sub-functional block is to find the IsA relation among two words based on the deep natural language understanding analysis and to extract the word pair of the IsA relation. The extracted pairs of IsA relations are then stored in the pre-defined data structures. + +- Time/location information extraction: + +The function of the time/location information extraction sub-functional block is to extract the information related with time/location from the sentences and documents that are necessary to generate answers in the QA system. Time information is the relative, absolute, or abstract time expressions that are present in the sentences. The purpose of the function is to extract the relevant time expressions from the natural sentences and find the relations among them. Location information is the absolute geological location or relative geological location of the words that are present in the natural sentences. The function of the location extraction is to extract the location information and understand the relation of the location expressions in the sentences. + +- Continuous linguistic knowledge acquisition and extension: + +The function of the continuous linguistic knowledge acquisition and extension sub-functional block is to provide the functionality for the IQAS to be able to answer to the user's questions reflecting continuously changing real-world environments. This functionality is composed of two functions: the updating function of natural language understanding (NLU) analysis results and the framework function to learn and extend language knowledge based on the continuous updates. The updating function of NLU results includes a metadata saving sub-module which saves intermediate results to produce language analysis results. Another sub-module of the updating function of NLU, the updating sub-module analyses the saved metadata again based on the new language knowledge and then updates the existing language analysis results based on the re-analysed results of the metadata. + +The framework function to learn and extend language knowledge based on the continuous updates includes the sub-modules such as the plug-in function for the continuous learning, the workflow function, the document collecting function, the continuous engine function, the monitoring function for the system learning process and the workbench for learning verification function. + +![Figure 4 – Logical block diagram of the linguistic knowledge learning and extraction function. The diagram shows a flow starting from 'Natural language understanding result' (cylinder) pointing to a dashed box labeled 'Linguistic knowledge learning and extraction function'. Inside this box, the flow goes to 'Continuous linguistic knowledge extraction framework' (rectangle), which is influenced by 'Time/location information extraction' and 'Semantic relation extraction' (rectangles). The output of the framework is 'Continuous NLU result update' (rectangle).](9ae17964ddd9b814c7d905b1af2fddf2_img.jpg) + +``` +graph LR; A[(Natural language understanding result)] --> B[Linguistic knowledge learning and extraction function]; subgraph B; B1[Time/location information extraction]; B2[Semantic relation extraction]; B3[Continuous linguistic knowledge extraction framework]; B4[Continuous NLU result update]; B1 --> B3; B2 --> B3; B3 --> B4; end; +``` + +F.746.3(15)\_F04 + +Figure 4 – Logical block diagram of the linguistic knowledge learning and extraction function. The diagram shows a flow starting from 'Natural language understanding result' (cylinder) pointing to a dashed box labeled 'Linguistic knowledge learning and extraction function'. Inside this box, the flow goes to 'Continuous linguistic knowledge extraction framework' (rectangle), which is influenced by 'Time/location information extraction' and 'Semantic relation extraction' (rectangles). The output of the framework is 'Continuous NLU result update' (rectangle). + +Figure 4 – Logical block diagram of the linguistic knowledge learning and extraction function + +## 8.2 Question analysis + +The question analysis functional block analyses questions which are input in a natural language by a user; it understands the user's intention and recognizes various information on the answers that should be presented as the output of the IQAS system. The question analysis functional block consists of the following functional sub-blocks: + +- **Question classification:** defines question types based on the question answering strategy and classifies questions. +- **Question decomposition:** decomposes the question based on structural and semantic information and recognizes question types of sub-questions and the relations among them. +- **Question hypothesis generation:** generates multiple paraphrased questions with the same meaning of the user's question. The generated paraphrased questions are input to the QA engine and candidate answers are extracted. After integrated ranking of the candidate answers, the final answer is presented to the user. The ranking of the paraphrased question hypotheses are decided by evaluating meaning preservation, grammar preservation and the general expressiveness of the questions. +- **Question focus and answer type recognition:** recognizes the words or phrases that can replace the candidate answer as a question focus and answer type. The answer types are of lexical answer type and semantic answer type. +- **Keyword extraction:** recognizes the salient keywords which match with the title of the encyclopaedia. If any ambiguity occurs in matching them, it is resolved by comparing similarities. +- **Generation of question intermediate format:** generates intermediate question formats which are intermediate forms between the user's natural language questions and formal knowledge base queries such as SPARQL and structured query language (SQL). Question intermediate formats are represented as a logical equation with answer variables and answer restrictions extracted from the natural language question. +- The user's knowledge level is inferred by analysing the user's question sentence, domain-specific vocabularies in use and the individualized language usage of the user. + +The workflow of question analysis is shown in Figure 5. + +![Flowchart of the question analysis workflow. The process starts with 'Question' at the top, leading to 'Question classification'. This leads to 'Question decomposition', which contains sub-steps: 'Question decomposition', 'Classification of sub-question type', and 'Recognition of sub-question relation'. This leads to 'Question paraphrase generation'. From 'Question paraphrase generation', an arrow points to a 'Question analysis engine' box. The 'Question analysis engine' contains sub-steps: 'Question focus detection', 'Answer type recognition', and 'Keyword extraction and linking'. From the 'Question analysis engine', an arrow points to 'Question intermediate format generation', which then points to 'Question analysis result'.](0f985b39edc1d52ba3600c438bc8f0a5_img.jpg) + +``` + +graph TD + Question[Question] --> QC[Question classification] + QC --> QD[Question decomposition] + subgraph QD [Question decomposition] + QD1[Question decomposition] + QD2[Classification of sub-question type] + QD3[Recognition of sub-question relation] + end + QD --> QPG[Question paraphrase generation] + QPG --> QAE[Question analysis engine] + subgraph QAE [Question analysis engine] + QFD[Question focus detection] + AT[Answer type recognition] + KE[Keyword extraction and linking] + end + QAE --> QIFG[Question intermediate format generation] + QIFG --> QAR[Question analysis result] + +``` + +F.746.3(15)\_F05 + +Flowchart of the question analysis workflow. The process starts with 'Question' at the top, leading to 'Question classification'. This leads to 'Question decomposition', which contains sub-steps: 'Question decomposition', 'Classification of sub-question type', and 'Recognition of sub-question relation'. This leads to 'Question paraphrase generation'. From 'Question paraphrase generation', an arrow points to a 'Question analysis engine' box. The 'Question analysis engine' contains sub-steps: 'Question focus detection', 'Answer type recognition', and 'Keyword extraction and linking'. From the 'Question analysis engine', an arrow points to 'Question intermediate format generation', which then points to 'Question analysis result'. + +**Figure 5 – Workflow of question analysis** + +In Figure 6, an example of an intermediate question format is presented with a natural language question in Korean "X-seon ul balgyeon han yi saram-un bangsanung sidae-lul yeoreotda. Nobel Mulihak Sang-ul batun yi Gwahakja-nun nuguinka? (The X-rays that this person discovered opened up the era of radiology. Who is this scientist who received the Nobel prize in physics?)" + +In Table 1, the interfaces among sub-blocks of the question analysis functional block are defined. + +| | | Word | Part of speech | Named entity | Syntactic relation | Semantic role | | +|------------------------------|-----------|-----------|---------------------------|--------------------------------------|-----------------------------|--------------------|----------------------------------| +| | Variable | X | | | | | | +| Question intermediate format | Condition | Predicate | Type | | | | Type predicate semantic frame | +| | | Argument0 | X | – | – | – | | +| | | Argument1 | Saram (man) | Saram/NNG | | – | | +| | | Predicate | Balgyeon ha (discover do) | Balgyeon/NNG + ha/XSV | | | Lexical predicate semantic frame | +| | | Argument0 | X | – | – | NP_SBJ ARG0 | | +| | | Argument1 | X seon (ray) | X/SL + seon/NNG | – | VP_MOD ARG1 | | +| | | | Word | Part of speech | Named entity | Syntactic relation | Semantic role | +| | Variable | Y | | | | | | +| | Condition | Predicate | Type | | | | Type predicate semantic frame | +| | | Argument0 | Y | – | – | – | | +| | | Argument1 | Gwahakja (scientist) | Gwahak/NNG + ja/XSNG | Gwahakja/CV_OCCUPATION | – | | +| | | Predicate | Bat (receive) | Bat/VV | | | Lexical predicate semantic frame | +| | | Argument0 | Y | – | – | VP_MOD ARG0 | | +| | | Argument1 | Nobel Mulihak Sang | Nobel/NNP+muli/NNG +hak/XSN+sang/XSN | Nobel Mulihak Sang/CV_PRIZE | NP_OBJ ARG1 | | + +F.746.3(15)\_F06 + +**Figure 6 – Example of an intermediate question format** + +**Table 1 – Interface of question analysis functional sub-blocks** + +| Functions | Input | Output | +|--------------------------------------------|----------------------------------------------------------------------------|------------------------------------------------------------------------------| +| Question classification | Questions | Question type | +| Question decomposition | Question with question type attached | Sub-question and relations among sub-questions | +| Question hypothesis generation | Question, decomposed sub-questions | Question hypothesis with reliability | +| Question focus and answer type recognition | Question, answer type dictionary | Question focus and answer type | +| Keyword extraction | Question, title lists of encyclopaedia | Title, title identifier (ID) of encyclopaedia | +| Generation of intermediate question format | NLP analysed question, information extracted, answer types, question focus | Question semantic frame list representing answer constraints in logical form | + +## 8.3 Candidate answer generation + +The candidate answer generation module performs the index and search functions on the document collections and generates candidate answers using search results and various databases. The module then determines the ranking of the candidate answers by using methods such as type coercion, time/location inferences and calculation of similarities among sentences. The candidate answer generation functions involve sub-functions such as real-time information search, real-time candidate answer generation and multi-source/text-based/KB-based candidate answer generation. + +- **Real-time information search:** The target of the real-time information search includes the general document collections, as well as the open user-created encyclopaedia such as Wikipedia. When searching for the information in the open user-created encyclopaedia, the IQAS also uses the unstructured and semi-structured information. In the open user-created encyclopaedia, the main text and definitions are unstructured information and section title document, category document and info-box document are semi-structured information. The information search module consists of an indexing stage and a searching stage. The indexing function takes the unstructured text as input and generates multiple indexed databases. The searching function searches for similar documents from the indexed databases based on the question analysis results. It provides as a result document contents, rankings and weights. +- **KB-based answer generation:** This sub-function generates KB questions which are specified by each KB based on the question intermediate format produced during the question analysis stage. It then searches each KB and generates candidate answers. The KB-specific questions are generated by the following steps: The arguments of the question intermediate format are mapped onto the classes or instances defined in the specific KB; The predicate of the question intermediate format is mapped to the property defined in the specific KB; The KB-specific questions are searched for and generated in the specific KB, and finally the candidate answers are generated. + +## 8.4 Answer inference and generation + +The function of the answer inference and generation module is to decide upon and generate the best answer from among the candidate answers by measuring reliabilities based on multiple knowledge sources of different types. The inference and generation module performs feature normalization and ranking of the candidate answers by learning the reliability thresholds according to the question types. The best answer is synthesized and finally presented to the user as the output of the question answering service. + +## Bibliography + +- [b-ITU-T H.760] Recommendation ITU-T H.760 (2009), *Overview of multimedia application frameworks for IPTV services.* +- [b-ITU-T H.780] Recommendation ITU-T H.780 (2012), *Digital signage: Service requirements and IPTV-based architecture.* +- [b-ITU-T Y.101] Recommendation ITU-T Y.101 (2000), *Global Information Infrastructure terminology: Terms and definitions.* +- [b-FrameNet] FrameNet. +<> + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | General tariff principles | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Terminals and subjective and objective assessment methods | +| Series Q | Switching and signalling | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.746.4-201703-I_PDF-E/4636adff5682a064f0ae5f13a1d464a6_img.jpg b/marked/F/T-REC-F.746.4-201703-I_PDF-E/4636adff5682a064f0ae5f13a1d464a6_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..0e4b98b797eb95ae782df065efb81c9c8d9417ac --- /dev/null +++ b/marked/F/T-REC-F.746.4-201703-I_PDF-E/4636adff5682a064f0ae5f13a1d464a6_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0097984dfc1a0d7b71d0835ad3b97b32460eee34637d3c6c36fa6075ebd96113 +size 29040 diff --git a/marked/F/T-REC-F.746.4-201703-I_PDF-E/54fabc351eda5228d2fa28cd9ba07971_img.jpg b/marked/F/T-REC-F.746.4-201703-I_PDF-E/54fabc351eda5228d2fa28cd9ba07971_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9a17d4d78e119b1babfa993a367295bc1249ffa5 --- /dev/null +++ b/marked/F/T-REC-F.746.4-201703-I_PDF-E/54fabc351eda5228d2fa28cd9ba07971_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:af52dd402c9a99251ed14dc64ff606a402a4ea6bb03eb3f9b7fa33d7d6abc26d +size 134639 diff --git a/marked/F/T-REC-F.746.4-201703-I_PDF-E/853f59c89931a666c07903b31d098277_img.jpg b/marked/F/T-REC-F.746.4-201703-I_PDF-E/853f59c89931a666c07903b31d098277_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3fc4abd1c5d64e481903a1243c7aef1940d7affb --- /dev/null +++ b/marked/F/T-REC-F.746.4-201703-I_PDF-E/853f59c89931a666c07903b31d098277_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3eb853430b4ba5850633af1f3c2db548d31d2308bf8ea9deec6b2e8699b8f8a0 +size 103648 diff --git a/marked/F/T-REC-F.746.4-201703-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg b/marked/F/T-REC-F.746.4-201703-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..529e4f72f14cdf7f41f56cfbfb26655bf8181c59 --- /dev/null +++ b/marked/F/T-REC-F.746.4-201703-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:abb9e3ec6dfc8364cedc39a68cf796e9d3ff5529c79a1811eedeb12c6d6e7583 +size 3787 diff --git a/marked/F/T-REC-F.746.4-201703-I_PDF-E/bd671b21db63e6fdb2196e9b18502aac_img.jpg b/marked/F/T-REC-F.746.4-201703-I_PDF-E/bd671b21db63e6fdb2196e9b18502aac_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..bdbeaac2f1a15092eb9d83674145668d57001af0 --- /dev/null +++ b/marked/F/T-REC-F.746.4-201703-I_PDF-E/bd671b21db63e6fdb2196e9b18502aac_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:07f98b966b7c805de750289ee1d5fa8b2000d8d941884eec320b247afc0eedd9 +size 91700 diff --git a/marked/F/T-REC-F.746.4-201703-I_PDF-E/eefe19c5e14dc4d6c316b7f7fbb7d7d7_img.jpg b/marked/F/T-REC-F.746.4-201703-I_PDF-E/eefe19c5e14dc4d6c316b7f7fbb7d7d7_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..94db38695704db205c75c8c2d0b7e6e0a0596709 --- /dev/null +++ b/marked/F/T-REC-F.746.4-201703-I_PDF-E/eefe19c5e14dc4d6c316b7f7fbb7d7d7_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0873414f78b50b4de2217bc19964ae40a5b0b787934861d7f9295464021d6b60 +size 45947 diff --git a/marked/F/T-REC-F.746.4-201703-I_PDF-E/raw.md b/marked/F/T-REC-F.746.4-201703-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..27a70f98862def8d56566c6d78f63ee78037deb0 --- /dev/null +++ b/marked/F/T-REC-F.746.4-201703-I_PDF-E/raw.md @@ -0,0 +1,417 @@ + + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.746.4** + +(03/2017) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Requirements for deployment of information- +centric networks** + +Recommendation ITU-T F.746.4 + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.746.4 + +## Requirements for deployment of information-centric networks + +## Summary + +Recommendation ITU-T F.746.4 describes the scenarios and requirements for deployment of information-centric networks (ICNs). The deployment supports flexible methods for deploying various ICN instances on a single network service provider (NSP) network, deploying an ICN instance over multiple NSP networks, and interoperating between different ICN instances. The framework requires decoupling of the data plane and the control plane; in addition, one or more switches and controllers supporting such deployment are required. The software and hardware resources in the switches and controllers are required to be virtualized and to be used to create ICN instances. There exist appropriate inter-ICN service provider interfaces to facilitate interoperations of two ICN instances. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.746.4 | 2017-03-01 | 16 | 11.1002/1000/13179 | + +## Keywords + +Deployment, information-centric networks, interoperation, software-defined networking. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2017 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|------------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined within this document ..... | 1 | +| 4 Abbreviations and acronyms ..... | 1 | +| 5 Convention..... | 2 | +| 6 Background..... | 2 | +| 7 Requirements for the deployment of ICNs..... | 3 | +| 7.1 General requirements..... | 3 | +| 7.2 Requirements for the switch supporting the deployment..... | 3 | +| 7.3 Requirements for the controller supporting the deployment..... | 3 | +| 8 A framework for deployment of an ICN ..... | 4 | +| 8.1 Switch supporting the deployment ..... | 4 | +| 8.2 Controller supporting the deployment..... | 5 | +| 9 Security considerations ..... | 5 | +| Annex A – Scenarios of ICN deployment ..... | 6 | +| A.1 Scenario of a single ICN instance over multiple NSP networks ..... | 6 | +| A.2 Scenario of multiple ICN instances on a single NSP network ..... | 7 | +| A.3 Scenario of interoperation between ICN instances ..... | 8 | +| Bibliography..... | 9 | + + + +# Recommendation ITU-T F.746.4 + +## Requirements for deployment of information-centric networks + +# 1 Scope + +This Recommendation describes the scenarios and requirements for deployment of information-centric networks (ICNs). The deployment supports flexible methods for deploying various ICN instances on a single network service provider (NSP) network, deploying an ICN instance over multiple NSP networks, and interoperating between different ICN instances. + +This Recommendation describes scenarios and common requirements for the deployment of ICNs. It also presents the framework for the deployment. This Recommendation covers the following: + +- scenarios of the deployment of an ICN; +- requirements of the deployment of an ICN and its key components; +- framework for the deployment of an ICN. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T Y.3001] Recommendation ITU-T Y.3001 (2011), *Future networks: Objectives and design goals*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following term defined elsewhere: + +**3.1.1 future network** [ITU-T Y.3001]: A network able to provide services, capabilities, and facilities difficult to provide using existing network technologies. A future network is either: + +- a) A new component network or an enhanced version of an existing one, or +- b) A heterogeneous collection of new component networks or of new and existing component networks that is operated as a single network. + +## 3.2 Terms defined within this document + +None. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-------|-----------------------------------| +| API | Application Programming Interface | +| CONET | Content Networking | +| FN | Future Network | + +| | | +|---------|---------------------------------------| +| ICN | Information-Centric Network | +| NDN | Named Data Networking | +| NetInf | Network of information | +| NSP | Network Service Provider | +| PURSUIT | Publish–Subscribe Internet Technology | +| SDN | Software-Defined Networking | + +# 5 Convention + +In this Recommendation: + +- The expression "**is required to**" indicates a requirement which must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The expression "**is recommended**" indicates a requirement which is recommended but not absolutely required. Thus this requirement need not be present to claim conformance. + +# 6 Background + +In recent years, future networks (FNs) as described in [ITU-T Y.3001] have attracted considerable attention from both academia and industry. One FN objective is data awareness, which enables retrieval of data regardless of their location. To achieve this objective, the concept of information-centric networks (ICNs) has been proposed together with various ICN architectures [b-ITU-T Y.3033], [b-Dannewitz], [b-Jacobson], [b-Särelä], [b-Seskar], [b-Detti]. The next major challenge is to convince network service providers (NSPs) around the globe to adopt the concept of ICN and to deploy ICN-enabled networks. + +To achieve this goal, many prototypes and testbeds have been developed. In general, these proof-of-concept systems can be implemented by two approaches: overlay and non-overlay. With the first approach, some special overlay nodes are deployed to deal with ICN protocol-related operations on top of an existing physical network. For example, overlay-based systems have been developed for named data networking (NDN) [b-Jacobson] and network of information (NetInf) [b-Dannewitz]. + +In contrast, in the non-overlay approach, ICN functions are implemented in routers directly. For instance, the prototype of Publish–Subscribe Internet Technology (PURSUIT) [b-Särelä] has been developed based on the Click modular router [b-Kohler], which allows network designers to create their own routers based on general-purpose personal computer hardware. Besides that, some ICN projects, such as MobilityFirst [b-Seskar] and content networking (CONET) [b-Detti], developed their prototypes based on software-defined networking (SDN), which provides programmability to networks. + +Despite the importance of these prototypes, the deployment of ICN is still an open question. One obvious reason is that there are too many ICN candidates, which have different design principles and various technical details in terms of naming, routing, and so on. Taking content access as an example, while most ICN architectures are based on the assumption that a user will initiate a request and expect to access the content object as soon as possible, the PURSUIT project is based on the publish–subscribe model, in which both publisher and subscriber register on the ICN, and the publisher can push content whenever it is generated. Clearly, these two approaches can be applied by different types of application to optimize their performance. + +As each ICN architecture has its own pros and cons and is best suited to some particular application requirements, various ICN architectures will be adopted. On the other hand, different ICN service + +providers can also implement their own ICN instances to provide specific services for users. For instance, one ICN service provider provides streaming service and another ICN service provider provides instant message service. Thus, various ICN instances will coexist on top of the same physical network. + +Nevertheless, existing ICN prototypes only focus on the implementation of individual ICN architectures. Neither the coexistence of multiple ICN instances in the same physical network nor the requirement for interoperation of different ICN instances is considered. + +This Recommendation focuses on the issue of the deployment of ICNs, including the scenarios, requirements and framework. + +# **7 Requirements for the deployment of ICNs** + +## **7.1 General requirements** + +GEN-01: The deployment framework is required to decouple the data plane and the control plane. One or multiple switches and controllers supporting such deployment are required. + +GEN-02: The deployment framework is required to virtualize the resources of network infrastructure to facilitate ICN implementation and coexistence. That is, the software and hardware resources in switches and controllers are required to be virtualized and to be provided to ICN instances. + +GEN-03: The deployment framework is required to define appropriate inter-ICN service provider interfaces to facilitate interoperations between two ICN instances. + +GEN-04: The deployment framework is required to provide basic general functions, including name-based routing, in-network cache management and content-based security. + +## **7.2 Requirements for the switch supporting the deployment** + +DS-01: A switch is required to support general ICN-related operations and to process received packets according to the defined rules configured by control programs (i.e., ICN instances). The general ICN-related operations include name-based routing, in-network caching and content-based security. + +DS-02: A switch is required to provide the following primitive functions: forwarding packets and caching traversed content. + +DS-03: A switch is required to support virtualization for the implementation and coexistence of multiple ICN instances in the data plane. That is, the forwarding table and storage of the switch require allocation to different ICN instances to guarantee the required service quality. Packets belonging to different ICN instances require delivery to the appropriate control program and forwarding to appropriate interfaces. + +DS-04: A switch is required to provide interfaces for ICN instances to manipulate the allocated resources, including the forwarding table and storage. + +DS-05: A switch is recommended to provide the general ICN-related functions as modules. + +DS-06: A switch is recommended to support interoperation, including re-encapsulating packets and sending requests according to the rules configured by the control programs. + +## **7.3 Requirements for the controller supporting the deployment** + +DC-01: A controller is required to provide interfaces for an ICN service provider to deploy their own ICN instances in the control plane with full control of the physical resources. + +DC-02: A controller is required to support virtualization for the coexistence of multiple ICN instances. + +DC-03: A controller is required to support the registration of an ICN service provider. + +DC-04: A controller is required to support the registration of switches which support the deployment. + +DC-05: A controller is required to provide interfaces between different ICN instances to mutually exchange control information. + +DC-06: A controller is recommended to provide general ICN-related functions as modules, including routing manager, content manager, storage manager, security manager and interoperation manager. + +# 8 A framework for deployment of an ICN + +This clause addresses the capabilities of the switch and controller. + +Figure 8-1 shows the details of the switch and controller framework supporting the deployment. + +![Figure 8-1: A switch and controller framework supporting the deployment. The diagram shows a Controller and a Switch, both supporting the deployment. The Controller contains multiple ICN instances (1 to N) with internal managers (Inter-operation, Security, Storage, Content, Routing) and an ICN instances manager. The Switch is divided into Software and Hardware layers. The Software layer contains an ICN instance management API and multiple ICN instances (1 to N) with internal APIs (Inter-operation, Forwarding table, Storage management). The Hardware layer contains Inter-operation logic, Forwarding management logic, Storage management logic, Forwarding table, Storage, and a Switching element. A secure channel connects the Controller's ICN instances manager to the Switch's ICN instance management API.](54fabc351eda5228d2fa28cd9ba07971_img.jpg) + +The diagram illustrates the framework for deployment of an ICN. It consists of two main components: a Controller and a Switch, both supporting the deployment. + +**Controller:** The controller is represented by a server icon and contains multiple ICN instances (ICN instance 1 to ICN instance N). Each ICN instance contains a set of managers: Inter-operation manager, Security manager, Storage manager, Content manager, and Routing manager. All ICN instances are connected to a central ICN instances manager. The ICN instances manager is connected to the Switch via a Secure channel. + +**Switch:** The switch is represented by a switch icon and is divided into Software and Hardware layers. + +**Software Layer:** The software layer contains an ICN instance management API. Below this API, there are multiple ICN instances (ICN instance 1 to ICN instance N). Each ICN instance contains a set of APIs: Inter-operation API, Forwarding table API, and Storage management API. The ICN instance management API is connected to the ICN instances manager in the Controller. + +**Hardware Layer:** The hardware layer contains the following elements: Inter-operation logic, Forwarding management logic, Storage management logic, Forwarding table, Storage, and a Switching element. The Switching element is connected to the Forwarding table and Storage. The Forwarding management logic and Storage management logic are connected to the Forwarding table and Storage respectively. The Inter-operation logic is connected to the Forwarding management logic and Storage management logic. + +Figure 8-1: A switch and controller framework supporting the deployment. The diagram shows a Controller and a Switch, both supporting the deployment. The Controller contains multiple ICN instances (1 to N) with internal managers (Inter-operation, Security, Storage, Content, Routing) and an ICN instances manager. The Switch is divided into Software and Hardware layers. The Software layer contains an ICN instance management API and multiple ICN instances (1 to N) with internal APIs (Inter-operation, Forwarding table, Storage management). The Hardware layer contains Inter-operation logic, Forwarding management logic, Storage management logic, Forwarding table, Storage, and a Switching element. A secure channel connects the Controller's ICN instances manager to the Switch's ICN instance management API. + +Figure 8-1 – A switch and controller framework supporting the deployment + +## 8.1 Switch supporting the deployment + +The switch, consisting of the hardware and software systems, is responsible for forwarding packets belonging to different ICN instances. To support various ICN-related operations, including local caching decision and name recognition, there are six hardware elements in the hardware system: switching element, forwarding table, storage, forwarding management logic, storage management + +logic and interoperation logic. These hardware resources are shared by different ICN instances. Such hardware resources are virtualized into different slices, each of which is dedicated to a particular ICN instance. + +The switch provides interfaces, including forwarding table application programming interfaces (APIs), storage management APIs and interoperation APIs, to ICN instances to manage the resources allocated to them. The forwarding table API is used to instruct the forwarding management logic to manipulate one or more forwarding tables for a particular ICN instance. The storage management API is responsible for managing local caching policy in storage management logic or caching content in the content storage under the control of the controller. The interoperation API is designed to guide the interoperation logic to encapsulate packets or perform protocol translation received from an incompatible user. + +## **8.2 Controller supporting the deployment** + +The controller is responsible for managing virtualized physical network resources for each ICN instance. The controller provides five functions for ICN service providers, as shown in Figure 8-1. The content manager is responsible for maintaining information about content sources and destinations. The routing manager is used to maintain the topology of this ICN instance and perform ICN-related routing algorithms. The storage manager is used to set the policies for managing physical content storage. The security manager is responsible for providing secure service. The interoperation manager is used to guide the interoperation with other ICN instances. Each ICN service provider can deploy its own ICN instance according to the specification of a particular ICN architecture by selecting appropriate modules to instantiate. + +# **9 Security considerations** + +This Recommendation has some security considerations similar to those pertinent to data aware networking and aligns with the security requirements specified in [b-ITU-T Y.3033]. The channels between controller and switch and that among controllers are required to be secured and also have good privacy protection. + +# Annex A + +## Scenarios of ICN deployment + +(This annex forms an integral part of this Recommendation.) + +A general scenario for the deployment of an ICN is shown in Figure A.1. + +Each NSP network supporting the deployment consists of two types of networking elements, i.e., switch and controller. For example, in Figure A.1 there are two NSP networks (i.e., NSP1 and NSP2). + +Multiple ICN instances can coexist and each of them is owned by a particular operator, which is defined as an ICN service provider. Different ICN instances may adopt different ICN architectures. + +Each ICN instance has its own group of users. A user of one ICN service provider may move to a geographical area that their ICN service provider does not cover. For example, a user of ICN service provider 1 may move to a place covered only by ICN service provider 4. + +![Figure A.1: An overview of ICN deployment. The diagram shows three layers: ICN users at the top, ICN instances in the middle, and NSP networks at the bottom. ICN users are connected to ICN instances via solid vertical arrows. ICN instances are connected to NSP networks via dashed vertical arrows. NSP networks are connected to each other via solid horizontal arrows. A legend at the bottom defines the symbols: a switch icon for 'Switch supporting the deployment', a controller icon for 'Controller supporting the deployment', a solid line for 'Data channel', a dashed line for 'Control channel', and a dash-dot line for 'Virtual channel between ICN service providers'.](853f59c89931a666c07903b31d098277_img.jpg) + +The diagram illustrates the three-tier architecture of ICN deployment. The top tier, labeled 'ICN users', contains four groups of devices (laptop, server, smartphone) in purple, yellow, orange, and blue ovals. The middle tier, labeled 'ICN instances', contains four corresponding service provider ovals. The bottom tier, labeled 'NSP networks', shows two network clusters, NSP1 and NSP2, each with multiple switches (cylinder icons with red 'X') and two central controllers (server rack icons). Solid lines represent data channels between users and instances, and between switches. Dashed lines represent control channels between controllers. Dash-dot lines represent virtual channels between different ICN service providers across the NSP networks. + +Legend: + +- Switch supporting the deployment (cylinder icon with red 'X') +- Controller supporting the deployment (server rack icon) +- Data channel (solid line) +- Control channel (dashed line) +- Virtual channel between ICN service providers (dash-dot line) + +Figure A.1: An overview of ICN deployment. The diagram shows three layers: ICN users at the top, ICN instances in the middle, and NSP networks at the bottom. ICN users are connected to ICN instances via solid vertical arrows. ICN instances are connected to NSP networks via dashed vertical arrows. NSP networks are connected to each other via solid horizontal arrows. A legend at the bottom defines the symbols: a switch icon for 'Switch supporting the deployment', a controller icon for 'Controller supporting the deployment', a solid line for 'Data channel', a dashed line for 'Control channel', and a dash-dot line for 'Virtual channel between ICN service providers'. + +F.746.4(17)\_FA.1 + +Figure A.1 – An overview of ICN deployment + +### A.1 Scenario of a single ICN instance over multiple NSP networks + +As ICN instances are implemented on top of NSP networks to provide content delivery service, one common implementation scenario for the deployment is when one ICN instance is implemented over multiple NSP networks. To guarantee service quality, one ICN service provider can rent resources from multiple NSP networks to deploy its own ICN and to optimize content delivery efficiency. + +For instance, ICN service provider 3 in Figure A.2 is built on the networks of both NSP1 and NSP2. A user associated with ICN service provider 3 needs not be aware of the underlying NSP networks. When the user moves from one location on the NSP1 network to another in NSP2, it gets uninterrupted service from ICN Service Provider 3. + +To facilitate an ICN instance across multiple NSP networks, both the data channels and control channels to allow the interconnection are provided. + +The data channel is used for data transmission. The control channel is used to exchange control information, e.g., routing and interconnection information. With such information, ICN service provider 3 sets up communications across multiple NSP networks and provides unified service to its users. + +![Diagram of a single ICN instance across multiple NSP networks.](eefe19c5e14dc4d6c316b7f7fbb7d7d7_img.jpg) + +The diagram illustrates a single ICN instance spanning two network service provider (NSP) domains, NSP1 and NSP2, and an ICN service provider (ICNSP3). At the top, two laptops are connected to a central oval labeled 'ICNSP3'. Below this, two separate network clouds represent NSP1 and NSP2. Each cloud contains a 'Controller' (server icon), a 'Switch' (cylinder icon), and two routers (cylinder icons with red 'X' marks). A dashed line labeled 'Control channel' connects the controllers of NSP1 and NSP2. A solid line labeled 'Data channel' connects the routers of NSP1 and NSP2. Dashed lines also connect the laptops to the controllers of NSP1 and NSP2. The label 'F.746.4(17)\_FA.2' is in the bottom right corner. + +Diagram of a single ICN instance across multiple NSP networks. + +**Figure A.2 – A single ICN instance deployed over multiple NSP networks** + +### A.2 Scenario of multiple ICN instances on a single NSP network + +Since different ICN architectures have their own pros and cons, multiple ICN instances employing different ICN architectures can coexist on the same NSP network. + +One NSP network supports multiple ICN service providers. Each of them creates its own ICN instance to provide optimized services to its own users. For example, as shown in Figure A.3, ICN service provider 1 may choose NDN to provide services with the aim of optimizing video streaming, while ICN service provider 2 may adopt PURSUIT to provide services to efficiently facilitate online social networking. + +To allow the creation of multiple ICN instances on a single NSP network, a set of interfaces is provided such that ICN service providers can perform ICN architecture-related operations (name-based routing, in-network caching, etc.). The traffic belonging to different ICN instances is processed independently. + +![Diagram of multiple ICN instances on a single NSP network.](4636adff5682a064f0ae5f13a1d464a6_img.jpg) + +The diagram shows two ICN service providers, 'ICN service provider 1' and 'ICN service provider 2', connected to a single NSP network cloud labeled 'NSP1'. Inside the NSP1 cloud, there is a 'Controller' (server icon), a 'Switch' (cylinder icon), and two routers (cylinder icons with red 'X' marks). Dashed lines connect 'ICN service provider 1' to the controller and routers. Another set of dashed lines connects 'ICN service provider 2' to the same controller and routers. The label 'F.746.4(17)\_FA.3' is in the bottom right corner. + +Diagram of multiple ICN instances on a single NSP network. + +**Figure A.3 – Multiple ICN instances deployed on a single NSP network** + +### A.3 Scenario of interoperation between ICN instances + +A user of one ICN service provider may need to access services from another ICN service provider, where the two ICN service providers adopt different ICN architectures as shown in Figure A.4. + +One example is that a user in ICN service provider 1 needs a content object which is owned by a server in ICN service provider 2 as shown in Figure A.4 a). Another example is that a user of ICN service provider 1 moves to a geographical area that is not covered by the underlying NSP1 of ICN service provider 1 as shown in Figure A.4 b). + +There are two ways to support interoperation. The first is tunnelling, with which a user of one ICN service provider can connect to its ICN instance via tunnels established in another ICN instance. The second is to perform protocol translation in the border switch. + +![Figure A.4 – Interoperation of two ICN instances. The diagram is divided into two parts: (a) Within the same NSP and (b) Across different NSPs. Part (a) shows a User and a Server connected to ICN service provider 1 and ICN service provider 2 respectively, which are both connected to a single NSP1 containing a Controller and Switches. Part (b) shows a User connected to ICN service provider 1 and another User connected to ICN service provider 4. ICN service provider 1 is connected to NSP1, and ICN service provider 4 is connected to NSP2. NSP1 and NSP2 are connected via a Control channel and a Data channel.](bd671b21db63e6fdb2196e9b18502aac_img.jpg) + +The diagram illustrates two scenarios for ICN interoperation. +Part (a), titled 'a) Within the same NSP', shows a 'User' and a 'Server' connected to 'ICN service provider 1' and 'ICN service provider 2' respectively. Both providers are connected to a single network domain labeled 'NSP1'. Inside NSP1, there is a 'Controller' and two 'Switch' nodes. A dashed line connects the User to the Server, passing through both service providers and into the NSP1 infrastructure. +Part (b), titled 'b) Across different NSPs', shows a 'User' connected to 'ICN service provider 1' and another 'User' connected to 'ICN service provider 4'. 'ICN service provider 1' is connected to 'NSP1', which contains a 'Controller' and 'Switch' nodes. 'ICN service provider 4' is connected to 'NSP2', which also contains a 'Controller' and 'Switch' nodes. A 'Control channel' (dashed line) and a 'Data channel' (solid line) connect the controllers of NSP1 and NSP2. A curved arrow indicates a connection between the two Users, spanning across the different service providers and network domains. + +Figure A.4 – Interoperation of two ICN instances. The diagram is divided into two parts: (a) Within the same NSP and (b) Across different NSPs. Part (a) shows a User and a Server connected to ICN service provider 1 and ICN service provider 2 respectively, which are both connected to a single NSP1 containing a Controller and Switches. Part (b) shows a User connected to ICN service provider 1 and another User connected to ICN service provider 4. ICN service provider 1 is connected to NSP1, and ICN service provider 4 is connected to NSP2. NSP1 and NSP2 are connected via a Control channel and a Data channel. + +**Figure A.4 – Interoperation of two ICN instances** + +F.746.4(17)\_FA.4 + +# Bibliography + +- [b-ITU-T Y.3033] Recommendation ITU-T Y.3033 (2014), *Framework of data aware networking for future networks*. +- [b-Dannewitz] Dannewitz, C. (2009). [NetInf: An information-centric design for the future Internet](#), In: *Proceedings of the 3rd GI/ITG KuVS Workshop on The Future Internet*. +- [b-Detti] Detti A, Blefari-Melazzi N, Salsano S, Pomposini M (2011). [CONET: A content centric inter-networking architecture](#). In: *ICN 11 Proceedings of the ACM SIGCOMM Workshop on Information-centric Networking*, Toronto, ON, 2011-08-19, pp. 50–55. New York, NY: ACM. +- [b-Jacobson] Jacobson V, Smetters DK, Thornton JD, Plass MF, Briggs NH, Braynard RL (2009). Networking named content. In: *CoNEXT '09 Proceedings of the 5th International Conference on Emerging Networking Experiments and Technologies*, Rome, 2009-12-01/04, pp. 1–12. New York, NY: ACM. +- [b-Kohler] Kohler E, Morris R, Chen B, Jannotti J, Kaashoek MF (2000). [The Click modular router](#). *ACM Transactions on Computer Systems*, **18**(3), pp. 263–297. +- [b-Särelä] Särelä M, Rinta-aho T, Tarkoma S (2008). RTFM: [Publish/subscribe internetworking architecture](#), In: Cunningham P, Cunningham M, eds. *ICT-Mobile Summit 2008 Conference Proceedings*, 8 pp. Dublin: IIMC International Information Management Corporation Ltd. +- [b-Seskar] Seskar I, Nagaraja K, Nelson S, Raychaudhuri D (2011). MobilityFirst future internet architecture project. In: *AINTEC '11 Proceedings of the 7th Asian Internet Engineering Conference*, Bangkok, Thailand, 2011-09-09/11, pp. 1–3. New York, NY: ACM. + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.746.5-201712-I_PDF-E/1a827b10290f33d4fec04d0e8ef7a897_img.jpg b/marked/F/T-REC-F.746.5-201712-I_PDF-E/1a827b10290f33d4fec04d0e8ef7a897_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..47f99794860b996e4ad3c9471c44b0331bab33ee --- /dev/null +++ b/marked/F/T-REC-F.746.5-201712-I_PDF-E/1a827b10290f33d4fec04d0e8ef7a897_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:21002a06949b2bd8300a06e3b965be928571f129a4248e006a46f2e4effe32f7 +size 105024 diff --git a/marked/F/T-REC-F.746.5-201712-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.746.5-201712-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..22dc7ff0fb30f976cc1e576c45d2b7e65fa5ddc7 --- /dev/null +++ b/marked/F/T-REC-F.746.5-201712-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aa14933e776f7d0be8b2d60b81cbb6c1ee49deb5080e0281da05296f48189055 +size 6859 diff --git a/marked/F/T-REC-F.746.5-201712-I_PDF-E/4e4be0bd8b235167902f2c03e41da651_img.jpg b/marked/F/T-REC-F.746.5-201712-I_PDF-E/4e4be0bd8b235167902f2c03e41da651_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9c0dd555f11990f181ecfa2e115d895b62b5d0b9 --- /dev/null +++ b/marked/F/T-REC-F.746.5-201712-I_PDF-E/4e4be0bd8b235167902f2c03e41da651_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3be5ef7162fc3b14e968b0ce4c58ae7d065ae12def08beae64c1686b9747156d +size 25215 diff --git a/marked/F/T-REC-F.746.5-201712-I_PDF-E/54fabc351eda5228d2fa28cd9ba07971_img.jpg b/marked/F/T-REC-F.746.5-201712-I_PDF-E/54fabc351eda5228d2fa28cd9ba07971_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d1020b72febe07ce9c96aba8252e543b1febd1ef --- /dev/null +++ b/marked/F/T-REC-F.746.5-201712-I_PDF-E/54fabc351eda5228d2fa28cd9ba07971_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3b17587a474ac32362fddeb247f6a42dd09b67a7581587138c7a1a85795198d3 +size 49654 diff --git a/marked/F/T-REC-F.746.5-201712-I_PDF-E/5a4e62bead259c258d069fd3663ea670_img.jpg b/marked/F/T-REC-F.746.5-201712-I_PDF-E/5a4e62bead259c258d069fd3663ea670_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d8f618e80c518fd41a6e2e632bca3df6d9410fac --- /dev/null +++ b/marked/F/T-REC-F.746.5-201712-I_PDF-E/5a4e62bead259c258d069fd3663ea670_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ca20ea960dcb8873129fd7a2b78e2d089d39286e7c5918dd2138886c28273408 +size 115041 diff --git a/marked/F/T-REC-F.746.5-201712-I_PDF-E/75f0cb39f1cd165dfe4a6aa6c4d9388d_img.jpg b/marked/F/T-REC-F.746.5-201712-I_PDF-E/75f0cb39f1cd165dfe4a6aa6c4d9388d_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..35f89a419c71f6a0a88b5b4412b583db2e50c6ca --- /dev/null +++ b/marked/F/T-REC-F.746.5-201712-I_PDF-E/75f0cb39f1cd165dfe4a6aa6c4d9388d_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bbfe7ec1f63fe4366b9f9d7f7e8f642d72cfe39da399b6ac58b114e4e67e2013 +size 72193 diff --git a/marked/F/T-REC-F.746.5-201712-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg b/marked/F/T-REC-F.746.5-201712-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..295be625831e70b1753860f6c0627a4454159087 --- /dev/null +++ b/marked/F/T-REC-F.746.5-201712-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a8db4745f0bc3d464d43641580c01bbb77ec1ee95d1b0f4cf6d6ca1427faf62b +size 17821 diff --git a/marked/F/T-REC-F.746.5-201712-I_PDF-E/e6df2733626a85205c1db682e6259c46_img.jpg b/marked/F/T-REC-F.746.5-201712-I_PDF-E/e6df2733626a85205c1db682e6259c46_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..4e2ed4b0b7bd1b52779de5111f98bb0408e86f42 --- /dev/null +++ b/marked/F/T-REC-F.746.5-201712-I_PDF-E/e6df2733626a85205c1db682e6259c46_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc37d30bd2220a982aec20751cbed2d2e586003acf9e5c1b299c5392a8a40a3d +size 34737 diff --git a/marked/F/T-REC-F.746.5-201712-I_PDF-E/raw.md b/marked/F/T-REC-F.746.5-201712-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..edf3e513a9c13693c9dc4b872ad5d8b838e799ae --- /dev/null +++ b/marked/F/T-REC-F.746.5-201712-I_PDF-E/raw.md @@ -0,0 +1,554 @@ + + +I n t e r n a t i o n a l T e l e c o m m u n i c a t i o n U n i o n + +# **ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +# **F.746.5** + +(12/2017) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Framework for a language learning system +based on speech and natural language +processing (NLP) technology** + +Recommendation ITU-T F.746.5 + +**ITU-T** + +![ITU logo: A globe with a red lightning bolt and the text 'ITU International Telecommunication Union'.](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a stylized globe with a red lightning bolt striking across it. To the right of the globe, the text "ITU" is written in a bold, blue font, and below it, "International Telecommunication Union" is written in a smaller, blue font. + +ITU logo: A globe with a red lightning bolt and the text 'ITU International Telecommunication Union'. + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.746.5 + +# Framework for a language learning system based on speech and natural language processing (NLP) technology + +## Summary + +Recommendation ITU-T F.746.5 describes functional requirements and detailed functions for the framework of a language learning system based on speech and natural language processing technology. It provides a framework for a language learning system that will serve as a reference framework for language learning systems to be developed and used as low cost tools in many educational situations. This Recommendation defines the features, general requirements and functionality to support the language learning system based on speech and natural language processing (NLP) technology. The scope covers a high-level description of architecture, terminals, servers, interface and clients. . + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.746.5 | 2017-12-14 | 16 | 11.1002/1000/13427 | + +## Keywords + +Language learning, natural language processing, speech interface, speech recognition. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2018 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|---------------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Introduction..... | 3 | +| 6.1 Basic concept of speech interface and NLP technology ..... | 3 | +| 6.2 Advanced technology for dialogue-based speech interface ..... | 4 | +| 7 Requirements for language learning system..... | 5 | +| 8 Functional components and interfaces of a language learning system..... | 5 | +| 8.1 Speech recognition module ..... | 6 | +| 8.2 Natural language processing module..... | 6 | +| 8.3 Speech synthesis module ..... | 7 | +| 8.4 Pronunciation fluency evaluation module ..... | 7 | +| 8.5 Dialogue processing module ..... | 8 | +| 8.6 Dialogue understanding evaluation module ..... | 9 | +| 8.7 Dialogue knowledge learning module..... | 9 | +| 8.8 Language learning module ..... | 10 | + + + +# Recommendation ITU-T F.746.5 + +## Framework for a language learning system based on speech and natural language processing (NLP) technology + +## 1 Scope + +This Recommendation presents an overview of the framework for a language learning system based on speech and natural language processing (NLP) technology. It describes the features, general requirements and functionality, which is a framework to support language-learning systems. The scope covers a high-level description of architecture, devices, servers and clients. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this document. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this document are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T F.746.3] Recommendation ITU-T F.746.3 (2015), *Intelligent question answering service framework*. + +[ITU-T H.703] Recommendation ITU-T H.703 (2016), *Enhanced user interface framework for IPTV terminal devices*. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 co-reference resolution** [ITU-T F.746.3]: A function that detects the preceding referents of the pronouns which replace the noun phrases of the input sentences. + +**3.1.2 named entity recognition** [ITU-T F.746.3]: A function that recognizes named entities such as PLO which are people, locations and organizations from the sentences. The PLO can be decomposed into more specific named entities depending on the applications. + +**3.1.3 natural language processing** [ITU-T F.746.3]: A method that analyses text in natural languages through several processes such as part-of- speech recognition, syntactic analysis and semantic analysis. + +**3.1.4 part-of-speech recognition** [ITU-T F.746.3]: A function that recognizes parts of speech (POS) in the sentences and assigns relevant POS tags considering contextual meaning of the target sentences. + +**3.1.5 semantic analysis** [ITU-T F.746.3]: A function that recognizes the semantic relations among the words around predicates that exist in the same sentence. The semantic analysis function then generates a semantic predicate-argument structure (PAS). + +**3.1.6 speech** [ITU-T H.703]: Speech is the vocalized form of human communication. + +**3.1.7 speech recognition** [ITU-T H.703]: A kind of user interface for translation of spoken words into text. + +**3.1.8 syntactic analysis** [ITU-T F.746.3]: A function that analyses sentence structures and generates dependency relation among words based on dependency grammars. + +### **3.2 Terms defined in this Recommendation** + +This Recommendation defines the following terms: + +**3.2.1 dialogue-based speech interface:** An interface based on speech, especially dialogues between the user and the device or system. + +**3.2.2 dialogue act:** The user's intention or purpose of the utterances in a dialogue. Example: request for information, command for action, agreement. + +## **4 Abbreviations and acronyms** + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-----|---------------------------------------------------------| +| DA | Dialogue Act | +| DNN | Deep Neural Network | +| HCI | Human Computer Interaction | +| ICT | Information and Communication Technology | +| IT | Information Technology | +| LLS | Language Learning System Based on Speech/NLP Technology | +| NE | Named Entity | +| NLP | Natural Language Processing | +| PC | Personal Computer | +| POS | Part of Speech | +| SMS | Short Message Service | +| SVM | Support Vector Machines | +| TTS | Text to Speech | + +## **5 Conventions** + +The following conventions are used in this Recommendation: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted, if conformance to this Recommendation is to be claimed. +- The keywords "is prohibited from" indicate a requirement which must be strictly prohibited, if conformance to this Recommendation is to be claimed. +- The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. +- The keywords "is not recommended" indicate a requirement which is not recommended but which is not specifically prohibited. Thus, conformance with this Recommendation can still be claimed even if this requirement is present. +- The keywords "can optionally" indicate an optional requirement which is permissible, without implying any sense of being recommended. This term is not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with this Recommendation. + +# 6 Introduction + +Language learning requires a lot of time and cost for individual learners who wish to learn foreign languages, especially when personalized education is needed for each learner with a different level of learning capabilities. The ratio of students to teachers should be minimized to have an effective learning situation for individualized learning. Advances in information and communication technology (ICT), especially those in speech/language areas have made the language learning experience less expensive and more effective for individualized learning. + +Speech interface and natural language processing (NLP) technology is an advanced technology that allows a computer to understand what a person says, and facilitates exchange of information in a smooth conversation with the person. Speech interface/NLP technology will likely be combined with other technologies in many application areas such as national defence, medical services, etc. Moreover, speech interface/NLP technology can also be applied to language learning systems for conversation training. + +NLP is a core technology for human computer interaction (HCI) and a basis for knowledge and information services. It combines syntactic analysis and semantic analysis to understand human languages. NLP technology is used for understanding a user's speech in dialogue practices in various language learning scenarios. + +More advanced dialogue-based speech interface technology is also being developed. The basic dialogue processing flow is shown in Figure 3. The technology recognizes and understands users' speech and generates appropriate responses in limited dialogue situations. The core dialogue processing technology is applied to foreign language learning systems that simulate one-on-one conversation training. + +This Recommendation provides requirements, architecture and functions for a language learning system based on speech/NLP technology. + +## 6.1 Basic concept of speech interface and NLP technology + +Speech interface and NLP technology are the next generation interface that allows a computer to understand what a person says and also facilitates the exchange of information in a smooth conversation with the person. Figure 1 shows a basic speech interface and Figure 2 shows some basic NLP modules. Recently, speech interface has become one of the essential elements in the information technology (IT) industry such as intelligent robots, telematics, the next generation personal computer (PC), and digital-home. Speech interface/NLP technology will likely be combined with other technologies in many application areas such as national defence, medical services, etc. Henceforth, speech interface/NLP technology will grow into a multimodal interface with a united input device such as voice, pen, mouse and gestures. A high-level interactive speech interface will soon be attainable by incorporating circumstantial information and the speaker's intention. Moreover, speech interface/NLP technology plays the role of a core technology for mobile web information services, and can also be applied to language learning systems for conversation training. + +NLP is a core technology for HCI and the basis for knowledge and information services. It combines syntactic analysis and semantic analysis to understand human languages. NLP technology is used for understanding a user's speech in dialogue practices in various language learning scenarios. + +![Figure 1: Basic speech interface flowchart showing two parallel processes: Input speech to Speech recognition to Output text, and Input text to Speech synthesis to Output speech.](a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg) + +``` + +graph LR + subgraph Top + A1[Input speech] --> B1[Speech recognition] --> C1[Output text] + end + subgraph Bottom + A2[Input text] --> B2[Speech synthesis] --> C2[Output speech] + end + +``` + +Figure 1: Basic speech interface flowchart showing two parallel processes: Input speech to Speech recognition to Output text, and Input text to Speech synthesis to Output speech. + +F.746.5(17)\_F01 + +**Figure 1 – Basic speech interface** + +![Figure 2: Basic NLP flows with processing modules. It shows a flow from Input speech to Part of speech analysis to Named entity recognition to Syntactic analysis to Semantic analysis. Below the flow, the sentence 'I am a bus driver' is processed through various linguistic representations, leading to the final semantic interpretation: 'Occupation: speaker: bus driver'.](e6df2733626a85205c1db682e6259c46_img.jpg) + +I am a bus driver + +→ I/pn am/be a/ia bus/n driver/n. + +→ I/pn am/be a/ia bus/n [transportation] driver/n [occupation]. + +→ (I) am a (bus driver). + +→ Occupation: speaker: bus driver + +Figure 2: Basic NLP flows with processing modules. It shows a flow from Input speech to Part of speech analysis to Named entity recognition to Syntactic analysis to Semantic analysis. Below the flow, the sentence 'I am a bus driver' is processed through various linguistic representations, leading to the final semantic interpretation: 'Occupation: speaker: bus driver'. + +F.746.5(17)\_F02 + +**Figure 2 – Basic NLP flows with processing modules** + +## 6.2 Advanced technology for dialogue-based speech interface + +More advanced technology for dialogue-based speech interface has also been developed. Figure 3 shows the basic dialogue processing flow. The technology recognizes and understands users' speech and generates appropriate responses in limited dialogue situations. The core technology is applicable to foreign language learning systems that simulate one-on-one conversation training. The core technology will also be used for natural speech interface of information services on mobile devices (smartphones, navigation terminals, etc.) or wearable devices such as smart glasses and smart watches. Using this technology, smartphones will accelerate various information services in mobile environments such as voice search, voice short message service (SMS), speech enabled virtual agents, etc. For example, the conversational speech processing technology can be applied to a high quality language learning system, and the resulting system can assess the pronunciation of students and leads the conversation to help upgrade speaking skills. + +![Figure 3: Dialogue processing flowchart. It shows 'User speech' entering a 'Dialogue understanding' block. This block is connected to a 'Language analysis DB', 'Semantics DB', 'Dialogue pattern', and 'DA learning information' DBs. Below these, a 'Dialogue understanding training module' is connected to a 'Dialogue intention training corpus' and feeds into the 'Dialogue understanding' block.](54fabc351eda5228d2fa28cd9ba07971_img.jpg) + +``` + +graph TD + subgraph Dialogue_processing [Dialogue processing] + direction TB + subgraph DBs [DBs] + LA[Language analysis DB] + S[Semantics DB] + DP[Dialogue pattern] + DAL[DA learning information] + end + DU[Dialogue understanding] + DUTM[Dialogue understanding training module] + DIC[Dialogue intention training corpus] + US[User speech] + DU <--> LA + DU <--> S + DU <--> DP + DU <--> DAL + DIC --> DUTM + DUTM --> DU + US --> DU + end + +``` + +Figure 3: Dialogue processing flowchart. It shows 'User speech' entering a 'Dialogue understanding' block. This block is connected to a 'Language analysis DB', 'Semantics DB', 'Dialogue pattern', and 'DA learning information' DBs. Below these, a 'Dialogue understanding training module' is connected to a 'Dialogue intention training corpus' and feeds into the 'Dialogue understanding' block. + +F.746.5(17)\_F03 + +**Figure 3 – Dialogue processing flow** + +## 7 Requirements for language learning system + +The general requirements for a language learning system (LLS) based on speech/NLP technology are: + +- LLS is required to provide speech recognition function for spontaneous speech. +- LLS is recommended to provide dialogue processing function for spontaneous utterances. +- LLS is recommended to provide evaluation function for spontaneous speech. +- LLS is recommended to provide error correction function for target language grammars. +- LLS is recommended to provide pronunciation modelling function for target language. +- LLS is recommended to provide expression learning function for target language. +- LLS is recommended to provide pronunciation fluency feature extraction function for target language. +- LLS is recommended to provide automatic evaluation function for pronunciation fluency for target language. + +![Functional architecture of language learning system diagram](5a4e62bead259c258d069fd3663ea670_img.jpg) + +The diagram illustrates the functional architecture of a language learning system, organized into two main horizontal sections. The top section, enclosed in a dashed box, contains the 'Speech recognition' module on the left, which includes 'Sound model learning', 'Pre-processing', 'Feature extraction', 'Language model learning', and 'Input/output management'. To the right of this are 'Speech synthesis', 'Dialogue processing', 'Language learning', 'Dialogue knowledge learning', 'Pronunciation fluency evaluation', and 'Dialogue understanding evaluation'. Arrows show a flow from 'Speech recognition' to 'Speech synthesis' and 'Dialogue processing', and from 'Dialogue processing' to 'Dialogue knowledge learning' and 'Dialogue understanding evaluation'. 'Language learning' is a central node receiving input from 'Speech synthesis' and 'Dialogue knowledge learning', and sending output to 'Pronunciation fluency evaluation'. The bottom section, also in a dashed box, is labeled 'Natural language processing' and contains 'Natural language understanding' (with 'POS recognition', 'Named entity recognition', 'Syntactic analysis', 'Semantic analysis', and 'Co-reference resolution') and 'Learning and extraction of linguistic knowledge' (with 'Time/location information extraction', 'Semantic relation extraction', and 'Continuous linguistic knowledge acquisition/extraction'). Arrows show a flow from 'Natural language understanding' to 'Continuous linguistic knowledge acquisition/extraction'. A large vertical arrow points from the bottom section to the top section, specifically to the 'Language learning' node. A small label 'F.746.5(17)\_F04' is in the bottom right corner. + +Functional architecture of language learning system diagram + +**Figure 4 – Functional architecture of language learning system** + +## 8 Functional components and interfaces of a language learning system + +Figure 4 shows the functional components of a language learning system. The essential modules of the functional components are as follows: + +- Speech recognition module: supports speech recognition function for the user's input. +- Natural language processing module: supports natural language processing on the language learning system. +- Speech synthesis module: supports speech synthesis on the terminal for the system output. +- Pronunciation fluency evaluation module: supports pronunciation fluency evaluation function for the language learning system. + +- Dialogue processing module: supports dialogue management and generation functions for the language learning system. +- Dialogue understanding evaluation module: supports evaluation of the user's understanding of the dialogue. +- Dialogue knowledge learning module: supports incremental knowledge learning on the dialogue. +- Language learning core module: supports core language learning functions such as grammar learning and dialogue expression learning. + +### 8.1 Speech recognition module + +The function of the speech recognition module is to generate output text based on the user's speech signals. The speech recognition module is composed of the following sub-functions: pre-processing, sound model learning, feature extraction, input/output management and language model learning sub-functions. + +#### 8.1.1 Pre-processing function + +The pre-processing function extracts a feature vector and removes noise from the noisy speech input. In this module, distributed noises with a high energy level and active noises with low energy level are removed to improve the quality of sound. + +#### 8.1.2 Sound model learning function + +The sound model learning functional entity produces a speaker independent acoustic model from a large speech database to recognize the general user's speech based on a deep neural network (DNN). This function also includes the process of building a speech database, and adaptation training for the specific speech environment, to make the speaker independent acoustic model optimal in the application environment. The sound model learning functional entity consists of a memory and a processor that generates sound-model state sets corresponding to multiple set training speech data. It then generates a multi-set state cluster from the sound-model state sets, and sets the multi-set training speech data as an input node, and the multi-set state cluster as an output node in order to learn a DNN structured parameter. When a user's speech and characteristic information are received by a user interface, the processor recognizes the user's speech on the basis of the learned DNN structured parameter by setting a sound-model state set corresponding to the characteristic information of the user's speech as an output node input/output management function. + +#### 8.1.3 Input/output management function + +The input/output management function provides the audio input interface function which processes the speech data input from the microphone, and the output interface function which recognizes speech data and sends back the recognition result to the user. + +#### 8.1.4 Language model learning function + +Language model learning functional entity performs training of word N-gram language model that defines the search network for continuous speech recognition. For language model training, text corpus of the recognition target should be prepared. The text corpus goes through a cleaning and refinement process first, and then language model training is performed on the clean text corpus. The language model is trained on the limited word list and the utterance dictionary is produced for the target word list. The language model database that stores a language model is used to convert the extracted feature vector into a text with a decoder. + +### 8.2 Natural language processing module + +The NLP module consists of a natural language understanding function and a linguistic knowledge learning/extraction function. The natural language understanding function is illustrated in Figure 5. + +It analyses the user's speech using the technologies of semantic level analysis such as semantic entity recognition and co-reference resolution, in addition to the well-known part-of-speech recognition, named entity (NE) recognition and syntactic analysis for sentence structures. The linguistic knowledge learning/extraction function extracts semantic relations among words and time/location information in the documents, and provides the framework to continuously acquire linguistic knowledge and to extend it to be able to deal with new knowledge. Natural language understanding function consists of a part-of-speech recognition sub-function, named entity recognition sub-function, syntactic analysis sub-function, semantic entity recognition sub-function and co-reference resolution sub-function. Learning and extraction of linguistic knowledge function is composed of semantic relation extraction sub-function, time/location information extraction sub-function and continuous linguistic knowledge acquisition and extension sub-function. + +![Flowchart of the Natural language understanding function showing a sequential process: POS analysis, NE analysis, Word sense recognition, Chunking, Dependency analysis, and Semantic role labelling.](4e4be0bd8b235167902f2c03e41da651_img.jpg) + +``` + +graph LR + A[POS analysis] --> B[NE analysis] + B --> C[Word sense recognition] + C --> D[Chunking] + D --> E[Dependency analysis] + E --> F[Semantic role labelling] + +``` + +Flowchart of the Natural language understanding function showing a sequential process: POS analysis, NE analysis, Word sense recognition, Chunking, Dependency analysis, and Semantic role labelling. + +F.746.5(17)\_F05 + +**Figure 5 – Natural language understanding function** + +### 8.3 Speech synthesis module + +Also known as text-to-speech (TTS) module, the speech synthesis module generates speech signals from the text input by text normalization and morphological analysis, intonation target (i.e., pitch contour) generation and speech signal generation using language analysis and intonation prediction information for the input sentences. + +### 8.4 Pronunciation fluency evaluation module + +The pronunciation fluency evaluation module evaluates the user's spontaneous speech input using the fluency feature model. The fluency evaluation consists of the analytic evaluation and the holistic evaluation for each evaluation item. The total scores of the analytic evaluation is the addition of each evaluation item score with different weights. The holistic evaluation is based on the impression of the fluency of the user by the evaluators. The final pronunciation fluency evaluation result is decided by selecting one of two evaluation methods. + +The pronunciation fluency evaluation is performed with the speech recognition module whereby the utterance verification function verifies the user's speech based on multi-event detection information and a predefined utterance verification model. To verify the user's utterance in a speech recognition system the following processes are used: + +- a noise processor configured to process the noise of an input speech signal; +- a feature extractor configured to extract features of speech data obtained through the noise processing; +- an event detector configured to detect events of the plurality of speech features occurring in the speech data using the noise-processed data and the data of the extracted features; +- a decoder configured to perform speech recognition using a plurality of pre-set speech recognition models for the extracted feature data; and +- an utterance verifier configured to calculate confidence measurement values in units of words and sentences using the information on the plurality of events detected by the event detector and a pre-set utterance verification model, and perform utterance verification according to the calculated confidence measurement values. + +The event detector includes a detector configured to detect features such as a noise level, a breathing sound, an interjection, a repetition of a phrase, hesitation, an exclamation, sound stretching, laughing, speech rate, and an incorrect pronunciation from the noise-processed data and the extracted feature data. + +The utterance verification model includes a word-level utterance verifier and sentence-level utterance verifier. + +The word-level utterance verifier calculates word-specific confidence scores of words of sections in which the events have occurred and words for which no event has occurred using the specific anti-model and the filler model pre-set in a database. + +The sentence-level event application utterance verifier calculates the confidence scores of the sentences using the statistical method, such as support vector machines (SVM) model, which has been pre-set in a database and trained by the statistical model such as SVM with training data reflecting a feature of a natural language. For sentence-level utterance verifier, the sentence structure and meaning analyser analyses the sentence structure and the meaning of the speech recognition result sentences using a morpheme analyser such as POS tagger and named entity analyser. + +### 8.5 Dialogue processing module + +The **dialogue processing module** performs dialogue act understanding function to express the dialogue intention/act of the utterances by analysing user's utterance strings and understanding their meaning. + +The **dialogue act training function** automatically trains dialogue patterns and statistical classification training information based on the domain dialogue corpus annotated with the dialogue acts to understand the participating user's intention. + +The **dialogue management function** analyses the n-best dialogue acts and meaning expressions and decides the best dialogue act, which is used to generate relevant system dialogue and manage the dialogue situations. + +The **dialogue generation function** selects the correct system response templet for the user's intention and decides meaningful information values to fill in the slots of the selected templet. Finally, it generates the appropriate system dialogue sentences. + +**Dialogue management and generation function** provides the following sub-functions: + +- **User intention analysis function**, to understand the user's intention correctly for the current topic +- **Auxiliary topic handling function**, to provide responses if the topic is outside the main topic and guide the dialogue towards the appropriate topic flow +- **Topic management knowledge search function**, to find relevant response patterns, slots or task knowledge to achieve the objective of the current topic +- **Dialogue history management function**, to store changing tasks, slots and intentions and monitor the intention flows of the dialogue +- **Next utterance recommendation function**, to suggest the next intention and utterances of the user according to the ongoing dialogue flow +- **System reply generation function**, to provide responses according to the user's intention +- **Dialogue management log record function**, to record the dialogue flow, the history change or search conditions. This log information is used to check the errors in the system and dialogue knowledge + +### 8.6 Dialogue understanding evaluation module + +The dialogue understanding evaluation module illustrated in Figure 6 evaluates the user's understanding of the ongoing topic by comparing user's utterances to the dialogue model of the system. It uses the dialogue understanding function and dialogue management function to search the dialogue knowledge. This function is part of the language learning module as an evaluation function. The system provides feedback, such as a better example to the user while evaluating the dialogue understanding to correct his responses. + +The response from the system differs according to the user's utterance determining if the topic is relevant or not. If the user's response contains grammatical errors or wrong expressions, the system provides the correct information for language learning purpose. + +The system guides the dialogue if the user responds with a different topic by restoring the ongoing topic of the dialogue. It can also change the topic to achieve the final goal of the dialogue and for the language learning purpose. + +![Flowchart of the language learning by dialogue processing function. The process starts with 'User's utterance' and proceeds through a series of decision diamonds: 'Topic relevant?', 'dialogue flow correct?', 'grammatical error?', and 'error feedback?'. Each 'No' branch leads to a 'Dialogue processing' box containing specific responses like 'Non topic response', 'Topic restoration', or 'Error feedback'. Each 'Yes' branch leads to the next decision diamond. The final 'Yes' branch leads to a 'Topic response' box. The diagram is labeled F.746.5(17)_F06.](75f0cb39f1cd165dfe4a6aa6c4d9388d_img.jpg) + +``` +graph TD; A[User's utterance] --> B{Topic relevant?}; B -- No --> C[Dialogue processing]; B -- Yes --> D{dialogue flow correct?}; D -- No --> E[Dialogue processing]; D -- Yes --> F{grammatical error?}; F -- No --> G[Dialogue processing]; F -- Yes --> H{error feedback?}; H -- No --> G; H -- Yes --> G; subgraph C [Dialogue processing]; C1[Non topic response]; C2[Topic restoration]; C3[Error feedback]; end; subgraph E [Dialogue processing]; E1[Error feedback]; E2[Topic restoration]; end; subgraph G [Dialogue processing]; G1[Topic response]; G2[Error feedback]; G3[Topic response]; end; +``` + +Flowchart of the language learning by dialogue processing function. The process starts with 'User's utterance' and proceeds through a series of decision diamonds: 'Topic relevant?', 'dialogue flow correct?', 'grammatical error?', and 'error feedback?'. Each 'No' branch leads to a 'Dialogue processing' box containing specific responses like 'Non topic response', 'Topic restoration', or 'Error feedback'. Each 'Yes' branch leads to the next decision diamond. The final 'Yes' branch leads to a 'Topic response' box. The diagram is labeled F.746.5(17)\_F06. + +Figure 6 – Language learning by dialogue processing function + +### 8.7 Dialogue knowledge learning module + +The dialogue knowledge-learning module adds new dialogue knowledge in the dialogue management database by pseudo situation simulation and incremental learning of situation states. The sub-functions are automatic extraction function of conversation scenarios and incremental dialogue learning sub-functions. + +Automatic extraction of conversation scenarios sub-function: serves to collect big data dialogues among people from the scenarios of movies and dramas, and performs the speaker boundary recognition, relevance assessment and topic classification. + +Incremental dialogue learning sub-function: identifies the topic of the dialogue scenarios, augments with the slot information, intention and subjects according to the topic, and produces automatically or by human validation the dialogue knowledge that can be used for the dialogue system. + +### 8.8 Language learning module + +The language learning module, illustrated in Figure 7, takes the user's speech as input and automatically detects grammatical errors, corrects them and then sends the corrected sentence to the user as a feedback to improve his/her competence of foreign languages. The language-learning module consists of the grammatical error correction function and the expression learning function. The grammatical error correction function detects errors included in the user's spoken sentences and corrects them. On the other hand, the expression learning function provides feedback of the grammatical error and suggests better expressions to the user. + +- Grammatical error detection sub-function: detects grammatical errors in the user's spoken sentences. The grammatical errors are limited to those defined in the system. Multiple errors can be detected in one sentence. Various error detection models are used for the grammatical error detection. +- Grammatical error correction sub-function: provides the correction information for detected grammatical errors. The error correction function produces grammatical error correction information by using the POS of the error words and context information around the words. +- Error selection and correction sub-function: selects the best error candidate among those from different error detection models and generates error correction information on the selected candidate error. +- Error feedback providing sub-function: informs the user of the grammatical errors and provides better expressions. The explicit/implicit error feedbacks are used accordingly depending on the situation of the error occurrences. + +![Flowchart of the Language learning module showing three main stages: Grammatical error detection, Grammatical error correction, and Grammatical error feedback.](1a827b10290f33d4fec04d0e8ef7a897_img.jpg) + +``` +graph TD + Input --> ED1[Error detection by answer candidates] + Input --> ED2[Error detection by correction memory] + Input --> ED3[Dialogue base error detection] + Input --> ED4[Error detection by N-gram] + Input --> ED5[Error detection by machine learning] + + ED1 --- KB1[(Answer set based error correction KB)] + ED2 --- EM1[(Error correction memory)] + ED3 --- DK1[(Dialogue context error correction knowledge)] + ED4 --- ND1[(N-gram data)] + ED5 --- GEDM1[(Grammatical error detection model)] + + subgraph GED [Grammatical error detection] + ED1 + ED2 + ED3 + ED4 + ED5 + KB1 + EM1 + DK1 + ND1 + GEDM1 + end + + GEDM1 --> CCG[Correction candidate generation] + + subgraph GEC [Grammatical error correction] + CCG --> BCS([Best candidate selector]) + BCS <--> LMV[LM based verification] + end + + LMV --- LMD[(Language model DB)] + + subgraph GEF [Grammatical error feedback] + BCS --> CFG[Correction feedback generation] + CFG <--> FC([Feedback controller]) + GEDM2[(Grammatical error detection model)] + end +``` + +The flowchart illustrates the Language learning module's architecture. It begins with an 'Input' arrow pointing to five parallel error detection sub-functions: 'Error detection by answer candidates', 'Error detection by correction memory', 'Dialogue base error detection', 'Error detection by N-gram', and 'Error detection by machine learning'. Each sub-function is associated with a specific knowledge base or model: 'Answer set based error correction KB', 'Error correction memory', 'Dialogue context error correction knowledge', 'N-gram data', and 'Grammatical error detection model'. These five components are grouped under the label 'Grammatical error detection'. The output of the 'Error detection by machine learning' sub-function flows into 'Correction candidate generation'. This is followed by a 'Best candidate selector' (oval), which interacts bidirectionally with 'LM based verification'. The 'LM based verification' component is linked to a 'Language model DB'. The output of the 'Best candidate selector' flows into 'Correction feedback generation', which interacts bidirectionally with a 'Feedback controller' (oval). This final stage, including 'Correction feedback generation', 'Feedback controller', and a 'Grammatical error detection model', is grouped under the label 'Grammatical error feedback'. + +Flowchart of the Language learning module showing three main stages: Grammatical error detection, Grammatical error correction, and Grammatical error feedback. + +F.746.5(17)\_F07 + +Figure 7 – Language learning module + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.746.6-201712-I_PDF-E/08441fa90c5fd11994626f662ac13f19_img.jpg b/marked/F/T-REC-F.746.6-201712-I_PDF-E/08441fa90c5fd11994626f662ac13f19_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..6a13b1391861526dca636d69ffdb10a7e6613a36 --- /dev/null +++ b/marked/F/T-REC-F.746.6-201712-I_PDF-E/08441fa90c5fd11994626f662ac13f19_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e65c7608e5733dfe0ba3fe6c945ee3199dc431708f7c74e99943a3486128396a +size 77724 diff --git a/marked/F/T-REC-F.746.6-201712-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.746.6-201712-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9002ee68e2b9ebf63d27d39ba8c73bdafb0e6be6 --- /dev/null +++ b/marked/F/T-REC-F.746.6-201712-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8df701536c25ab8246d2868de3c20c8a4e25168bb52c5eb9078278ab1b5b4fd8 +size 6859 diff --git a/marked/F/T-REC-F.746.6-201712-I_PDF-E/7a0db9703b68b3d06cdaeefc084c0006_img.jpg b/marked/F/T-REC-F.746.6-201712-I_PDF-E/7a0db9703b68b3d06cdaeefc084c0006_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9588f96c15f541922dd595471038debf74f11cd9 --- /dev/null +++ b/marked/F/T-REC-F.746.6-201712-I_PDF-E/7a0db9703b68b3d06cdaeefc084c0006_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9a7c4eca3d3b6ffe65b443a231c7ba3dbc302dae2798a4db8db70c8abe6f46bc +size 39432 diff --git a/marked/F/T-REC-F.746.6-201712-I_PDF-E/af7916c89a458fdab6c3f443217388ae_img.jpg b/marked/F/T-REC-F.746.6-201712-I_PDF-E/af7916c89a458fdab6c3f443217388ae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..059ada5b5d9a430ac82374c159da07ef08f52c4f --- /dev/null +++ b/marked/F/T-REC-F.746.6-201712-I_PDF-E/af7916c89a458fdab6c3f443217388ae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e7f89671a71c5cfcaca9fc15634d63446f21120bc19b7963c15a5ee7eae59242 +size 79409 diff --git a/marked/F/T-REC-F.746.9-201905-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.746.9-201905-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9002ee68e2b9ebf63d27d39ba8c73bdafb0e6be6 --- /dev/null +++ b/marked/F/T-REC-F.746.9-201905-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8df701536c25ab8246d2868de3c20c8a4e25168bb52c5eb9078278ab1b5b4fd8 +size 6859 diff --git a/marked/F/T-REC-F.746.9-201905-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg b/marked/F/T-REC-F.746.9-201905-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..71999e8e57e2d6441db0854d1711d8833dd58883 --- /dev/null +++ b/marked/F/T-REC-F.746.9-201905-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c488218da142d61628204f83092de694b2e83c675aded4784d4079be8c179ef2 +size 28973 diff --git a/marked/F/T-REC-F.746.9-201905-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg b/marked/F/T-REC-F.746.9-201905-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..33bc7c77d29d3663efc123af9d4f3106853effe6 --- /dev/null +++ b/marked/F/T-REC-F.746.9-201905-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1542b65851f385acedd668ae26cff0549c11f69c1a626fff34344e73e502b033 +size 50391 diff --git a/marked/F/T-REC-F.746.9-201905-I_PDF-E/5b8a756d9a71c35f17db8bcb90b438a3_img.jpg b/marked/F/T-REC-F.746.9-201905-I_PDF-E/5b8a756d9a71c35f17db8bcb90b438a3_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c236d74c2918dd65c3c394a6753dd5a16d95797c --- /dev/null +++ b/marked/F/T-REC-F.746.9-201905-I_PDF-E/5b8a756d9a71c35f17db8bcb90b438a3_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:033f8c95e7f29cb0059d3f5b423295e0c9cb42dbdb68070449d81124d58f8eae +size 27996 diff --git a/marked/F/T-REC-F.746.9-201905-I_PDF-E/5e92d9e8e9ce204e405bff2367f88176_img.jpg b/marked/F/T-REC-F.746.9-201905-I_PDF-E/5e92d9e8e9ce204e405bff2367f88176_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9a48883c1051e32fb1161582366c1e8551076497 --- /dev/null +++ b/marked/F/T-REC-F.746.9-201905-I_PDF-E/5e92d9e8e9ce204e405bff2367f88176_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:45ea5a2a8071f12b96ed5a8c45fb3d41b10058ee24adb361d4415c0b834c3325 +size 17773 diff --git a/marked/F/T-REC-F.746.9-201905-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg b/marked/F/T-REC-F.746.9-201905-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..11f93d9af0f598379bbf0819f2a2567a2017ef30 --- /dev/null +++ b/marked/F/T-REC-F.746.9-201905-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e8998289c3406db513253de8671a4c24394d4575338003be4a9acec6dd4f5552 +size 111956 diff --git a/marked/F/T-REC-F.746.9-201905-I_PDF-E/ff0952ef692c9d960ce5f6708bcc9711_img.jpg b/marked/F/T-REC-F.746.9-201905-I_PDF-E/ff0952ef692c9d960ce5f6708bcc9711_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..bede9702dd7699a0f27d6b25ee5293adb953cfb5 --- /dev/null +++ b/marked/F/T-REC-F.746.9-201905-I_PDF-E/ff0952ef692c9d960ce5f6708bcc9711_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0e9a4a402b82dd01c83b62d3dfdce7fb649b1d4f383cded54aea8704a9efa2d1 +size 45891 diff --git a/marked/F/T-REC-F.746.9-201905-I_PDF-E/raw.md b/marked/F/T-REC-F.746.9-201905-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..15d2139a4effd0cece23ad34a646cf6927bdbb98 --- /dev/null +++ b/marked/F/T-REC-F.746.9-201905-I_PDF-E/raw.md @@ -0,0 +1,588 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.746.9** + +(05/2019) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Requirements and architecture for indoor +conversational robot systems** + +Recommendation ITU-T F.746.9 + +ITU-T + +![ITU logo](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) features a globe with a red lightning bolt striking across it. To the right of the globe, the text "International Telecommunication Union" is written in blue, with "ITU" in a larger, bold font above it. + +ITU logo + +International +Telecommunication +Union + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +## Recommendation ITU-T F.746.9 + +# Requirements and architecture for indoor conversational robot systems + +## Summary + +Recommendation ITU-T F.746.9 describes the requirements and architecture for an indoor conversational robot system. This Recommendation defines different functions to support indoor conversational robot systems. The scope of this Recommendation is focused on architecture, terminals, servers and interfaces between modules. It also describes different functions including a variable-distance speech preprocessing function, user identifying speech recognition function, multitask dialogue processing function and speech synthesis function. Additionally, it describes interfaces between modules for indoor conversational robot systems. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.746.9 | 2019-05-14 | 16 | 11.1002/1000/13916 | + +### Keywords + +AI robot, dialogue processing, speech interface + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2019 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|------------|------------------------------------------------------------------------------------------------------------------------------------|----| +| 1 | Scope..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 1 | +| 4 | Abbreviations and acronyms ..... | 2 | +| 5 | Conventions ..... | 2 | +| 6 | Introduction..... | 3 | +| 6.1 | Basic concept of indoor conversational robot ..... | 3 | +| 7 | Requirements of indoor conversational robot systems..... | 3 | +| 8 | Functional components and interfaces for indoor conversational robot systems ..... | 4 | +| 8.1 | Variable-distance speech preprocessing module..... | 5 | +| 8.2 | User identifying speech recognition module..... | 7 | +| 8.3 | Multitask dialogue processing module..... | 8 | +| 8.4 | Speech synthesis module..... | 9 | +| 9 | Application programme interfaces (API) for indoor conversational robot systems..... | 9 | +| 9.1 | APIs to communicate between client and server..... | 9 | +| 9.2 | APIs for preprocessing module ..... | 10 | +| 9.3 | APIs for automatic speech recognition module..... | 11 | +| 9.4 | APIs for dialogue module..... | 12 | +| Appendix I | – Evaluation method for measuring performance of an indoor conversational robot system in variable distant noisy environments..... | 14 | +| I.1 | Evaluation of speech recognition in variable distant noisy environments .... | 14 | +| I.2 | Evaluation of call success in noisy environments ..... | 15 | +| I.3 | Evaluation of conversations of specific purposes in indoor noisy environments ..... | 15 | +| | Bibliography..... | 16 | + + + +# Requirements and architecture for indoor conversational robot systems + +# 1 Scope + +This Recommendation provides an overview of the framework for indoor conversational robot systems. It describes the features, general requirements, architecture and functionality, which are the framework to support indoor conversational robot systems. The scope is focused on architecture, terminals, servers, interface and communication protocols between servers and clients in a high-level approach. This Recommendation will allow users of the indoor conversational robot system to be able to experience convenient information services in various areas with a user-friendly speech-dialogue interface. + +# 2 References + +None. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 co-reference resolution** [b-ITU-T F.746.3]: A function that detects the preceding referents of the pronouns which replace the noun phrases of the input sentences. + +**3.1.2 named entity recognition** [b-ITU-T F.746.3]: A function that recognizes named entities such as PLO which are people, organizations and places from the sentences. The PLO can be decomposed to more specific named entities depending on the applications. + +**3.1.3 natural language processing** [b-ITU-T F.746.3]: A method that analyses text in natural languages through several processes such as part-of- speech recognition, syntactic analysis and semantic analysis. + +**3.1.4 part-of-speech recognition** [b-ITU-T F.746.3]: A function that recognizes parts of speech (POS) in the sentences and assigns relevant POS tags considering contextual meaning of the target sentences. + +**3.1.5 semantic analysis** [b-ITU-T F.746.3]: A function that recognizes the semantic relations among the words around predicates that exist in the same sentence. The semantic analysis function then generates a semantic predicate-argument structure (PAS). + +**3.1.6 speech** [b-ITU-T H.703]: Speech is the vocalized form of human communication. + +**3.1.7 speech recognition** [b-ITU-T H.703]: A kind of user interface for translation of spoken words into text. + +**3.1.8 syntactic analysis** [b-ITU-T F.746.3]: A function that analyses sentence structures and generates dependency relations among words based on dependency grammars. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 dialogue-based speech interface**: An interface based on speech, especially dialogues between a user and a device or system. + +**3.2.2 dialogue act:** The user's intention or purpose of the utterances in a dialogue, for example, a request for information, command for action or agreement. + +# **4 Abbreviations and acronyms** + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|---------------------------------------------------------| +| AI | Artificial Intelligence | +| ASR | Automatic Speech Recognition | +| DA | Dialogue Act | +| DB | Database | +| DNN | Deep Neural Network | +| EER | Equal Error Rate | +| FAR | False Alarm Rate | +| FRR | False Rejection Rate | +| HCI | Human Computer Interaction | +| ICT | Information and Communication Technology | +| IPTV | Internet Protocol Television | +| IT | Information Technology | +| LLS | Language Learning System based on speech/NLP technology | +| LSTM | Long Short Term Memory | +| NE | Named Entity | +| NLP | Natural Language Processing | +| PC | Personal Computer | +| POS | Part Of Speech | +| RT | Reverberation Time | +| SMS | Short Message Service | +| SNR | Signal to Noise Ratio | +| SVM | Support Vector Machines | +| TTS | Text To Speech | +| VAD | Voice Activity Detection | +| VoD | Video on Demand | + +# **5 Conventions** + +The following conventions are used in this Recommendation: + +- The keywords "is required to" indicate a requirement that must be strictly followed and from which no deviation is permitted, if conformance to this Recommendation is to be claimed. +- The keywords "is prohibited from" indicate a requirement that must be strictly prohibited, if conformance to this Recommendation is to be claimed. +- The keywords "is recommended" indicate a requirement that is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. + +- The keywords "is not recommended" indicate a requirement that is not recommended but which is not specifically prohibited. Thus, conformance with this Recommendation can still be claimed even if this requirement is present. +- The keywords "can optionally" indicate an optional requirement that is permissible, without implying any sense of being recommended. This term is not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with this Recommendation. + +# 6 Introduction + +These days the application of artificial intelligence (AI) technology has expanded further into areas such as robotics. Indoor conversational robot systems are one of the AI applications equipped with speech interfaces that enable users to have a dialogue with robots. The key technologies in such a multimedia system are speech recognition technology for variable distance speech and multitask dialogue processing technology which analyses the resulting natural language text from the speech recognition, understands the meaning of the spoken dialogues and generates the best relevant responses. In this way, the intelligent robot can perform the service for the human user while having a conversation with him/her. + +The indoor conversational robot can be applied to various areas. It can serve people as a kiosk robot to guide customers or to introduce different services in the private and public sectors. It can also serve as a conversational helper for everyday life for individuals living alone, e.g. an elderly person. It can provide people with recipes through question and answer speech dialogues. It can be used as a personal assistant robot to manage a schedule. The last use-case example is a robot guide which provides dialogue-based information service on domestic multimedia content such as that for Internet Protocol TV (IPTV) or video on demand (VoD). + +In this document, we present the requirements, architecture and functions for indoor conversational robot systems. + +## 6.1 Basic concept of indoor conversational robot + +An indoor conversational robot is a type of intelligent agent that has the capability to have a dialogue with people in a room. The purpose of the dialogue between a robot and people is for the robot to provide various services to people. To provide such intelligent services the core technologies include variable-distance speech recognition and multitask dialogue processing technology. The robot should first be able to recognize speech spoken by a person in a noisy room who may walk around. It means the distance of the speech recognition target is not determined but variable. After the speech recognition is successful, the robot then should be able to understand the result of the speech recognition that is in text form. It should process the text that is in a natural language to find the meaning of the speech in terms of different task dimensions. It also performs the dialogue processing by dialogue management and dialogue status tracking techniques to find out the speaker's intention and purpose of the dialogue while adapting to the speaker's conversational style. In the long run, the robot, or the intelligent agent achieves his goal by providing the appropriate service after figuring out the conversation. + +# 7 Requirements of indoor conversational robot systems + +The general requirements for an indoor conversational robot system (ICR) are described as follows. + +- An ICR is required to provide a speech recognition function for spontaneous speech. +- An ICR is required to be able to recognize the predefined activation command spoken by the user. + +- An ICR is recommended to provide a keyword registration function for the user. +- An ICR is recommended to provide a dialogue processing function for spontaneous utterances. +- An ICR is recommended to be able to understand the user's intention by analysing the speaker's speech acts. +- An ICR is recommended to provide a conversation tracking function to understand the interactive dialogues between the user and the system. +- An ICR is recommended to provide a dialogue generation function to perform the interactive dialogues between the user and the system. +- An ICR is recommended to be able to identify pre-validated users. +- An ICR is recommended to be able to recognize the location of the user and change its position towards the user. +- An ICR is recommended to be able to identify pre-validated users, recognize the location of the users and the predefined activation command at the same time. +- An ICR is recommended to recognize speech in the noisy environment with reverberation. +- An ICR is recommended to process dialogues demanding multiple tasks. +- An ICR is recommended to be able to adapt to the user through conversational experiences through incremental learning. + +# **8 Functional components and interfaces for indoor conversational robot systems** + +The functional components of indoor conversational robot systems, presented in Figure 1, are composed of the following modules: + +- variable-distance speech preprocessing module; +- user identifying speech recognition module; +- multitask dialogue processing module; +- speech synthesis module (TTS: text to speech). + +![Figure 1: Indoor conversational robot system architecture diagram. The diagram shows a User (represented by a person icon) and a ROBOT (represented by a cloud icon). The User sends speech input (e.g., 'Romeo, where are you Romeo?', 'Tony, what's the traffic like on the way to the shopping centre?', 'Albert, you'd better take a bus, and you need to be back for a meeting at 2.') which is affected by noise, reverberation, and distance. The speech is processed by a Variable-distance speech preprocessing module (containing Beam forming, Sound separation, and Noise reverberation removal). The output is then processed by a User identifying speech recognition module (containing Keyword, Speaker, and Speech recognition). The result is sent to a Multi-task dialogue processing module (containing Dialogue generation, Semantic analysis, Dialogue manager, and Dialogue state tracking). The Dialogue manager interacts with a Web cloud. The output of the Dialogue manager is processed by a TTS (Text-to-Speech) module, which generates the robot's response (e.g., 'Albert, you'd better take a bus, and you need to be back for a meeting at 2.'). The User is labeled as 'Variable-distance indoor noise/reverberation'.](cfef993dcc8fb513de79eb1f93cf26ae_img.jpg) + +F.746.9(19)\_F01 + +Figure 1: Indoor conversational robot system architecture diagram. The diagram shows a User (represented by a person icon) and a ROBOT (represented by a cloud icon). The User sends speech input (e.g., 'Romeo, where are you Romeo?', 'Tony, what's the traffic like on the way to the shopping centre?', 'Albert, you'd better take a bus, and you need to be back for a meeting at 2.') which is affected by noise, reverberation, and distance. The speech is processed by a Variable-distance speech preprocessing module (containing Beam forming, Sound separation, and Noise reverberation removal). The output is then processed by a User identifying speech recognition module (containing Keyword, Speaker, and Speech recognition). The result is sent to a Multi-task dialogue processing module (containing Dialogue generation, Semantic analysis, Dialogue manager, and Dialogue state tracking). The Dialogue manager interacts with a Web cloud. The output of the Dialogue manager is processed by a TTS (Text-to-Speech) module, which generates the robot's response (e.g., 'Albert, you'd better take a bus, and you need to be back for a meeting at 2.'). The User is labeled as 'Variable-distance indoor noise/reverberation'. + +**Figure 1 – Indoor conversational robot system** + +In the variable-distance speech preprocessing module, various reverberations of the indoor environments and noise inflicted variable-distance speech is cleaned and the reliable speech recognition is made possible. + +In the user identifying speech recognition module, the pre-validated user calls the robot with predefined words which is waiting for a command. The robot is then activated, recognizes the specific user and starts a dialogue with the user. + +In the multitask dialogue processing module, the robot understands the information in the natural language text which is the speech recognition result. The robot has the conversation with the user by adaptive dialogue management, generates responses and expands or changes the dialogue topic appropriately depending on the different tasks. + +In the speech synthesis module, the result of the dialogue generation turns into the sound of the robot. That is the well-known text-to-speech module in the general speech interface. + +![Figure 2: Functional architecture of indoor conversational robot system. This is a linear flow diagram showing four sequential steps: Speech preprocessing, Speech recognition, Dialogue processing, and Speech synthesis. The steps are connected by arrows from left to right.](5e92d9e8e9ce204e405bff2367f88176_img.jpg) + +F.746.9(19)\_F02 + +Figure 2: Functional architecture of indoor conversational robot system. This is a linear flow diagram showing four sequential steps: Speech preprocessing, Speech recognition, Dialogue processing, and Speech synthesis. The steps are connected by arrows from left to right. + +**Figure 2 – Functional architecture of indoor conversational robot system** + +## 8.1 Variable-distance speech preprocessing module + +In the variable-distance speech preprocessing module, various reverberations of the indoor environments and noise inflicted variable-distance speech is cleaned and the reliable speech recognition is made possible. The following functions are included in the variable-distance speech preprocessing module as essential functions: + +- Sound localization function that is robust for reverberation – This function detects the direction of the sound source with noise and reverberation; +- Noise reverberation-removing function – This function removes or minimizes noise and reverberations from the sound source using methods such as the beamforming algorithm that keeps the direction information and minimizes noise from the sound. Reverberation suppression is performed which is independent of the array structure of the microphone. +- Sound separation function – This function separates voice features from multichannel input sound signal information of frequency, time and channel to remove background noise. The resulting voice signal proceeds to a speech recognizer as the input signal. +- Sound feature compensation function – This function changes from distorted voice feature to clean voice feature using inference by deep neural network (DNN) sound modelling for better speech recognition. + +For reliable variable-distance speech recognition, the database (DB) information on the items in the following table is required. + +**Table 1 – Database information for variable-distance speech recognition for indoor conversational robot system** + +| DB information type | Information subtype | Example | +|----------------------------|-------------------------------------|------------------------------------| +| Language information | | | +| | Language type | English, French, Korean | +| Speaker information | | | +| | Gender | Male, Female | +| | Age | 17, 25, 60th, 30th | +| | Origin | Macao, Italy, Southern France | +| Room acoustic information | | | +| | Surrounding noise type | Point noise source/ambient noise | +| | Noise type | Music, babble | +| | Speech-to-noise ratio | 100 vs. 1 | +| Microphone information | | | +| | Model name | MT600 | +| | Type of microphone | Dynamic, condenser, mems | +| | Channel type | Single/multiple | +| | Number of microphones | | +| | Microphone array configuration type | linear, circular | +| Speech information | | | +| | Location of the speaker | Distance and angle from microphone | +| | Speech contents | Words, linked words, sentences | +| | Speech style | Reading, spontaneous | +| File information | | | +| | File type | Raw, wav | +| | Sampling rate | 8 kHz, 16 kHz, 44.1 kHz, 48 kHz | +| | File name structure | Speech1-raw-16 kHz | + +## 8.2 User identifying speech recognition module + +The user identifying speech recognition module performs the following processes: the pre-validated user calls the robot with predefined words which is waiting for a command. The robot is then activated, recognizes the specific user, and starts a dialogue with the user. The function of the speech recognition module is to generate output text from the user's speech signals. The following functions are used for the user identifying speech recognition module. + +- Speech DB construction – This function, for the indoor home environment, variable distance multichannel speech DB collection, is needed using microphone arrays for the speech recognition module. The collected speech utterances is arranged to include silent sections for end-point detection. + +![Diagram of a multichannel microphone array speech DB collection system. The system consists of a multichannel mic array connected to a multichannel amplifier, which is connected to a USB interface, which is connected to a PC for multichannel recording.](ff0952ef692c9d960ce5f6708bcc9711_img.jpg) + +The diagram illustrates a multichannel microphone array speech DB collection system. On the left, a 'Multichannel mic array' is shown, consisting of a circular base with multiple microphones arranged around it. Arrows point from the microphones to a 'Multichannel amplifier' in the center, which is a rack-mounted unit with many small black modules. From the amplifier, arrows point to a 'USB interface' unit, which is a small white device with several ports. Finally, an arrow points from the USB interface to a 'PC for multichannel recording', which is a desktop computer with a monitor displaying a landscape image. + +Diagram of a multichannel microphone array speech DB collection system. The system consists of a multichannel mic array connected to a multichannel amplifier, which is connected to a USB interface, which is connected to a PC for multichannel recording. + +F.746.9(19)\_F03 + +**Figure 3 – Example of multichannel microphone array speech DB collection system** + +- End-point detection function – This function detects the voice signal from the input signal. This function is used as a preprocess for speech recognition, speaker verification or speaker identification. It consists of voice activity detection (VAD) stage and smoothing stage. Voice activity detection (VAD) is to classify the input signal into voice and non-voice using DNN. The result of VAD is run through a smoothing process by a hangover technique and returns the end-point of each sentence. +- Keyword extraction function – This function, which utilizes a DNN sound model, is robust against background noise. Keywords are registered by the user for activating robots and used as training data for keyword extraction and speech recognition. +- Speaker identification function – This function, which utilizes DNN feature analysis, is robust against background noise. Speaker identification is a type of classification task to detect a speaker of the utterance from a certain speaker group. Recent AI technologies such as DNN improves the performance of classification tasks. +- Automatic speech recognition function (ASR) – This function generates output text from the user's speech signals. The ASR function should be scalable to multi-robot environments as the example architecture in Figure 4 shows. Training data for ASR is composed of that recorded in similar environments with noise and variable distance. + +![Diagram of a speech interface for multiple conversational robots. On the left, three user silhouettes are shown with speech waveforms. The top user has a microphone icon, the middle user has a red ear icon, and the bottom user has a computer monitor icon. Arrows labeled 'PCM' and 'N-best results' point from these users to a central 'Load balancer' (represented by a server icon with a database symbol). The load balancer has arrows pointing to three 'Speech recognizer' units (represented by server icons). A 'Client' icon (microphone) is also shown with an arrow pointing to the load balancer. The diagram is labeled 'F.746.9(19)_F04' in the bottom right corner.](4801720824e4b5e2361a5564f91cfb70_img.jpg) + +Diagram of a speech interface for multiple conversational robots. On the left, three user silhouettes are shown with speech waveforms. The top user has a microphone icon, the middle user has a red ear icon, and the bottom user has a computer monitor icon. Arrows labeled 'PCM' and 'N-best results' point from these users to a central 'Load balancer' (represented by a server icon with a database symbol). The load balancer has arrows pointing to three 'Speech recognizer' units (represented by server icons). A 'Client' icon (microphone) is also shown with an arrow pointing to the load balancer. The diagram is labeled 'F.746.9(19)\_F04' in the bottom right corner. + +F.746.9(19)\_F04 + +**Figure 4 – Speech Interface for multiple conversational robots** + +The user identifying speech recognition module employs the deep neural network (DNN) model in most functions in the module. In general, a deep neural network (DNN) model shows good performance in the generation or classification task. However, DNN fundamentally has non-linear characteristics, and therefore it is difficult to interpret how a result from inputs given to a black box model has been derived. To explain the hidden process, a linear approximation of a DNN model which is a non-linear function is performed with the following operations consecutively: an operation of expanding an input to a neuron of the DNN into a polynomial; a second operation of approximating the neuron of the DNN with a Taylor series in parallel with the polynomial expansion of the input; and a third operation of classifying the polynomially-expanded input and the Taylor-series approximated neuron as a polynomial of input signals and a polynomial of weights. + +## 8.3 Multitask dialogue processing module + +The multitask dialogue processing module performs the following functions: the robot understands the information in the natural language text which is the speech recognition result. The robot has the conversation with the user by adaptive dialogue management, generates responses and expands or changes the dialogue topic appropriately depending on the different tasks. The following functions are included in the multitask dialogue processing module. + +- Semantic analysis of dialogues – Semantic analysis is the essential function for understanding the user's intention of the utterance. It should be capable of multitask semantic tagging which is used for the analysis. Speech act and dialogue act analysis models are employed for the dialogue understanding function. A slot filling model for dialogue understanding is constructed based on DNN using a large corpus of user dialogues. Dialogue topic expansion and modification to improve the dialogue model are also included in the semantic analyser function. +- Dialogue state tracking function – This function detects the state of each turn of the user/robot dialogue such as "question", "answer" or "making commands" which are robust for dialogue topic expansion and modification. DNN-based dialogue state tracking technology is used for dialogue state recognition. In Figure 5 an example dialogue state tracker is presented which uses long short-term memory (LSTM) networks, a recently developed neural network. + +![Diagram of a dialogue state tracker architecture. It shows a sequence of four LSTM units. Each unit receives a 'Dialogue-act list' as input at time steps t=1, t=2, t=3, and t=L. Each LSTM unit outputs a 'Goal label'. The units are connected sequentially from left to right.](33ed1f9b27c7c21c797aa928b0f06851_img.jpg) + +Diagram of a dialogue state tracker architecture. It shows a sequence of four LSTM units. Each unit receives a 'Dialogue-act list' as input at time steps t=1, t=2, t=3, and t=L. Each LSTM unit outputs a 'Goal label'. The units are connected sequentially from left to right. + +F.746.9(19)\_F05 + +**Figure 5 – Example of dialogue state tracker** + +- Dialogue management function – This function manages the dialogue between the user and the robot based on the intention of the user. The dialogue management function should be capable of expanding and modifying dialogue topics. Various reinforcement learning methods based on DNN are employed for better performance of dialogue management. +- Dialogue generation function – This function produces a robot utterance with a correct dialogue act from the input dialogue act of the human user. Based on a statistical model the dialogue generation function is task adaptable for multiple dialogue topics. + +## **8.4 Speech synthesis module** + +The speech synthesis module performs the following functions: the result of the dialogue generation turns into the sound of the robot. This is the well-known text-to-speech module in the general speech interface. The speech synthesis module generates speech signals from the text input by text normalization and morphological analysis, intonation target (i.e., pitch contour) generation and speech signal generation using language analysis and intonation prediction information for the input sentences. + +# **9 Application programme interfaces (API) for indoor conversational robot systems** + +In this clause, APIs for different modules of a robot system are defined: communication between client and server, preprocessing module, speech recognition module and dialogue processing module. + +## **9.1 APIs to communicate between client and server** + +The APIs to communicate between the client and server consists of methods to make a new connection to the server, to recognize recorded multichannel signals, to get an ASR or dialogue result for a given input signal and to get a response from the robot to the current user input in wave format. + +**Table 2 – APIs for communication between server and client** + +| | | +|-----------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Constructor | | +| Constructor and description | | +| ConversationalRobot() | | +| Default constructor | | +| Methods | | +| Modifier and type | Method and description | +| int | openConnection(String id, String ipAddress, Integer port) | +| | Make a new connection to ASR/dialogue server
id: Unique identifier of client
ipAddress: ip address of ASR/Dialogue server
port: TCP port number to connect server | + +**Table 2 – APIs for communication between server and client** + +| | | +|-----------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| | Returns 0 on success or ERROR code otherwise | +| void | closeConnection() | +| int | setParameters(String key, String value)
Set control parameters of either client or server including number of channels, kind of input signal, domain of conversation etc
key: parameter name
value: parameter value
Returns 0 on success or ERROR code otherwise | +| String | getParameters(String key)
Get control parameters of either client or server
key: parameter name
Returns value corresponding to key | +| int | Recognize(Array> sig, int dim, int len)
Recognize recorded multichannel signal
dim: number of microphone channels
len: length of signal
Returns status of recognizer | +| String | getResult(int kind)
Get ASR or dialogue result for given input signal so far
kind: kind of result: ASR or Dialogue
Returns result of ASR, dialogue, etc in text format | +| Array | getResponse()
Get response of robot to current user input in wave format, if any
Returns waveform data which can be played in | +| String | getStatus(int kind)
Get current status of ASR or dialogue system
kind: kind of status
Returns the current status of engine in text format | + +## 9.2 APIs for preprocessing module + +The APIs for the preprocessing module creates resources for the preprocessing and processing of a block of multichannel audio signals for the use of speech recognition in the next step. + +**Table 3 – APIs for preprocessing module** + +| | | +|-----------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Constructor | | +| Constructor and description | | +| CPreproc() | | +| Default constructor | | +| Methods | | +| Modifier and type | Method and description | +| int | createPreproc(int ch, int fs)
Create resources for preprocessing
ch: number of channels
fs: sampling frequency of input signal
Returns 0 on success, error code otherwise | + +**Table 3 – APIs for preprocessing module** + +| | | +|--------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| void | freePreproc()
Free allocated resources | +| int | processBlock(Array> insig, int len, Array& outsig)
Preprocess a block of muti-channel audio signals
insig: multichannel speech signals
len: length of input signals
outsig: output single channel signal
Returns length of output signals | +| int | setParameters(String key, String value)
Set control parameters of preprocessing module
key: parameter name
value: parameter value
Returns 0 on success or error code otherwise | +| String | getParameters(String key)
Get control parameters of preprocessing module
key: parameter name
Returns value corresponding to the key | + +## 9.3 APIs for automatic speech recognition module + +This module includes key-word spotting, speaker identification, language identification and speech recognition. It returns result of the ASR module in a text format. The result can have more than one hypothesis. + +**Table 4 – APIs for ASR module** + +| | | +|-----------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Constructor | | +| Constructor and description | | +| CASR() | | +| Default constructor | | +| Methods | | +| Modifier and type | Method and description | +| int | createASR(int ch, int fs)
Create resources for ASR
ch: number of channels
fs: sampling frequency of input signal
Returns 0 on success, error code otherwise | +| void | freePreproc()
Free allocated resources | +| int | processBlock(Array> insig, int len)
Process a block of multichannel audio signals for key-wordspotting, speaker identification, language identification, speech recognition etc
insig: multichannel speech signals
len: length of input signals | + +**Table 4 – APIs for ASR module** + +| | | +|---------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| | Returns status of engine | +| int | setParameters(String key, String value)
Set control parameters of ASR module
key: parameter name
value: parameter value
Returns 0 on success or error code otherwise | +| String | getParamters(String key)
Get control parameters of ASR module
key: parameter name
Returns value corresponding to key | +| Array | getResult(int kind)
Get result of ASR module so far
kind: a kind of result requested
Returns result of ASR module in text format. The result can have more than one hypothesis | + +## 9.4 APIs for dialogue module + +The APIs for the dialogue module create resources for dialogue processing and processing text for the multiple candidate results. It returns the result of the dialogue module in a text format. The result can have more than one hypothesis as with the speech recognition module. + +**Table 5 – APIs for dialogue module** + +| | | +|-----------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Constructor | | +| Constructor and description | | +| CDialog() | | +| Default constructor | | +| Methods | | +| Modifier and type | Method and description | +| int | createDialog(int lang)
Create resources for dialogue
lang: language id
Returns 0 on success, error code otherwise | +| void | freeDialog()
Free allocated resources | +| int | processDialog(Array nbest)
Process a text of nbest hypothesis
nbest: contents of user utterance
Returns status of engine | +| int | set Parameters(String key, String value)
Set control parameters of dialogue module
key: parameter name
value: parameter value
Returns 0 on success or error code otherwise | + +**Table 5 – APIs for dialogue module** + +| | | +|---------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| String | getParameters(String key)
Get control parameters of dialogue module
key: parameter name
Returns value corresponding to key | +| Array | getResult(int kind)
Get result of dialogue module so far
kind: a kind of result requested
Returns result of dialogue module in text format. The result can have more than one hypothesis | + +## Appendix I + +## Evaluation method for measuring performance of an indoor conversational robot system in variable distant noisy environments + +(This appendix does not form an integral part of this Recommendation.) + +In this appendix the evaluation method for evaluating an indoor conversational robot system is introduced. The evaluation method consists of the three methods given in clauses I.1, I.2 and I.3, each of which evaluates different function. + +### I.1 Evaluation of speech recognition in variable distant noisy environments + +The speech recognition performance is evaluated in a home-like space with various noise sources where the user speaks given sentences. Word error rate (WER) is the evaluation metric for the speech recognition performance. WER is measured by comparing the sentences spoken by the user with those recognized by the speech recognition system. The WER calculation formula is as follows [b-Klakow]. + +$$\text{WER} = (S + D + I)/N$$ + +$S$ = number of substituted words + +$D$ = number of deleted words + +$I$ = number of inserted words + +$N$ = total number of words in reference text + +Noise sources that are considered are home environments such as TV, audio sounds, voice babble and noise from home appliances. Noise volume is decided based on the signal to noise ratio (SNR) by comparing it with the user's voice volume. The performance of the speech recognition function is measured after noise sources are produced according to three conditions as SNR 15 dB, 10 dB, 5 dB. The size of the space, reverberation feature, location of the robot, location of the sound source (relative distance and direction of the sound source and robot) and location of the noise source (relative distance and direction of the noise source and robot) are considered for the evaluation of the speech recognition function as presented in Figure A.1 as an example. Reverberation time (RT) is a measure of the time required for the sound to "fade away" in an enclosed area after the source of the sound has stopped. The RT60 metric is used for reverberation feature of the space which expresses the degree of reverberation. For the RT60 metric, 15~20 ms is used to mimic general home environments. RT60 is defined as the time it takes for the sound pressure level to reduce by 60 dB, measured after the generated test signal is abruptly ended. + +![Diagram illustrating the example configuration of microphones, speech, and noise source. A blue circle represents the Robot. A yellow circle represents the Noise source at 3m, -45°. Three red circles represent Speech sources: one at 1m, 0°; one at 5m, 0°; and one at 5m, 45°. Dashed lines connect the robot to each source. A legend on the right identifies the symbols: yellow circle for Noise source, red circle for Speech source, and blue circle for Robot. The diagram is labeled F.746.9(19)_FA.1.](5b8a756d9a71c35f17db8bcb90b438a3_img.jpg) + +Diagram illustrating the example configuration of microphones, speech, and noise source. A blue circle represents the Robot. A yellow circle represents the Noise source at 3m, -45°. Three red circles represent Speech sources: one at 1m, 0°; one at 5m, 0°; and one at 5m, 45°. Dashed lines connect the robot to each source. A legend on the right identifies the symbols: yellow circle for Noise source, red circle for Speech source, and blue circle for Robot. The diagram is labeled F.746.9(19)\_FA.1. + +**Figure A.1 – Example configuration of microphones, speech and noise source** + +### I.2 Evaluation of call success in noisy environments + +Call success rate measures the performance of a robot who recognizes the voice call successfully and gets ready for the response. False rejection rate (FRR), false alarm rate (FAR), and equal error rate (EER) are used as call success rate metrics. FRR is defined as the rate that a robot fails to recognize the predefined word or phrases a user speaks. Therefore, the call success rate is $[1 - \text{FRR}]$ . + +$$\text{FRR} = (\text{number of missed calls}) / (\text{total calls tried})$$ + +False recognition voice calls which are not correct call words or phrases are defined by FAR. + +$$\text{FAR} = (\text{number of false recognized calls}) / (\text{total calls tried})$$ + +Recognized calls are accepted or rejected based on the threshold value that was predefined by the statistical model. The performance of FRR and FAR is decided depending on the threshold tuning state. In general, FAR increases if FRR is improved. Considering this general case, the performance can be measured using EER where FRR and FAR are set to the same value. The selection of FRR, FAR, EER can be decided according to the application purpose of the robot. + +### I.3 Evaluation of conversations of specific purposes in indoor noisy environments + +Conversations are evaluated by the conversation success rate which measures the degree of achievement of the defined missions that a robot and a human user try to complete through speech conversations. The conversation success rate is measured by the number of successful missions among the total missions that were achieved within the defined number of turns of the conversation. + +An example would be the case where the number of conversation turns are limited to within 5 turns. The mission of "Make the dental reservation with a dentist at 3 p.m." is regarded as successful if the robot finishes the mission within 5 turns. + +## Bibliography + +- [b-ITU-T F.746.3] Recommendation ITU-T F.746.3 (2015), *Intelligent question answering service framework*. +- [b-ITU-T H.703] Recommendation ITU-T H.703 (2016), *Enhanced user interface framework for IPTV terminal devices*. +- [b-Klakow] Klakow, D. & Peters, J., 2002, *Testing the correlation of word error rate and perplexity*. Speech Communication. Vol.38, No.1; pp. 19–28. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.747.10-202201-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.747.10-202201-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1e32f9b8f1141bed243c16e02a6cd1411cbb87f1 --- /dev/null +++ b/marked/F/T-REC-F.747.10-202201-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c395c4bbde83b1e232652a43f51538278c116f51f710c37e20a073f7333726c8 +size 6134 diff --git a/marked/F/T-REC-F.747.10-202201-I_PDF-E/4e4be0bd8b235167902f2c03e41da651_img.jpg b/marked/F/T-REC-F.747.10-202201-I_PDF-E/4e4be0bd8b235167902f2c03e41da651_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..407605a37a654c616146d4fefde404b62932742e --- /dev/null +++ b/marked/F/T-REC-F.747.10-202201-I_PDF-E/4e4be0bd8b235167902f2c03e41da651_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:94e46d85bf477136ef175d51e4cddbb9b92bea9a5b9dce1acc8014589a4eb136 +size 53834 diff --git a/marked/F/T-REC-F.747.10-202201-I_PDF-E/4ee27dbf5ef12e7b58b0ef0937bc5a5e_img.jpg b/marked/F/T-REC-F.747.10-202201-I_PDF-E/4ee27dbf5ef12e7b58b0ef0937bc5a5e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..334665eb912f3c32879a06edfa4e9256ed808c5a --- /dev/null +++ b/marked/F/T-REC-F.747.10-202201-I_PDF-E/4ee27dbf5ef12e7b58b0ef0937bc5a5e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:75ccd634a4cb8020528fee09535290d7e78e3503f297ab0f0321dc20588a7951 +size 59730 diff --git a/marked/F/T-REC-F.747.10-202201-I_PDF-E/75f0cb39f1cd165dfe4a6aa6c4d9388d_img.jpg b/marked/F/T-REC-F.747.10-202201-I_PDF-E/75f0cb39f1cd165dfe4a6aa6c4d9388d_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3b652f8b678e06a83f776fb6354b64ba8cfd15cf --- /dev/null +++ b/marked/F/T-REC-F.747.10-202201-I_PDF-E/75f0cb39f1cd165dfe4a6aa6c4d9388d_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1702deaa9377b73e616f602719036c980785ab9acd477983ae1280b6ff5c2e84 +size 96614 diff --git a/marked/F/T-REC-F.747.10-202201-I_PDF-E/b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg b/marked/F/T-REC-F.747.10-202201-I_PDF-E/b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..37731ec38f0e3df5f9ce4ab373538793a7508122 --- /dev/null +++ b/marked/F/T-REC-F.747.10-202201-I_PDF-E/b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:84f951cc17a5371fffcf62f9258268dceae22a913682eb3ed85a64ac8124f866 +size 41348 diff --git a/marked/F/T-REC-F.747.11-202212-I_PDF-E/053f1077d592e6622cd21dc4bb4cb366_img.jpg b/marked/F/T-REC-F.747.11-202212-I_PDF-E/053f1077d592e6622cd21dc4bb4cb366_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c46cb0309718748bf7e3837f7913ea0a5250e80a --- /dev/null +++ b/marked/F/T-REC-F.747.11-202212-I_PDF-E/053f1077d592e6622cd21dc4bb4cb366_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ae031d4f772606128d73ae8463093ed56d369db30153cdc219f0787cdf5bf4b7 +size 46342 diff --git a/marked/F/T-REC-F.747.11-202212-I_PDF-E/159f15d60777adb6b16c91d84b32c32f_img.jpg b/marked/F/T-REC-F.747.11-202212-I_PDF-E/159f15d60777adb6b16c91d84b32c32f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..6315c1d3a941e00f6b02fe6a81bedde8b8b8950f --- /dev/null +++ b/marked/F/T-REC-F.747.11-202212-I_PDF-E/159f15d60777adb6b16c91d84b32c32f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0ed11f74bf3883fcb302deedc9c3a81c65a3f55e5a7cc17d5b1671c1b8f8a96f +size 27588 diff --git a/marked/F/T-REC-F.747.11-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.747.11-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ac295b1b0c7f14b1c5a33ef6682b1b017c812477 --- /dev/null +++ b/marked/F/T-REC-F.747.11-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37aed4fc114a39843438175326e0b7e1d09a9d6075723d82fea32040fd560ee4 +size 7192 diff --git a/marked/F/T-REC-F.747.11-202212-I_PDF-E/78f163db583eb0c6d013e6f8ffe6d7d8_img.jpg b/marked/F/T-REC-F.747.11-202212-I_PDF-E/78f163db583eb0c6d013e6f8ffe6d7d8_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a63f5d0fa7fb8b1717d88deb1b25cba21f26bf19 --- /dev/null +++ b/marked/F/T-REC-F.747.11-202212-I_PDF-E/78f163db583eb0c6d013e6f8ffe6d7d8_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f59bbbe2a0843b263f1f8d4798d2ad51b11e8d45f7a306ea7cd2540bc8579601 +size 9208 diff --git a/marked/F/T-REC-F.747.11-202212-I_PDF-E/832a3f95cc4469630aca33d36af8602c_img.jpg b/marked/F/T-REC-F.747.11-202212-I_PDF-E/832a3f95cc4469630aca33d36af8602c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9e444f8176209e44a499b9f6b25ae36f04f37fb1 --- /dev/null +++ b/marked/F/T-REC-F.747.11-202212-I_PDF-E/832a3f95cc4469630aca33d36af8602c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:00486b7bfe59859a299f43798180aeb6498b247461be931831d0943747d7e946 +size 31785 diff --git a/marked/F/T-REC-F.747.11-202212-I_PDF-E/df6babe297323feb1575ba89f5cf3b09_img.jpg b/marked/F/T-REC-F.747.11-202212-I_PDF-E/df6babe297323feb1575ba89f5cf3b09_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..28c77ae601624719a06ac182a90f874a57af1be4 --- /dev/null +++ b/marked/F/T-REC-F.747.11-202212-I_PDF-E/df6babe297323feb1575ba89f5cf3b09_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d1141229b6e24c71770039363889dd5570972433f67911ca46780a020377c258 +size 14814 diff --git a/marked/F/T-REC-F.747.11-202212-I_PDF-E/eb22a8740f7c6a0f6ee98f16d99ed8b9_img.jpg b/marked/F/T-REC-F.747.11-202212-I_PDF-E/eb22a8740f7c6a0f6ee98f16d99ed8b9_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..25c35d96b4b388ece4ba70c96fccefefd51fd56f --- /dev/null +++ b/marked/F/T-REC-F.747.11-202212-I_PDF-E/eb22a8740f7c6a0f6ee98f16d99ed8b9_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fee6210101ff4d1b44cce1f29ff6d18b5a01a6a73353e8777ed0f53b08ec9062 +size 36680 diff --git a/marked/F/T-REC-F.747.11-202212-I_PDF-E/raw.md b/marked/F/T-REC-F.747.11-202212-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..954876a349c29371f470817faff46237cee98b89 --- /dev/null +++ b/marked/F/T-REC-F.747.11-202212-I_PDF-E/raw.md @@ -0,0 +1,405 @@ + + +# Recommendation **ITU-T F.747.11 (12/2022)** + +SERIES F: Non-telephone telecommunication services +Multimedia services + +## --- **Requirements for intelligent surface-defect detection service in industrial production lines** + +![ITU logo](390120de4fe440c42fea8154fcaad334_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international teletext service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.747.11 + +# Requirements for intelligent surface-defect detection service in industrial production lines + +## Summary + +Intelligent surface-defect detection service in industrial production lines refers to the accurate product defect location determination, high-speed classification of defect types, real-time output and transmission of visual and auditory information to ensure the quality of industrial products. Compared with inspection carried out manually by workers, the intelligent surface-defect detection service can improve efficiency and consistency and reduce manual operations in dangerous areas. This Recommendation specifies requirements for intelligent surface-defect detection service in industrial production lines, including performance requirements, application requirements and functional requirements. To provide an effective surface-defect detection service, it is required to fulfil three important parts. Firstly, it is important to ensure the accuracy location determination and classification. Secondly, the inference efficiency of the service is also required to satisfy the real-time settings. Last but not least, the service is required to adapt to the typical application scenarios in industrial production line inspection tasks. + +Recommendation ITU-T F.747.11 provides related requirements for intelligent surface-defect detection service in industrial production lines. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.747.11 | 2022-12-14 | 16 | 11.1002/1000/15192 | + +## Keywords + +Industrial production line, intelligent, multimedia video, surface-defect detection. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2023 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation ..... | 1 | +| 4 Abbreviations and acronyms ..... | 1 | +| 5 Conventions ..... | 2 | +| 6 Overview of the intelligent surface-defect detection service ..... | 2 | +| 7 Requirements of the intelligent surface-defect detection service..... | 3 | +| 7.1 Performance requirements ..... | 3 | +| 7.2 Application requirements..... | 4 | +| 7.3 Functional requirements..... | 4 | +| 8 Reference architecture..... | 4 | +| 8.1 Intelligent device ..... | 4 | +| 8.2 Edge device..... | 5 | +| 8.3 ISD server ..... | 5 | +| 8.4 Application scenario..... | 6 | +| Appendix I – Use cases of ISD service ..... | 7 | +| I.1 Provide detection service for steel and steel products..... | 7 | +| I.2 Provide detection service for electronic devices ..... | 7 | +| Bibliography ..... | 9 | + + + +# Recommendation ITU-T F.747.11 + +## Requirements for intelligent surface-defect detection service in industrial production lines + +## 1 Scope + +This Recommendation specifies requirements for intelligent surface-defect detection (ISD) service in industrial production lines, including performance requirements, application requirements and functional requirements. The reference architecture, multiservice management and development tools in general machine vision-based platform are out of the scope of this Recommendation. + +The scope of this Recommendation includes: + +- performance requirements of ISD service; +- application requirements of ISD service; +- functional requirements of ISD service. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T F.748.21] Recommendation ITU-T F.748.21 (2022), *Requirements and framework for feature-based distributed intelligent systems*. + +# 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following term defined elsewhere: + +**3.1.1 machine vision** [b-ITU-T F.748.16]: Signal processing to acquire, process and interpret an image or video to support visual analysis for applications, such as automatic inspection, process control and guidance. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following term: + +**3.2.1 surface-defect detection**: Surface-defect detection refers to the identification of defects on the surface of industrial products, including the location determination, classification and measurement of different types of defects. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|-------------------------------| +| CPU | Central Processing Unit | +| FPGA | Field Programmable Gate Array | + +| | | +|-----|--------------------------------------| +| GPU | Graphics Processing Unit | +| ISD | Intelligent Surface-defect Detection | +| mAP | mean Average Precision | +| PCB | Printed Circuit Boards | +| SoC | System on Chips | + +## 5 Conventions + +In this Recommendation: + +- The keywords "**is required to**" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "**is recommended**" indicate a requirement which is recommended but which is not absolutely required. Thus this requirement needs not be present to claim conformance. +- These terms are not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +# 6 Overview of the intelligent surface-defect detection service + +Surface defects are mainly caused by occasional machine failure, operation error or production environment problems in the production process. For surface defects, images captured by cameras can be directly used for detection. Under the influence of various factors, the types of surface defect are diverse. Taking the surface of aluminium product as an example, there are various types of defect such as paint bubble, erasing, orange peel and so on. To identify the defect type and defect area, overkill rate and miss rate are two important metrics for an intelligent surface-defect detection service. Overkill rate is measured by the number of defects incorrectly identified while the miss rate is measured by the number of defects not identified. the tolerance of the miss rate is stricter than the tolerance of the overkill rate. + +Different operation measures are required to be taken for different defect types. Some defects have little impact on quality and can be accepted by the purchaser if the product is offered at a lower price, while some defects can be repaired. Products with a defect with a serious quality impact cannot be sold. Therefore, in practical application, it is recommended to classify products and distinguish between normal samples without defects and defective samples. Further, it is recommended to locate defects and mark the specific location of defects, which is collectively referred to as defect detection in this Recommendation. + +In the real production environment, the imaging quality of some images is poor and there is a lot of noise and interference. It is difficult to obtain good detection efficiency in these scenes by implementing traditional image processing algorithm or the manual feature extraction method. AI has entered a period of rapid development, and deep learning has made significant breakthroughs in the field of computer vision. In image classification, target detection, image segmentation and other computer vision tasks, AI has a good effect. The application of computer vision based on deep learning in surface-defect detection has become a major development trend. + +The ISD service in industrial production lines refers to accurate products defect location determination, high-speed prediction of defect types, real-time outputs and transmission of visual and auditory information to ensure the quality of industrial products. Compared with the inspection carried out manually by workers, the ISD service can improve efficiency and consistency and reduce manual operations in dangerous areas. + +# **7 Requirements of the intelligent surface-defect detection service** + +## **7.1 Performance requirements** + +- a) Applications include but are not limited to detecting surface defects on electronic devices, steel, and so on. The shapes and scales of defects vary in different scenarios, so the service is recommended to carry out classification, object detection, semantic segmentation and low-level vision tasks, including identification of defects type, location by overlaid rectangular box or defect outline, and measurement of defect area. +- b) Noisy annotations are inevitable even for human experts, thus image-level acceptance metrics, such as overkill rate and miss rate, are usually adopted in industrial applications. Besides, the performance requirements vary in different industries. Tasks specified as having a strict control on the miss rate are normally allowed to meet a relatively relaxed overkill rate requirement, and system efficiency is recommended to be considered with the spatial size of the input image and the universal workflow. + +#### **7.1.1 Steel or steel products** + +It is recommended to have a 5-10% overkill rate, but the miss is strictly controlled under 0.1%. The processing time is recommended to be between 0.5 s and 1 s. However, in some exterior steel accessories, such as seat slide rail detection, although scratches and bumps still need to be detected, the requirements will be relaxed compared with the aforementioned standard. + +#### **7.1.2 Electronic devices** + +The application for electronic devices is highly associated with the industries described in clauses 7.1.2.1 to 7.1.2.4. + +##### **7.1.2.1 Battery industry** + +Since surface elevation is essential for identifying pit-like defects and the widely used black/white photography further increases the difficulty of identification, the overkill and miss rates are recommended to be below 5% and 1% respectively. Input images are recommended to retain high resolutions to better locate any tiny defects. + +##### **7.1.2.2 Magnetic core and magnetic ring** + +For magnetic core applications, the runtime is recommended to be less than 100 ms per image and the overkill and miss rates are recommended to both be lower than 1%. For magnetic ring, the target overkill and miss rates are recommended to be less than 0.5% and 1%, respectively. The recommended input image size is $2\,000 \times 2\,000$ pixels, much larger than that of magnetic core, e.g., $500 \times 800$ pixels, thus the systematic runtime is recommended to be less than 500 ms and independent stations can work in parallel. + +##### **7.1.2.3 Display industry** + +The overkill and miss rates are recommended to be controlled under 5% and 0.1%, respectively. The efficiency requirements are conditioned on the image size. For example, the defect detection system is recommended to recognize tiny cracks, breakages and dirt on an $8\,000 \times 16\,000$ pixel image within 500 ms, hence additional plug-in algorithms and modules are not accessible due to the strict runtime constraint and limited computation resources. + +##### **7.1.2.4 Semiconductor industry** + +The semiconductor industry usually cannot tolerate any missed defective products, while the overkill rate can be even relaxed to 20% because the major production purpose lies in the quantity growth and a compulsory manual check will follow the machine inspection. Also, images are rather small. The runtime is recommended to be 15 ms on $400 \times 400$ pixel images. + +## 7.2 Application requirements + +- a) The ISD service is required to provide flexible and extensible service interfaces. +- b) The ISD service is recommended to adapt to the actual input in diverse scenarios according to the specific use case. +- c) The defect detection results are recommended to be delivered with low latency to the manufacturing executive system. +- d) The output of the service is recommended to include both visual description (defect location and types) and the synchronous audio messages (normal and alarm) if needed. + +## 7.3 Functional requirements + +- a) The data pre-processing algorithms deployed in edge devices should be efficient to remove unrelated information for detection. +- b) The surface-defect algorithms are required to support classification, prediction, regression and clustering. +- c) The AI models are required to be deployed in various embedded platform, such as the graphic processing unit (GPU), central processing unit (CPU), system on chips (SoC) and field programmable gate array (FPGA). + +## 8 Reference architecture + +The ISD service introduces a cooperation mechanism among the terminal devices, edge devices and servers shown in Figure 1. The images captured by the front-end need to be transferred to the edge devices and GPU server to finish high-speed inference and the final visual detection results and the corresponding auditory information will be updated in the real-time application system. + +![Figure 1 – Illustration of the ISD service. The diagram shows a flow from an Intelligent device to an Edge device, then to an ISD server, and finally to an Application. The Edge device contains De-noising, Threshold segmentation, and Morphological processing. The ISD server contains Classification, Location, and Segmentation. The Application contains Steel and steel products, Electronic devices, Clothes, and .....](053f1077d592e6622cd21dc4bb4cb366_img.jpg) + +``` +graph LR; ID[Intelligent device] --> ED[Edge device]; ED --> IS[ISD server]; IS --> App[Application]; subgraph ED [Edge device]; ED1[De-noising]; ED2[Threshold segmentation]; ED3[Morphological processing]; end; subgraph IS [ISD server]; IS1[Classification]; IS2[Location]; IS3[Segmentation]; end; subgraph App [Application]; App1[Steel and steel products]; App2[Electronic devices]; App3[Clothes]; App4[.....]; end; +``` + +F.747.11(22) + +Figure 1 – Illustration of the ISD service. The diagram shows a flow from an Intelligent device to an Edge device, then to an ISD server, and finally to an Application. The Edge device contains De-noising, Threshold segmentation, and Morphological processing. The ISD server contains Classification, Location, and Segmentation. The Application contains Steel and steel products, Electronic devices, Clothes, and ..... + +**Figure 1 – Illustration of the ISD service** + +## 8.1 Intelligent device + +Intelligent devices are mainly image acquisition devices combined with AI chips to complete image/video acquisition. The image acquisition devices generally include cameras, lenses and light sources. The camera is an indispensable part of a defect detection service. Its function is to convert the optical image generated by the lens into the corresponding analogue or digital signal through the imaging sensors, and transmit the information to the upper computers for processing through the interface between the cameras and the computers. For surface-defect detection, linear array cameras and planar array cameras are commonly used. Cameras are combined with AI chips, using the algorithms within the cameras to set the shoot parameters. The algorithms can automatically identify each kind of environment, and adjust the camera settings, in order to make shooting achieve excellent results. When there is a poor shooting environment, to the algorithms should activate shoot and + +advanced features such as automatic high dynamic range (HDR), thus further enhancing the image quality. + +## **8.2 Edge device** + +The output of intelligent devices should be pre-processed. The pre-processing of multimedia mainly includes image de-noising, threshold segmentation, morphological processing and so on. In the process of image collection, transmission and processing, the image quality will be degraded and various noises will be generated due to the influence of the cleanliness of the surface of the detected object, the cleanliness of the industrial lens, the internal noise of the sensitive components and lighting changes. Image noise has a great influence on subsequent image processing, which affects each link of image processing and the output result. Therefore, before surface detection, de-noising algorithms are used to remove the noises of the detected image. Image threshold segmentation algorithms are used to compress the amount of data and simplify the steps of image analysis and processing, and this is an essential image pre-treatment process before image analysis, feature extraction and pattern recognition. Image morphology processing algorithms are used to analyse the structure and shape characteristics of the image, so as to facilitate the processing of the subsequent defect detection process. + +In support of the requirements in clause 7.1, the edge device provides: + +- support of image de-noising, image threshold segmentation and image morphological processing; +- support of feature [ITU-T F.748.21] extraction, such as human designed features and convolution neural network features. + +## **8.3 ISD server** + +ISD servers have the capability to solve three problems of surface-defect detection: "what is the defect", "where is the defect", and "how large is the defect". "What is the defect" corresponds to the classification task, "where is the defect" corresponds to the locating task and "how large is the defect" corresponds to the segmentation task. + +In support of the requirements of clauses 7.1 and 7.2, the ISD server provides: + +- support of defect classification, defect location and defect segmentation; +- support of model training and model inference. + +The model performance is required to be evaluated before deployment; the model performance can be mean average precision (mAP). + +### **8.3.1 Defect classification** + +In the real industrial production environment, defect classification in a complex environment becomes a difficult task due to the huge variety in object shape, size, texture, colour, background, layout and imaging illumination. The classification network based on a convolutional neural network with a powerful feature extraction ability is used in surface-defect classification. + +In support of the requirements of clauses 7.1 and 7.2, defect classification includes: + +- identifying common industrial defect types including wrinkles, pits and scratches; +- supporting multiple algorithms including machine learning algorithms; +- supporting the coordination of edge devices and ISD servers for time limitation. + +### **8.3.2 Defect location** + +The purpose of defect localization is to obtain accurate location and category information of the target. It not only obtains the types of defects in the image, but also gives the specific location of the defects. + +In support of the requirements of clause 7.1, defect location includes: + +- support for the identification of the location of defects by overlaid rectangular boxes. + +### **8.3.3 Defect segmentation** + +The surface-defect detection segmentation network transforms the task of surface-defect detection into semantic segmentation and even instance segmentation of the defect and the normal area. It can not only segment the defect area, but also obtain the location, category and corresponding geometric attributes of the defects including length, width, area, contour and centre location. + +In support of the requirements of clause 7.1, defect segmentation includes: + +- support for the identification of defect outlines and measurement of defect area. + +## **8.4 Application scenario** + +Compared with traditional image processing methods, which are divided into multiple steps and links to process defect detection tasks, a deep learning based ISD service unify these into end-to-end feature extraction and classification. There are different types of defect detection in different industries and fields. The intelligent defect detection service in different application scenarios should correspond to real surface-defect data. + +In support of the requirements of clauses 7.1 and 7.2, the application provides: + +- support of flexible and extensible service interfaces; +- support of both visual description including defect location and types and synchronous audio messages including normal and alarm. + +## Appendix I + +### Use cases of ISD service + +(This appendix does not form an integral part of this Recommendation.) + +### I.1 Provide detection service for steel and steel products + +The defects, i.e., cracks and pits on steel products, shown in Figure I.1, are identical to the background region, and the reflective surface might even fail the defect detection system equipped with a normal lighting system. In consequence, the lighting scheme is recommended to be designed carefully according to the sample surface property, because the system efficacy is the major concern of the related industries. The requirements of steel instruments such as accelerators, rollers and so on, will focus on the scratches and bumps of the work piece, because they may risk safety. + +![Figure I.1: Two grayscale images of steel surfaces. The left image shows a surface with numerous small pits and a few larger cracks. The right image shows a surface with a series of parallel, slightly curved lines, which are cracks or scratches.](eb22a8740f7c6a0f6ee98f16d99ed8b9_img.jpg) + +Figure I.1: Two grayscale images of steel surfaces. The left image shows a surface with numerous small pits and a few larger cracks. The right image shows a surface with a series of parallel, slightly curved lines, which are cracks or scratches. + +**Figure I.1 – Illustrations of defects (crack and pit) on steel products** + +### I.2 Provide detection service for electronic devices + +A detection service for electronic devices locates and identifies surface defects including cracks, wrinkles, pits, scratches and other possible anomalies. + +Surface defects in the battery industry mainly include wrinkles, pits and scratches, as shown in Figure I.2, thus detection- and segmentation-based techniques can handle these issues. + +![Figure I.2: Three grayscale images of battery components. The left image shows a surface with a prominent horizontal wrinkle. The middle image shows a surface with a small, dark pit, indicated by a red arrow. The right image shows a surface with a horizontal scratch.](832a3f95cc4469630aca33d36af8602c_img.jpg) + +Figure I.2: Three grayscale images of battery components. The left image shows a surface with a prominent horizontal wrinkle. The middle image shows a surface with a small, dark pit, indicated by a red arrow. The right image shows a surface with a horizontal scratch. + +**Figure I.2 – Illustration of defects (wrinkle, pit and scratch) in the battery industry** + +The magnetic-core-related industries have much smaller input images, e.g., $500 \times 800$ pixels, with clear boundaries between the fore- and background regions. The individual defect specification has been well established and multidefect identification is not common. The acceptance standard changes only if multidefect scenarios need to be considered. In contrast to the magnetic-core-related industries, the majority of the magnetic ring community requires simultaneous multidefect + +identification including cracks, waterlogging and impurities that usually cause a slight abnormal surface appearance. + +![Figure I.3: A close-up image of a dark, textured surface, likely a magnetic core or ring, showing a prominent vertical crack running down the center.](78f163db583eb0c6d013e6f8ffe6d7d8_img.jpg) + +Figure I.3: A close-up image of a dark, textured surface, likely a magnetic core or ring, showing a prominent vertical crack running down the center. + +**Figure I.3 – Illustration of defects (crack) in the magnetic core or ring industry** + +The display industry focuses on defect detection at the display edges, thus the foreground region can be located based on the colour contrast. However, some high-quality display producers might have demanding inspection standards in extreme cases. Defects including crack and breakage are shown in Figure I.4. + +![Figure I.4: A grayscale image showing a close-up of a display edge. A vertical crack is visible on the left side, and there is a dark, irregular shape on the right side indicating breakage or a defect.](df6babe297323feb1575ba89f5cf3b09_img.jpg) + +Figure I.4: A grayscale image showing a close-up of a display edge. A vertical crack is visible on the left side, and there is a dark, irregular shape on the right side indicating breakage or a defect. + +**Figure I.4 – Illustration of defects (crack and breakage) in the display industry** + +As a main product of the semiconductor industry, the PCB-sub board has been widely used in electronic products. The randomness and variety of the defective products are the key issues for the defect detection system. In contrast to the other settings, the query sample is always provided with a perfect template sample for better identifying the potential defects on the query sample. Besides, to achieve better performance, the design drawings can be also leveraged to locate the regions of interest on various PCB-sub boards. The defects including lost, impurity and pits are shown in Figure I.5. + +![Figure I.5: A composite grayscale image showing three different types of defects in the semiconductor industry. The left side shows a circular feature with a small dark spot (lost or impurity). The middle side shows a rectangular feature with a dark, irregular shape (impurity or pit). The right side shows a series of horizontal lines with small dark spots (pits or impurities).](159f15d60777adb6b16c91d84b32c32f_img.jpg) + +Figure I.5: A composite grayscale image showing three different types of defects in the semiconductor industry. The left side shows a circular feature with a small dark spot (lost or impurity). The middle side shows a rectangular feature with a dark, irregular shape (impurity or pit). The right side shows a series of horizontal lines with small dark spots (pits or impurities). + +**Figure I.5 – Illustrations of defects (lost, impurity and pit) in the semiconductor industry** + +## Bibliography + +- [b-ITU-T F.748.16] Recommendation ITU-T F.748.16 (2022), *Requirements for machine vision-based applications and services in smart manufacturing*. + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.747.12-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.747.12-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ac295b1b0c7f14b1c5a33ef6682b1b017c812477 --- /dev/null +++ b/marked/F/T-REC-F.747.12-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37aed4fc114a39843438175326e0b7e1d09a9d6075723d82fea32040fd560ee4 +size 7192 diff --git a/marked/F/T-REC-F.747.12-202212-I_PDF-E/bd671b21db63e6fdb2196e9b18502aac_img.jpg b/marked/F/T-REC-F.747.12-202212-I_PDF-E/bd671b21db63e6fdb2196e9b18502aac_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c91f1efcbeb962f0d25d0c9264b78c35322a8be0 --- /dev/null +++ b/marked/F/T-REC-F.747.12-202212-I_PDF-E/bd671b21db63e6fdb2196e9b18502aac_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5fe2cb61b975549bde4999bd741651c648c54a349ba2fe07d88ad6872fd69eb4 +size 81051 diff --git a/marked/F/T-REC-F.747.12-202212-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.747.12-202212-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..bd43137f115fdea174350d6169415ebf4ef5d680 --- /dev/null +++ b/marked/F/T-REC-F.747.12-202212-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4d2bf77f0afe85d0166989c6c87611a2579cf7831fb2d313fe2f868780a7eb0c +size 144241 diff --git a/marked/F/T-REC-F.747.13-202309-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/F/T-REC-F.747.13-202309-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c1629394c02f697f60ae954ad36bf274fa8082cb --- /dev/null +++ b/marked/F/T-REC-F.747.13-202309-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3763acd7a0ed56aa30f3f90468c1fb0cb54573eb9045bca3eddf8976b4f0f41d +size 7219 diff --git a/marked/F/T-REC-F.747.13-202309-I_PDF-E/10781f43062bf3e9601a1e086710556c_img.jpg b/marked/F/T-REC-F.747.13-202309-I_PDF-E/10781f43062bf3e9601a1e086710556c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..8d82ae61d9b9245fbf213e9abe7278af50c01635 --- /dev/null +++ b/marked/F/T-REC-F.747.13-202309-I_PDF-E/10781f43062bf3e9601a1e086710556c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8dcecde09fcc7fa886c0d39acafcd12fcb167d390a5b4b8525ebf08c2b722b71 +size 53249 diff --git a/marked/F/T-REC-F.747.13-202309-I_PDF-E/724c7777b608e53be38b12b6fb3c43bc_img.jpg b/marked/F/T-REC-F.747.13-202309-I_PDF-E/724c7777b608e53be38b12b6fb3c43bc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..696b708fd6f9b152fe4a5355c570829f1612248b --- /dev/null +++ b/marked/F/T-REC-F.747.13-202309-I_PDF-E/724c7777b608e53be38b12b6fb3c43bc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b95e1e6cad85090213d841f063a180df4f3d3926844e5d009b48033541e1564b +size 158698 diff --git a/marked/F/T-REC-F.747.13-202309-I_PDF-E/daa4a6fa7e2ba1954258f86b4928eb32_img.jpg b/marked/F/T-REC-F.747.13-202309-I_PDF-E/daa4a6fa7e2ba1954258f86b4928eb32_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1d852613c8447d87b67cb49036097818c43cfb89 --- /dev/null +++ b/marked/F/T-REC-F.747.13-202309-I_PDF-E/daa4a6fa7e2ba1954258f86b4928eb32_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6df50826d70eccf182ece01cbef9eca0daa06346fba4a010142f5c1027f80dbb +size 53136 diff --git a/marked/F/T-REC-F.747.13-202309-I_PDF-E/e9314c83043183351ed74908e9bf2f90_img.jpg b/marked/F/T-REC-F.747.13-202309-I_PDF-E/e9314c83043183351ed74908e9bf2f90_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..575b01654265ea1f892ef0b1edc6e5e4092277d9 --- /dev/null +++ b/marked/F/T-REC-F.747.13-202309-I_PDF-E/e9314c83043183351ed74908e9bf2f90_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3247480dadf182396c0cd56a2a5bb11d0c8544e1dbece3f76fad3fd046e4f0c1 +size 70398 diff --git a/marked/F/T-REC-F.747.16-202503-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/F/T-REC-F.747.16-202503-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..25f8fbfaef317942ed3102c74b40a7d42f80608b --- /dev/null +++ b/marked/F/T-REC-F.747.16-202503-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:83cab7dcabfe939b223d97242d8523e461968d21f7c5432693c30a64f2d95521 +size 7050 diff --git a/marked/F/T-REC-F.747.16-202503-I_PDF-E/0cc86fe8fc37b0edc9581f2af9459a52_img.jpg b/marked/F/T-REC-F.747.16-202503-I_PDF-E/0cc86fe8fc37b0edc9581f2af9459a52_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..bcb8fe167b3f0d0e3953e49b7658c1740ccf0c8c --- /dev/null +++ b/marked/F/T-REC-F.747.16-202503-I_PDF-E/0cc86fe8fc37b0edc9581f2af9459a52_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:363b8e0664d1bb8c9802671d09aae81a273b1b3ee2018c3c8477f43f8e34544d +size 35788 diff --git a/marked/F/T-REC-F.747.16-202503-I_PDF-E/5860ad6bd2a2dd8d1ab12864b8f90f37_img.jpg b/marked/F/T-REC-F.747.16-202503-I_PDF-E/5860ad6bd2a2dd8d1ab12864b8f90f37_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..27f13114914af30a3d0197609f9b780a6992c3a1 --- /dev/null +++ b/marked/F/T-REC-F.747.16-202503-I_PDF-E/5860ad6bd2a2dd8d1ab12864b8f90f37_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e7dbd4fc0053c6bf0a8c81faa41bbfb8ac41641ae954b991d1c8876300cf43fb +size 99062 diff --git a/marked/F/T-REC-F.747.16-202503-I_PDF-E/624d9669faa18991d525fea5f0e03269_img.jpg b/marked/F/T-REC-F.747.16-202503-I_PDF-E/624d9669faa18991d525fea5f0e03269_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7b1bc2dcd2560b7cfe94b9e8b4c8b8df1ad77c6c --- /dev/null +++ b/marked/F/T-REC-F.747.16-202503-I_PDF-E/624d9669faa18991d525fea5f0e03269_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c6fbaee33b6600b27bcc8d78215f16f6204a474db33500c6eacb9762206b6962 +size 34051 diff --git a/marked/F/T-REC-F.747.16-202503-I_PDF-E/c222a9006d6d60a8d81e6ffbfc0e74ad_img.jpg b/marked/F/T-REC-F.747.16-202503-I_PDF-E/c222a9006d6d60a8d81e6ffbfc0e74ad_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..36db3ad249e1d58b79cefc3c79819a0fa4011bbd --- /dev/null +++ b/marked/F/T-REC-F.747.16-202503-I_PDF-E/c222a9006d6d60a8d81e6ffbfc0e74ad_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e4d2cf4b9ead0d92ac9bbed91ab29568dc75789094a9e9d2c28655feb594a8dc +size 53690 diff --git a/marked/F/T-REC-F.747.16-202503-I_PDF-E/dd330f8b8f6c16eae20c3a676b4eb804_img.jpg b/marked/F/T-REC-F.747.16-202503-I_PDF-E/dd330f8b8f6c16eae20c3a676b4eb804_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..e8af4558f13fd751eebb9ef9c0a5a83f5c3a5b97 --- /dev/null +++ b/marked/F/T-REC-F.747.16-202503-I_PDF-E/dd330f8b8f6c16eae20c3a676b4eb804_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:451d480a3cef86e38f1a562ac18850802ae8a3ce2aeb2e164746b230e6f92064 +size 39052 diff --git a/marked/F/T-REC-F.747.9-201703-I_PDF-E/10953d657a5f47fdc829a800419dd370_img.jpg b/marked/F/T-REC-F.747.9-201703-I_PDF-E/10953d657a5f47fdc829a800419dd370_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..74d3dbf271d06e87d2be2c1f7c1701b1ed1322e1 --- /dev/null +++ b/marked/F/T-REC-F.747.9-201703-I_PDF-E/10953d657a5f47fdc829a800419dd370_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:59dd2a2a2c7f696e417cb66c05778a57ef175d2355d24a6fc8e17731b106f5cc +size 16137 diff --git a/marked/F/T-REC-F.747.9-201703-I_PDF-E/16152cf1d84aea10848758f51a91ff6a_img.jpg b/marked/F/T-REC-F.747.9-201703-I_PDF-E/16152cf1d84aea10848758f51a91ff6a_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..273254f7ad647929fd978087010e4568fffe2c43 --- /dev/null +++ b/marked/F/T-REC-F.747.9-201703-I_PDF-E/16152cf1d84aea10848758f51a91ff6a_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9d1ee77cd888a7a347a97104fbc132e2ab2b0e7c64f75c22b1fc208df06ca7ff +size 33363 diff --git a/marked/F/T-REC-F.747.9-201703-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.747.9-201703-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9002ee68e2b9ebf63d27d39ba8c73bdafb0e6be6 --- /dev/null +++ b/marked/F/T-REC-F.747.9-201703-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8df701536c25ab8246d2868de3c20c8a4e25168bb52c5eb9078278ab1b5b4fd8 +size 6859 diff --git a/marked/F/T-REC-F.747.9-201703-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg b/marked/F/T-REC-F.747.9-201703-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..fe26e954fc92b3648c11dfa6ee5a2c6479b5d4d0 --- /dev/null +++ b/marked/F/T-REC-F.747.9-201703-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e390961c4e3340897c27a73777c9f82f142d45ea49184c92bc16c07a1be2dc20 +size 190508 diff --git a/marked/F/T-REC-F.747.9-201703-I_PDF-E/8fbdfc3d17fb1dae7b2d8f5a287fa9fc_img.jpg b/marked/F/T-REC-F.747.9-201703-I_PDF-E/8fbdfc3d17fb1dae7b2d8f5a287fa9fc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..567c9dbe849ac2861ab038626ab1c00dc4ec6242 --- /dev/null +++ b/marked/F/T-REC-F.747.9-201703-I_PDF-E/8fbdfc3d17fb1dae7b2d8f5a287fa9fc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ae6efb5e6d65dcb433737cc38d376e7835c6f7e0d1d081fbce895ac23b6ad96c +size 118061 diff --git a/marked/F/T-REC-F.747.9-201703-I_PDF-E/bb6d33498937738ff5dac8d91c9ebaad_img.jpg b/marked/F/T-REC-F.747.9-201703-I_PDF-E/bb6d33498937738ff5dac8d91c9ebaad_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..27bad515d22de4f9c5f27977c8389ff4d56699d2 --- /dev/null +++ b/marked/F/T-REC-F.747.9-201703-I_PDF-E/bb6d33498937738ff5dac8d91c9ebaad_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:794a6fec2e9a8755a61152d59dbf130b1cbb3b92ff9d583897db0973c45bbd9d +size 21034 diff --git a/marked/F/T-REC-F.747.9-201703-I_PDF-E/c092f712a80ce3310c5e29d0fa0e454a_img.jpg b/marked/F/T-REC-F.747.9-201703-I_PDF-E/c092f712a80ce3310c5e29d0fa0e454a_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d24564576cf9aef02c8a2c2e04e8d4dae579583d --- /dev/null +++ b/marked/F/T-REC-F.747.9-201703-I_PDF-E/c092f712a80ce3310c5e29d0fa0e454a_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5d0e5c9b1ef014d97faa1370fa6f65535524fd448d5827f6d61deb0c440c6210 +size 10441 diff --git a/marked/F/T-REC-F.747.9-201703-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg b/marked/F/T-REC-F.747.9-201703-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..671c56dd0fba6074fcf3079ca36b6304fa4d37a6 --- /dev/null +++ b/marked/F/T-REC-F.747.9-201703-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:343eacb402ed47e95a897cf7831e36bfc68331f3aadd495120f181b02fd1e7e8 +size 117869 diff --git a/marked/F/T-REC-F.747.9-201703-I_PDF-E/d5a837fa4f4675e5ee596003cf55985c_img.jpg b/marked/F/T-REC-F.747.9-201703-I_PDF-E/d5a837fa4f4675e5ee596003cf55985c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..94deaed4480180af70b2588422f6c2150fe3e99b --- /dev/null +++ b/marked/F/T-REC-F.747.9-201703-I_PDF-E/d5a837fa4f4675e5ee596003cf55985c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9bba05ee25849d2e9b0970fcf25f6ac34c97fcfd0c1bbb803b681e3707ac9b6d +size 25216 diff --git a/marked/F/T-REC-F.747.9-201703-I_PDF-E/raw.md b/marked/F/T-REC-F.747.9-201703-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..76db9288b64b93a705eee222260f3c6a4d62c33e --- /dev/null +++ b/marked/F/T-REC-F.747.9-201703-I_PDF-E/raw.md @@ -0,0 +1,498 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.747.9** + +(03/2017) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +# --- **Requirements and architecture for energy management services** + +Recommendation ITU-T F.747.9 + +**ITU-T** + +![ITU logo: A globe with a red lightning bolt, next to the text 'International Telecommunication Union'.](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +ITU logo: A globe with a red lightning bolt, next to the text 'International Telecommunication Union'. + +## ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +### **TELEGRAPH SERVICE** + +| | | +|-----------------------------------------------------------------|-------------| +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | + +### **MOBILE SERVICE** + +| | | +|--------------------------------------------------------|-------------| +| Mobile services and multdestination satellite services | F.110–F.159 | +|--------------------------------------------------------|-------------| + +### **TELEMATIC SERVICES** + +| | | +|-------------------------------------------|-------------| +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | + +## **MESSAGE HANDLING SERVICES** + +F.400–F.499 + +### **DIRECTORY SERVICES** + +F.500–F.549 + +### **DOCUMENT COMMUNICATION** + +| | | +|--------------------------------------|-------------| +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | + +### **DATA TRANSMISSION SERVICES** + +F.600–F.699 + +## **MULTIMEDIA SERVICES** + +**F.700–F.799** + +## **ISDN SERVICES** + +F.800–F.849 + +## **UNIVERSAL PERSONAL TELECOMMUNICATION** + +F.850–F.899 + +## **ACCESSIBILITY AND HUMAN FACTORS** + +F.900–F.999 + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.747.9 + +# Requirements and architecture for energy management services + +## Summary + +Recommendation ITU-T F.747.9 describes requirements, scenarios and functional architecture for user-side energy management service (EMS) where energy consumption equipment with heterogeneous metering and control capacities coexist. In user-side EMS systems, energy consumption data for both single device and device group need to be collected; these devices can be controlled by coarse-grained (on-off state) and fine-grained instructions. Furthermore, such operations are executed manually or automatically. The user-side EMS system makes energy-saving decisions based on multiple factors, such as the demands and optimization policies from users, pricing strategies from suppliers, government subsidies, device status. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.747.9 | 2017-03-01 | 16 | 11.1002/1000/13180 | + +## Keywords + +Heterogeneous devices, metering and control, user-side energy management. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2017 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 1 | +| 5 Conventions ..... | 2 | +| 6 Background..... | 2 | +| 7 General requirements for user-side EMS ..... | 3 | +| 7.1 Requirements for device management ..... | 3 | +| 7.2 Requirements for data acquisition ..... | 3 | +| 7.3 Requirements for policies management ..... | 3 | +| 7.4 Requirements for energy consumption analysis..... | 3 | +| 7.5 Requirements for energy-saving decisions implementation..... | 4 | +| 8 Functional architecture for user-side EMS ..... | 4 | +| 8.1 Supported energy-metering devices ..... | 5 | +| 8.2 Supported energy control device ..... | 6 | +| 9 Security considerations ..... | 7 | +| Annex A – Scenarios for user-side EMS ..... | 8 | +| A.1 User-side EMS for domestic users ..... | 8 | +| A.2 User-side EMS for public buildings ..... | 10 | +| A.3 User-side EMS for factories ..... | 11 | +| Bibliography..... | 12 | + + + +# Recommendation ITU-T F.747.9 + +# Requirements and architecture for energy management services + +## 1 Scope + +This Recommendation describes requirements, scenarios and functional architecture for user-side energy management service (EMS) where energy consumption equipment with heterogeneous metering and control capacities coexist. This Recommendation covers the following: + +- scenarios of user-side EMS; +- requirements of user-side EMS; +- functional architecture for user-side EMS. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T Y.4252] Recommendation ITU-T Y.4252/Y.2064 (2014), *Energy saving using smart objects in home networks*. + +[ITU-T Y.4409] Recommendation ITU-T Y.4409 (2015), *Requirements and architecture of the home energy management system and home network services*. + +# 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 home energy management system** [ITU-T Y.4409]: A computer system comprising a software platform providing basic support services and a set of applications providing the functionality needed for the effective operation of home equipment, such as home appliances and storage batteries, so as to assure adequate security of energy supply at minimum cost. + +**3.1.2 smart object** [ITU-T Y.4252]: An object which is aware of its characteristic, context and situation. It shares and processes information, such as its identity, current location, physical properties and the information it senses from its surroundings, while performing object-to-object communications. + +### 3.2 Terms defined in this Recommendation + +None. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-----|---------------------------| +| AMR | Automatic Meter Reading | +| EMS | Energy Management Service | + +| | | +|------|-------------------------------| +| HEMS | Home Energy Management System | +| IoT | Internet of Things | + +## 5 Conventions + +In this Recommendation: + +- The expression "**is required to**" indicates a requirement which must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The expression "**is recommended**" indicates a requirement which is recommended but not absolutely required. Thus, this requirement need not be present to claim conformance. + +# 6 Background + +There are growing concerns about energy efficiency and emissions reduction among people from all aspects of life, which raises the demand for energy management for customers. With the development of information technology (e.g., Internet of things (IoT) [ITU-T Y.4252]), various smart objects, including sensors and actuators, can interconnect with one another to monitor and change the status of energy consumption equipment, which provides methods for energy management. Thus, user-side energy management service (EMS) is used to help consumers improve their energy efficiency as well as reduce their energy costs. + +The user-side EMS provides raw data to users as well as processed information of energy consumption by collecting and monitoring energy consumption data from various equipment. It also makes energy-saving decisions based on multiple factors, such as user demands, supplier's pricing strategies, and optimization policies. Some other factors, such as government subsidies, device status and environmental conditions can also affect energy-saving decisions. For example, usage related information, such as the number of customers shopping in a mall, the occupancy rate of a hotel, the number of students enrolled in a school, or attendance at a stadium, can affect energy-decisions, e.g., which lights and air conditioners need to be turned on/off. + +In recent years, various EMS systems have been developed and standardized. [ITU-T Y.4252] describes requirements and capabilities for saving energy by using smart objects in home networks. The home energy management system (HEMS) [ITU-T Y.4409] supports energy efficiency and reduction of energy consumption by monitoring and controlling devices. All these existing standards consider only smart objects which can be monitored and controlled automatically and remotely. + +However, not all existing energy-consumption devices are smart objects. They are not only different in the type of consumed energy (e.g., electricity, gas, coal), but also different in the capacity of metering and control. For example, in practice, an electricity meter can measure the amount of electric energy consumed by one circuit, which may be wired with multiple electrical equipment located in a home or in a factory. It can also be wired with individual electrical equipment. Thus, both coarse-grained and fine-grained data of consumed electric energy are collected manually or through automatic meter reading (AMR). Also, many electrical devices can only be turned on and off manually, e.g., a light bulb with on-off switches. Some other electrical devices can be fine-tuned, e.g., temperature setting of an air conditioner. Furthermore, smart appliances can be controlled automatically and remotely. + +Considering the cost of adding metering units for individual energy consumption equipment and the difficulties of updating all traditional devices to smart objects, various heterogeneous devices with different capacities of metering and control must be supported in a practical user-side EMS system. In such an EMS system, energy consumption data for both single device and device group need to be collected, and these devices can be controlled by coarse-grained (on-off state) and fine-grained instructions. Furthermore, such operations are executed manually or automatically. + +This Recommendation describes the requirements and scenarios of user-side EMS as well as the functional architecture for user-side EMS with heterogeneous devices. + +# **7 General requirements for user-side EMS** + +Several basic requirements of user-side EMS are given, including device management, data acquisition, policies management, energy consumption analysis, and decisions implementation. + +### **7.1 Requirements for device management** + +DM-01: The user-side EMS is required to register and deregister energy-consumption devices, metering devices and controlling devices. + +DM-02: The user-side EMS is required to maintain attribute information of energy-consumption devices, meter devices and control devices. + +DM-03: The user-side EMS is required to associate the energy-consumption devices with the corresponding metering device(s) and control device(s). + +### **7.2 Requirements for data acquisition** + +DA-01: The user-side EMS is required to collect and store the energy consumption data for each energy consuming device or device group. + +DA-02: The user-side EMS is required to provide the raw data as well as pre-processed data for further energy consumption calculation, optimization, and device control. + +DA-03: The user-side EMS is required to support at least one kind of mechanism (i.e., manual or automatic) of metering data collection. + +DA-04: The user-side EMS is recommended to implement a metering adaptation layer which supports various data collection mechanisms, such as remote and manual meter reading. + +DA-05: The user-side EMS is recommended to include the consumption of electricity, natural gas, wood, coal, gasoline, diesel and other energy resources as energy consumption data, if applicable. + +DA-06: The user-side EMS is recommended to collect environmental data periodically. + +### **7.3 Requirements for policies management** + +PM-01: The user-side EMS is recommended to fetch and store energy consumption demands and optimization policies of users. + +PM-02: The user-side EMS is recommended to fetch and store the pricing policies of supply side. + +### **7.4 Requirements for energy consumption analysis** + +EA-01: The user-side EMS is required to process the collected energy consumption data and provide the statistical energy consumption report to the user and the supplier if required. + +EA-02: The user-side EMS is recommended to provide an energy efficiency evaluation. For domestic users and public buildings, this evaluation is based on the amount of energy used, the number of people, and the environmental conditions. For a factory, the evaluation takes the amount of production into consideration. + +EA-03: The user-side EMS is recommended to provide data analysis such as statistical analysis, energy consumption prediction. + +EA-04: The user-side EMS is recommended to generate an optimal energy-saving solution based on all the data and information collected, the policies related, and the demands of user. + +EA-05: The user-side EMS is recommended to keep track of the change in energy consumption for one device or device group after applying each control command. + +### **7.5 Requirements for energy-saving decisions implementation** + +ED-01: The user-side EMS is required to have the capability for users to implement energy-saving decisions manually. + +ED-02: The user-side EMS is recommended to implement the energy-saving decisions automatically. + +ED-03: The user-side EMS is recommended to implement controlling adaptation layer which supports various control mechanisms, such as remote and manual control. + +## **8 Functional architecture for user-side EMS** + +The functional architecture of user-side EMS is shown in Figure 8-1. + +As shown in Figure 8-1, the user-side EMS interacts with EMS clients, supply-side EMS and various energy-consumption devices. There are six modules of user-side EMS which interact with other systems and provide energy management for users. + +- 1) The device management module registers and deregisters various energy consumption equipment and other devices (i.e., meters and controllers). The relationships among the energy consumption equipment, metering and control devices are maintained. The attribute information of energy consumption equipment, metering devices, and control devices is also maintained. +- 2) The metering data management module records real-time energy consumption data for each energy consuming device or device group. The consumed data of all types of energies, including gas, electricity, coal and other energy resources, are collected and submitted to the metering data management module manually or through AMR. +- 3) The control management module generates control instructions which are executed manually or automatically. +- 4) The strategy module interacts with EMS clients and maintains user-side strategies. The EMS client collects user demands, including the energy consumption demand, efficiency goals, and energy-saving strategies, and sends them to this module. +- 5) The supply-side price module keeps a record of the pricing policies and energy plan provided by the supply-side EMS. +- 6) The energy consumption data processing and optimization module performs the energy consumption statistics and makes energy-saving decisions. The statistical results are sent to supply-side EMS which can be used to help the energy supplier make energy supply plan and pricing policies. The results also are sent to the EMS client for displaying. + +![Figure 8-1: A framework of user-side EMS. The diagram shows the interaction between Supply-side EMS, User-side EMS, and EMS client. The User-side EMS contains a central 'Energy consumption processing and optimization' block connected to 'Supply-side price', 'Strategy', 'Metering data management', 'Device management', and 'Control management'. The 'Metering data management' and 'Control management' blocks connect to 'Metering adaption sublayer' and 'Controlling adaption sublayer' respectively. These sublayers connect to 'Metre' and 'Controller' blocks, which in turn connect to 'Device(groups)'. The 'EMS client' contains 'User interaction' and 'Display' blocks. The 'Supply-side EMS' contains 'Price management and energy plan' and 'Energy consumption statistics' blocks. Arrows indicate data flow: 'Pricing' from Supply-side EMS to User-side EMS; 'Statistics' from User-side EMS to Supply-side EMS; 'Pricing input' from Supply-side price to Energy consumption processing and optimization; 'Strategies input' from Strategy to Energy consumption processing and optimization; 'User demands' from User interaction to Energy consumption processing and optimization; 'Statistics' from Energy consumption processing and optimization to Display; 'Energy consumption data' from Metering adaption sublayer to Energy consumption processing and optimization; 'Device registration' from Device management to Metering adaption sublayer; 'Controlling' from Controlling adaption sublayer to Energy consumption processing and optimization. The 'Metering adaption sublayer' has 'Manual reading' and 'Automatic reading' paths to 'Metre' blocks. The 'Controlling adaption sublayer' has 'Manual control' and 'Automatic control' paths to 'Controller' blocks. The 'Energy consumption equipment' label is at the bottom.](cfef993dcc8fb513de79eb1f93cf26ae_img.jpg) + +Figure 8-1: A framework of user-side EMS. The diagram shows the interaction between Supply-side EMS, User-side EMS, and EMS client. The User-side EMS contains a central 'Energy consumption processing and optimization' block connected to 'Supply-side price', 'Strategy', 'Metering data management', 'Device management', and 'Control management'. The 'Metering data management' and 'Control management' blocks connect to 'Metering adaption sublayer' and 'Controlling adaption sublayer' respectively. These sublayers connect to 'Metre' and 'Controller' blocks, which in turn connect to 'Device(groups)'. The 'EMS client' contains 'User interaction' and 'Display' blocks. The 'Supply-side EMS' contains 'Price management and energy plan' and 'Energy consumption statistics' blocks. Arrows indicate data flow: 'Pricing' from Supply-side EMS to User-side EMS; 'Statistics' from User-side EMS to Supply-side EMS; 'Pricing input' from Supply-side price to Energy consumption processing and optimization; 'Strategies input' from Strategy to Energy consumption processing and optimization; 'User demands' from User interaction to Energy consumption processing and optimization; 'Statistics' from Energy consumption processing and optimization to Display; 'Energy consumption data' from Metering adaption sublayer to Energy consumption processing and optimization; 'Device registration' from Device management to Metering adaption sublayer; 'Controlling' from Controlling adaption sublayer to Energy consumption processing and optimization. The 'Metering adaption sublayer' has 'Manual reading' and 'Automatic reading' paths to 'Metre' blocks. The 'Controlling adaption sublayer' has 'Manual control' and 'Automatic control' paths to 'Controller' blocks. The 'Energy consumption equipment' label is at the bottom. + +**Figure 8-1 – A framework of user-side EMS** + +In clauses 8.1 and 8.2, an outline of supported metering and control devices is described. + +### 8.1 Supported energy-metering devices + +The acquisition of energy consumption data provides the input data for user-side EMS. Two steps are involved: energy consumption metering and data collection. + +Metering devices can be classified based on the number of monitored devices and the method of metering data collection. + +According to the number of monitored devices, there are two types of metering devices: + +- 1) Meters which measure the energy consumption data of one device. For example, with the emergence of sockets with metering, the consumed power of a specific device can be metered independently. +- 2) Meters which measure the energy consumption data of one device group (a set of devices). For example, one conventional electricity meter can only meter the total power consumption for one ordinary house, which means that all the household electrical appliances are wired with the same circuit and considered as a whole. + +According to the method of metering data collection, there are two types of metering devices: + +- 1) Meters whose data must be collected manually. +- 2) Meters which support AMR. + +Figure 8-2 shows a conventional electricity meter, a conventional gas meter and a socket with metering, the measurement data of which is collected manually. + +![Figure 8-2: Three images showing traditional metering devices: an ordinary electricity meter, an ordinary gas meter, and a smart socket with metering capabilities.](d5a837fa4f4675e5ee596003cf55985c_img.jpg) + +Figure 8-2: Three images showing traditional metering devices: an ordinary electricity meter, an ordinary gas meter, and a smart socket with metering capabilities. + +**Figure 8-2 – Ordinary electricity meter, ordinary gas meter and socket with metering** + +Two meters supporting remote meter reading are shown in Figure 8-3. + +![Figure 8-3: Two images showing modern metering devices that support AMR: a single-phase electricity meter and a gas meter.](10953d657a5f47fdc829a800419dd370_img.jpg) + +Figure 8-3: Two images showing modern metering devices that support AMR: a single-phase electricity meter and a gas meter. + +**Figure 8-3 – Electricity meter and gas meter supporting AMR** + +The shortcomings of collecting data manually include the high cost of human resources and inefficiency of data collection. However, updating all meters to support AMR is very expensive. In a home, there are various household energy-consumption devices such as light bulbs, chargers, air conditioners/fans, refrigerator, electric/gas water heater, and gas stove. It is unnecessary to add meters for each device. Thus, in practice, one metering device is often used to measure the energy consumption data of one device group and most metering devices are still needed to collect data manually. + +For user-side EMS, the device diversity in metering capacity and data collection method must be fully taken into account. + +### **8.2 Supported energy control device** + +An energy control device can prevent or change the transmission of energy and thus change the status of one or multiple energy-consumption devices. + +Energy control devices can be classified based on the number of controlled devices, the method of control and the granularity of control. + +According to the number of controlled devices, there are two types of energy control devices. + +- 1) Energy control device which controls one device. For example, one power switch controls one light bulb, i.e., one-to-one mode. +- 2) Energy control device which controls one set of devices. For example, one power switch controls one set of light bulbs, i.e., one-to-many mode. + +According to the method of control, there are two types of energy control devices. + +- 1) Energy control devices which are manually operated. For example, one conventional power switch and one gas switch shown in Figure 8-4. +- 2) Energy control devices which can be operated automatically and remotely. For example, a user controls the on-off state of the smart sockets and smart appliances remotely as shown in Figure 8-5. + +According to the granularity of control, there are two types of energy control devices: + +- 1) Energy control devices which control the "on" and "off" states of the energy-consumption equipment. +- 2) Energy control devices which fine-tune the consumed energy of energy-consumption equipment. For example, adjusting the temperature setting of a room changes the consumed energy of the air conditioner in that room. + +![Conventional power switch and gas switch](bb6d33498937738ff5dac8d91c9ebaad_img.jpg) + +The image shows two conventional energy control devices. On the left is a white NAIBO NBDZ47-63 C32 air circuit breaker with blue handles and technical specifications printed on it. On the right is a brass gas valve with a yellow handle. + +Conventional power switch and gas switch + +**Figure 8-4 – Conventional power switch and gas switch** + +![Smart socket and gas meter](c092f712a80ce3310c5e29d0fa0e454a_img.jpg) + +The image shows two smart energy control devices. On the left is a white smart wall outlet with a single hole and a small indicator light. On the right is a grey gas meter with a digital display and a blue button, labeled 'GAS METER'. + +Smart socket and gas meter + +**Figure 8-5 – Smart socket and gas switch gas meter supporting remote control** + +## **9 Security considerations** + +It is recommended that the security requirements of [b-ITU-T Y.2201], [b-ITU-T Y.2701], and applicable ITU-T X-series, ITU-T Y-series and ITU-T M-series security Recommendations be taken into consideration, which includes access control, authentication, data confidentiality, communications security, data integrity, availability and privacy. + +## Annex A + +## Scenarios for user-side EMS + +(This annex forms an integral part of this Recommendation.) + +In the domestic usage case, the user-side EMS is used to serve individual families or houses which are the basic units of human society. User-side EMS can not only save energy but also enhance public awareness of energy savings. + +In the case of user-side EMS for public buildings, the application area of user-side EMS is expanded, from inside a building or house for a family, to an architectural complex. + +In the case of user-side EMS for a factory, various types of energy resources, such as electricity, gas, coal and heat, are involved. + +There exist great opportunities to achieve greater energy efficiency in these scenarios. + +The study of other scenarios is encouraged and welcomed. + +### A.1 User-side EMS for domestic users + +The scenario of user-side EMS for a domestic user is shown in Figure A.1. + +![Diagram illustrating the scenario of an Energy Management System (EMS) for a domestic user. The central component is the User-side EMS, which interacts with an EMS client (displaying a computer monitor) and receives Pricing and Statistics from the Supply-side EMS. The User-side EMS is connected to various household devices via dashed lines: an Electrical range hood and Gas cooker in the kitchen; a Lighting system and Household appliances in the living room; and an Electric bath heater, Electric water meter, and Washing machine in the bathroom. These devices are further connected to various meters and sockets: Conventional electric meter, Conventional gas meter, Smart electric meter, Conventional electric meter, Socket with meter, Conventional power meter, Conventional gas switch, Smart socket, and Conventional socket with switch. The diagram also includes three photographs of a home interior showing the kitchen, living room, and bathroom.](33ed1f9b27c7c21c797aa928b0f06851_img.jpg) + +The diagram illustrates the scenario of an Energy Management System (EMS) for a domestic user. At the center is the **User-side EMS**, represented by a black tower icon. It has bidirectional connections to an **EMS client** (a desktop computer) labeled **Interaction**, and to the **Supply-side EMS** (represented by a power line tower) labeled **Pricing** and **Statistics**. The User-side EMS is connected via dashed lines to various household components. In the kitchen, it connects to an **Electrical range hood** and a **Gas cooker**. In the living room, it connects to a **Lighting system** and **Household appliances**. In the bathroom, it connects to an **Electric bath heater**, an **Electric water meter**, and a **Washing machine**. These components are further linked to various meters and sockets: **Conventional electric meter**, **Conventional gas meter**, **Smart electric meter**, **Conventional electric meter**, **Socket with meter**, **Conventional power meter**, **Conventional gas switch**, **Smart socket**, and **Conventional socket with switch**. The diagram includes three photographs of a home interior: a kitchen with a gas cooker and range hood, a living room with a TV and lighting system, and a bathroom with a washing machine and electric bath heater. The label **F.747.9(17)\_FA.1** is at the bottom right. + +Diagram illustrating the scenario of an Energy Management System (EMS) for a domestic user. The central component is the User-side EMS, which interacts with an EMS client (displaying a computer monitor) and receives Pricing and Statistics from the Supply-side EMS. The User-side EMS is connected to various household devices via dashed lines: an Electrical range hood and Gas cooker in the kitchen; a Lighting system and Household appliances in the living room; and an Electric bath heater, Electric water meter, and Washing machine in the bathroom. These devices are further connected to various meters and sockets: Conventional electric meter, Conventional gas meter, Smart electric meter, Conventional electric meter, Socket with meter, Conventional power meter, Conventional gas switch, Smart socket, and Conventional socket with switch. The diagram also includes three photographs of a home interior showing the kitchen, living room, and bathroom. + +**Figure A.1 – Scenario of EMS for domestic user** + +In a home, there are two common types of energy consumption equipment, namely electric equipment and gas equipment. The electric equipment includes lights, decorative lamps, television, washing machine, refrigerator, air conditioner, electric water heater, electric stove, kettle and many others. The typical gas equipment includes gas water heater, gas stove and gas heater. + +The user-side EMS collects and processes energy consumption data of these devices to provide raw data as well as processed information of energy consumption to home users. The EMS system helps users to understand how their own behaviours affect the usage of energy. + +The EMS also provides users with guidelines or suggestions about how to improve energy efficiency as well as save energy. The pricing policies of supply side (e.g., the power companies and gas companies) can affect these decisions. For instance, in order to reduce energy demand during peak hours, power companies usually introduce a step-tariff, i.e., charging higher prices at peak hours and lower prices at off-peak hours to regulate energy consumption. Thus, if a washing machine has a timer or can be controlled remotely, the EMS may suggest doing laundry at off-peak hours. The user does this by setting the timer manually or the EMS starting the washing machine remotely at the right time. + +Another typical scenario is related to the energy-saving decision for bathing. For an electric water heater, in conjunction with a socket with metering, the consumed energy required for a bath can be metered and submitted to the EMS manually. Similarly, the consumed gas required for a bath can also be collected. Then, the EMS can calculate the fee paid for one bath based on the consumed gas or electric power and recommend appropriate money-saving decisions to the user. + +### A.2 User-side EMS for public buildings + +A public building is a general term for civil architecture which covers all of the buildings used for social activities rather than for an individual family. It mainly includes large buildings (e.g., office premises, shopping malls, hotels, hospitals), transportation hubs (e.g., airports, bus stations) and sports buildings (e.g., public gymnasiums, stadiums). As shown in Figure A.2, the application area of the user-side EMS for public buildings is expanding from inside a building or house for a family as the EMS for domestic users, to an architectural complex. + +![Figure A.2: Application area of the user-side EMS for public buildings. The diagram shows three stages of expansion: 'Inside of a building' (a floor plan), 'The whole of a building' (a 3D model of a single building), and 'The architectural complex' (a 3D model of multiple buildings). Arrows indicate the progression from left to right. A small label 'F.747.9(17)_FA.2' is visible below the third stage.](16152cf1d84aea10848758f51a91ff6a_img.jpg) + +The diagram illustrates the expansion of the user-side EMS application area for public buildings. It consists of three stages connected by arrows: 1. 'Inside of a building': A 2D floor plan showing various rooms. 2. 'The whole of a building': A 3D perspective view of a single building. 3. 'The architectural complex': A 3D perspective view of multiple buildings of different heights and shapes, situated on a green base. A small label 'F.747.9(17)\_FA.2' is located below the third stage. + +Figure A.2: Application area of the user-side EMS for public buildings. The diagram shows three stages of expansion: 'Inside of a building' (a floor plan), 'The whole of a building' (a 3D model of a single building), and 'The architectural complex' (a 3D model of multiple buildings). Arrows indicate the progression from left to right. A small label 'F.747.9(17)\_FA.2' is visible below the third stage. + +**Figure A.2 – Application area of the user-side EMS for public buildings** + +Figure A.3 shows one scenario of the user-side EMS for public buildings. In order to create a comfortable environment, public buildings are generally equipped with adequate lighting and air-conditioning facilities. For example, in northern China, a central heating system is often employed in winter. Typically, this central heating system is charged by the size of the building regardless of the amount of actual consumed energy. The Ministry of Housing and Urban-Rural Development of China is now vigorously promoting the mandatory installation of meters for heating in new constructions, and also is forcing the installation of meters in existing buildings, meaning people need to pay according to the actual amount of consumed heat. Thus, reducing energy consumption and the cost will become an important task for public building management. + +The consumed energy for a building is always varying. For example, after normal working hours, some people may still remain in the building to work. In this case, the building manager can continue using the central heating system or using air conditioners in the appropriate rooms in the building. The building manager can manually count which rooms are still in use and offer this information to the EMS which will make decisions to save energy. + +![Figure A.3: Scenario of the user-side EMS for public buildings. The diagram shows a central 'User-side EMS' unit connected to various building systems and external entities. At the top, 'Supply-side EMS' (represented by a power line image) connects via 'Pricing' and 'Statistics' arrows. To the right, an 'EMS client' (represented by a computer monitor) connects via 'Interaction' arrows. Below the central unit, a dashed line connects to a 'Split air conditioning power supply system' (represented by an AC unit image), which in turn connects to a 'Conventional electric meter'. Another dashed line connects to a 'Lighting system' (represented by a light fixture image), which connects to another 'Conventional electric meter'. A solid arrow points from the central unit to an 'Office electricity' source (represented by a wall outlet image), which connects to a 'Conventional socket with switch'. A dashed line connects to a 'Centre air conditioning power supply system' (represented by a large AC unit image), which connects to a 'Smart electric meter'. Another dashed line connects to a 'Centre air conditioning water supply system' (represented by a water meter image), which connects to a 'Conventional water meter'. A dashed line connects to a 'Central heating system' (represented by a boiler image), which connects to a 'Smart heat meter' and a 'Smart valve'. The central unit is also connected to a central image of a building interior with arrows pointing to 'Office electricity' and 'Lighting system'. A label 'F.747.9(17)_FA.3' is in the bottom right corner.](8fbdfc3d17fb1dae7b2d8f5a287fa9fc_img.jpg) + +Figure A.3: Scenario of the user-side EMS for public buildings. The diagram shows a central 'User-side EMS' unit connected to various building systems and external entities. At the top, 'Supply-side EMS' (represented by a power line image) connects via 'Pricing' and 'Statistics' arrows. To the right, an 'EMS client' (represented by a computer monitor) connects via 'Interaction' arrows. Below the central unit, a dashed line connects to a 'Split air conditioning power supply system' (represented by an AC unit image), which in turn connects to a 'Conventional electric meter'. Another dashed line connects to a 'Lighting system' (represented by a light fixture image), which connects to another 'Conventional electric meter'. A solid arrow points from the central unit to an 'Office electricity' source (represented by a wall outlet image), which connects to a 'Conventional socket with switch'. A dashed line connects to a 'Centre air conditioning power supply system' (represented by a large AC unit image), which connects to a 'Smart electric meter'. Another dashed line connects to a 'Centre air conditioning water supply system' (represented by a water meter image), which connects to a 'Conventional water meter'. A dashed line connects to a 'Central heating system' (represented by a boiler image), which connects to a 'Smart heat meter' and a 'Smart valve'. The central unit is also connected to a central image of a building interior with arrows pointing to 'Office electricity' and 'Lighting system'. A label 'F.747.9(17)\_FA.3' is in the bottom right corner. + +**Figure A.3 – Scenario of the user-side EMS for public buildings** + +### A.3 User-side EMS for factories + +A large portion of energy is consumed by factories such as steel, mining, petroleum, chemicals, metallurgy, tobacco, food processing, and textile manufacturing. + +Due to a lack of technologies and money, the energy consumption information of many factories, especially small and medium-sized enterprises, is still collected and managed manually. As for many large-scale and energy-intensive industries, the user-side EMS is still in its early stage, + +The EMS for a factory manages multiple energy resources. For example, in a coal mine, there are many devices (e.g., central pump, winches, shearers, coal feeders, belt conveyors and air compressors) which are located at various locations and are powered by different resources (e.g., coal, electricity and gas). Diversified data must be collected. + +Further, the energy-saving decisions are influenced by policies such as government subsidies or other incentive mechanisms, and mandatory regulations. The security of the EMS related data and information is also critical for factory owners. + +The EMS is helpful for finding the composition of energy use and predicting the trend of energy consumption. Furthermore, the accurate data collected by the EMS can be used by energy suppliers such as power companies to improve the efficiency, reliability, and sustainability of the production and distribution of electricity. It can also be used by regulators to make better energy related policies. + +## Bibliography + +- [b-ITU-T Y.2201] Recommendation ITU-T Y.2201 (2009), *Requirements and capabilities for ITU-T NGN*. +- [b-ITU-T Y.2701] Recommendation ITU-T Y.2701 (2007), *Security requirements for NGN release 1*. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.11-202008-I_PDF-E/5a24ac755b962fd5f0183f13de0726de_img.jpg b/marked/F/T-REC-F.748.11-202008-I_PDF-E/5a24ac755b962fd5f0183f13de0726de_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3857bfa3ecf733396c3c1e578088301835a53af5 --- /dev/null +++ b/marked/F/T-REC-F.748.11-202008-I_PDF-E/5a24ac755b962fd5f0183f13de0726de_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6fff3632a2666dcb5e8a80416e8b187e3f51649bd98595fda331d78216094459 +size 28011 diff --git a/marked/F/T-REC-F.748.11-202008-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg b/marked/F/T-REC-F.748.11-202008-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..529e4f72f14cdf7f41f56cfbfb26655bf8181c59 --- /dev/null +++ b/marked/F/T-REC-F.748.11-202008-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:abb9e3ec6dfc8364cedc39a68cf796e9d3ff5529c79a1811eedeb12c6d6e7583 +size 3787 diff --git a/marked/F/T-REC-F.748.11-202008-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg b/marked/F/T-REC-F.748.11-202008-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..27b5e2bccab2ae77a3087b717389276b363b3409 --- /dev/null +++ b/marked/F/T-REC-F.748.11-202008-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7bbdd7c816290d05da024a9d498ed3e004c2ec887b7cd21e2cc7bd23adca3e10 +size 35415 diff --git a/marked/F/T-REC-F.748.11-202008-I_PDF-E/e6df2733626a85205c1db682e6259c46_img.jpg b/marked/F/T-REC-F.748.11-202008-I_PDF-E/e6df2733626a85205c1db682e6259c46_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..5f76ed99f77972b9de0d849b857128e754b2652f --- /dev/null +++ b/marked/F/T-REC-F.748.11-202008-I_PDF-E/e6df2733626a85205c1db682e6259c46_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:47ca7bd711f9e932a12c6a9c8b763d1d0df6aa32d8572beb566bb2721ae8d8ed +size 20939 diff --git a/marked/F/T-REC-F.748.12-202106-I_PDF-E/raw.md b/marked/F/T-REC-F.748.12-202106-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..b8fd543383716ad04dda6fd5f6a9851902259e56 --- /dev/null +++ b/marked/F/T-REC-F.748.12-202106-I_PDF-E/raw.md @@ -0,0 +1,507 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.748.12** + +(06/2021) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Deep learning software framework evaluation +methodology** + +Recommendation ITU-T F.748.12 + +ITU-T + +![ITU logo](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with a white grid pattern, overlaid by a stylized 'ITU' text in blue. + +ITU logo + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.12 + +# Deep learning software framework evaluation methodology + +## Summary + +A deep learning software framework provides an easy and fast way for manufactures to develop their own artificial intelligence (AI) applications. However, different frameworks show different performances under different scenarios. Recommendation ITU-T F.748.12 helps to evaluate deep learning software frameworks to help manufactures take full advantage of certain frameworks and avoid the disadvantages of others. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.748.12 | 2021-06-13 | 16 | 11.1002/1000/14681 | + +## Keywords + +Deep learning software framework, ease of use, ecological construction, evaluation, performance, security and stability, supported architecture, underlying optimization. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2021 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|--------------------------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 1 | +| 5 Conventions ..... | 2 | +| 6 Industrial realization of deep learning software framework in artificial intelligence .. | 2 | +| 6.1 Application architecture based on deep learning software framework ..... | 2 | +| 6.2 Hardware ..... | 3 | +| 6.3 Compilers ..... | 3 | +| 6.4 Frameworks ..... | 3 | +| 6.5 Fundamental applications..... | 3 | +| 6.6 Industry applications ..... | 3 | +| 7 Indicating requirements and evaluation methods of deep learning software framework..... | 3 | +| 7.1 Requirements overview ..... | 3 | +| 7.2 Classic deep learning models ..... | 4 | +| 7.3 Specific indicating items and evaluation methods for training framework.... | 5 | +| 7.4 Specific indicating items and evaluation methods for inference framework..... | 7 | + + + +# Recommendation ITU-T F.748.12 + +## Deep learning software framework evaluation methodology + +# 1 Scope + +This Recommendation provides the evaluation methodology for deep learning software framework. It addresses the following subjects: + +- a) Ecological construction of deep learning software framework; +- b) Ease of use of deep learning software framework; +- c) Performance of deep learning software framework; +- d) The supported architecture of deep learning software framework; +- e) Underlying optimization of deep learning software framework; +- f) Security and stability of deep learning software framework. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation does not use any terms defined elsewhere. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 deep learning:** A representation learning method, used to model high-level abstractions in data through the use of model architectures, which are composed of multiple nonlinear transformations. + +**3.2.2 deep learning software framework:** A tool that uses a set of pre-built and optimized components to define a model to achieve the encapsulation of artificial intelligence algorithms, data calls, and the use of computing resources. + +**3.2.3 deep learning model:** A deep learning algorithm used to solve a specific task, usually refers to the computational graph structure information and parameter information used to represent the deep learning algorithm. + +**3.2.4 model compression:** A mechanism to reduce the size of the model through algorithms such as pruning, quantification, regularization, knowledge distillation, and conditional calculation. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +ASIC            Application-specific integrated circuit + +| | | +|------|--------------------------------| +| CPU | Central processing unit | +| FPGA | Field-programmable gate array | +| GPU | Graphics processing unit | +| LLVM | Low level virtual machine | +| NNEF | Neural network exchange format | +| ONNX | Open neural network exchange | +| SGD | Stochastic gradient descent | + +# 5 Conventions + +The following conventions are used in this Recommendation: + +- The keywords "is required to" indicate a requirement that must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The keywords "is recommended" indicate a requirement that is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. + +# 6 Industrial realization of deep learning software framework in artificial intelligence + +## 6.1 Application architecture based on deep learning software framework + +Figure 6-1 illustrates the architecture of applications based on the deep learning software framework. + +![Figure 6-1: Architecture of applications based on deep learning software framework. The diagram shows five layers: Industry applications (Product, Application, Platforms and services), Fundamental applications (Intelligent speech, Natural language processing, Computer vision, Etc.), Frameworks (Training: TensorFlow, MXNet, Caffe, PyTorch, CNTK, Paddle Paddle; Inferencing: TensorFlow lite, Caffe2go, CoreML, Paddle-lite, TensorRT), Compilers (LLVM CUDA, Intel inference engine, TVM, XLA, ONNX, NNEF, LLVM), and Hardware (Servers: CPU, GPU; Mobile terminals: ASIC, FPGA). Arrows indicate flow from Algorithms to Frameworks and from Data to Frameworks.](07b17a620c75522d53916a11e12d1bff_img.jpg) + +The diagram illustrates the architecture of applications based on the deep learning software framework, organized into five horizontal layers: + +- Industry applications:** The top layer, containing three categories: Product, Application, and Platforms and services. +- Fundamental applications:** The second layer, containing four categories: Intelligent speech, Natural language processing, Computer vision, and Etc. +- Frameworks:** The third layer, which is the central part of the architecture. It is divided into two sub-sections: + - Training:** Includes TensorFlow, MXNet, Caffe, PyTorch, CNTK, Paddle Paddle, and others. + - Inferencing:** Includes TensorFlow lite, Caffe2go, CoreML, Paddle-lite, TensorRT, and others. + This layer receives input from **Algorithms** (via a "Packaging" arrow) and **Data** (via a "Processing" arrow). +- Compilers:** The fourth layer, labeled "Intermediate representation". It includes LLVM CUDA, Intel inference engine, TVM, XLA, ONNX, NNEF, and LLVM. An arrow points from the Frameworks layer down to this layer. +- Hardware:** The bottom layer, divided into two categories: + - Servers:** Includes CPU and GPU. + - Mobile terminals:** Includes ASIC and FPGA. + An arrow points from the Compilers layer down to this layer. + +Figure 6-1: Architecture of applications based on deep learning software framework. The diagram shows five layers: Industry applications (Product, Application, Platforms and services), Fundamental applications (Intelligent speech, Natural language processing, Computer vision, Etc.), Frameworks (Training: TensorFlow, MXNet, Caffe, PyTorch, CNTK, Paddle Paddle; Inferencing: TensorFlow lite, Caffe2go, CoreML, Paddle-lite, TensorRT), Compilers (LLVM CUDA, Intel inference engine, TVM, XLA, ONNX, NNEF, LLVM), and Hardware (Servers: CPU, GPU; Mobile terminals: ASIC, FPGA). Arrows indicate flow from Algorithms to Frameworks and from Data to Frameworks. + +F.748.12(21)\_F6-1 + +Figure 6-1 – Architecture of applications based on deep learning software framework + +## 6.2 Hardware + +Hardware includes central processing unit (CPU), graphics processing unit (GPU), application-specific integrated circuit (ASIC), field-programmable gate array (FPGA), and other chips, as well as servers, mobiles and embedded terminal devices constructed by these chips. + +## 6.3 Compilers + +Existing intermediate representations, including low level virtual machine (LLVM) CUDA, Intel Inference Engine, TVM, XLA, open neural network exchange (ONNX), neural network exchange format (NNEF), etc. + +## 6.4 Frameworks + +The frameworks are used to shield the underlying hardware and implement the AI algorithms, including the training framework and inference framework. The training framework includes TensorFlow, PyTorch, Caffe, PaddlePaddle, MXNet, etc. The inference framework includes Tensor RT, TensorFlow Lite, Caffe2go, Paddle Mobile, etc. + +## 6.5 Fundamental applications + +The fundamental applications mainly indicate the deep learning technology for each application scenario, such scenarios include computer vision, natural language processing, intelligent speech, etc. + +## 6.6 Industry applications + +The industry applications mean applying deep learning technology to specific vertical industries. + +# 7 Indicating requirements and evaluation methods of deep learning software framework + +## 7.1 Requirements overview + +This Recommendation separately formulates related requirements for the training framework and inference framework of deep learning. For a training framework, it covers five aspects including ecological construction, ease of use, performance, supported architecture, security and stability. For inference framework, it covers four aspects including ease of use, performance, underlying optimization, security and stability. + +The training framework indicating system includes the elements listed in Table 7-1. + +**Table 7-1 – Training framework indicating system and specific indicator items** + +| Indicating system | Specific indicator item | +|-------------------------|----------------------------------------------------| +| Ecological construction | Interface | +| | Core developers and contributors | +| | The situation in which the issues are solved | +| Ease of use | Model building and conversion | +| | Secondary development based on high-level language | +| | Custom extension | +| | Cross-platform | +| | Model library support | +| | Tutorial, documentation, and training materials | +| | Dynamic graph and static graph | + +**Table 7-1 – Training framework indicating system and specific indicator items** + +| Indicating system | Specific indicator item | +|------------------------|-------------------------------------------| +| | Stability | +| | Debuggability | +| Performance | Model library operating behaviour | +| | Operating behaviour of a customized model | +| | Hardware acceleration support | +| Supported architecture | CPU/FPGA | +| | Single GPU/multi-GPU | +| | Distributed training | +| | Virtual environment support | +| | Operating system support | +| Security and stability | Usage of third-party library | +| | Data security | + +The inference framework indicating system includes the elements listed in Table 7-2. + +**Table 7-2 – Inference framework indicating system and specific indicator items** + +| Indicating system | Specific indicator item | +|-------------------------|-------------------------------------------| +| Ease of use | Model optimizing functionality | +| | Universal model representation | +| | Cross-platform | +| Performance | Inference speed | +| | Run-up time | +| | System resources occupation | +| | Energy consumption | +| Underlying optimization | Support for different underlying hardware | +| | Optimization for instruction set | +| Security and stability | Model encryption | + +## 7.2 Classic deep learning models + +Evaluation for a deep learning software framework needs some classic test examples. Table 7-3 lists some classic deep learning models in different test scenarios and corresponding metrics, which are recommended when doing the evaluation. + +**Table 7-3 – Different test scenarios and classic models with related dataset and quality objectives** + +| Test scenarios | Dataset | Quality objectives | Reference model | +|---------------------------------|-----------|-----------------------|--------------------------| +| Image classification | ImageNet | 74.90% classification | ResNet-50 v1.5 | +| Object detection (light-weight) | COCO 2017 | 21.2% mAP | SSD (ResNet-34 backbone) | + +**Table 7-3 – Different test scenarios and classic models with related dataset and quality objectives** + +| Test scenarios | Dataset | Quality objectives | Reference model | +|---------------------------------|--------------------|----------------------------------------|--------------------------------| +| Object detection (heavy-weight) | COCO 2017 | 0.377 Box min AP,
0.339 Mask min AP | Mask R-CNN | +| Pre-trained model | Wikipedia | Related to specific tasks | BERT | +| Translation (recurrent) | WMT English-German | 21.8 BLEU | Neural machine translation | +| Translation (non-recurrent) | WMT English-German | 25.0 BLEU | Transformer | +| Recommendation | MovieLens-20M | 0.635 HR@10 | Neural collaborative filtering | +| Reinforce learning | Pro games | 40.00% move prediction | Mini Go | + +## 7.3 Specific indicating items and evaluation methods for training framework + +### 7.3.1 Ecological construction + +#### 7.3.1.1 Interface + +Refers to the programming languages that the training framework supports. + +#### 7.3.1.2 Core developers and contributors + +Refers to the number of core developers and contributors, activeness of the framework, especially the number of watch, star, fork, pull request, core developers, and contributors in GitHub. + +#### 7.3.1.3 The situation in which issues are solved + +Refers to the mechanism about solving issues and the operating situation. + +### 7.3.2 Ease of use + +#### 7.3.2.1 Model building and conversion + +Refers to the consistency of application programming interface (API), model definition structure, model storage format, and model converters. + +#### 7.3.2.2 Secondary development based on high-level language + +Refers to high-level languages the training framework supports. Weighting the popularity of each language to calculate the average support for high-level languages of the training framework. + +#### 7.3.2.3 Custom extension + +Refers to support and easy-used tools to develop extensive functionality, such as custom op, custom network layer, and added lines of code to realize one extensive functionality. + +#### 7.3.2.4 Cross-platform + +Refers to different supported platforms of the framework, such as Linux, Windows, Android, iOS, cloud platform, and so on. + +#### 7.3.2.5 Model library support + +Refers to supported deep learning models, including CNN, RNN, and models that are used in industrial scenarios. The number of models and the covered types of scenes are required to be included. + +#### **7.3.2.6 Tutorial documentation and training materials** + +Tutorial documentation and training materials are required to be declared, including whether or not official documents, community documents and material from cooperative training institutions exist. If they exist, the number of documents, completeness, updates, and languages that are used to write the documents are also required to be included. + +#### **7.3.2.7 Dynamic graph and static graph** + +Refers to code execution mechanism of the training framework, such as a dynamic or a static graph. + +#### **7.3.2.8 Stability** + +Refers to the mechanism of improving the stability of the training framework. + +#### **7.3.2.9 Debuggability** + +Refers to the flexibility and effectiveness of debugging mechanisms. + +### **7.3.3 Performance** + +#### **7.3.3.1 Model library operating performance** + +Refers to the execution time and memory usage of running classic models from the official model library with different batch sizes. + +#### **7.3.3.2 Operating performance of a customized model** + +Refers to concurrency, stability, and scale of features. For concurrency, evaluations are recommended to be taken to get the specific calculating quantities executed simultaneously and the interactive efficiency of the framework. For stability, evaluations are recommended to be taken to get statistic data of the failures during a-week-time by continuously executing the tasks based on the training framework. For a scale of features, evaluations are recommended to be taken to get the biggest throughput ability of the training framework and level it with its ability. + +#### **7.3.3.3 Hardware acceleration support** + +Refers to the optimization of the training framework for different underlying hardware, such as CPU, GPU, FPGA, dedicated AI accelerators, and so on. + +### **7.3.4 Supported architecture** + +#### **7.3.4.1 CPU/FPGA** + +Refers to the performance of classic models executed on CPU or FPGA, including training time and memory usage. + +#### **7.3.4.2 Single GPU/multi-GPU** + +Refers to the synchronized stochastic gradient descent (SGD) performance of classic models and other sophisticated model optimization methods executed on a single GPU and multi-GPU. + +#### **7.3.4.3 Distributed training** + +Refers to synchronized SGD behaviour of classic models and other sophisticated model optimization methods executed on the same machine and different machines. + +#### **7.3.4.4 Virtual environment support** + +Refers to the applicability of the training framework to a virtual environment, virtual machine, and so on. + +#### **7.3.4.5 Operating systems support** + +Refers to the applicability of the training framework to multiple operating systems such as Linux, Windows, and macOS. + +### **7.3.5 Security and stability** + +#### **7.3.5.1 Usage of third-party library** + +Refers to the third-party library used in the training framework. + +#### **7.3.5.2 Data security** + +Refers to whether encrypted data is supported in training and whether an encryption tool is provided for the framework. + +## **7.4 Specific indicating items and evaluation methods for inference framework** + +### **7.4.1 Ease of use** + +#### **7.4.1.1 Model optimizing functionality** + +Refers to the ability of model compression, accelerate algorithm components and hyper-parameter optimization components. + +#### **7.4.1.2 Universal model representation** + +Refers to whether the inference framework supports the universal format of the deep learning model, such as ONNX. + +#### **7.4.1.3 Cross-platform** + +Refers to the compatibility of framework to different platforms such as Linux, Windows, Android, iOS, cloud platform, etc. + +### **7.4.2 Performance** + +#### **7.4.2.1 Inference speed** + +Refers to the inference time by executing the classic deep learning models using inference framework under different energy consumption. + +#### **7.4.2.2 Run-up speed** + +Refers to the run-up time by executing the classic deep learning models using inference framework based on different scales of data. + +#### **7.4.2.3 System resources occupation** + +Refers to storage consumption, CPU memory usage, and GPU memory usage by executing deep learning models of different scales. + +#### **7.4.2.4 Energy consumption** + +Refers to cost energy in one hour by executing deep learning models of different scales. + +### **7.4.3 Underlying optimization** + +#### **7.4.3.1 Support for different underlying hardware** + +Refers to the ability of inference framework supporting different underlying hardware such as FPGA, ASIC, GPU, and so on. + +#### **7.4.3.2 Optimization for instruction set** + +Refers to optimization of the supported instruction set of the inference framework, such as advanced RISC machine (ARM), INTEL instruction set. + +### **7.4.4 Security and stability** + +#### **7.4.4.1 Model encryption** + +Refers to the encryption strategy of the inference framework for the trained inference model. + + + +# SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.13-202106-I_PDF-E/raw.md b/marked/F/T-REC-F.748.13-202106-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..93ae52643d9a8d17753deaa076a66ac88c14a3f3 --- /dev/null +++ b/marked/F/T-REC-F.748.13-202106-I_PDF-E/raw.md @@ -0,0 +1,665 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.748.13** + +(06/2021) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Technical framework for a shared machine +learning system** + +Recommendation ITU-T F.748.13 + +ITU-T + +![ITU logo](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with white lines representing latitude and longitude, and a stylized 'ITU' text in blue. + +ITU logo + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multidestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +## Recommendation ITU-T F.748.13 + +# Technical framework for a shared machine learning system + +## Summary + +Recommendation ITU-T F.748.13 defines the roles, technical and security requirements of a shared machine learning system, and provides technical architectures, functional components, and processing procedures of the shared machine learning system in the centralized and decentralized modes. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.748.13 | 2021-06-13 | 16 | 11.1002/1000/14682 | + +## Keywords + +Encrypted data, multi-party computation, shared machine learning, trusted execution environment. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2021 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|------|------------------------------------------------------------------------------------------------------------------------------------------|----| +| 1 | Scope ..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 1 | +| 4 | Abbreviations and acronyms ..... | 1 | +| 5 | Conventions ..... | 2 | +| 6 | Overview of the shared machine learning system ..... | 2 | +| 7 | Roles in the shared machine learning system ..... | 3 | +| 7.1 | Overview ..... | 3 | +| 7.2 | Data provider ..... | 3 | +| 7.3 | Computation platform ..... | 3 | +| 7.4 | Result receiver ..... | 4 | +| 7.5 | Task initiator..... | 4 | +| 8 | Technical requirements of the shared machine learning system ..... | 4 | +| 8.1 | Basic functional requirements ..... | 4 | +| 8.2 | Scalability requirements ..... | 5 | +| 8.3 | Reliability requirements ..... | 5 | +| 8.4 | Compatibility requirements ..... | 6 | +| 8.5 | Performance requirements ..... | 6 | +| 8.6 | Usability requirements..... | 6 | +| 9 | Security requirements of the shared machine learning system..... | 6 | +| 9.1 | Authentication requirements ..... | 6 | +| 9.2 | Access control requirements..... | 7 | +| 9.3 | Security auditing requirements..... | 7 | +| 9.4 | Data security requirements ..... | 7 | +| 9.5 | Privacy protection requirements..... | 7 | +| 10 | Technical architecture, functional components, and processing procedure of the shared machine learning system in centralized mode..... | 8 | +| 10.1 | Technical architecture of the shared machine learning system in the centralized mode..... | 8 | +| 10.2 | Functional components of the shared machine learning system in the centralized mode..... | 9 | +| 10.3 | Processing procedures of the shared machine learning system in the centralized mode..... | 10 | +| 11 | Technical architecture, functional components, and processing procedure of the shared machine learning system in decentralized mode..... | 11 | +| 11.1 | Technical architecture of the shared machine learning system in the decentralized mode ..... | 11 | + +| | Page | +|--------------------------------------------------------------------------------------------------------|-------------| +| 11.2 Functional components of the shared machine learning system in the decentralized mode ..... | 11 | +| 11.3 Processing procedures of the shared machine learning system in the decentralized mode ..... | 12 | +| Appendix I – Use cases for shared machine learning systems ..... | 13 | +| I.1 Use case: Improving modules for recognizing telecom frauds using data from multiple networks ..... | 13 | +| I.2 Intelligent credit and risk control use case ..... | 14 | +| I.3 Intelligent marketing use case ..... | 14 | +| Bibliography..... | 15 | + +# Technical framework for a shared machine learning system + +# 1 Scope + +This Recommendation defines a shared machine learning (SML) system, and provides its roles, technical and security requirements, technical architectures, functional components, and processing procedures of the shared machine learning system in the centralized and decentralized modes. Use cases for shared machine learning systems are provided in the appendix as well. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 machine learning** [b-ITU-T Y.3172]: Processes that enable computational systems to understand data and gain knowledge from it without necessarily being explicitly programmed. + +**3.1.2 trusted execution environment** [b-ITU-T J.1201]: A secure area of the main processor in an IBB-capable cable STB and TV to ensure that sensitive data is stored, processed and protected in an isolated and trusted environment. It offers isolated safe execution of authorized security software providing end-to-end security by enforcement of protected execution of authenticated code, confidentiality, authenticity, privacy, system integrity and data access rights. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 shared machine learning**: A machine learning paradigm enabling aggregation of multi-party data information and multi-party data privacy protection for scenarios where multiple data providers and the computation platform do not trust each other. + +**3.2.2 multi-party computation**: A subfield of cryptography with the goal of creating methods for parties to jointly compute a function over their inputs while keeping those inputs private. + +**3.2.3 remote attestation**: A method by which a host (client) authenticates its hardware and software configuration to a remote host (server). The goal of remote attestation is to enable a remote system (challenger) to determine the level of trust in the integrity of the platform of another system (attestator). + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations: + +| | | +|-----|-----------------------------------| +| API | Application Programming Interface | +| QPS | Queries Per Second | +| SML | Shared Machine Learning | +| SMS | Short Messaging Service | +| TEE | Trusted Execution Environment | +| WOE | Weight of Evidence | + +# 5 Conventions + +The following conventions are used in this Recommendation: + +- The keywords "is required to" indicate a requirement that must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The keywords "is prohibited from" indicate a requirement that must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The keywords "is recommended" indicate a requirement that is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. +- The keywords "is not recommended" indicate a requirement that is not recommended but which is not specifically prohibited. Thus, conformance with this Recommendation can still be claimed even if this requirement is present. +- The keyword "functions" is defined as a collection of functionalities. + +# 6 Overview of the shared machine learning system + +With shared machine learning (SML) systems, multi-party participants share encrypted data or exchange parameters to protect data security and privacy. The encrypted data or exchanged parameters of each party are gathered to train a shared machine learning model to make the best use of data. Shared machine learning models keep training to achieve self-optimization and multi-party participants or others who have the authorization to access the model, can input information to get outputs or predictions based on shared values. Shared machine learning systems can be applied in media applications but are not limited to media-related applications. + +Shared machine learning includes centralized and decentralized modes. The centralized mode is a solution for multi-party data encryption sharing and fusion learning in a trusted execution environment (TEE). The decentralized mode is a solution for sharing and learning by multiple participants based on secure multi-party computing by exchanging non-original data that does not reveal privacy. + +For centralized mode, the data is collected and trained in a trusted third party. It can be applied to scenarios where participants would like to share data that does not involve privacy, have low service access costs, and also be applied to complex computing scenarios. It can also support all the algorithms, cluster deployment and centralized computing. + +For decentralized mode, it uses multi-party computation methods to exchange data or parameters to complete the training tasks. It can be applied to scenarios where participants need strict data privacy protection or in scenarios where participants have a large amount of local data since different parties' exchange parameters, but not real data when the machine learning model is training. Multi-party computation is suitable for simple calculations since it can support fewer algorithms compared to the trusted execution environment solution. + +# 7 Roles in the shared machine learning system + +## 7.1 Overview + +The shared machine learning system is described in Figure 1. + +![Diagram illustrating the roles in the shared machine learning system. The roles are Task initiator, Computation platform, Result receiver, Data provider A, and Data provider B. The Task initiator sends a Task request to the Computation platform. The Computation platform sends a Computation result to the Result receiver. The Computation platform interacts with Data provider A and Data provider B via Computation interaction. Data provider A and Data provider B exchange Data.](7a0db9703b68b3d06cdaeefc084c0006_img.jpg) + +``` +graph TD; TI[Task initiator] -- "Task request" --> CP[Computation platform]; CP -- "Computation result" --> RR[Result receiver]; CP <--> |"Computation interaction"| DP_A[Data provider A]; CP <--> |"Computation interaction"| DP_B[Data provider B]; DP_A <--> |"Data exchange"| DP_B; RR <--> |"Computation result"| DP_A; RR <--> |"Computation result"| DP_B; +``` + +Diagram illustrating the roles in the shared machine learning system. The roles are Task initiator, Computation platform, Result receiver, Data provider A, and Data provider B. The Task initiator sends a Task request to the Computation platform. The Computation platform sends a Computation result to the Result receiver. The Computation platform interacts with Data provider A and Data provider B via Computation interaction. Data provider A and Data provider B exchange Data. + +**Figure 1 – Roles in the shared machine learning system** + +In a shared machine learning system, there are several roles, including task initiator, data provider, computation platform, and result receiver. Note that Figure 1, only shows two data providers, but in the real deployment, it can have multiple data providers. A party can act as multiple roles. For example, one of the data providers can be a task initiator and / or result receiver. + +## 7.2 Data provider + +A data provider owns the data and provides the data as input to the computation platform or the other data provider. Note that the data providers also have computation capabilities. + +In the centralized mode, the data is encrypted and transmitted from the data provider to the computation platform. + +In the decentralized mode, the data is processed as secret or parameters and then exchanged between the data providers, under the coordination of the computation platform. + +In any mode, the exchanged data is in a form of a calculation factor, not the original data itself. And the computation platform or the data providers shall not infer any information about the original data from the calculation factors. + +## 7.3 Computation platform + +The computation platform receives the task request from the task initiator and sends the encrypted result to the result receiver. It interacts with the data providers to perform the computation tasks. + +In the centralized mode, the computation platform receives the encrypted data from the data providers, decrypts the encrypted data, and performs the computation in the trusted execution environment. + +In the decentralized mode, the computation platform divides the computation task and coordinates the data providers to exchange parameters or secrets between each other and perform computation tasks individually. + +The computation platform integrates computation algorithms. + +## **7.4 Result receiver** + +In the centralized mode, when a computation task is completed, the computation platform will send the encrypted result to the result receiver, which can decrypt and get the final result. + +In the decentralized mode, the data providers send the computation results to the result receiver, and the result receiver merges the received computation results and gets the final result. + +The result receiver can be run by the data providers, or the task initiator, or the computation platform. + +## **7.5 Task initiator** + +The task initiator initiates the computation task with the computation platform. + +The task initiator can be run by the computation platform or the data providers. + +# **8 Technical requirements of the shared machine learning system** + +## **8.1 Basic functional requirements** + +### **8.1.1 Data management functional requirements** + +The data management functional requirements for the shared machine learning system include: + +- The shared machine learning system is required to support the management and display of metadata information. +- The shared machine learning system is required to support data usage authorization function. +- The shared machine learning system is required to support data alignment function. +- The shared machine learning system is recommended to support data usage behaviour auditing. +- The shared machine learning system is recommended to support encrypted data storage. + +### **8.1.2 Algorithm management functional requirements** + +The algorithm management functional requirements for the shared machine learning system include: + +- The shared machine learning system is required to support at least one mainstream machine learning training and prediction algorithm that can achieve privacy protection, such as linear regression, logistic regression, tree model, deep neural network, graph neural network. +- The shared machine learning system is required to support the parameter adjustment function to obtain better training results. +- The shared machine learning system is required to support at least one data segmentation method of horizontal segmentation or vertical segmentation and is recommended to support both data segmentation methods at the same time. +- The shared machine learning system is required to support the model of resisting semi-honest attacks. +- The shared machine learning system is recommended to support the data privacy intersecting function to ensure that the samples can be aligned when the data is segmented vertically. +- The shared machine learning system is recommended to support secure data analysis functions, such as maximum and minimum statistics, mean variance statistics. + +- The shared machine learning system is recommended to support machine learning pre-processing and algorithm effect evaluation functions, including data set division, missing value filling, weight of evidence (WOE), feature correlation evaluation, and other capabilities to improve the performance of machine learning models. + +### **8.1.3 Computation management functional requirements** + +The computation management functional requirements for the shared machine learning system include: + +- The shared machine learning system is required to provide basic task management capabilities, such as task creation and cancellation. +- The shared machine learning system is recommended to provide rich task management capabilities, such as task progress monitoring, task queuing, historical task information tracking. +- The shared machine learning system is recommended to support distributed resource management and task scheduling capabilities. + +## **8.2 Scalability requirements** + +The scalability requirements for the shared machine learning system include: + +- The shared machine learning system is recommended to have good scalability and to allow new functional components to be added according to business needs. +- The shared machine learning system is recommended to support users to access the system through application programming interface (API). + +## **8.3 Reliability requirements** + +The reliability requirements for the shared machine learning system include: + +- The shared machine learning system is required to guarantee the system availability and to avoid being shut down by incorrect input data. +- The shared machine learning system is required to have the ability to perform automatic disaster recovery after failures (such as server failure, hard disk failure, network failure, shutdown, restart), including data backup and restore, etc. +- The shared machine learning system is recommended to support clustering of training and prediction and support cross-computer room disaster recovery, so that the service has failover and disaster tolerance capabilities, thereby improving the system availability. +- The shared machine learning system is recommended to provide operational and maintenance capabilities such as grayscale release, monitoring and alarm, and link tracking, to improve system reliability. + +## **8.4 Compatibility requirements** + +The compatibility requirements for the shared machine learning system include: + +- The shared machine learning system is required to be compatible with the other machine learning algorithms. +- The shared machine learning system is recommended to achieve backward compatibility during the upgrade process, including compatibility between system modules, and compatibility between systems and files. +- The shared machine learning system is recommended to support the deployment requirements of different environments, such as cloud environment, virtual machine, physical machine. +- The shared machine learning system is recommended to support normal operation on a variety of mainstream trusted execution environments for the trusted execution environment solution. + +## **8.5 Performance requirements** + +The performance requirements for the shared machine learning system include: + +- The shared machine learning system is recommended to support distributed training to increase the amount of data supported by system training and reduce training time. +- The shared machine learning system is recommended to have the capability to improve prediction queries per second (QPS) through horizontal expansion. + +## **8.6 Usability requirements** + +The usability requirements for the shared machine learning system include: + +- The shared machine learning system is required to provide complete deployment and operation instructions to facilitate users to understand, access, and use the system. +- The shared machine learning system is recommended to provide a platform interface to reduce the cost of user learning and usage. +- The shared machine learning system is recommended to provide an algorithm development framework so that users can develop custom algorithms that meet security constraints based on the framework. +- The shared machine learning system is recommended to adopt a system design to reduce the interruption to users during system upgrades. + +# **9 Security requirements of the shared machine learning system** + +## **9.1 Authentication requirements** + +The authentication requirements for the shared machine learning system include: + +- The shared machine learning system is required to support identity authentication functions for the users who access the shared machine learning system. Users include data providers, system users, and result receivers. +- The shared machine learning system is required to support certificate authentication for the data providers and result receivers. +- The shared machine learning system is recommended to support two or more combinations of technologies (password verification, mailbox verification, short messaging service (SMS) verification, etc.) to achieve user identity authentication for system users. + +## **9.2 Access control requirements** + +The access control requirements for the shared machine learning system include: + +- The shared machine learning system is required to support authorization mechanism for the data users to guarantee authorization of their data to participate in the shared machine learning. +- The shared machine learning system is recommended to support complete roles and authorization systems. The shared machine learning system is required to give explicit roles and access rights after the system user logs in the system. +- The shared machine learning system is recommended to require the user to re-authenticate or reactivate the session when the session is idle according to the training policy. + +## **9.3 Security auditing requirements** + +The security auditing requirements for the shared machine learning system include: + +- The shared machine learning system is required to support functions of log recording and log auditing for the data operations when users access the shared machine learning system. +- The shared machine learning system is recommended to support the history backtracking and auditing function for the main operations to the system. + +## **9.4 Data security requirements** + +The data security requirements for the shared machine learning system include: + +- The shared machine learning system is required to support the function of encrypted transmission and storage of important data. +- The processed data is required not to contain sensitive data required by relevant regulation. +- The shared machine learning system is required to support the limiting of transmission of data. +- The shared machine learning system is required to support secure transmission protocols or secure transmission channels to ensure the security and reliability of data transmission links and prevent being attacked. +- The shared machine learning system is required to ensure the confidentiality, integrity, and availability of data characteristics and samples, and the data is prohibited from being obtained illegally by unauthorized users. +- The shared machine learning system is required to destroy the trusted execution environment after completing the computation tasks in the centralized mode. + +## **9.5 Privacy protection requirements** + +The privacy protection requirements for the shared machine learning system include: + +- The shared machine learning system is required to ensure that data related to user privacy is not leaked to other data providers, coordinators, or users. +- The shared machine learning system is required to provide a mechanism to avoid being located or roughly located to an individual in any form from the information exchanged between the data providers. +- The shared machine learning system is required to allow data providers to use reasonable encryption methods in order to ensure that other parties other than the recipient can infer user privacy data from the encrypted interactive data to prevent possible attacks, such as brute force cracking, reasoning attacks, and so on. + +- In the secure multi-party computing solution, the shared machine learning system is required to ensure that the user privacy-related data does not go out from the local storage, and only random numbers or parameters are exchanged between different data providers. + +# 10 Technical architecture, functional components, and processing procedure of the shared machine learning system in centralized mode + +## 10.1 Technical architecture of the shared machine learning system in the centralized mode + +In a centralized mode of a shared machine learning system, local data from multi-party participants is processed, encrypted, and transmitted to a trusted execution environment to train the shared model. Since the data in each party is encrypted, the data cannot be seen by the platform or other participants. When multi-party participants or others have the authorization to access the shared model, they can get access to it by using application programming interfaces (APIs) to input information and to get outputs or predictions based on shared values. + +Before providing data, participants (data providers) verify the correctness and validity of the software running in the trusted execution environment through remote attestation. After that, an encryption mechanism is negotiated between the participants and the trusted execution environment. Then, participants encrypt data using negotiated mechanism and send the ciphertext to the trusted execution environment. The encryption mechanism can be a symmetric encryption mechanism or asymmetric encryption mechanism. + +To ensure the computed model will not be leaked, the trained model should be encrypted after the whole computation is completed. + +The technical architecture is described in Figure 2. + +![Figure 2 – Technical architecture of the shared machine learning system in the centralized mode. The diagram shows a central 'Computation platform' containing a 'TEE' (Trusted Execution Environment). The TEE includes an 'Encryption module', a 'Shared learning module', two 'Decryption modules', and two 'Authentication modules'. A 'Task initiator' sends a 'Task request' to the platform. Two 'Data provider' units, 'Data provider A' and 'Data provider B', are shown. Each provider contains a 'Data' store, a 'Process and encryption module', and an 'Authentication module'. Data providers send 'Encrypted data' to the TEE's decryption modules. The TEE's authentication modules send 'Encryption key' information to the decryption modules. The TEE's shared learning module processes the data. The TEE's encryption module sends an 'Encrypted model' to the 'Result receiver'. Data providers also receive 'Negotiated encryption mechanism software signature' from the TEE's authentication modules.](16c1175b5f05a4b55e6d396fc51b15b3_img.jpg) + +The diagram illustrates the technical architecture of the shared machine learning system in centralized mode. At the center is a 'Computation platform' containing a 'TEE' (Trusted Execution Environment). The TEE is composed of several modules: an 'Encryption module' at the top, a 'Shared learning module' in the middle, two 'Decryption modules' below it, and two 'Authentication modules' at the bottom. A 'Task initiator' on the left sends a 'Task request' to the platform. On the left and right are 'Data provider A' and 'Data provider B' respectively. Each data provider contains a 'Data' storage, a 'Process and encryption module', and an 'Authentication module'. Data providers send 'Encrypted data' to the TEE's decryption modules. The TEE's authentication modules send 'Encryption key' information to the decryption modules. The TEE's shared learning module processes the data. The TEE's encryption module sends an 'Encrypted model' to the 'Result receiver' at the top. Data providers also receive 'Negotiated encryption mechanism software signature' from the TEE's authentication modules. + +Figure 2 – Technical architecture of the shared machine learning system in the centralized mode. The diagram shows a central 'Computation platform' containing a 'TEE' (Trusted Execution Environment). The TEE includes an 'Encryption module', a 'Shared learning module', two 'Decryption modules', and two 'Authentication modules'. A 'Task initiator' sends a 'Task request' to the platform. Two 'Data provider' units, 'Data provider A' and 'Data provider B', are shown. Each provider contains a 'Data' store, a 'Process and encryption module', and an 'Authentication module'. Data providers send 'Encrypted data' to the TEE's decryption modules. The TEE's authentication modules send 'Encryption key' information to the decryption modules. The TEE's shared learning module processes the data. The TEE's encryption module sends an 'Encrypted model' to the 'Result receiver'. Data providers also receive 'Negotiated encryption mechanism software signature' from the TEE's authentication modules. + +F.748.13(21)\_F02 + +**Figure 2 – Technical architecture of the shared machine learning system in the centralized mode** + +The technical architecture of the shared machine learning system in the centralized mode is mainly composed of a computation platform and multiple data providers. The computation platform is composed of authentication modules, decryption modules, encryption modules and a shared learning module. Each data provider consists of a process and encryption module, an authentication module, and data. The modules in the technical architecture are functional components. + +The data of each data provider is processed and encrypted by the encryption key obtained from the authentication module, then the encrypted data is uploaded to the trusted execution environment of the computation platform. The authentication module of the computation platform decrypts the encrypted data by using the decryption key, and then sends the decrypted data to the shared learning module. The shared learning module performs the shared machine learning operation on the decrypted data from multiple data providers. + +## **10.2 Functional components of the shared machine learning system in the centralized mode** + +The computation platform is mainly composed of authentication modules, decryption modules, encryption modules, and a shared learning module, all of which exist in a trusted execution environment (TEE). + +- Authentication module: this module is responsible for negotiating the encryption mechanism with each data provider and providing the decryption key to the decryption module. The decryption key is used for decrypting the encrypted data uploaded by the data provider to the platform. The authentication module of the computation platform is responsible for signing the software code running in the trusted execution environment and supporting the data provider to check the software code. +- Decryption module: this module is responsible for decrypting the encrypted data uploaded by the data provider based on the decryption key. The decryption key can be an asymmetric key or a symmetric key. +- Shared learning module: this module is responsible for data training based on the decrypted data from multiple data providers to output a shared model. +- Encryption module: this module is responsible for encrypting the computed model after the whole computation is completed. + +The data provider is mainly composed of data, process and encryption modules, and authentication modules. There are at least two data providers. + +- Data: the data provided by the data provider can be processed, encrypted, and uploaded to the computation platform. The encrypted data can be encrypted raw data, calculation factors, or any other data needed to be protected when transmitting among participants and the computation platform. +- Process and encryption module: this module is used to process and encrypt the data based on the encryption mechanism negotiated with the computation platform, and then the encrypted data is uploaded to the computation platform. +- Authentication module: the authentication module of the data provider is used to realize remote authentication of the data provider and the computation platform, including negotiating the encryption mechanism with the computation platform, and checking the code signature. + +The task initiator sends the computation task request to the computation platform and initiates the computation task. + +The result receiver receives the encrypted result from the computation platform and gets the final result. + +## 10.3 Processing procedures of the shared machine learning system in the centralized mode + +In the centralized mode, any data provider or the computation platform can initialize computational tasks, then the computation platform creates a trusted execution environment. The local data provided by each data provider can be processed, encrypted, and uploaded to the computation platform. The platform decrypts the received encrypted data sent by each data provider in the trusted execution environment and performs model training based on the decrypted data to obtain a shared model. The data processing, encryption, decryption, and training steps can be repeated multiple times. Finally, the trusted execution environment is destroyed to ensure data security and privacy. + +The processing procedures are described in Figure 3, and the processing procedures are recommended: + +![Sequence diagram of processing procedures in centralized mode](1a827b10290f33d4fec04d0e8ef7a897_img.jpg) + +The diagram illustrates the interaction between five entities: Task initiator, Data provider A, Computation platform, Data provider B, and Result receiver. The sequence of steps is as follows: + +- 1A. Download tools and deploy (Data provider A to Computation platform) +- 1B. Download tools and deploy (Data provider B to Computation platform) +- 2A. Data encryption and authorization (Data provider A to Data provider A) +- 2B. Data encryption and authorization (Data provider B to Data provider B) +- 3A. Upload encrypted data (Data provider A to Computation platform) +- 3B. Upload encrypted data (Data provider B to Computation platform) +4. Initiate computation task (Task initiator to Computation platform) +5. Create a trusted execution environment (Computation platform to Computation platform) +6. Decrypt data and verify access rights (Computation platform to Computation platform) +7. Perform computations (Computation platform to Computation platform) +8. Deliver encrypted computation result (Computation platform to Result receiver) +9. Destroy the trusted execution environment (Computation platform to Computation platform) + +F.748.13(21)\_F03 + +Sequence diagram of processing procedures in centralized mode + +**Figure 3 – Processing procedures of the shared machine learning system in the centralized mode** + +The technical process procedure of the shared machine learning system in the centralized mode are as follows: + +- Step 1: data providers download tools from the computation platform and deploy the tools. +- Step 2: data providers perform the data preparation, including data encryption, data authorization. +- Step 3: data providers upload the encrypted data to the computation platform. +- Step 4: task initiator initiates computation tasks on the computation platform, including the model to be trained, and the algorithms. +- Step 5: the computation platform creates a trusted execution environment. +- Step 6: the computation platform decrypts the encrypted data in the trusted execution environment. +- Step 7: the computation platform performs computations with the decrypted data in the trusted execution environment to obtain a computation result. +- Step 8: the computation platform delivers the computation result to the result receiver. +- Step 9: the computation platform destroys the trusted execution environment and the data in it. + +Note that the processing procedures above are normal procedures. In the real deployment, the orders and details of each procedure can be different. For example, the task initiator can initiate the computation in the first step. The data providers can upload encrypted data to the computation platform repeatedly. The modelling data or trained model can also be sealed up and stored outside the computation platform to be re-used later. + +### 11 Technical architecture, functional components, and processing procedure of the shared machine learning system in decentralized mode + +#### 11.1 Technical architecture of the shared machine learning system in the decentralized mode + +In the decentralized mode of the shared machine learning system, each party needs to deploy a learning module locally and transmit data to the local learning module. Learning modules among different data providers exchange parameters by using different encryption methods to achieve data sharing without exchanging raw data to protect data privacy. Computation platform assists to trigger the update of the learning module for each participant and coordinate relationships among different parties. + +The technical architecture is described in Figure 4. + +![Figure 4 – Technical architecture of the shared machine learning system in the decentralized mode](a706c91f074ac2840c161a3d4a7c0f91_img.jpg) + +The diagram shows a decentralized machine learning architecture. At the top is a 'Task initiator' box sending a 'Task request' arrow down to a 'Computation platform' box. Inside the 'Computation platform' is a 'Controlling module'. From the 'Computation platform', two 'Trigger and coordinate' arrows point left and right to 'Data provider A' and 'Data provider B' respectively. Each Data provider contains a 'Data' cylinder and a 'Learning module' box. A double-headed arrow between the two Learning modules is labeled 'Exchange random numbers or encrypted parameters'. Below the Data providers, arrows labeled 'Encrypted result' point down to a 'Result receiver' box. Data provider B also shows a 'Model 2' cylinder. + +Figure 4 – Technical architecture of the shared machine learning system in the decentralized mode + +**Figure 4 – Technical architecture of the shared machine learning system +in the decentralized mode** + +The technical architecture of the shared machine learning system in the decentralized mode is mainly composed of a computation platform, a result receiver, a task initiator, and multiple data providers. + +#### 11.2 Functional components of the shared machine learning system in the decentralized mode + +The computation platform mainly includes a controlling module, which divides computation tasks to different data providers, and coordinates with the learning modules in the data providers. Then the learning modules in each data provider exchange random numbers or encrypted parameters to perform the shared machine learning operation. + +The data provider is mainly composed of data and learning modules. There are at least two data providers. + +- Data: the data in the data provider is provided to the local learning module for machine learning. The data provider ensures that the users' privacy is not leaked. + +Rec. ITU-T F.748.13 (06/2021) 11 + +- Learning module: this module is used to receive machine learning tasks issued by the computation platform and performs the shared machine learning operation based on the local data and random numbers or encrypted parameters exchanged from other data providers. + +The result receiver receives the encrypted result from the data providers and gets the final result. + +The task initiator sends the task request to the computation platform. + +## 11.3 Processing procedures of the shared machine learning system in the decentralized mode + +In the decentralized mode, data providers perform shared learning operations by exchanging information that would not leak the users' privacy based on an established multi-party computation protocol. + +The recommended processing procedures of the decentralized solution are described in Figure 5. + +![Sequence diagram of shared machine learning in decentralized mode involving Task initiator, Data provider A, Computation platform, Data provider B, and Result receiver. Steps include tool download, local data preparation, task initiation, task coordination, data loading, multi-party computation data exchange, calculation, and result merging.](8592a32c2fdf17c1e562f0ba6b7e8e1a_img.jpg) + +``` + +sequenceDiagram + participant TI as Task initiator + participant DPA as Data provider A + participant CP as Computation platform + participant DPB as Data provider B + participant RR as Result receiver + + CP-->>DPA: 1A. Download tools + CP-->>DPB: 1B. Download tools + Note over DPA: 2A. Deploy tools and +prepare local data + Note over DPB: 2B. Deploy tools and +prepare local data + TI->>CP: 3. Initiate computation task + CP->>DPA: 4A. Coordinate and deliver +computation task + CP->>DPB: 4B. Coordinate and deliver +computation task + Note over DPA: 5A. Load local data + Note over DPB: 5B. Load local data + DPA<->DPB: 6. Exchange data based on +multi-party computation protocol + Note over DPA: 7A. Perform calculation +and get result + Note over DPB: 7B. Perform calculation +and get result + DPA->>RR: 8A. Output the +dense state result + DPB->>RR: 8B. Output the +dense state result + Note over RR: 9. Merge the +dense state results and get +the final result + +``` + +F.748.13(21)\_F05 + +Sequence diagram of shared machine learning in decentralized mode involving Task initiator, Data provider A, Computation platform, Data provider B, and Result receiver. Steps include tool download, local data preparation, task initiation, task coordination, data loading, multi-party computation data exchange, calculation, and result merging. + +**Figure 5 – Processing procedures of the shared machine learning system in the decentralized mode** + +The technical processing procedures of shared machine learning system are as follows: + +- Step 1: each data provider downloads tools from the computation platform. +- Step 2: each data provider deploys the downloaded tools and prepares the local data. +- Step 3: the task initiator initiates computation tasks to the computation platform. +- Step 4: the computation platform coordinates and delivers the computation tasks to the data providers. +- Step 5: the data providers load local data for computation. +- Step 6: the data providers exchange random numbers or encrypted parameters through multiple interactions based on multi-party computation protocol. +- Step 7: the data providers perform dense state calculation and get dense state results. +- Step 8: the data providers output the dense state result to the result receiver. +- Step 9: the result receiver merges the dense state results and gets the final result. + +# Appendix I + +## Use cases for shared machine learning systems + +(This appendix does not form an integral part of this Recommendation.) + +Shared machine learning can be used in different types of services that need data sharing to train shared models and need data security as well. + +This appendix describes use cases for shared machine learning systems to illustrate its concept and technical architectures. + +## **I.1 Use case: Improving modules for recognizing telecom frauds using data from multiple networks** + +It is valuable and important to detect and prevent telecom frauds in time and automatically. Usually, deceivers and being-deceived people may be in the same or different communication networks when they are in communication. Therefore, to train modules in recognizing telecom frauds would require close cooperation between multiple network operators, but based on the laws and regulations, the data of the communication networks are protected, separated, and are not allowed to be opened and exchanged. + +The training mode of the shared machine learning can help overcome data access restrictions of multiple networks. In this training mode, the communication networks keep their raw data and just upload necessary desensitized information (such as data labels and index, not the data itself) to the model controlling platform (see Figure I.1). Without loss of generality, the model controlling platform can use the desensitized information to arrange the module training tasks to the local training platforms in the communication networks respectively. The module training tasks are performed by the module training executors in the local training platforms with the local raw data. As the training progresses, the local training platform can exchange and synchronize the parameters of the being-trained modules with that in the model controlling platform. + +The trained modules for detecting telecom frauds can be deployed and performed in multiple communication networks respectively. + +In this case, the network operators can use shared machine learning mechanisms and relevant systems to train and perform modules for detecting telecom frauds, and without directly sharing their raw data. + +The overall framework for detecting telecom frauds using data from multiple networks is described in Figure I.1. + +![Diagram of the overall framework for detecting telecom fraud using data from multiple networks. At the top is a 'Model controlling platform based on SML' containing four modules: 'Trusted execution environment', 'Module training controller', 'Module parameter synchronizer', and 'Data information collaborator'. Below this platform are three 'Local training platform' boxes, labeled 'Network A', 'Network B', and 'Network C'. Each local platform contains four modules: 'Trusted execution environment', 'Module training executor', 'Module parameter synchronizer', and 'Raw data'. Double-headed vertical arrows connect the 'Model controlling platform' to each of the three 'Local training platform' boxes.](b6671cfafda3820aafe9a24fa7a4d8c7_img.jpg) + +F.748.13(21)\_FI.1 + +Diagram of the overall framework for detecting telecom fraud using data from multiple networks. At the top is a 'Model controlling platform based on SML' containing four modules: 'Trusted execution environment', 'Module training controller', 'Module parameter synchronizer', and 'Data information collaborator'. Below this platform are three 'Local training platform' boxes, labeled 'Network A', 'Network B', and 'Network C'. Each local platform contains four modules: 'Trusted execution environment', 'Module training executor', 'Module parameter synchronizer', and 'Raw data'. Double-headed vertical arrows connect the 'Model controlling platform' to each of the three 'Local training platform' boxes. + +**Figure I.1 – Overall framework for detecting telecom fraud using data from multiple networks** + +## I.2 Intelligent credit and risk control use case + +Shared machine learning can be used in intelligent credit scenarios in the financial field. Encrypted users' credit data from different parties are gathered to train a shared risk control model by a trusted third party. Therefore, data providers, such as banks, financial institutions, and other related parties can use the shared model to check the credit qualification of individuals to determine the credit amount. The shared learning technology promotes the intelligent checking process and assists to reduce on-line human credit checking costs. Additionally, shared machine learning in this scenario can support user-side authorization and privacy data protection to realize better risk control. + +## I.3 Intelligent marketing use case + +Shared learning can provide a precise users' equity strategy and a secure sharing environment that improves risk identification. For example, in a car insurance scenario, shared machine learning can significantly improve the differentiated equity of car insurance. It can assist insurance companies to develop better sales strategies before users purchasing insurance under the conditions of users' authorization. Through the car owners' information, users with different risk levels can be subdivided. Further, accurate image and risk analysis can be carried out for the owners to realize the precise users' equity strategy. + +# Bibliography + +- [b-ITU-T J.1201] Recommendation ITU-T J.1201 (2019), *Functional requirements of a smart TV operating system.* +- [b-ITU-T Y.3172] Recommendation ITU-T Y.3172 (2019), *Architectural framework for machine learning in future networks including IMT-2020.* + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.14-202203-I_PDF-E/raw.md b/marked/F/T-REC-F.748.14-202203-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..fe74586c9915bd1d888baff88831537866aec375 --- /dev/null +++ b/marked/F/T-REC-F.748.14-202203-I_PDF-E/raw.md @@ -0,0 +1,478 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.748.14** + +(03/2022) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Requirements and evaluation methods of +non-interactive 2D real-person digital human +application systems** + +Recommendation ITU-T F.748.14 + +ITU-T + +![ITU logo](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with a white grid pattern, overlaid by a stylized 'ITU' text in blue. + +ITU logo + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.14 + +# Requirements and evaluation methods of non-interactive 2D real-person digital human application systems + +## Summary + +Recommendation ITU-T F.748.14 specifies requirements and evaluation methods for non-interactive two-dimensional (2D) real-person digital human application systems, in terms of image, voice, movement, display, etc. It can be used to guide relevant parties to test, select or evaluate a non-interactive 2D real-person digital human application system. The evaluation methodology can reflect the current state of a non-interactive 2D real-person digital human application system by providing meaningful comparison dimensions. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.748.14 | 2022-03-16 | 16 | 11.1002/1000/14961 | + +## Keywords + +2D real-person, audio and video matching degree, character fidelity, non-interactive virtual human. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2022 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-------------------------------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Overview ..... | 2 | +| 7 Requirements and evaluation methods ..... | 2 | +| 7.1 Overview of evaluation methods ..... | 2 | +| 7.2 Image ..... | 2 | +| 7.3 Speech..... | 3 | +| 7.4 Animation ..... | 4 | +| 7.5 Multimodal input ..... | 7 | +| 7.6 Multimodal output ..... | 7 | +| 7.7 Overall evaluation ..... | 8 | +| Appendix I – Use case for non-interactive 2D real-person digital human application systems..... | 9 | +| I.1 Virtual news anchor ..... | 9 | +| Bibliography..... | 10 | + + + +# Recommendation ITU-T F.748.14 + +## Requirements and evaluation methods of non-interactive 2D real-person digital human application systems + +## 1 Scope + +This Recommendation specifies requirements and evaluation methods for non-interactive two-dimensional (2D) real-person digital human (i.e., intelligent-driven digital human) application systems, for which it addresses: + +- an image module; +- a speech module; +- an animation module; +- a multimodal output module. + +This Recommendation applies to the quality evaluation of a non-interactive 2D real-person digital human application system. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T F.748.15] Recommendation ITU-T F.748.15 (2022), *Framework and metrics for digital human application systems*. + +# 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 application system** [b-ISO/IEC TR 10032]: A collection of application processes which utilizes the services provided by the human-computer interface, communications facility, and data management system to perform the processing necessary to meet the requirements of the information system. + +**3.1.2 digital human** [ITU-T F.748.15]: A computer application that integrates the technologies of computer graphics, computer vision, intelligent speech and natural language processing. It can be used for digital content generation and human-computer interaction to help improve content production efficiency and user experience. + +**3.1.3 intelligent driven digital human** [ITU-T F.748.15]: A digital human driven by a computer system to automatically complete a series of actions by technical means, including text-driven, audio-driven and video-driven. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 2D digital human:** A digital human whose figure is a planar image and whose graphic content only contains information about horizontal and vertical dimensions. + +**3.2.2 2D real person digital human:** A 2D digital human that is trained by extracting the voice, lip shape, facial expression and other features from the video of a real person and using the techniques of speech, lip shape, facial expression synthesis and deep learning. + +**3.2.3 non-interactive digital human:** A digital human without the ability to interact, which can only read and execute commands. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +2D two-Dimensional + +# 5 Conventions + +This Recommendation uses the following conventions: + +- The phrase "is required to" indicates a requirement that must be strictly followed and from which no deviation is permitted if conformity to this Recommendation is to be claimed. +- The phrase "is recommended" indicates a requirement that is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformity. + +# 6 Overview + +Compared to a three-dimensional digital human, interactive digital human, a non-interactive 2D real-person image digital human has the advantages of high fidelity, low cost, less resource consumption and more mature technology, and is widely used in news broadcasting, lecturing scenes, etc., and at the time of publication is the most widely used type of digital human. A use case of non-interactive 2D real-person image digital human application systems is provided in Appendix I. + +# 7 Requirements and evaluation methods + +## 7.1 Overview of evaluation methods + +The evaluation methods for non-interactive 2D real-person digital human application systems include material inspection and technical test that are detailed as follows: + +- 1) material inspection: checking the documents that describe or certify the indicators; +- 2) technical test: making the target of evaluation produce specific results using predetermined methods or tools, and comparing and verifying the results of the operation results with those expected. + +## 7.2 Image + +#### 7.2.1 Quality of digital human image + +For the specification of this metric, see clause 7.2.1 of [ITU-T F.748.15]. + +Requirement: The digital human image is required to be free from the damage condition mentioned in clause 7.1.1 of [ITU-T F.748.15]. + +Evaluation method: Mainly by means of technical tests. Use ordinary text data sets to synthesize digital human video, and check the image of digital human for any of the following defects: + +- serious distortion; +- serious mosaic; + +- obvious frame skip; +- other damage. + +### 7.3 Speech + +#### 7.3.1 Speech accuracy + +For the specification of this metric, see clause 7.3.1 of [ITU-T F.748.15]. + +Requirement: The accuracy of text to speech is recommended to meet the requirements of users. + +Evaluation method: Mainly by means of technical tests. Construct a text test set containing ordinary text and special text, generate corresponding videos, count the number of mispronounced characters (the number of mispronounced words recorded in English) and the number of mispronounced sentences, and calculate the pronunciation word correct rate and pronunciation sentence correct rate. + +#### 7.3.2 Rhythm accuracy + +For the specification of this metric, see clause 7.3.2 of [ITU-T F.748.15]. + +Requirement: Rhythm accuracy is recommended to meet the requirements of users. + +Evaluation method: Mainly by means of technical tests. Use ordinary text test data, input a paragraph of text to generate corresponding video, record the text pause and the wrap break position, compare them with the standard answer, count the number of examples with correct pause taking each sentence as an example, and calculate the rhythm correct rate. + +#### 7.3.3 Speech comfort + +For the specification of this metric, see clause 7.3.3 of [ITU-T F.748.15]. + +Requirement: Speech comfort is recommended to meet the requirements of users. + +Evaluation method: Mainly by means of technical tests. The videos generated by the system are uniformly distributed to $K$ ( $K > 10$ ) testers with professional background knowledge, scored subjectively according to the subjective scoring rules in Table 1 for speech comfort, and finally averaged. + +**Table 1 – Subjective scoring rules for speech comfort** + +| Evaluation dimension | Description | 5 | 4 | 3 | 2 | 1 | +|-------------------------------------|-----------------------------------------------------------------------------------------|------------------------------|-----------------------------------------------------------|--------------------|-----------------------|-------------------| +| Pronunciation and intonation | Is the pronunciation standard as a whole? | Quite standard | Relatively standard | Basically standard | Individually standard | Quite nonstandard | +| | Is pronunciation clear? | Quite clear | Relatively clear | Basically clear | Not quite clear | Quite unclear | +| | Is word segmentation and sentence pause proper? | Quite proper | Relatively proper | Basically proper | Not quite proper | Quite improper | +| | Is tone and intonation natural? | Quite natural | Relatively natural | Basically natural | Not quite natural | Quite unnatural | +| | Is accent and pronunciation proper? | Quite proper | Relatively proper | Basically proper | Not quite proper | Quite improper | +| | Is speech speed expression proper? | Quite proper | Relatively proper | Basically proper | Not quite proper | Quite improper | +| Fluency and coherence | Is speech expression fluent? | Quite proper | Relatively proper | Basically proper | Not quite proper | Quite improper | +| Emotional fullness | According to the semantics and content of the text, is the emotional expression proper? | Quite proper | Relatively proper | Basically proper | Not quite proper | Quite improper | +| Anthropomorphic comfort | Is voice as anthropomorphic as a real human? | Completely indistinguishable | Relatively similar and slightly different from real voice | Basically similar | Not quite the same | Quite different | +| | When listening to the sound, do you feel happy? | Quite happy | Relatively happy | Just so so | Not quite happy | Quite unhappy | +| | Are you willing to have this voice serve you? | Quite willing | Relatively willing | Just so so | Not quite willing | Quite unwilling | + +### 7.4 Animation + +#### 7.4.1 Movement fit + +For the specification of this metric, see clause 7.4.1 of [ITU-T F.748.15]. + +Requirement: Movement fit is required to meet the requirements listed in Table 2. + +Evaluation method: Mainly by means of technical tests. Ordinary text data set is used to synthesize digital human video, and the conformity checked of the movement fit of digital human according to the movement types of digital human and their fit reflection. + +**Table 2 – Movement types of digital human and their fit reflection** + +| No. | Movement type | Fit reflection | +|-----|-----------------------------------------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| 1 | Lip movement |

a) As a speech organ, lips can automatically generate lip animation parameters according to input language information (speech or text).

  1. 1) The mouth shape meets the requirement of synchronous phoneme pronunciation at single frame time.
  2. 2) The mouth shape can satisfy the influence (coupling) of phonemes at several adjacent frames on the mouth shape at present.

b) As a semantic expression channel, lip movements can independently generate a reasonable mouth shape according to inner expression needs (e.g., emotion or intention).

  1. 1) basic lip units include lip pull-up, sharp mouth corner pull-up, mouth corner tightening, mouth corner pull-down, lower lip pull-down, lower lip push-up, pouting, mouth stretching, lip eversion, lip tightening, lip squeezing, mouth opening and lip sucking.
| +| 2 | Eyebrow and eyelid movement |

a) As a non-semantic expression channel, eyebrow and eyelid movements can independently display and imitate the physiological needs (e.g., blink) of real humans; when a digital human speaks, the eyebrow and eyelid movement conforms to the time-sequence rhythm characteristics of speech.

b) As a semantic expression channel, eyebrow and eyelid movement can autonomously display emotion or intention consistent with the inner state.

  1. 1) The basic expression movement unit involving eyebrows and eyelids includes inner eyebrow rise, inner eyebrow fall and frown.
  2. 2) The basic expression movement unit involving eyelid include upper eyelid rise, cheek lift and outer eye orbicularis tightening, inner eye orbicularis tightening, eye closing, blink and wink.
| +| 3 | Eyeball movement |

a) As a physiological need, the eyeballs can autonomously imitate the eyeball rotation of real humans.

b) As a semantic expression channel, eyeball movement can autonomously reflect the inner state (e.g., emotion or intention).

| +| 4 | Head rotation movement |

a) As a non-semantic expression channel, head rotation movement can autonomously express and imitate the physiological movement of real humans; when a digital human speaks, the head rotation conforms to the time-sequence rhythm of speech.

b) As a semantic expression channel, the head rotation movement can express the semantic information that meets the needs of the scene autonomously, e.g., nodding and shaking the head.

  1. 1) The basic head rotation movement unit includes head left-turning, head right-turning, head rise, head lowering, head left-leaning, head right-leaning, head forward-leaning, head backward-leaning, head up-and-down-shaking, head left-and-right shaking, head raise and left/right-leaning.
| +| 5 | Upper body movement [including torso joint (rotation) movement, big arm, forearm and palm movement] |

a) As a rhythmic movement, the upper body movement conforms to the time-sequence rhythm characteristics of speech when a digital human speaks.

b) As an indicative functional movement, digital human can express the role of character relationship, spatial position, time sequence, abstract concepts, etc. independently through upper body movement.

| + +**Table 2 – Movement types of digital human and their fit reflection** + +| No. | Movement type | Fit reflection | +|-----|------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| | |

c) As a symbolic functional movement, digital humans can independently gesticulate the attribute or behaviour of the entity through the upper body movement ratio to depict the corresponding entity or movement.

d) As a figurative functional movement, digital humans can independently construct a space through upper body movement to represent an abstract concept.

e) As an operational functional movement, digital humans can operate objects effectively, naturally, independently through upper body movement, and simulate related production and life of real human.

| +| 6 | Lower body movement (including movement of thighs, calves and soles of feet) |

a) As a balanced functional movement, digital humans can simulate the lower body muscles of real human beings against the gravity of the earth, and effectively and naturally maintain the balance of the body.

b) As a displacement functional action, digital humans can simulate lower body movement (e.g., walking, running and jumping) of real humans realizing body displacement, and can naturally switch various movement modes, and at the same time, can also reflect inner emotional state (e.g., walking or running posture differs slightly with differences in emotional state).

| +| 7 | Whole body movement |

With multi-modal movements coordinated, the whole body (including lips, eyebrows and eyelids, eyeballs, head rotation, upper body limbs and lower body limbs) works together to complete expression function or perform a certain function.

| + +#### **7.4.2 Movement comfort** + +For the specification of this metric, see clause 7.4.2 of [ITU-T F.748.15]. + +Requirement: Movement comfort is recommended to meet the requirements of users. + +Evaluation method: Mainly by means of technical tests. The videos generated by the system are uniformly distributed to $K$ ( $K > 10$ ) testers with professional background knowledge, scored subjectively according to the subjective scoring rules for movement comfort in Table 3, and finally averaged. + +**Table 3 – Subjective scoring rules of movement comfort** + +| Evaluation dimension | Description | 5 | 4 | 3 | 2 | 1 | +|----------------------------------|-----------------------------------------------|---------------|--------------------|-------------------|-------------------|-------------------| +| Mouth matching degree | Does the mouth shape match the pronunciation? | Perfectly so | Relatively so | Basically so | Not quite so | Completely not so | +| Naturalness of facial expression | Is facial expression natural? | Quite natural | Relatively natural | Basically natural | Not quite natural | Quite unnatural | +| Naturalness of body movement | Are the body movements natural? | Quite natural | Relatively natural | Basically natural | Not quite natural | Quite unnatural | + +### **7.5 Multimodal input** + +#### **7.5.1 Multimodal input mode** + +For the specification of this metric, see clause 7.6.1 of [ITU-T F.748.15]. + +Requirement: Multimodal input mode is required to support at least one output types, such as text, speech, image and touch. + +Evaluation method: Mainly by means of technical tests. Verify this by a live demonstration of the supported input modes. + +### **7.6 Multimodal output** + +#### **7.6.1 Real time factor** + +For the specification of this metric, see clause 7.6.1 of [ITU-T F.748.15]. + +Requirement: Real time factor is recommended to meet the requirements of users. Video resolution is inversely proportional to the time of synthesis. The lower the resolution is, the shorter the time of synthesis required. + +Evaluation method: Mainly by means of technical tests. Use ordinary text data set to synthesize digital human video, record the time from the start of execution to the completion of video generation and the time of synthesized video, and calculate the real time factor. + +#### **7.6.2 Smoothness** + +For the specification of this metric, see clause 7.6.2 of [ITU-T F.748.15]. + +Requirement: Smoothness is recommended to meet the requirements of users. + +Evaluation method: Mainly by means of material inspection. Check the number of frames per second of the video synthesized by the system. + +#### **7.6.3 Picture accuracy** + +For the specification of this metric, see clause 7.6.3 of [ITU-T F.748.15]. + +Requirement: Picture accuracy is recommended to meet the requirements of users. + +Evaluation method: Mainly by means of technical tests. Use ordinary text data set to synthesize digital human video, observe the output video of the system (video duration >600 s), count the number of frames with lag or frame skip, and calculate the real time factor. + +#### **7.6.4 Audio and video matching degree** + +For the specification of this metric, see clause 7.6.4 of [ITU-T F.748.15]. + +Requirement: Audio and video matching degree is recommended to meet the requirements of users. + +Evaluation method: Mainly by means of technical tests. Use ordinary text data set to synthesize digital human video, observe the output video of the system (video duration >600 s), count the number of mismatches such as lip-synchronization error, audio advance and delay, and calculate the audio and video matching rate. + +#### **7.6.5 Multimodal output mode** + +For the specification of this metric, see clause 7.7.5 of [ITU-T F.748.15]. + +Requirement: Multimodal output mode is required to support at least one output types, such as mobile phone, television, projection, light-emitting diode display, naked-eye stereo, virtual reality and augmented reality display. + +Evaluation method: Mainly by means of technical tests. Verify this by a live demonstration of the supported output modes. + +## 7.7 Overall evaluation + +#### 7.7.1 Character fidelity + +This metric is used to examine whether the system can show the fidelity equivalent to or indistinguishable from human. + +Requirement: Character fidelity is recommended to meet the requirements of users. + +Evaluation method: Mainly by means of technical tests. Prepare 100 segments of live video and synthesized virtual human video with a duration of 15 s, and try to ensure that the virtual person and real person are consistent in every detail (e.g., clothing, hairstyle and scene). Select $M$ segments of live videos and $N$ segments of synthesized virtual human videos, to be judged by $K$ ( $K \geq 10$ ) testers with professional background knowledge. Record the number of times, $F_1$ , when a tester misjudges the virtual person as the real person, and the number of times, $F_2$ , when a tester misjudges the real person as the virtual person, and calculate the false acceptance rate and the false rejection rate. The specific calculation formulae are shown in Equations 7-1 and 7-2. Results with a false rejection rate less than 10% are regarded as valid. + +$$R_1 = \frac{F_1}{N} \times 100\% \quad (7-1)$$ + +where + +$R_1$ : False acceptance rate + +$F_1$ : The number of times testers misidentified the synthetic video as real + +$N$ : The number of synthetic videos. + +$$R_2 = \frac{F_2}{M} \times 100\% \quad (7-2)$$ + +where + +$R_2$ : False rejection rate + +$F_2$ : The number of times the testers misjudged the real-person video as a synthetic + +$M$ : The number of videos recorded by real people. + +## Appendix I + +### Use case for non-interactive 2D real-person digital human application systems + +(This appendix does not form an integral part of this Recommendation.) + +## I.1 Virtual news anchor + +The production of traditional video news is time-consuming and costly. On the one hand, the timeliness of video news broadcasting is poor due to many processes such as on-site recording and post-editing. On the other hand, the production of news videos requires the cooperation of anchors, photographers and other media workers. Especially in the era of information globalization, if a news report in multiple languages is involved, it requires the participation of broadcasting and hosting professionals in different languages. + +The emergence of digital human technology makes it possible to produce news and video content quickly and efficiently. The technology can learn from the recorded videos of real anchors and "clone" virtual "dopes" with the same broadcasting ability, i.e., virtual anchors. The only need is to input text, the virtual anchor can generate news broadcast video in a short time, can realize multilingual broadcasting, and support 7 days a week, 24 h per day non-stop work. Especially in the special period of the pandemic, the emergence of virtual anchors meets the needs of journalists working remotely, supports efficient news production and reduces the risk of infection of journalists. The workflow of a virtual news anchor system is shown in Figure I.1. + +![Flowchart of the virtual news anchor system workflow. The process starts with two initial steps: 1. Choose a news anchor image and 2. Set the background and logo, which are combined with a plus sign. These lead to step 3: Enter the news text to be broadcast. This is followed by step 4: Edit broadcast content (Speaking speed/polyphonic words/timbre/movement), step 5: Set video parameters (Opening/ending/subtitle), and finally step 6: Synthesize the news video. Arrows indicate the sequential flow from left to right.](9ae17964ddd9b814c7d905b1af2fddf2_img.jpg) + +``` +graph LR; subgraph Start [ ]; 1((1 +Choose a news +anchor image)); 2((2 +Set the background +and logo)); 1 --- Plus[+]; 2 --- Plus; end; Plus --> 3((3 +Enter the news +text to be broadcast)); 3 --> 4((4 +Edit broadcast +content (Speaking +speed/polyphonic +words/timbre/ +movement)); 4 --> 5((5 +Set video +parameters +(Opening/ending/ +subtitle)); 5 --> 6((6 +Synthesize the +news video)); +``` + +F.748.14(22) + +Flowchart of the virtual news anchor system workflow. The process starts with two initial steps: 1. Choose a news anchor image and 2. Set the background and logo, which are combined with a plus sign. These lead to step 3: Enter the news text to be broadcast. This is followed by step 4: Edit broadcast content (Speaking speed/polyphonic words/timbre/movement), step 5: Set video parameters (Opening/ending/subtitle), and finally step 6: Synthesize the news video. Arrows indicate the sequential flow from left to right. + +**Figure I.1 – The workflow of virtual news anchor system** + +The description of the workflow is as follows. The user: + +1. logs in to the virtual anchor system and selects an image from the news anchor image library; +2. sets the news video background and logo; +3. enters the news text to be broadcast in the text edit box; +4. edits broadcast content, including speaking speed, polyphonic words, timbre and movement; +5. sets video parameters, including opening, ending and subtitle; +6. generates the corresponding voice and animation according to the input text, and then synthesizes the news video. + +## Bibliography + +[b-ISO/IEC TR 10032] ISO/IEC TR 10032:2003, *Information technology – Reference model of data management*. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.15-202203-I_PDF-E/raw.md b/marked/F/T-REC-F.748.15-202203-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..966c7fdb06f57c06c4c8bfbb40cfc53b67dd9a84 --- /dev/null +++ b/marked/F/T-REC-F.748.15-202203-I_PDF-E/raw.md @@ -0,0 +1,767 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.748.15** + +(03/2022) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Framework and metrics for digital human +application systems** + +Recommendation ITU-T F.748.15 + +ITU-T + +![ITU logo](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with a white grid pattern, overlaid by a stylized 'ITU' text in blue. + +ITU logo + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +## **TELEGRAPH SERVICE** + +| | | +|-----------------------------------------------------------------|-------------| +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | + +## **MOBILE SERVICE** + +| | | +|--------------------------------------------------------|-------------| +| Mobile services and multdestination satellite services | F.110–F.159 | +|--------------------------------------------------------|-------------| + +## **TELEMATIC SERVICES** + +| | | +|-------------------------------------------|-------------| +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | + +## **MESSAGE HANDLING SERVICES** + +| | | +|--|-------------| +| | F.400–F.499 | +|--|-------------| + +## **DIRECTORY SERVICES** + +| | | +|--|-------------| +| | F.500–F.549 | +|--|-------------| + +## **DOCUMENT COMMUNICATION** + +| | | +|--------------------------------------|-------------| +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | + +## **DATA TRANSMISSION SERVICES** + +| | | +|--|-------------| +| | F.600–F.699 | +|--|-------------| + +## **MULTIMEDIA SERVICES** + +| | | +|--|--------------------| +| | F.700–F.799 | +|--|--------------------| + +## **ISDN SERVICES** + +| | | +|--|-------------| +| | F.800–F.849 | +|--|-------------| + +## **UNIVERSAL PERSONAL TELECOMMUNICATION** + +| | | +|--|-------------| +| | F.850–F.899 | +|--|-------------| + +## **ACCESSIBILITY AND HUMAN FACTORS** + +| | | +|--|-------------| +| | F.900–F.999 | +|--|-------------| + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.15 + +# Framework and metrics for digital human application systems + +## Summary + +Recommendation ITU-T F.748.15 specifies a framework for digital human application systems, and proposes corresponding subjective and objective metrics for the dimensions of image, speech, animation, interactive processing and multimodal input/output. This Recommendation can be used to guide relevant parties to test, select or evaluate a digital human application system. The metrics can reflect the current state of the digital human application system by providing meaningful comparison dimensions. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.748.15 | 2022-03-16 | 16 | 11.1002/1000/14962 | + +## Keywords + +Animation module, digital human, framework and metrics, image module, interactive processing module, multimodal input module, multimodal output module, speech module. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2022 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-----------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Framework for digital human application system ..... | 2 | +| 7 Metrics and normative description ..... | 3 | +| 7.1 Overview ..... | 3 | +| 7.2 Image ..... | 4 | +| 7.3 Speech..... | 5 | +| 7.4 Animation ..... | 6 | +| 7.5 Interactive processing..... | 8 | +| 7.6 Multimodal input..... | 12 | +| 7.7 Multimodal output..... | 12 | +| Appendix I – Use case for digital human application system..... | 14 | +| I.1 Virtual customer service..... | 14 | +| Bibliography..... | 16 | + + + +# Recommendation ITU-T F.748.15 + +## Framework and metrics for digital human application systems + +# 1 Scope + +This Recommendation specifies a framework and metrics for digital human (i.e., intelligent-driven digital human) application systems, and addresses: + +- a) a framework; +- b) metrics for an image module; +- c) metrics for a speech module; +- d) metrics for an animation module; +- e) metrics for an interactive processing module; +- f) metrics for a multimodal input module; +- g) metrics for a multimodal output module. + +This Recommendation applies to quality evaluation for digital human application systems. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +# 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 application system** [b-ISO/IEC TR 10032]: A collection of application processes which utilizes the services provided by the human-computer interface, communications facility, and data management system to perform the processing necessary to meet the requirements of the information system. + +**3.1.2 text-to-speech synthesis (TTS)** [b-ITU-T P.10]: A TTS process generates a speech signal from text codes. It is usually composed of the two parts: + +- a language-dependent text processing part (the high level processing part), which generates from the character string (by reading rules, vocabulary and semantic analysis) a set of phonetic, prosodic, etc., parameters which are used by: +- an acoustical signal generating part, the synthesizer itself, which generates the audible speech. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 digital human:** A computer application that integrates the technologies of computer graphics, computer vision, intelligent speech and natural language processing. It can be used for digital content generation and human-computer interaction to help improve content production efficiency and user experience. + +**3.2.2 intelligent driven digital human:** The digital human driven by a computer system to automatically complete a series of actions by technical means, including text-driven, audio-driven and video-driven. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +2D two-Dimensional + +3D three-Dimensional + +ASR Automatic Speech Recognition + +NLP Natural Language Processing + +TTS Text To Speech + +# 5 Conventions + +This Recommendation uses the following conventions: + +- The phrase "is required to" indicates a requirement that must be strictly followed and from which no deviation is permitted if conformity to this Recommendation is to be claimed. +- The phrase "is recommended" indicates a requirement that is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformity. + +# 6 Framework for digital human application system + +This Recommendation divides the digital human system into six parts: image; speech; animation; interactive processing; multimodal input; and multimodal output. The first four parts are closely related to digital human characters themselves, and the last two are the attributes that the digital human needs to have for presentation or interaction with multimodal input and output devices, as shown in Figure 1. + +![Figure 1 – Framework for digital human application system. The diagram shows a flow from 'Image' to 'Speech' to 'Animation' to 'Multimodal output' to 'User'. A dashed box labeled 'Digital human application system' contains 'Interactive processing' (which includes 'Understanding and decision making' and 'Perception and recognition'), 'Multimodal input', and 'Animation'. Arrows show the flow from 'User' to 'Multimodal input' to 'Perception and recognition' to 'Understanding and decision making' to 'Animation'.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +The diagram illustrates the framework for a digital human application system. It consists of several modules and their interactions: + +- Image**: A module that provides the visual representation of the digital human. +- Speech**: A module that generates voice output. +- Animation**: A module that generates character movement and facial expressions. +- Multimodal output**: A module that combines the outputs from Image, Speech, and Animation for the User. +- User**: The external entity interacting with the system. +- Digital human application system**: A dashed box containing optional modules: + - Interactive processing**: A sub-module containing 'Understanding and decision making' and 'Perception and recognition'. + - Multimodal input**: A module that receives input from the User. + +Flow: Image → Speech → Animation → Multimodal output → User. User → Multimodal input → Perception and recognition → Understanding and decision making → Animation. + +F.748.15(22) + +Figure 1 – Framework for digital human application system. The diagram shows a flow from 'Image' to 'Speech' to 'Animation' to 'Multimodal output' to 'User'. A dashed box labeled 'Digital human application system' contains 'Interactive processing' (which includes 'Understanding and decision making' and 'Perception and recognition'), 'Multimodal input', and 'Animation'. Arrows show the flow from 'User' to 'Multimodal input' to 'Perception and recognition' to 'Understanding and decision making' to 'Animation'. + +NOTE – The dashed box indicates that this section is optional + +**Figure 1 – Framework for digital human application system** + +- Image module: Providing the image of a digital human. Character styles include cartoon, realistic and ultra-realistic. Based on the different dimensions of figure resources, image modules can be divided into two-dimensional (2D) and three-dimensional (3D) digital human types; +- Speech module: A module that can generate a character voice corresponding to the text. +- Animation module: A module that can generate character animation based on text, including facial expression and body movement. Based on the different driving methods, animation modules can be divided into real-human driving and intelligent driving types. +- Interactive processing module: A module that enables a digital human to perceive, recognize, understand and make decisions, i.e., to recognize user input information through intelligent technologies such as speech recognition and semantic recognition, comprehend user intention on this basis and decide the subsequent speech and animation of the digital human according to the current intention of the user, so as to drive the digital human to start a new round of interaction. +- Multimodal input module: A module that is used to receive input from the user. +- Multimodal output module: A module that presents the output to the user. + +According to the different ways of interaction with the user, digital human application systems can be divided into interactive and non-interactive categories. At present, a non-interactive digital human mainly refers to a broadcast digital human excluding an interactive processing module and multimodal input module. + +# 7 Metrics and normative description + +## 7.1 Overview + +This Recommendation divides metrics of a digital human application system into basic and optional types. Basic metrics are required to be met by all digital human application systems. Optional metrics of a digital human application system can be selectively met according to its actual situation. + +## 7.2 Image + +#### 7.2.1 Quality of digital human image (basic metric) + +The quality of a digital human image is the subjective evaluation its visual feeling. A digital human image is required to be free from the following defects. + +For a 2D character image: + +- serious distortion; +- a serious mosaic; +- obvious frame jump; +- other damage. + +For a 3D character image: + +- serious deformation; +- serious interpenetration; +- points that are not welded; +- broken surface; +- other damage. + +#### 7.2.2 Fineness (optional metric) + +Fineness is the degree of detail of a digital human image provided by the system, which is a special metric for a 3D digital human. + +Its specification is recommended to include the following information: + +- the number of character model surfaces; +- the modelling of facial details of the character, e.g., eyeballs, lenses, eyelashes, internal structure of oral cavity and hair; +- the wiring structure of character face; +- the resolution of character mapping. + +#### 7.2.3 Image comfort (optional metric) + +Image comfort is the degree to which the digital human image provided by the system makes users feel comfortable physiologically. This metric is subjective. According to the image quality of the digital human, the user gives a subjective score on a Likert scale1 to evaluate quality, from 1 worst to 5 best. See Table 1 for specific scoring rules. + +**Table 1 – Subjective scoring rules for image comfort** + +| Evaluation dimension | Description | 5 | 4 | 3 | 2 | 1 | +|----------------------|-----------------------------------------------|---------------|--------------------|-------------------|-------------------|-------------------| +| Favourability | Do you like the design of this image? | Very much | Relatively | Just so so | Not very much | Dislike very much | +| Naturalness | Is the image natural? | Quite natural | Relatively natural | Basically natural | Not quite natural | Quite unnatural | +| Use vision | Are you willing to have this image serve you? | Quite willing | Relatively willing | Just so so | Not quite willing | Quite unwilling | + +1 Likert scale is a psychological response scale, which is often used in questionnaires. + +### 7.3 Speech + +#### 7.3.1 Speech accuracy (basic metric) + +Speech accuracy of a system concerns the effectiveness of synthesis of speech from text. Speech errors include missing sounds, excessive pronunciation, phoneme errors and intonation errors. The corresponding performance metrics are derived as follows. + +The pronunciation word correct rate and the pronunciation sentence correct rate can be calculated by Equations 7-1 and 7-2, respectively. + +$$R_{\text{pwc}} = \left(1 - \frac{E_{\text{w}}}{N_{\text{w}}}\right) \times 100\% \quad (7-1)$$ + +where + +$R_{\text{pwc}}$ : Pronunciation word correct rate; + +$E_{\text{w}}$ : The number of mispronounced words (the sum of words with various speech errors); + +$N_{\text{w}}$ : The total number of words in the text. + +$$R_{\text{psc}} = \left(1 - \frac{E_{\text{s}}}{N_{\text{s}}}\right) \times 100\% \quad (7-2)$$ + +where + +$R_{\text{psc}}$ : Pronunciation sentence correct rate; + +$E_{\text{s}}$ : The number of mispronounced sentences; + +$N_{\text{s}}$ : The total number of sentences in the text. + +#### 7.3.2 Rhythm accuracy (basic metric) + +Rhythm accuracy concerns the effectiveness of synthesis of system speech. Rhythm includes sentence pause, pitch, length, volume, accent position and focus position, which correspond to natural speech rules such as focus pronunciation, question intonation and exclamatory sentence intonation. Only pauses are examined here. The corresponding performance metrics are derived as follows. + +The rhythm accuracy can be calculated by Equation 7-3. + +$$A_{\text{r}} = \frac{N_{\text{c}}}{N} \times 100\% \quad (7-3)$$ + +where + +$A_{\text{r}}$ : Rhythm accuracy; + +$N_{\text{c}}$ : The number of test samples with correct pauses; + +$N$ : The total number of test samples. + +#### 7.3.3 Speech comfort (basic metric) + +Speech comfort is the degree to which the speech synthesized by the system makes the user feel comfortable physiologically. This metric is subjective. According to the speech quality of digital human, the user gives a subjective score on a Likert scale to evaluate the quality, from 1 worst to 5 best. See Table 2 for specific scoring rules. + +**Table 2 – Subjective scoring rules for speech comfort** + +| Evaluation dimension | Description | 5 | 4 | 3 | 2 | 1 | +|-------------------------------------|-----------------------------------------------------------------------------------------|------------------------------|-----------------------------------------------------------|--------------------|-----------------------|-------------------| +| Pronunciation and intonation | Is the pronunciation standard as a whole? | Quite standard | Relatively standard | Basically standard | Individually standard | Quite nonstandard | +| | Is pronunciation clear? | Quite clear | Relatively clear | Basically clear | Not quite clear | Quite unclear | +| | Is word segmentation and sentence pause proper? | Quite proper | Relatively proper | Basically proper | Not quite proper | Quite improper | +| | Is tone and intonation natural? | Quite natural | Relatively natural | Basically natural | Not quite natural | Quite unnatural | +| | Is accent and pronunciation proper? | Quite proper | Relatively proper | Basically proper | Not quite proper | Quite improper | +| | Is speech speed expression proper? | Quite proper | Relatively proper | Basically proper | Not quite proper | Quite improper | +| Fluency and coherence | Is speech expression fluent? | Quite proper | Relatively proper | Basically proper | Not quite proper | Quite improper | +| Emotional fullness | According to the semantics and content of the text, is the emotional expression proper? | Quite proper | Relatively proper | Basically proper | Not quite proper | Quite improper | +| Anthropomorphic comfort | Is voice as anthropomorphic as a real human? | Completely indistinguishable | Relatively similar and slightly different from real voice | Basically similar | Not quite the same | Quite different | +| | When listening to the sound, do you feel happy? | Quite happy | Relatively happy | Just so so | Not quite happy | Quite unhappy | +| | Are you willing to have this voice serve you? | Quite willing | Relatively willing | Just so so | Not quite willing | Quite unwilling | + +### 7.4 Animation + +#### 7.4.1 Movement fit (basic metric) + +Movement fit relates to a digital human in the system and the current context. The movement types of a digital human and their fit reflection are shown in Table 3. + +**Table 3 – Movement types of a digital human and their fit reflection** + +| No. | Movement type | Fit reflection | +|-----|-----------------------------------------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| 1 | Lip movement |

a) As a speech organ, lips can automatically generate lip animation parameters according to input language information (speech or text).

  1. 1) The mouth shape meets the requirement of synchronous phoneme pronunciation at single frame time.
  2. 2) The mouth shape can satisfy the influence (coupling) of phonemes at several adjacent frames on the mouth shape at present.

b) As a semantic expression channel, lip movements can independently generate a reasonable mouth shape according to inner expression needs (e.g., emotion or intention).

  1. 1) Basic lip units include lip pull-up, sharp mouth corner pull-up, mouth corner tightening, mouth corner pull-down, lower lip pull-down, lower lip push-up, pouting, mouth stretching, lip eversion, lip tightening, lip squeezing, mouth opening and lip sucking.
| +| 2 | Eyebrow and eyelid movement |

a) As a non-semantic expression channel, eyebrow and eyelid movements can independently display and imitate the physiological needs (e.g., blink) of real humans; when a digital human speaks, the eyebrow and eyelid movement conforms to the time-sequence rhythm characteristics of speech.

b) As a semantic expression channel, eyebrow and eyelid movement can autonomously display emotion or intention consistent with the inner state.

  1. 1) The basic expression movement unit involving eyebrows and eyelids includes inner eyebrow rise, inner eyebrow fall and frown.
  2. 2) The basic expression movement unit involving eyelid include upper eyelid rise, cheek lift and outer eye orbicularis tightening, inner eye orbicularis tightening, eye closing, blink and wink.
| +| 3 | Eyeball movement |

a) As a physiological need, the eyeballs can autonomously imitate the eyeball rotation of real humans.

b) As a semantic expression channel, eyeball movement can autonomously reflect the inner state (e.g., emotion or intention).

| +| 4 | Head rotation movement |

a) As a non-semantic expression channel, head rotation movement can autonomously express and imitate the physiological movement of real humans; when a digital human speaks, the head rotation conforms to the time-sequence rhythm of speech.

b) As a semantic expression channel, the head rotation movement can express the semantic information that meets the needs of the scene autonomously, e.g., nodding and shaking the head.

  1. 1) The basic head rotation movement unit includes head left-turning, head right-turning, head rise, head lowering, head left-leaning, head right-leaning, head forward-leaning, head backward-leaning, head up-and-down-shaking, head left-and-right shaking, head raise and left/right-leaning.
| +| 5 | Upper body movement (including torso joint (rotation) movement, big arm, forearm and palm movement) |

a) As a rhythmic movement, the upper body movement conforms to the time-sequence rhythm characteristics of speech when a digital human speaks.

b) As an indicative functional movement, digital humans can express the role of character relationship, spatial position, time sequence, abstract concepts, etc. independently through upper body movement.

| + +**Table 3 – Movement types of a digital human and their fit reflection** + +| No. | Movement type | Fit reflection | +|-----|------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| | |

c) As a symbolic functional movement, digital humans can independently gesticulate the attribute or behaviour of the entity through the upper body movement ratio to depict the corresponding entity or movement.

d) As a figurative functional movement, digital humans can independently construct a space through upper body movement to represent an abstract concept.

e) As an operational functional movement, digital humans can operate objects effectively, naturally, independently through upper body movement and simulate related production and life of real humans.

| +| 6 | Lower body movement (including movement of thighs, calves and soles of feet) |

a) As a balanced functional movement, digital humans can simulate the lower body muscles of real human beings against the gravity of the earth, and effectively and naturally maintain the balance of the body.

b) As a displacement functional action, digital human can simulate the lower body movement (e.g., walking, running and jumping) of real human realizing body displacement, and can naturally switch various movement modes, and at the same time, can also reflect inner emotional state (e.g., walking/running posture slightly differs with difference of emotional state).

| +| 7 | Whole body movement | With multi-modal movements coordinated, the whole body (including lips, eyebrows and eyelids, eyeballs, head rotation, upper body limbs and lower body limbs) works together to complete expression function or perform a certain function. | + +#### 7.4.2 Movement comfort (basic metric) + +Movement comfort is the degree to which digital human movement in the system makes the user feel physiologically comfortable. This metric is subjective. According to the image quality of digital humans, the user gives a subjective score on a Likert scale to evaluate the quality, from 1 worst to 5 best. See Table 4 for specific scoring rules. + +**Table 4 – Subjective scoring rules for movement comfort** + +| Evaluation dimension | Description | 5 | 4 | 3 | 2 | 1 | +|----------------------------------|-----------------------------------------------|---------------|--------------------|-------------------|-------------------|-------------------| +| Mouth matching degree | Does the mouth shape match the pronunciation? | Perfectly so | Relatively so | Basically so | Not quite so | Completely not so | +| Naturalness of facial expression | Is facial expression natural? | Quite natural | Relatively natural | Basically natural | Not quite natural | Quite unnatural | +| Naturalness of body movement | Are the body movements natural? | Quite natural | Relatively natural | Basically natural | Not quite natural | Quite unnatural | + +### 7.5 Interactive processing + +#### 7.5.1 Interaction logic architecture of digital humans + +The interaction logic of digital humans consists of four modules: multi-modal input; perception and comprehension; multi-modal output; and manual module. The basic architecture of the system is shown in Figure 2 + +- Multi-modal input: This module is used to receive user input information, and supports various input modes such as text, speech, image and touch control. + +- Perception and comprehension: This module processes and perceives multi-modal input information through speech perception, visual perception and other technologies, uses natural language processing (NLP) and other technologies to comprehend the user's intention, and decides the subsequent speech and movement of the digital human according to the current intention of the user. +- Multi-modal output: This module is used to output interaction results through speech, animation or other forms. +- Manual module: This module is used to access and fulfil the manual request in real time of a user for service. A manual module can select the processing results of the perception and comprehension module, i.e., after the digital human acquires the input, the input information is transferred to manual processing without entering the intelligent perception and comprehension module; or a digital human perceives and comprehends the input information and hands it over to manual processing. + +![Figure 2 – Interaction logic architecture of digital humans. The diagram shows a flow from Multimodal input (Text, Speech, Image, Touch) to Perception and understanding (Speech perception, Visual perception, Natural language processing) and Manual. Both lead to Multimodal output (Speech, Animation, Other virtual display information).](bd0b93e7a46ede276d0a3b79ac487bd9_img.jpg) + +``` + +graph TD + subgraph Input [Multimodal input] + direction LR + T[Text] --- S[Speech] --- I[Image] --- Touch[Touch] + end + Input --> PU[Perception and understanding] + Input --> M[Manual] + subgraph PU [Perception and understanding] + direction LR + SP[Speech perception] --- VP[Visual perception] --- NLP[Natural language processing] + end + PU --> MO[Multimodal output] + M --> MO + subgraph Output [Multimodal output] + direction LR + Speech[Speech] --- Animation[Animation] --- OVDI[Other virtual display information] + end + +``` + +F.748.15(22) + +Figure 2 – Interaction logic architecture of digital humans. The diagram shows a flow from Multimodal input (Text, Speech, Image, Touch) to Perception and understanding (Speech perception, Visual perception, Natural language processing) and Manual. Both lead to Multimodal output (Speech, Animation, Other virtual display information). + +**Figure 2 – Interaction logic architecture of digital humans** + +#### 7.5.2 Interaction metrics + +##### 7.5.2.1 Automatic speech recognition accuracy (optional basic metric) + +Automatic speech recognition (ASR) accuracy is the performance of the ASR in a system according to the following metrics. + +The word recognition correct rate and the sentence recognition correct rate can be calculated by Equations 7-4 and 7-5, respectively. + +$$R_{\text{wrc}} = \left(1 - \frac{E_{\text{w}}}{N_{\text{w}}}\right) \times 100\% \quad (7-4)$$ + +where + +$R_{\text{wrc}}$ : Word recognition correct rate; + +$E_{\text{w}}$ : The number of wrongly recognized words; + +$N_{\text{w}}$ : The total number of words in the text. + +$$R_{\text{src}} = \left(1 - \frac{E_{\text{s}}}{N_{\text{s}}}\right) \times 100\% \quad (7-5)$$ + +where + +$R_{\text{src}}$ : Sentence recognition correct rate; + +$E_{\text{s}}$ : The number of wrongly recognized sentences; + +$N_{\text{s}}$ : The total number of sentences in the text. + +##### 7.5.2.2 Single-round dialogue accuracy (optional metric) + +This accuracy relates to the performance of the system in single-round dialogues with the user. The metric is derived as follows: + +The intention recognition rate and the precise recognition rate can be calculated by Equations 7-6 and 7-7, respectively. + +$$R_{\text{ir}} = \frac{M_1 + M_2}{N} \times 100\% \quad (7-6)$$ + +$$R_{\text{pr}} = \frac{M_1}{N} \times 100\% \quad (7-7)$$ + +where + +$R_{\text{ir}}$ : Intention recognition rate; + +$R_{\text{pr}}$ : Precise recognition rate; + +$M_1$ : The number of the precisely recognized dialogues; + +$M_2$ : The number of dialogues with correct fuzzy recommendation; + +$N$ : The total number of tested single-round dialogues. + +##### 7.5.2.3 Multi-round dialogue accuracy (optional metric) + +This accuracy relates to the performance of the system in multi-round dialogues with the user. The metric is derived as follows: + +The task completion rate and the context recognition rate can be calculated by Equations 7-8 and 7-9, respectively. + +$$R_{\text{tc}} = \frac{L_1}{N} \times 100\% \quad (7-8)$$ + +where + +$R_{\text{tc}}$ : Task completion rate; + +$L_1$ : The number of completed tasks (successful recognition of all rounds of multi-round dialogues in each test is regarded as task fulfilment); + +$N$ : The total number of tested multi-round dialogues. + +$$R_{\text{cr}} = \frac{L_2}{M} \times 100\% \quad (7-9)$$ + +where + +$R_{\text{cr}}$ : Context recognition rate; + +$L_2$ : The total number of successfully recognized rounds; + +$M$ : The total number of tested rounds. + +##### 7.5.2.4 Average response time of interaction (optional metric) + +This average relates to the response time of the system in a round of interaction with the user. Its metric is derived as follows: + +The average interaction response time can be calculated by Equation 7-10: + +$$\bar{T} = \frac{\sum_{i=1}^N (T_{\text{si}} - T_{\text{ei}})}{N} \quad (7-10)$$ + +where + +$\bar{T}$ : Average interaction response time; + +$i$ : Indicates the round of interaction; + +$T_{\text{si}}$ : The point at which the system starts to feedback (in seconds); + +$T_{ei}$ : The point at which user input ends (in seconds); + +$N$ : The total number of rounds of interaction tested (one round of interaction is from user input to system output). + +##### 7.5.2.5 Smoothness of interaction (optional metric) + +This metric relates to the smoothness and stability of system services and the number of concurrencies the system can carry. + +##### 7.5.2.6 Business risk identification (optional metric) + +This metric relates to the ability of the system to identify and avoid business risks such as fraud, malicious interference and suicide. + +##### 7.5.2.7 Business resolution rate (optional metric) + +The business resolution rate is the actual business solution ability of the system. Its metric is derived as follows: + +The business resolution rate can be calculated by Equation 7-11. + +$$R_{br} = \frac{M}{N} \times 100\% \quad (7-11)$$ + +where + +$R_{br}$ : Business resolution rate; + +$M$ : The number of test problems solved; + +$N$ : The total number of tests. + +##### 7.5.2.8 Emotion pacification ability (optional metric) + +This metric relates to the ability of the system to recognize the emotional change of the user and give timely and effective pacification. + +##### 7.5.2.9 Manual intervention rate (optional metric) + +This rate relates to the situation of manual intervention in the process of providing service to judge its independent business ability. Its metric is derived as follows: + +The manual intervention rate can be calculated by Equation 7-12: + +$$R_{mi} = \frac{M}{N} \times 100\% \quad (7-12)$$ + +where + +$R_{mi}$ : Manual intervention rate; + +$M$ : The number of manual interventions; + +$N$ : The total number of tests. + +##### 7.5.2.10 User interaction experience (optional metric) + +This metric is subjective. Based on the interaction experience of the participating system, the user gives a subjective score from two aspects such as dialogue interaction accuracy and style consistency with reference to Table 5. + +**Table 5 – Subjective metric of user experience** + +| Evaluation dimension | Description | 5 | 4 | 3 | 2 | 1 | +|-----------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------|-------------------------|-------------------------|------------------------|------------------------|---------------------------| +| Interaction accuracy | Can the system accurately answer user queries when consulted, e.g., frequently asked questions? | Completely accurate | Relatively accurate | Basically accurate | Not quite accurate | Completely inaccurate | +| | Under the task execution scenario, can the system accurately help the user complete the task according to the context? | Completely accurate | Relatively accurate | Basically accurate | Not quite accurate | Completely inaccurate | +| | Can the system give a satisfactory reply under the situation of chatting? | Completely satisfactory | Relatively satisfactory | Basically satisfactory | Not quite satisfactory | Completely unsatisfactory | +| | Can the system accurately recognize interruptions? | Completely accurate | Relatively accurate | Basically accurate | Not quite accurate | Completely inaccurate | +| Style consistency | In the process of interaction, does the digital human in the system keep the consistent style, which conforms to the personality of the set character? | Quite consistent | Relatively consistent | Basically consistent | Not quite consistent | Completely inconsistent | + +### **7.6 Multimodal input** + +#### **7.6.1 Multimodal input mode (basic metric)** + +This metric is used to assess the input types or modes supported by the system, such as text, speech, image and touch. + +### **7.7 Multimodal output** + +#### **7.7.1 Real time factor (optional metric)** + +This metric is basic in a broadcast virtual human system, which is used to access the real time factor in broadcast system, i.e., the ratio of video synthesis time to output video duration. + +#### **7.7.2 Smoothness (basic metric)** + +The number of frames per second is used to evaluate virtual human video generated by the system. + +#### **7.7.3 Picture accuracy (basic metric)** + +This metric is used to assess the accuracy of the picture when the system generates a video with a fixed number of frames. If errors such as frame skipping and jamming occur, the picture is regarded as inaccurate. + +#### **7.7.4 Audio and video matching degree (basic metric)** + +This metric is used to assess the matching degree of audio and video when the system generates video with a fixed duration (in seconds). In the case of redundant or missing mouth shape, the dislocation such as audio advance and delay is regarded as audio and video mismatch. + +#### **7.7.5 Multimodal output mode (basic metric)** + +This metric used to assess the output types or modes supported by the system, such as mobile phone, television, projection, light-emitting diode display, naked-eye stereo, virtual reality and augmented reality display. + +# Appendix I + +## Use case for digital human application system + +(This appendix does not form an integral part of this Recommendation.) + +## I.1 Virtual customer service + +At present, the labour cost of customer service continues to rise, and customers' requirements for service experience are becoming more stringent. Traditional technology cannot balance the scale and differentiated service, so new methods are needed to solve these contradictions. Virtual customer service, with realistic image, intelligent dialogue, intelligent personalized service, enabling customer service, release productivity while helping to improve user experience and help enterprises digital management. The workflow of virtual customer service system is shown in Figure I.1. + +![Figure I.1 – The workflow of virtual customer service system. The diagram shows a flowchart of the system's components and their interactions. The system is enclosed in a box labeled 'Virtual customer service system'. On the right, a vertical bar represents the 'Customer'. Arrows with circled numbers show the flow: 1) Customer to Multimodal input; 2) Multimodal input to Automatic speech recognition; 3) Automatic speech recognition to Natural language processing; 4) Natural language processing to Text to speech; 5) Text to speech to Movement parameter extraction module; 6) Text to speech to Audio and video fusion; 7) Text to speech to Movement parameter extraction module; 8) Movement parameter extraction module to Image module; 9) Image module to Audio and video fusion; 10) Audio and video fusion to Multimodal output; 11) Multimodal output to Customer. A feedback loop goes from Multimodal output back to Movement parameter extraction module.](3da1a07cb87051bf616c9876db958cf0_img.jpg) + +``` +graph LR + subgraph Virtual_customer_service_system [Virtual customer service system] + MPE[Movement parameter extraction module] + TTS[Text to speech] + NLP[Natural language processing] + ASR[Automatic speech recognition] + IM[Image module] + AVF[Audio and video fusion] + MO[Multimodal output] + MI[Multimodal input] + + MI -- 2 --> ASR + ASR -- 3 --> NLP + NLP -- 4 --> TTS + TTS -- 5 --> MPE + TTS -- 6 --> AVF + TTS -- 7 --> MPE + MPE -- 8 --> IM + IM -- 9 --> AVF + AVF -- 10 --> MO + MO -- 11 --> Customer + MO -- 5 --> MPE + end + Customer -- 1 --> MI +``` + +Figure I.1 – The workflow of virtual customer service system. The diagram shows a flowchart of the system's components and their interactions. The system is enclosed in a box labeled 'Virtual customer service system'. On the right, a vertical bar represents the 'Customer'. Arrows with circled numbers show the flow: 1) Customer to Multimodal input; 2) Multimodal input to Automatic speech recognition; 3) Automatic speech recognition to Natural language processing; 4) Natural language processing to Text to speech; 5) Text to speech to Movement parameter extraction module; 6) Text to speech to Audio and video fusion; 7) Text to speech to Movement parameter extraction module; 8) Movement parameter extraction module to Image module; 9) Image module to Audio and video fusion; 10) Audio and video fusion to Multimodal output; 11) Multimodal output to Customer. A feedback loop goes from Multimodal output back to Movement parameter extraction module. + +Figure I.1 – The workflow of virtual customer service system + +The description of the workflow follows: + +- 1) the customer gives voice commands to request customer service; +- 2) the multimodal input module transmits the collected customer voice to the ASR module; +- 3) the ASR module recognizes the customer's speech, converts it into text, and transmits it to the NLP module; +- 4) the NLP module performs speech comprehension and generates corresponding reply text, and transmits it to the TTS module; +- 5) the NLP module sends the generated reply text to the movement parameter extraction module; +- 6) the TTS module synthesizes the reply text into speech and transmits it to the audio and video fusion module; +- 7) the TTS module transmits the synthesized speech to the movement parameter extraction module; +- 8) the movement parameter extraction module extracts the movement parameters from the reply text, including body, facial expression and lip shape parameters, and send them to the image module; +- 9) the image module generates the corresponding video according to the extracted movement parameters, and send it to the audio and video fusion module; + +- 10) the audio and video fusion module synchronously superimposes the generated audio and video, and obtains the audio and video data corresponding to the virtual customer service reply statement, and sends it to multimodal output module; +- 11) the multimodal output module presents the corresponding audio and video to the customer as the virtual customer service's response – then the customer starts the next round dialogue. + +# Bibliography + +- [b-ITU-T P.10] Recommendation ITU-T P.10/G.100 (2017), *Vocabulary for performance, quality of service and quality of experience.* +- [b-ISO/IEC TR 10032] ISO/IEC TR 10032:2003, *Information technology – Reference model of data management.* + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.16-202205-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.748.16-202205-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1e32f9b8f1141bed243c16e02a6cd1411cbb87f1 --- /dev/null +++ b/marked/F/T-REC-F.748.16-202205-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c395c4bbde83b1e232652a43f51538278c116f51f710c37e20a073f7333726c8 +size 6134 diff --git a/marked/F/T-REC-F.748.16-202205-I_PDF-E/562f471e8153729557e6a4ee6343c32c_img.jpg b/marked/F/T-REC-F.748.16-202205-I_PDF-E/562f471e8153729557e6a4ee6343c32c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ffb123b56cba332d325fd0d867c83dfe4c3b8b28 --- /dev/null +++ b/marked/F/T-REC-F.748.16-202205-I_PDF-E/562f471e8153729557e6a4ee6343c32c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e7f4fa29c0a5ea4e052587160e4eb429f735c0d33d8d2c3b7b57e16804a10e8a +size 35520 diff --git a/marked/F/T-REC-F.748.16-202205-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg b/marked/F/T-REC-F.748.16-202205-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3bbe7dabef4fecfb7a6fff36224b606b7cf18602 --- /dev/null +++ b/marked/F/T-REC-F.748.16-202205-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a79610d8ec31af2e499cebe970a1fbff29feb965f601418ccce16a123a0afbd7 +size 126825 diff --git a/marked/F/T-REC-F.748.16-202205-I_PDF-E/fa859e4e468bfb2710a94527f2c504af_img.jpg b/marked/F/T-REC-F.748.16-202205-I_PDF-E/fa859e4e468bfb2710a94527f2c504af_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..731ce4e1c0d864c57d19fabebddd7c33d17fdf43 --- /dev/null +++ b/marked/F/T-REC-F.748.16-202205-I_PDF-E/fa859e4e468bfb2710a94527f2c504af_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:46b4941f45b2949b5a274024b3b04c34a6aaf8f1fd830663ea777746c3742708 +size 136746 diff --git a/marked/F/T-REC-F.748.16-202205-I_PDF-E/raw.md b/marked/F/T-REC-F.748.16-202205-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..f86b141c4a25898b5a758eeb8067bf3173ff5e29 --- /dev/null +++ b/marked/F/T-REC-F.748.16-202205-I_PDF-E/raw.md @@ -0,0 +1,621 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.748.16** + +(05/2022) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Requirements for applications and services in +smart manufacturing based on machine vision** + +Recommendation ITU-T F.748.16 + +ITU-T + +![ITU logo](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with a white grid pattern, overlaid by a stylized 'ITU' text in blue. + +ITU logo + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +## **TELEGRAPH SERVICE** + +| | | +|-----------------------------------------------------------------|-------------| +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | + +## **MOBILE SERVICE** + +| | | +|--------------------------------------------------------|-------------| +| Mobile services and multdestination satellite services | F.110–F.159 | +|--------------------------------------------------------|-------------| + +## **TELEMATIC SERVICES** + +| | | +|-------------------------------------------|-------------| +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | + +## **MESSAGE HANDLING SERVICES** + +F.400–F.499 + +## **DIRECTORY SERVICES** + +F.500–F.549 + +## **DOCUMENT COMMUNICATION** + +| | | +|--------------------------------------|-------------| +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | + +## **DATA TRANSMISSION SERVICES** + +F.600–F.699 + +## **MULTIMEDIA SERVICES** + +**F.700–F.799** + +## **ISDN SERVICES** + +F.800–F.849 + +## **UNIVERSAL PERSONAL TELECOMMUNICATION** + +F.850–F.899 + +## **ACCESSIBILITY AND HUMAN FACTORS** + +F.900–F.999 + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.16 + +# Requirements for applications and services in smart manufacturing based on machine vision + +## Summary + +Recommendation ITU-T F.748.16 presents the overview and requirements for applications and services in smart manufacturing based on machine vision (MV). It describes the basic concept, scenario and ecosystem of machine vision, and identifies several typical requirements, such as data acquisition, data preprocessing and data processing. This Recommendation also gives a reference model for MV-based applications and services in smart manufacturing. The general goals of standardization for the MV service are: + +- Defining the requirements of MV-based services and applications; +- Helping end users and providers to specify the MV tasks and solutions; +- Enhancing confidence in the MV ecosystem and opening new applications for MV systems. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.748.16 | 2022-05-22 | 16 | 11.1002/1000/14963 | + +## Keywords + +Machine vision, requirement, smart manufacturing. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2022 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|------------------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 1 | +| 5 Conventions ..... | 2 | +| 6 Overview of machine vision..... | 2 | +| 6.1 Basic concept of machine vision ..... | 2 | +| 6.2 Ecosystem..... | 3 | +| 6.3 End-to-end layered architecture ..... | 4 | +| 7 Requirements for MV-based applications and services..... | 5 | +| 7.1 Image and video data acquisition ..... | 5 | +| 7.2 Image and video pre-processing..... | 6 | +| 7.3 Image and video processing ..... | 6 | +| 7.4 Algorithm training..... | 7 | +| 7.5 Model evaluation..... | 7 | +| 7.6 Security..... | 8 | +| 7.7 Identification, authentication and authorization ..... | 8 | +| 7.8 Accounting and charging..... | 8 | +| 7.9 Open service environment..... | 9 | +| 7.10 Network ..... | 9 | +| 8 Reference model for supporting MV-based applications and services ..... | 9 | +| 8.1 High level reference model ..... | 9 | +| 8.2 Functional domain requirements ..... | 10 | +| Appendix I – Typical scenarios of applications based on machine vision ..... | 13 | +| I.1 Object sorting ..... | 13 | +| I.2 Quality assessment ..... | 13 | +| I.3 Robot guidance..... | 13 | +| I.4 Object measurement ..... | 13 | +| Bibliography..... | 15 | + + + +# Recommendation ITU-T F.748.16 + +# Requirements for applications and services in smart manufacturing based on machine vision + +# 1 Scope + +This Recommendation presents the basic concept of machine vision (MV) and identifies requirements for MV-based applications in smart manufacturing. This Recommendation also specifies a reference model of MV-based applications and services in smart manufacturing. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T Y.2060] Recommendation ITU-T Y.2060 (2012), *Overview of the Internet of things*. + +# 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 algorithm** [b-ISO/IEC 11557]: Set of rules for transforming the logical representation of data. + +**3.1.2 application** [b-ITU-T Y.101]: A structured set of capabilities, which provide value-added functionality supported by one or more services. + +**3.1.3 endpoint** [b-ITU-T H.350]: A logical device that provides video and/or voice media encoding/decoding, and signalling functions. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 machine vision**: Signal processing to acquire, process and interpret an image or video to support visual analysis for applications, such as automatic inspection, process control and guidance. + +**3.2.2 machine vision platform**: A set of capabilities integrated in a platform that provides basic and common functions for machine vision applications. + +**3.2.3 model**: An output created by an algorithm running on training data that can generate an inference or prediction, based on input data. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +5G Fifth-Generation mobile networks + +AI Artificial Intelligence + +| | | +|--------|----------------------------------------| +| LAN | Local Area Network | +| MV | Machine Vision | +| MVAS | Machine Vision Application and Service | +| OBS | Object Sorting | +| OM | Object Measurement | +| ONVIF | Open Network Video Interface Forum | +| QoS | Quality of Service | +| QV | Quality Verification | +| RG | Robot Guidance | +| ResNet | Residual Networks | +| RTSP | Real Time Streaming Protocol | +| SDK | Software Development Kit | +| SSD | Solid State Disk | +| VSSNet | Video Scale Selection Network | +| YOLO | You Only Look Once | + +## 5 Conventions + +In this Recommendation: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "is recommended" indicate a specification which is recommended but which is not absolutely required. Thus it need not be present to claim conformance. +- The keyword "can" indicate an optional specification which is permissible, without implying any sense of being recommended. This term is not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator or service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +# 6 Overview of machine vision + +### 6.1 Basic concept of machine vision + +Machine vision traditionally refers to the use of computer vision for object recognition and classification by extracting useful information about an object from images or video, in an industrial or practical application or in a process where it is necessary to execute a certain function or outcome. It is a branch of the combination of artificial intelligence (AI), image-processing and pattern-recognition techniques, the purpose of which is to replace human eyes for measurement and judgement by adding a set of vision systems to machines or automatic production lines. MV encompasses all industrial and non-industrial applications in which a combination of hardware and software provide operational guidance to devices in the execution of their functions based on the capture and processing of images. The major tasks performed by an MV system can be grouped into three processes: data acquisition, processing or analysis, and recognition or measurement. Various features of the objects are extracted, and final decisions are made using different image-processing algorithms and pattern-recognition techniques, respectively. + +### 6.2 Ecosystem + +Figure 6-1 depicts business roles which are relevant in a MV ecosystem and relationships between them. Four key roles are identified: MV service customer, MV resource provider, MV platform provider and MV service provider. + +![Diagram of business roles in a MV ecosystem](562f471e8153729557e6a4ee6343c32c_img.jpg) + +The diagram illustrates the hierarchy of business roles in a MV ecosystem. At the top is a rounded rectangle labeled 'MV service customers'. Below it is a larger rounded rectangle labeled 'MV service providers', which contains two sub-boxes: 'Application provider' and 'Other service providers'. Below that is a rounded rectangle labeled 'MV platform provider'. At the bottom is another large rounded rectangle labeled 'MV resource providers', which contains two sub-boxes: 'Device operator' and 'Other resource providers'. Vertical lines connect the top level to the second, the second to the third, and the third to the fourth. A small label 'F.748.16(22)' is located at the bottom right of the diagram. + +Diagram of business roles in a MV ecosystem + +**Figure 6-1 – Business roles in a MV ecosystem** + +#### 6.2.1 MV service provider + +The MV service provider utilizes capabilities provided by the MV platform provider in order to provide services to the MV service customer. The following sub-roles are identified (but not limited to): + +- Application provider: the role that is responsible for offering one or more applications to the MV service customer. +- Other MV service providers: the roles that provide other services to the MV service customer (e.g., providers of algorithm training). + +The MV service provider has a business relationship with the MV service customer and the MV platform provider. + +NOTE – An actor playing the role of MV service provider can also play the role of MV service platform provider. + +#### 6.2.2 MV resource provider + +The MV resource provider provides raw data or other necessary resources with the device operator to the MV platform provider according to the service logic. + +The following sub-roles are identified (but not limited to): + +- MV device operator: the role that is responsible for the implementation of devices and for exchanging information with the MV platform provider. For example, the operator who implements cameras for smart manufacturing can act as the MV device operator. +- Other resource providers: the roles that provide other resources to the MV platform provider (e.g., providers of computing, storage resources). + +The MV resource provider has a business relationship with the MV platform provider. + +NOTE – An actor playing the role of MV resource provider can also play the role of MV service provider and the MV platform provider. + +#### **6.2.3 MV platform provider** + +The MV platform provider is the role responsible for providing the following functions: + +- Integration of resources provided by MV resource providers. +- Support and control of the service integration and delivery functionalities. +- Offering to the MV service provider capabilities (including resource exposure) for supporting MV applications. + +The MV platform provider has a business relationship with both the MV resource provider and the MV service provider. + +NOTE – An actor playing the role of MV service platform provider can also play the role of MV service provider and MV resource provider. + +#### **6.2.4 MV service customer** + +The MV service customer is the subscriber of MV service. The MV service customer consumes services offered by the MV service provider. Organizations such as enterprises and factories are examples of MV service customers. + +The MV service customer has a business relationship with the MV service provider. + +### **6.3 End-to-end layered architecture** + +Smart manufacturing system is a typical IoT connection system. Figure 6-2 shows the end-to-end layered architecture of an MV system in smart manufacturing based on the IoT reference model [ITU-T Y.2060]. It is composed of four layers as well as the associated management. + +The four layers are as follows: + +- Application layer; +- Service support and application support layer; +- Network layer; +- Device layer. + +![Figure 6-2: Overview of the end-to-end layered architecture of an MV system in smart manufacturing. The diagram shows four layers: Application layer, Service support and application support layer, Network layer, and Device layer. The Application layer includes Object sorting, Quality assessment, Robot guidance, and Object measurement. The Service support and application support layer includes Smart manufacturing algorithm service, Smart manufacturing service management, Smart manufacturing algorithm/model library, and Smart manufacturing algorithm/model development. The Network layer includes Networking and transport, Wireless network access (Wireless LAN, 5G), and Wired access (Optical fibre). The Device layer includes Industrial cameras and Industrial controllers. A Management section on the right includes Security management and Device management. Vertical double-headed arrows connect the layers, and horizontal double-headed arrows connect the layers to the Management section. A zigzag line indicates a break in the connection between the Network and Device layers. The label F.748.16(22) is at the bottom right.](cfef993dcc8fb513de79eb1f93cf26ae_img.jpg) + +Figure 6-2: Overview of the end-to-end layered architecture of an MV system in smart manufacturing. The diagram shows four layers: Application layer, Service support and application support layer, Network layer, and Device layer. The Application layer includes Object sorting, Quality assessment, Robot guidance, and Object measurement. The Service support and application support layer includes Smart manufacturing algorithm service, Smart manufacturing service management, Smart manufacturing algorithm/model library, and Smart manufacturing algorithm/model development. The Network layer includes Networking and transport, Wireless network access (Wireless LAN, 5G), and Wired access (Optical fibre). The Device layer includes Industrial cameras and Industrial controllers. A Management section on the right includes Security management and Device management. Vertical double-headed arrows connect the layers, and horizontal double-headed arrows connect the layers to the Management section. A zigzag line indicates a break in the connection between the Network and Device layers. The label F.748.16(22) is at the bottom right. + +**Figure 6-2 – Overview of the end-to-end layered architecture of an MV system in smart manufacturing** + +The device layer includes many devices, such as industrial cameras, industrial controllers, sensors and mechanical arms. They complete the data and image acquisition of the manufacturing site, and control the sensors and mechanical arms to finish the corresponding operation based on the data processing results. Moreover, the requirements of time delay sensitivity are also different. + +The localization and wired connection of an MV system increasingly cannot meet the needs of application scenarios, so it is necessary to solve the problems of mobility and sharing through wireless access. The network layer includes two kinds of access methods, such as wireless access network (e.g., 5G and wireless LAN) and wired access (e.g., optical fibre) and transport network. + +The service support and application support layer includes smart manufacturing algorithm / model library, smart manufacturing algorithm / model development, smart manufacturing algorithm service and service management. The processing of various applications and image algorithm and learning algorithm modelling all demand strong computing power. + +The application layer provides applications such as object sorting, quality verification, robot guidance and component identification, etc. + +Management includes device management and security management. + +# 7 Requirements for MV-based applications and services + +### 7.1 Image and video data acquisition + +The image or video data acquisition part is generally composed of a light source, lens, digital camera and image acquisition card. The data acquisition process can be simply described as a digital camera that captures the target object and converts it into an image signal under the condition that the light + +source provides lighting, and finally transmits it to the image-processing part through the data acquisition card. In the design of the data acquisition part, we should take into account many issues, mainly in digital cameras, data acquisition cards and light sources. + +In addition, in some cases, e.g., in order to successfully navigate a mobile robot, obtaining detailed information about its direct environment is a major concern. The specifications for data acquisition are as follows: + +- DA-01: It is recommended that an MVAS support a high resolution of the collected images or videos. +- DA-02: It is recommended that an MVAS support real time performance of the collected images or videos. + +### 7.2 Image and video pre-processing + +Image pre-processing describes any type of processing that is performed on raw image data to begin with the next processing procedure. Pre-processing is the initial step which transforms the data into a format that will be more easily and effectively processed. Hence, it is important to have an effective pre-processing stage. Here are a lot more pre-processing techniques that will be used according to the need of a given system. The specifications for pre-processing are as follows: + +- IPP-01: It is recommended that an MVAS support image enhancement, such as removing noise or correcting the contrast in the image. +- IPP-02: It is recommended that an MVAS support thresholding such as removing the background containing any scenes, watermarks etc. +- IPP-03: It is recommended that an MVAS support characters being separate from one another. +- IPP-04: It is recommended that an MVAS support morphological pre-processing to enhance the characters in cases. + +### 7.3 Image and video processing + +#### 7.3.1 Recognition + +Recognition refers to a collection of related tasks (classification, detection) for identifying objects. In the real world, the objects in the image are aligned in different directions and such images are fed to the system. The scale of object will be varied when the distance is changed. The specifications for recognition are as follows: + +- IP-01: It is recommended that an MVAS support the function of selecting important information, eliminating redundant information, completing image classification and recognizing the object. +- IP-02: It is recommended that an MVAS support the function of adapting the variation of viewpoint, scale and deformation. +- IP-03: It is recommended that an MVAS support the function of predicting accurate values. + +#### 7.3.2 Inspection + +Inspection refers to the detection of various types of defects by inspecting parts for dozens of criteria and confirming quality. The specifications for inspection are as follows: + +- INS-01: It is recommended that an MVAS support extracting image features, including texture features, geometric shape features, colour features and etc. +- INS-02: It is required that MV support the further selection of features that are more conducive to classification from the feature set, that is, feature selection. + +- INS-03: It is required that MV support correctly classifying and recognizing the type of feature. + +#### 7.3.3 Positioning + +Positioning refers to a collection of related tasks to locate the position and direction of a component to identify whether the component position is within the specified tolerances or ensuring that the component is at the correct angle for verifying. The specifications for positioning are as follows: + +- POS-01: It is recommended that an MVAS support the detection of the positional relationship between the target and machine tool. +- POS-02: It is recommended that an MVAS support positioning accurately and instantly for factory automation. + +#### 7.3.4 Measurement + +Measurement is a task set to find a number that shows the size or amount of objects. When measuring objects, the processing technology of visual information mainly depends on image processing, which includes image filtering, image enhancement, edge extraction, refinement, feature extraction, image recognition and environment understanding. The specifications for measurement are as follows: + +- MEA-01: It is recommended that an MVAS support detecting the edge of an object; +- MEA-02: It is recommended that an MVAS support measurement, including but not limited to measurement of straight length, diameter, angle, radian and area; +- MEA-03: It is recommended that an MVAS support the results of measurement to meet accuracy and repeatability requirements; + +NOTE – Accuracy is the degree of conformance between a measurement of an observable quantity and a recognized standard or specification that indicates the true value of the quantity. Repeatability is the amount by which repeated measurements of the same quantity and same operating conditions conform. + +### 7.4 Algorithm training + +Before an algorithm can model a matter, it should learn the experience or environment represented by input data, which is called algorithm training. The specifications for algorithm training are as follows: + +- AT-01: It is recommended that an MVAS support algorithm training in a fully labelled dataset on classification and regression (supervised). +- AT-02: It is recommended that an MVAS support algorithm training in an unlabelled dataset on clustering, anomaly detection, association, auto-encoders etc. (fully unsupervised). +- AT-03: It is recommended that an MVAS support algorithm training in a hybrid of a labelled and unlabelled dataset to improve the accuracy and efficiency of the learning process (semi-supervised). + +### 7.5 Model evaluation + +The performance and accuracy of the model should be estimated before the model is deployed. Model evaluation can quickly find the problems in the process of model selection or training, and iteratively optimize the model. The specifications for model evaluation are as follows: + +- ME-01: It is recommended that an MVAS support the ability of identifying relationships and patterns between variables (accuracy). +- ME-02: It is recommended that an MVAS support the ability of understanding and measuring relevance (precision). +- ME-03: It is recommended that an MVAS support the ability of digesting new data and making predictions (generalization). + +- ME-04: It is recommended that an MVAS support the ability of storing knowledge gained while addressing one issue and applying it to a different but related issue (transferring). + +### 7.6 Security + +Data and systems providing an MVAS should be protected from unauthorized access, including preventing or reducing the probability of unauthorized access, use, disclosure, disruption, deletion, corruption, modification, inspection or recording. Information directly or indirectly contained in data involving individuals or organizations should not be disclosed and needs to be protected in the process of data collection, data storage, data query and analysis, data release, etc. The specifications for security are as follows: + +- SEC-01: It is recommended that an MVAS support mechanisms that protect against security threats, e.g., the discovery and registration of unauthorized endpoints, applications and services; +- SEC-02: It is recommended that an MVAS support policy setting (e.g., charging policies and data protection policies) on any request; +- SEC-04: It is recommended that an MVAS support protecting against security threats such as unnecessary service processing and/or fraudulent use of resources; +- SEC-05: It is recommended that an MVAS have capabilities to protect its infrastructure from malicious attacks, such as denial of service, eavesdropping, spoofing and tampering with messages; +- SEC-06: MVAS is required to support mechanisms for data confidentiality; +- SEC-07: MVAS is required to support mechanisms for data integrity; +- SEC-08: MVAS is required to support protection of the user's data, e.g., video capturing on the product line, profiles, presence, availability and location information. User data is required to be protected by revealing information only when valid authorization is provided. + +### 7.7 Identification, authentication and authorization + +Identification, authentication and authorization are methods for controlling users' access to their accounts. These methods ensure the traceability of behaviour and avoid unauthorized access The specifications for identification, authentication and authorization are as follows: + +- IAA-01: It is recommended that an MVAS support the authentication and authorization of endpoints accessing the system; +- IAA-02: It is recommended that an MVAS support the authentication and authorization of applications making use of resources; +- IAA-03: It is recommended that an MVAS support only authorized users being allowed to manage the system; +- IAA-04: It is recommended that an MVAS provide service access to authenticated users based on their access rights, profiles and policy; +- IAA-05: It is recommended that authentication and authorization, performed by the service provider, be processed securely. + +### 7.8 Accounting and charging + +Accounting and charging will enable the commercial operation of a system which offers services to be sold, in ways such as varying an algorithm or employing different security degrees. The specifications for accounting and charging are as follows: + +- A&C-01: An MVAS is required to enable all possible types of accounting arrangements, including transfer of accounting/charging information between providers; + +- A&C-02: It is recommended that an MVAS support various charging policies, including fixed rate charging, flow-based charging and event-base charging; +- A&C-03: It is recommended that an MVAS support mechanisms of accounting and charging management. + +### 7.9 Open service environment + +Using an open service generation and execution environment creates the need for rapid, effective and low-cost service development. This environment is a layered, modular and open business generation and execution environment. The specifications for an open service environment are as follows: + +- OSE-01: It is recommended that an MVAS support the flexible development of applications by MV service providers, application providers and other service providers; +- OSE-02: It is recommended that an MVAS support exposing capabilities via standard application interfaces (API); +- OSE-03: It is recommended that an MVAS support the integration of existing services with one another and with existing systems; +- OSE-04: It is recommended that an MVAS support application discovery; +- OSE-05: It is recommended that an MVAS support interfaces towards backend systems (e.g., charging, accounting, provisioning, operations and management, etc.); +- OSE-06: It is recommended that an MVAS support enabling deployment of any service by any authorized service provider; +- OSE-07: It is recommended that an MVAS support extensibility for future service and compatibility between these services. + +### 7.10 Network + +Networks carrying visual data should meet the specifications for MVAS quality of service. The specifications for the network are as follows: + +- NET-01: MVAS is required to support application-driven QoS; +- NET-02: MVAS is required to support QoS differentiation over various categories of traffic including packet-type flows and user designations; +- NET-03: MVAS is required to support performance measurement and management to ensure QoS; +- NET-04: It is recommended that an MVAS support QoS classes; +- NET-05: It is recommended that an MVAS support service/application priority. + +# 8 Reference model for supporting MV-based applications and services + +### 8.1 High level reference model + +The machine vision reference model is composed of seven functional domains: development domain, management domain, video acquisition domain, algorithm domain, service domain, capability exposure domain and security domain. + +![Figure 8-1: High-level reference model for supporting MV-based applications and services. The diagram is a hierarchical block diagram. At the top, the 'Application domain' includes Object sorting (OBS), Quality verification (QV), Robot guidance (RG), Object measurement (OM), and others. Below this, the 'Capabilities exposure domain' includes MV capabilities exposure service, Algorithm store, and MV portal management. The 'Management domain' includes Video equipment management and Camera management. The 'Service domain' includes Third MV algorithm service, OBS algorithm service, QV algorithm service, RG algorithm service, and OM algorithm service. The 'Algorithm domain' includes Algorithm library management, Third algorithm library, Model management (with OpenCV, YOLO, RCNN, and others), and Video acquisition domain (with Video storage function, Industrial camera SDK, and Video analysis processing function). The 'Development domain' includes Model acceleration tool, Algorithm visualization modeling (with Labeling and Modeling), Training, Evaluating, and Model compression tool. The 'Security domain' includes Data security, Application security, Platform security, and Equipment security.](fa859e4e468bfb2710a94527f2c504af_img.jpg) + +Figure 8-1: High-level reference model for supporting MV-based applications and services. The diagram is a hierarchical block diagram. At the top, the 'Application domain' includes Object sorting (OBS), Quality verification (QV), Robot guidance (RG), Object measurement (OM), and others. Below this, the 'Capabilities exposure domain' includes MV capabilities exposure service, Algorithm store, and MV portal management. The 'Management domain' includes Video equipment management and Camera management. The 'Service domain' includes Third MV algorithm service, OBS algorithm service, QV algorithm service, RG algorithm service, and OM algorithm service. The 'Algorithm domain' includes Algorithm library management, Third algorithm library, Model management (with OpenCV, YOLO, RCNN, and others), and Video acquisition domain (with Video storage function, Industrial camera SDK, and Video analysis processing function). The 'Development domain' includes Model acceleration tool, Algorithm visualization modeling (with Labeling and Modeling), Training, Evaluating, and Model compression tool. The 'Security domain' includes Data security, Application security, Platform security, and Equipment security. + +F.748.16(22) + +**Figure 8-1 – High-level reference model for supporting MV-based applications and services** + +### 8.2 Functional domain requirements + +The functions shown in Figure 8-1 are provided by MV applications on an MV platform. In order to reduce the cost and time to develop MV applications, an MV platform is expected to provide basic and common subfunctions for MV applications. For example, software functions of MV platform may include video data compression and camera control, and hardware functions may include video data storage (hard drive or SSD), camera and LAN interface to send camera data to other systems. In this subclause, the specifications of an MV platform are described. + +#### 8.2.1 Management domain + +The management domain consists of video equipment and camera management. The specifications for the management domain are as follows: + +- MD-01: It is recommended that an MV platform support remote control, maintenance, update, configuration and diagnosis of video equipment. +- MD-02: It is recommended that an MV platform support remote control, maintenance, update, configuration and diagnosis of cameras. + +#### 8.2.2 Video acquisition domain + +The video acquisition domain is responsible for accessing equipment, collecting equipment data and equipment data storage and management and industrial camera software development kit (SDK) management. The video signal output by the camera device is converted into binary digital information through a dedicated analogue and digital conversion device. Video acquisition involves multiple types of protocols, and the platform has supported video stream access management for protocols such as open network video interface forum (ONVIF) and real time streaming protocol (RTSP). The pictures saved in the video capture layer and the pushed videos can be used as the calculation data sources of other layers. The specifications for the video acquisition domain are as follows: + +- VAD-01: It is recommended that an MV platform support programmable control on the video acquisition device. +- VAD-02: It is recommended that an MV platform support configuration on video sampling information, such as frequency, resolution, etc. +- VAD-03: It is recommended that an MV platform support the pre-processing of video acquisition, such as compression, de-noise, etc. + +#### 8.2.3 Algorithm domain + +Machine vision applications are based on the analysis of video data and image data. Different applications and services need to establish different analysis models and match different algorithms. The algorithm domain provides algorithm library management and model management, supports common deep learning neural network algorithm models such as video scale selection network (VSSNet), residual networks (ResNet) and you only look once (YOLO), and supports third-party MV algorithm libraries to meet different business needs. The specifications for the algorithm domain are as follows: + +- AD-01: It is recommended that an MV platform support a multialgorithm framework, such as VSSNet, ResNet, YOLO etc. +- AD-02: It is recommended that an MV platform support management of different algorithms, such as algorithm scheduling, on-board, full lifecycle management, etc. +- AD-03: It is recommended that an MV platform support plug-in third party algorithms. + +#### 8.2.4 Service domain + +The service domain provides service management services such as object sorting algorithm service, quality verification algorithm service and robot guidance algorithm service. The platform customizes a batch of analysis task templates for MV scenarios, such as object analysis service templates and object inspection service templates. Users only need to set the template parameters to create an analysis task, which will be analysed and processed by the platform. The analysis task provides data source management, analysis task management, analysis engine, storage management, video analysis and image analysis functions. It may support parallel scheduling and intelligent flow control to ensure that massive code streams can be processed in a timely manner, and that the calculation factors are also in the optimal working state to ensure a prompt response of the entire link. The analysis task supports the storage of unstructured data and structured data after analysis, and provides management and control functions for computing tasks. The specifications for the service domain are as follows: + +- SD-01: It is recommended that an MV platform support algorithms for object sorting (OBS) services. +- SD-02: It is recommended that an MV platform support algorithms for quality verification (QV) services. +- SD-03: It is recommended that an MV platform support algorithms for robot guidance (RG) services. +- SD-04: It is recommended that an MV platform support algorithms for object measurement (OM) services. +- SD-05: It is recommended that an MV platform support algorithms for extended MV services. + +#### 8.2.5 Capability exposure domain + +Based on the exposure capabilities of the platform, an MV platform will open several interfaces. Open interface types may include an algorithm model interface, training service interface, analysis task interface and data service interface. The specifications for the capability exposure domain are as follows: + +- CED-01: It is recommended that an MV platform support an algorithm/model being called by third party MV applications (northbound). +- CED-02: It is recommended that an MV platform support calling an algorithm/model from third party MV platform (southbound). +- CED-03: It is recommended that an MV platform support algorithm training services being implemented from a third party MV services provider. + +#### **8.2.6 Security domain** + +The security domain guarantees data security, application security, platform security, access security and equipment security. The specifications for the security domain are as follows: + +- SecD-01: It is recommended that an MV platform support security protection for MV data. +- SecD-02: It is recommended that an MV platform support security protection for MV applications. +- SecD-03: It is recommended that an MV platform support security protection for itself. +- SecD-04: It is recommended that an MV platform support security protection for the MV endpoint. + +#### **8.2.7 Development domain** + +In this domain, a suite of set tools will be provided for MV-based model development such as model acceleration, model compression and model algorithm visualization including labelling, training, modelling and evaluating. The specifications for the development domain are as follows: + +- DD-01: It is recommended that an MV platform support labelling the training dataset. +- DD-02: It is recommended that an MV platform support optimizing the model. +- DD-03: It is recommended that an MV platform support model compression. +- DD-04: It is recommended that an MV platform support model acceleration. +- DD-05: It is recommended that an MV platform support the visualization of model development. +- DD-06: It is recommended that an MV platform support model compiling. +- DD-07: It is recommended that an MV platform support the model performance evaluation. + +## Appendix I + +### Typical scenarios of applications based on machine vision + +(This appendix does not form an integral part of this Recommendation.) + +Machine vision has been successfully applied in the field of industries, greatly improving the quality and productivity of products. The typical applications include object sorting, quality assessment, robot guidance and OM. + +### I.1 Object sorting + +In the application of MV, an object sorting application is based on identification and detection after an image is processed by an MV system, and object sorting is realized by a robot arm. Industrial production needs to be sorted according to the characteristics or/and quality requirements of the product, using an MV system for product data acquisition, data analysis and output results, and then through the robot, the corresponding materials or product can be placed in a fixed position, in order to achieve intelligent industrial production, modernization, automation. + +### I.2 Quality assessment + +Quality assessment is the application an MV system to simulate human vision by collecting images, storing and computing the images, and finally making quality judgements. MV can analyse and identify quality problems accurately. Machine vision quality assessment has been applied in many industries, in some specific applications such as product quality, assembly quality, print quality, food quality and so on. In the textile industry, MV is used to detect fabric quality. In the automotive industry, machine vision can be used to detect surface paint quality. In printing industries, MV can detect paper and print quality. In medical industries, MV can detect tablet and assembly quality. Surface detection is a common technique which refers to image quality as judged by face detection algorithms. While perceptual image quality usually depends on human observers, quality assessment depends on factors such as the face detection algorithm and performance measures such as recall and precision. + +### I.3 Robot guidance + +Robot guidance accurately detects the product and confirms its location. In semiconductor manufacturing, visual positioning is the most basic application in the MV industry. The problem is that adjustment pickers of chip position information are very difficult to control. MV can solve the need to pick up chips and bindings accurately. In specific operations, robot guidance achieves far greater speed and accuracy than manual positioning in tasks such as arranging parts on or off pallets, packaging parts off a conveyor belt, finding and aligning parts for assembly with other components, placing parts on a work shelf and removing parts from bins. In some cases, ultra-high-pixel images is required for detection on robot guidance. + +### I.4 Object measurement + +OM is the application of using a MV system for calculating the distances between two or more points in geometrical locations on an object and determines whether these measurements meet specifications. If not, the vision system sends a fail signal to the machine controller, triggering a reject mechanism that ejects the object from the line. In practice, a fix-mounted camera captures images of parts when passing the camera's field of view and the system software will calculate distances between various points in the images. Because many MV systems can measure object features within 0.0254 millimetres, a number of applications traditionally handled by contact gauging are possible. Gauging objects with machine vision has advantages as not only the product is measured without wear, but also the possibility of causing secondary damage to the product can be solved compared + +with traditional manual contact with the product for gauging purposes; this is a particularly obvious improvement in the level of precision manufacturing. Because of the requirements of high precision and high speeds, measurements of threads, drills, integrated circuit component pins, car parts, plugins, etc. are very common applications. + +## Bibliography + +- [b-ITU-T H.350] ITU-T H.350 (2011), *Directory services architecture for multimedia conferencing*. +- [b-ITU-T Y.101] ITU-T Y.101 (2000), *Global Information Infrastructure terminology: Terms and definitions*. +- [b-ISO/IEC 11557] ISO/IEC 11557:1992, *3,81 mm wide magnetic tape cartridge for information interchange – Helical scan recording – DDS-DC format using 60 m and 90 m length tapes*. + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.20-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.748.20-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..25f8fbfaef317942ed3102c74b40a7d42f80608b --- /dev/null +++ b/marked/F/T-REC-F.748.20-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:83cab7dcabfe939b223d97242d8523e461968d21f7c5432693c30a64f2d95521 +size 7050 diff --git a/marked/F/T-REC-F.748.20-202212-I_PDF-E/4e4be0bd8b235167902f2c03e41da651_img.jpg b/marked/F/T-REC-F.748.20-202212-I_PDF-E/4e4be0bd8b235167902f2c03e41da651_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..6cf2c973e9c0f3e7e8f209545c304e31d4d87f83 --- /dev/null +++ b/marked/F/T-REC-F.748.20-202212-I_PDF-E/4e4be0bd8b235167902f2c03e41da651_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9582f4c33a531db2151a0bb0c8d634fb5c6fc005481bec046b4ed6354fdf8c1b +size 82770 diff --git a/marked/F/T-REC-F.748.20-202212-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.748.20-202212-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..27a6f004ae0471d828be6cf422f3207eded123f7 --- /dev/null +++ b/marked/F/T-REC-F.748.20-202212-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4caa4c0677c643eefd89bacc86fc05ab311c23ab1746567aebec3037524f61b3 +size 177914 diff --git a/marked/F/T-REC-F.748.20-202212-I_PDF-E/raw.md b/marked/F/T-REC-F.748.20-202212-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..3d96d438e30668d1994e2f4ef5bad1a4763395be --- /dev/null +++ b/marked/F/T-REC-F.748.20-202212-I_PDF-E/raw.md @@ -0,0 +1,296 @@ + + +# Recommendation **ITU-T F.748.20 (12/2022)** + +SERIES F: Non-telephone telecommunication services +Multimedia services + +--- + +## **Technical framework for deep neural network model partition and collaborative execution** + +![ITU logo](390120de4fe440c42fea8154fcaad334_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +## Recommendation ITU-T F.748.20 + +# Technical framework for deep neural network model partition and collaborative execution + +## Summary + +The deep neural network (DNN) model inference process usually requires a large amount of computing resources and memory. Therefore, it is difficult for end devices to perform DNN models independently. An effective way to implement end-edge collaborative DNN execution is through DNN model partition, which can reduce latency and improve resource utilization at the same time. Recommendation ITU-T F.748.20 aims to specify the technical framework of DNN model partition and collaborative execution. First, it is necessary to predict the overall inference latency under the current system state according to different DNN partition strategies in advance. Then, it is necessary to choose the appropriate partition locations and collaborative execution strategy based on the equipment computation capabilities, network status and DNN model properties. Finally, the model collaborative execution is implemented and the resource allocation optimized in the meantime. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.748.20 | 2022-12-14 | 16 | 11.1002/1000/15197 | + +## Keywords + +Collaborative execution, deep neural network, model partition. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2023 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|------------------------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 1 | +| 5 Conventions ..... | 2 | +| 6 Overview of DNN model partition and collaborative execution..... | 2 | +| 7 Latency prediction model establishment of DNN model partition..... | 3 | +| 7.1 System state monitoring and parameter configuration ..... | 3 | +| 7.2 Modelling and analysis of latency prediction..... | 4 | +| 8 Partition strategy formulation of DNN model ..... | 4 | +| 8.1 Latency analysis of different partition strategies under diverse system state..... | 4 | +| 8.2 Partition position confirmation and resource allocation..... | 4 | +| 9 Collaborative execution of DNN model after DNN model partition ..... | 5 | +| 9.1 DNN model deployment based on partition strategy ..... | 5 | +| 9.2 Inference process of collaborative execution ..... | 5 | +| Appendix I – Typical scenarios for DNN model partition and collaborative execution ..... | 6 | +| I.1 Smart home ..... | 6 | +| I.2 Industry manufacture ..... | 6 | +| I.3 Intelligent transportation..... | 7 | + + + +## Technical framework for deep neural network model partition and collaborative execution + +# 1 Scope + +This Recommendation aims to specify the technical framework of deep neural network (DNN) model partition and collaborative execution, including the establishment of a latency prediction model, the formulation of a DNN model partition strategy, resource allocation optimization and the collaborative execution of the model, so as to meet the demand of DNN model inference and realize the efficient resource utilization between the end devices and the edge device. + +The scope of this Recommendation includes: + +- Overview of DNN model partition and collaborative execution; +- Latency prediction model establishment of a DNN model partition; +- Partition strategy formulation of a DNN model; +- Collaborative execution of a DNN model after DNN model partition. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T F.748.11] Recommendation ITU-T F.748.11 (2020), *Metrics and evaluation methods for deep neural network processor benchmark*. + +[ITU-T F.748.13] Recommendation ITU-T F.748.13 (2021), *Technical framework for the shared machine learning system*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +None. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 deep neural network model partition:** The deep neural network model is partitioned into two or more parts according to partition strategies, and each part selects its execution location (i.e., end/edge device). + +**3.2.2 deep neural network model inference:** All deep neural network layers are executed in sequence, and the inference result is obtained after completing the last deep neural network layer. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|--------|-------------------------------------| +| ADA | Adaboost | +| ANN | Artificial Neural Network | +| DNN | Deep Neural Network | +| DT | Decision Tree | +| GBRT | Gradient Boosting Regression Tree | +| KNN | K-Nearest Neighbour | +| KRR | Kernel Ridge Regression | +| LR | Linear Regression | +| MAE | Mean Absolute Error | +| MAPE | Mean Absolute Percentage Error | +| PII | Personally Identifiable Information | +| RANSAC | Random Sample Consensus regression | +| RF | Random Forest | +| SVM | Support Vector Machine | + +# 5 Conventions + +In this Recommendation: + +- The keywords "**is required to**" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "**is recommended**" indicate a requirement which is recommended but which is not absolutely required. Thus this requirement needs not be present to claim conformance. + +# 6 Overview of DNN model partition and collaborative execution + +The overall framework of DNN model partition and end-edge collaborative execution is shown in Figure 1. The overall framework consists of the edge device and multiple end devices, which includes the computation latency predictor module, network monitoring module, model processor module, DNN partitioner module, request processor module and deployment tool module. The deployment tool module is used to collect the training latency under different system configurations and control the CPU and memory usage of end devices and the edge device. The model processor module is used to analyse and extract the DNN model topology, static features of DNN layers, DNN output data size and so on. The computation latency predictor module is used to predict the computation latency of each DNN layer using the metrical data collected by the deployment tool module, which takes multiple static and dynamic system characteristics into account. The network monitoring module is used to monitor the network upload/download bandwidth in real time and to estimate the transmission latency of DNN layers. The DNN partitioner module is used to determine the appropriate partition locations for end-edge collaborative execution according to the feedback information of the model processor module, computation latency predictor module and network monitoring module. The request processor module is used to process the real-time request of DNN model inference from end devices, which controls the end devices and edge device to collaboratively complete the DNN inference according to the partition scheme. + +![Figure 1: The overall framework of DNN model partition and collaborative execution. The diagram shows three 'End device' units on the left, each containing a 'Collaborative inference client'. Each client consists of a 'Computation latency predictor', 'Model processor', 'Request processor' (with 'Predictor model execution'), 'DNN predictor', 'Network monitoring service', and a 'Deployment tool'. An 'APP' is connected to the 'Request processor' in each client. On the right is an 'Edge device' containing a 'Collaborative inference server'. The server includes a 'Request processor' (with 'Predictor model execution'), a 'Deployment tool', and a 'Computation latency predictor'. Dashed arrows indicate data flow from the 'Request processor', 'DNN predictor', and 'Deployment tool' of each client to the 'Request processor', 'Predictor model execution', and 'Deployment tool' of the edge server. The label 'F.748.20(22)' is at the bottom right.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +Figure 1: The overall framework of DNN model partition and collaborative execution. The diagram shows three 'End device' units on the left, each containing a 'Collaborative inference client'. Each client consists of a 'Computation latency predictor', 'Model processor', 'Request processor' (with 'Predictor model execution'), 'DNN predictor', 'Network monitoring service', and a 'Deployment tool'. An 'APP' is connected to the 'Request processor' in each client. On the right is an 'Edge device' containing a 'Collaborative inference server'. The server includes a 'Request processor' (with 'Predictor model execution'), a 'Deployment tool', and a 'Computation latency predictor'. Dashed arrows indicate data flow from the 'Request processor', 'DNN predictor', and 'Deployment tool' of each client to the 'Request processor', 'Predictor model execution', and 'Deployment tool' of the edge server. The label 'F.748.20(22)' is at the bottom right. + +**Figure 1 – The overall framework of DNN model partition and collaborative execution** + +# 7 Latency prediction model establishment of a DNN model partition + +## 7.1 System state monitoring and parameter configuration + +The DNN model consists of many types of neural network layer, such as the convolutional layer, the fully connected layer, the pooling layer and the activation layer. Each type of neural network layer has different parameter configurations and is required to establish different latency prediction models. The convolutional layer and fully connected layer are the most common layers. The main parameters of the convolutional layer include input size, kernel size, channel size, stride and padding. The main parameters of the fully connected layer include input size and output size. + +In addition to layer configuration, the system state also significantly affects the layer computation latency. It is required to monitor the dynamic system state, including CPU usage, free memory size, memory usage, network bandwidth, network topology and so on. These dynamic system states are required to be accurately and timely monitored, in order to profile the latency prediction model. + +## 7.2 Modelling and analysis of latency prediction + +It is required to establish the computation latency prediction model for each type of neural network layer. The input data of the latency prediction model includes the layer parameter configuration and system state. The output data of the latency prediction model is the layer computation delay given the current layer parameter configuration and system state. There are many latency prediction models, such as linear regression (LR), random sample consensus regression (RANSAC), kernel ridge regression (KRR), k-nearest neighbour (KNN), decision tree (DT), support vector machine (SVM), random forest (RF), adaboost (ADA), gradient boosting regression tree (GBRT) and artificial neural network (ANN). It is required to compare multiple latency prediction models and determine the most suitable model. The evaluation indicators include mean absolute error (MAE) and mean absolute percentage error (MAPE). + +# 8 Partition strategy formulation of a DNN model + +## 8.1 Latency analysis of different partition strategies under diverse system state + +The overall execution latency of each end device consists of three parts: the local computation latency on the end device, the communication latency between the end and edge devices, and the edge computation latency at the edge device. Different DNN model partition strategies can impact the overall execution latency. Specifically, the end device computation capability and the amount of DNN layers executed locally determine the local computation latency. The allocated communication resource amount and the intermediate data size determine the communication latency. The amount of the allocated edge computation resource and remaining DNN layers determine the edge computation latency. When the system state such as the amount of the allocated communication and computation resources changes, the partition strategies are required to change correspondingly, aiming to realize the trade-off between computation latency and communication latency. + +## 8.2 Partition position confirmation and resource allocation + +According to the latency analysis of different partition strategies under a diverse system state, it is required to jointly optimize a DNN partition strategy and resource allocation, so as to minimize latency or energy consumption. The DNN model partition strategy refers to divide the DNN model into two or more parts by deciding appropriate partition locations and selecting the execution location (i.e., end/edge device) for each DNN layer. DNN models can be divided into two categories: chain DNN and non-chain DNN. It is required to make different partition strategies for two types of DNN model. Resource allocation mainly involves the allocation of communication and computation resources of the edge device. + +The partition location is determined by the partition strategy. Considering the framework in Figure 1, multiple end devices cooperate with a resource-constrained edge device. Each end device is required to make its partition decision according to the computation capabilities of the end devices and the allocated edge resources. If the computation capability of the end device is sufficient, its DNN model is not partitioned and does not occupy edge resources, in which the overall DNN model is all executed by the end device. However, if the computation capability of the end device is insufficient, its DNN model will be partitioned and jointly executed by the end device and the edge device. + +Furthermore, the partition strategy is required to be handled jointly with resource allocation among end devices. Specifically, the edge device is required to efficiently allocate its computation resources to each end device for executing the offloaded DNN layers from end devices, and is required to allocate its communication resources to each end device for intermediate data transmission. + +# **9 Collaborative execution of a DNN model after DNN model partition** + +## **9.1 DNN model deployment based on partition strategy** + +After determining the DNN model partition strategy, the execution location of each DNN layer will be obtained, either at the edge device or on the end device. Due to the diversity and variation of DNN models, it is not practical to deploy all DNN models at the edge device in advance. + +There are two options to deploy the DNN models on demand based on the partition strategy. The first option is to deploy the entire required DNN model at the edge device. The second option is to deploy the DNN layers which are partitioned to the edge device for execution. The DNN model execution can be performed only after the DNN model or the required DNN layers are deployed. + +## **9.2 Inference process of collaborative execution** + +Each DNN layer is sequentially executed either at the edge device or on the end device according to the partition strategy [ITU-T F.748.13]. Each DNN layer receives the output of its direct antecedent layers as its own input, performs DNN execution and delivers its output to its direct subsequent layers. If the execution locations of the antecedent layer and its direct subsequent layer are different (i.e., the antecedent layer is executed at the edge device and its direct subsequent layer is executed on the end device, or the antecedent layer is executed on the end device and its direct subsequent layer is executed at the edge device), the output of the antecedent layer is required to be transmitted to its direct subsequent layer via communication links. If the execution locations of the antecedent layer and its direct subsequent layer are the same (i.e., both the antecedent layer and its direct subsequent layer are executed by the end device or the edge device), the output of the antecedent layer will be directly fed into its direct subsequent layer. Only when the subsequent layer receives the output of its direct antecedent layer can the subsequent layer be executed. The inference process completes when all DNN layers have been executed. + +## Appendix I + +## Typical scenarios for DNN model partition and collaborative execution + +(This appendix does not form an integral part of this Recommendation.) + +### I.1 Smart home + +Many smart home applications, such as home security and video surveillance, are part of our home life. These applications need to execute DNN models such as computer vision, video recognition and neural machine translation. However, most home devices, such as mobile phones, cameras and various sensors, are not capable of performing DNN models on their own, resulting in high execution latency and poor user experience. End-edge collaborative DNN execution is suitable to apply in this scenario. Edge servers and base stations are edge devices and cooperate with home devices to execute the DNN models, enabling to achieve lower execution latency and protect personally identifiable information (PII) in the meantime. + +### I.2 Industry manufacture + +The intelligent demand of industrial manufacture is increasing. Intelligence is the key to realize digital transformation of industrial manufacture. The industrial process requires data analysis and intelligent decision capabilities, which need to depend on various intelligent applications and corresponding DNN models. For example, intelligent monitoring, image recognition and other applications are applied to monitor the production safety situation and eliminate security risks. In order to guarantee the safety and timeliness of intelligent industrial production, it is required to implement collaborative DNN execution through DNN model partition. + +### I.3 Intelligent transportation + +![Diagram illustrating DNN model partition and collaborative execution in intelligent transportation applications. It shows three application categories (Traffic flow, Pedestrian management; Station checkpoint, Road tolls; Violation monitoring, Speed monitoring) connected to three end devices (Infrared sensing, Traffic checkpoint, Camera). These end devices are then connected to a central Edge device which contains various processing units: Video surveillance unit, Image recognition unit, Monitoring screens, Storage devices, Computing and decision systems, and Sensor analysis unit.](4e4be0bd8b235167902f2c03e41da651_img.jpg) + +The diagram illustrates the architecture of an intelligent transportation system. On the left, under the heading 'Applications', there are three groups of use cases: 'Traffic flow' and 'Pedestrian management'; 'Station checkpoint' and 'Road tolls'; and 'Violation monitoring' and 'Speed monitoring'. Each group is connected by an arrow to a corresponding 'End device' in the middle column. These end devices are 'Infrared sensing', 'Traffic checkpoint', and 'Camera', each enclosed in a red dashed box. From each end device, a dashed arrow points to a large 'Edge device' on the right, also enclosed in a red dashed box. The edge device contains a list of components: 'Video surveillance unit', 'Image recognition unit', 'Monitoring screens', 'Storage devices', 'Computing and decision systems', and 'Sensor analysis unit'. A small label 'F.748.20(22)' is located at the bottom right of the diagram. + +Diagram illustrating DNN model partition and collaborative execution in intelligent transportation applications. It shows three application categories (Traffic flow, Pedestrian management; Station checkpoint, Road tolls; Violation monitoring, Speed monitoring) connected to three end devices (Infrared sensing, Traffic checkpoint, Camera). These end devices are then connected to a central Edge device which contains various processing units: Video surveillance unit, Image recognition unit, Monitoring screens, Storage devices, Computing and decision systems, and Sensor analysis unit. + +**Figure I.1 – DNN model partition and collaborative execution in intelligent transportation application** + +Many intelligent traffic applications, such as traffic flow monitoring, pedestrian management, station intelligent checkpoint, road tolls, violation and speed monitoring, are common in our life. These applications need to execute DNN models realize image recognition, semantic segmentation, such as face recognition, licence plate recognition, behaviour monitoring and traffic flow management. + +Intelligent transportation applications require various kinds of intelligent end device, such as infrared sensing, traffic checkpoint and intelligent camera, which have some computing power. However, the current execution method depends on edge devices only. Edge devices have to independently process a huge amount of data, bringing great pressure to the hardware. Traditional DNNs do not fully perform in terms of the computing and analysis ability of end devices. Therefore, the system has a relative low collaboration efficiency and high execution latency. In order to fully utilize the computing ability of each device, DNN model partition and collaborative execution strategies are applied in allocating some DNN layers from edge devices to the aimed end device. End and edge devices handle computing and analysis tasks collaboratively, which improves the execution efficiency. By deploying suitable partition strategies for end-edge collaboration, the intelligent traffic system (shown in Figure I.1) satisfies the needs of intelligent transportation applications. + +Some typical intelligent transportation applications are as follows. a) Infrared sensors together with other sensors and large screens could monitor the vehicles and pedestrian flow, b) traffic checkpoint collaborates with image recognition unit capture and compute massive kinds of information on persons and vehicles for traffic management, c) cameras and storage devices, computing and decision systems collaborate to monitor violations and speed. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.21-202212-I_PDF-E/2236272d3b3db6e6363337f5a8db72f6_img.jpg b/marked/F/T-REC-F.748.21-202212-I_PDF-E/2236272d3b3db6e6363337f5a8db72f6_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..cd7b10b09475df9cb82f28067c5caf7d2f9692ec --- /dev/null +++ b/marked/F/T-REC-F.748.21-202212-I_PDF-E/2236272d3b3db6e6363337f5a8db72f6_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2f0859dbcb63873ef493cda131f8c07be79fcc7894df30c1c7917166e6889fa9 +size 87682 diff --git a/marked/F/T-REC-F.748.21-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.748.21-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..abda52cbb1047d6649f89e40c898a1c33bed7267 --- /dev/null +++ b/marked/F/T-REC-F.748.21-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6f3b7e07231c0f3e116c99c493bdda98ecd52313b6735faa6cea23815cf3162b +size 7190 diff --git a/marked/F/T-REC-F.748.21-202212-I_PDF-E/3f1987804d7d78bc3b3bc560e974280a_img.jpg b/marked/F/T-REC-F.748.21-202212-I_PDF-E/3f1987804d7d78bc3b3bc560e974280a_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ab91639a4c98a076513f54ebd635f158600ec4aa --- /dev/null +++ b/marked/F/T-REC-F.748.21-202212-I_PDF-E/3f1987804d7d78bc3b3bc560e974280a_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c624ce9e2d4105ee9793983863c8a2e16135d3af2f4bb4031013fd8a426197cd +size 88290 diff --git a/marked/F/T-REC-F.748.21-202212-I_PDF-E/acfc53eca625d62b38aa2563efa95c3e_img.jpg b/marked/F/T-REC-F.748.21-202212-I_PDF-E/acfc53eca625d62b38aa2563efa95c3e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..635169906cca154d0c2b461991bc5d4fa230b3fc --- /dev/null +++ b/marked/F/T-REC-F.748.21-202212-I_PDF-E/acfc53eca625d62b38aa2563efa95c3e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:211b69805fce3e077121e8c678acbfb51c734947d3f3f7a7932ed2f98a16d2b6 +size 15269 diff --git a/marked/F/T-REC-F.748.21-202212-I_PDF-E/af6be343f0c0a8f155f965dcf337b8af_img.jpg b/marked/F/T-REC-F.748.21-202212-I_PDF-E/af6be343f0c0a8f155f965dcf337b8af_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..68e9470074b4cd0f5ca7486b5f4c7d0651e41e58 --- /dev/null +++ b/marked/F/T-REC-F.748.21-202212-I_PDF-E/af6be343f0c0a8f155f965dcf337b8af_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:014fa996e182f7a9001a0d02c03fa9e0a3ecb3de03dea7d99fbd6f734723554b +size 26103 diff --git a/marked/F/T-REC-F.748.21-202212-I_PDF-E/raw.md b/marked/F/T-REC-F.748.21-202212-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..9d147af7e4132b05de499efc1abadd513c3c3578 --- /dev/null +++ b/marked/F/T-REC-F.748.21-202212-I_PDF-E/raw.md @@ -0,0 +1,717 @@ + + +# Recommendation **ITU-T F.748.21 (12/2022)** + +SERIES F: Non-telephone telecommunication services +Multimedia services + +--- + +## **Requirements and framework for feature-based distributed intelligent systems** + +![ITU logo](390120de4fe440c42fea8154fcaad334_img.jpg) + +The logo of the International Telecommunication Union (ITU), featuring the letters 'ITU' in blue inside a circular emblem with a globe and network lines. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessages service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.21 + +# Requirements and framework for feature-based distributed intelligent systems + +## Summary + +Recommendation ITU-T F.748.21 introduces a classification of features and framework for feature-based distributed intelligent systems relevant to intelligent scenarios, specifying the service requirement, functional requirements and security requirements for feature-based distributed intelligent systems. Use cases for such systems are also described. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.748.21 | 2022-12-14 | 16 | 11.1002/1000/15198 | + +## Keywords + +Feature classification, intelligent system, requirements. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2023 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|-------------|----------------------------------------------------------------------------|----| +| 1 | Scope ..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 1 | +| 4 | Abbreviations and acronyms ..... | 1 | +| 5 | Conventions ..... | 2 | +| 6 | Overview ..... | 2 | +| 7 | Classification of features in intelligent systems ..... | 3 | +| 7.1 | Signal domain features ..... | 3 | +| 7.2 | Latent domain features ..... | 6 | +| 7.3 | Semantic feature ..... | 7 | +| 8 | Service requirements of feature-based distributed intelligent systems..... | 8 | +| 8.1 | General requirement ..... | 8 | +| 8.2 | Data acquisition ..... | 8 | +| 8.3 | Pre-processing ..... | 8 | +| 8.4 | Data representation ..... | 9 | +| 8.5 | Data transmission ..... | 9 | +| 8.6 | Feature analysis ..... | 9 | +| 8.7 | Management ..... | 9 | +| 9 | Framework of feature-based intelligent systems ..... | 9 | +| 9.1 | Front-end equipment domain ..... | 9 | +| 9.2 | Data domain..... | 10 | +| 9.3 | Algorithm domain ..... | 10 | +| 9.4 | Service domain ..... | 10 | +| 9.5 | Application domain ..... | 11 | +| 9.6 | Development domain..... | 11 | +| 9.7 | Management domain ..... | 11 | +| 9.8 | Security domain..... | 12 | +| Appendix I | – Use cases of feature-based intelligent applications ..... | 14 | +| I.1 | Smart city..... | 14 | +| I.2 | ITS ..... | 14 | +| I.3 | Content rating ..... | 14 | +| I.4 | Smart manufacturing ..... | 14 | +| I.5 | Smart agriculture ..... | 14 | +| Appendix II | – Performance of intelligent tasks ..... | 15 | +| II.1 | Smart city..... | 15 | + +| | Page | +|---------------------------------------------------------|-------------| +| II.2 Smart manufacturing ..... | 15 | +| Appendix III – Relation between tasks and features..... | 17 | +| Bibliography..... | 18 | + +# Recommendation ITU-T F.748.21 + +## Requirements and framework for feature-based distributed intelligent systems + +# 1 Scope + +This Recommendation introduces a classification of features and framework for feature-based distributed intelligent systems relevant to intelligent scenarios, specifying the service requirement, functional requirements, performance requirements and security requirements for feature-based distributed intelligent systems. + +The scope of this Recommendation includes: + +- Classification of features in intelligent systems; +- Framework of feature-based distributed intelligent systems; +- Functional requirements for feature-based distributed intelligent systems; +- Security requirements for feature-based distributed intelligent systems; +- Use cases of feature-based intelligent applications. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +# 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 front-end equipment** [b-ITU-T K.142]: Outdoor camera and associated ancillary equipment. + +**3.1.2 personally identifiable information** [b-ISO/IEC 29100]: Any information that (a) can be used to identify the PII principal to whom such information relates, or (b) is or might be directly or indirectly linked to a PII principal. + +NOTE – To determine whether a PII principal is identifiable, account should be taken of all the means which can reasonably be used by the privacy stakeholder holding the data, or by any other party, to identify that natural person. + +### 3.2 Terms defined in this Recommendation + +None. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +AI Artificial Intelligence + +| | | +|-------|-----------------------------------------------| +| BRIEF | Binary Robust Independent Elementary Features | +| FAST | Features from Accelerated Segment Test | +| FDIS | Feature-based Distributed Intelligent Systems | +| FTP | File Transfer Protocol | +| HOG | Histogram Of Gradient | +| IoT | Internet of Things | +| IPC | Internet Protocol Camera | +| ITS | Intelligent Transportation System | +| PII | Personally Identifiable Information | +| RoI | Region of Interest | +| RTP | Real Time Transport Protocol | +| RTSP | Real Time Streaming Protocol | +| SIFT | Scale-Invariant Feature Transform | +| SURF | Speeded-Up Robust Features | +| UDP | User Datagram Protocol | +| YOLO | You Only Look Once | + +# 5 Conventions + +In this Recommendation: + +- The keywords "**is required to**" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "**is recommended**" indicate a requirement which is recommended but which is not absolutely required. Thus this requirement needs not be present to claim conformance. +- The keyword "**may**" indicates an optional requirement which is permissible, without implying any sense of being recommended. This term is not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +# 6 Overview + +A feature-based intelligent system extracts and transmits features instead of original data, i.e., video, audio, text and point cloud. Compared with the intelligent systems based on original multimedia data, which require a large amount of storage and transmission bandwidth, and limited technological solutions use multimedia coding to reduce data volume; it could save bandwidth by extracting effective information from multimedia data. + +In some latency sensitive intelligent systems, e.g., autonomous driving and industrial detection and control, it is crucial to process the data in real-time. Using features in addition to original data might allow fast searching and other AI tasks. A feature-based intelligent system is able to reduce latency in some scenarios. + +With the recent advance of hardware technologies, the front-end equipment is capable of performing more calculations than before, and part of the computation load can be moved to the front end, rather than having most works done on the cloud side, with the front-end equipment performing only as data collectors and data transmitters. The end-to-end systems in the intelligent system can be divided into + +feature extraction and feature-based intelligent applications, the former deployed at the front end and the latter deployed at the back end, and the feature, i.e., an intermediate layer of the neural network, is extracted and communicated to offload the computing resource of the cloud side, balancing the computational load at the back end especially when the number of devices grows explosively. Figure 6-1 shows an end-to-end pipeline supporting intelligent applications based on features. + +![Figure 6-1: An end-to-end pipeline of feature-based application. The diagram shows four sequential boxes: Data acquisition, Feature extraction, Feature transmission, and Intelligent applications, connected by arrows. The reference F.748.21(22) is noted below the last box.](acfc53eca625d62b38aa2563efa95c3e_img.jpg) + +``` +graph LR; A[Data acquisition] --> B[Feature extraction]; B --> C[Feature transmission]; C --> D[Intelligent applications]; +``` + +F.748.21(22) + +Figure 6-1: An end-to-end pipeline of feature-based application. The diagram shows four sequential boxes: Data acquisition, Feature extraction, Feature transmission, and Intelligent applications, connected by arrows. The reference F.748.21(22) is noted below the last box. + +**Figure 6-1 – An end-to-end pipeline of feature-based application** + +The fundamental process of feature-based distributed intelligent systems is as follows. First, the original data is fed into the feature extraction module, the data is processed and the different types of features are extracted according to the application configuration. Then the feature is compressed and packed for transmission and storage. Later, the intelligent application module receives the encoded bitstream and decodes it into original data or features. Last, the AI task is performed based on the selected feature. Use cases are presented in Appendix I. + +## **7 Classification of features in intelligent systems** + +Features are widely used in machine vision, image processing and audio recognition, and so on. They are designed or learned to describe information or properties of the whole or partial content of an image, a video or an audio segment. + +Various types of features are used in feature-based distributed intelligent systems (FDISs) to support AI tasks. The features are classified as signal domain features, latent domain features and semantic features. + +For each classification, some of the commonly used features in AI tasks are listed in this section. The features extracted and transmitted will be selected and configured according to the combination of AI tasks supporting intelligent applications. The mapping of the AI task and features supporting the given AI task can be found in Appendix III. + +The signal domain feature focuses on the colour and texture structure of an image, e.g., colour histogram, key points, edges and shapes, or motions and optical flows of a video, or signal properties of an audio segment, e.g., temporal shape, energy and spectral shape. The latent domain features describe the key information related to the semantics of the raw data. The semantic features describe semantic information, e.g., object, classification, velocity and action. + +### **7.1 Signal domain features** + +Based on the types of raw data from which the features are extracted, the signal domain features can be classified into two types: visual features and audio features. Visual features are extracted from the visual data such as images, videos or point cloud. Audio features are extracted from the audio signal. + +#### **7.1.1 Visual features** + +##### **7.1.1.1 Colour features** + +Colour features are extensively used for the extraction of visual features and are typically used in colour index-based image retrieval. Colour features are invariant to the translation or viewing angle. There are large numbers of colour features which are discussed in the literature including colour percentage, colour correlogram, colour histogram, colour coherence vector (CCV), and colour moments. Among these, colour moment is the simplest and most effective feature. + +Colour features are extracted by calculating the statistical information of the pixels' colour distribution in the image. The features are represented in the vector format, in which each dimension represents one type of statistical information, such as value counts, mean value, standard derivation and so on. + +##### **7.1.1.2 Edge features** + +Edge features indicate the area where the image brightness changes sharply or has discontinuities. Edge features are used to mark the boundaries of objects or surfaces, which can significantly reduce the amount of data recommended to be processed and also filter out information that may be regarded as relevant areas. + +Edge features are typically represented in the format of an image mask that has value in each position indicating whether a pixel belongs to an edge or not. The popular methods to detect edge features are Canny and Sobel. + +##### **Canny** + +Canny edge detector is an edge detection operator that uses a multistage algorithm to detect a wide range of edges in images. + +##### **Sobel** + +Sobel operator is a discrete differentiation operator, computing an approximation of the gradient of the image intensity function. + +##### **7.1.1.3 Corner features** + +Corner features are also known as interest point features, which indicate the intersection of two edges. Corner features have well-defined positions and can be robustly detected. + +Corner features are represented in the format of a vector that records the spatial position of the corner or interest point. A bundle of corner features can be used as an index of an image in the image retrieval task. The common methods to detect corner features are Harris corner detection and features from accelerated segment test (FAST) corner detection. + +###### **Harris corner detection** + +The Harris corner detector is a corner detection operator for extracting corners and inferring features of an image. It takes the differential of the corner score into account with reference to direction. + +###### **FAST: Features from accelerated segment test** + +FAST is a corner detection method. It is faster than many other well-known feature extraction methods, such as the difference of Gaussians (DoG) used by the scale-invariant feature transform (SIFT) and Harris detectors. The FAST corner detector is suitable for real-time video processing applications given its high-speed performance. + +##### **7.1.1.4 Feature descriptions** + +For an object in an image, descriptions of key points or interesting points can be used to identify the object while in image retrieval. To form a reliable feature description that is agnostic to image scale, rotation and illumination change, a feature description with complex information such as key points, scale and gradients is designed. + +The feature descriptions are represented in a data structure consisting of spatial position, scale and description vector and so on. Typically, a bundle of feature descriptions is used together to identify a single object or a whole image. The common feature description generation methods are SIFT, speeded-up robust features (SURF), histogram of gradient (HOG), and binary robust independent elementary features (BRIEF). + +###### SIFT + +SIFT is a classical feature with the property of invariance to scale, rotation, luminance and viewport. Its robustness makes it widely used in image retrieval, object detection and image coding. + +SIFT is defined to describe the keypoint inside an image by performing the following operations: + +###### *i) Keypoint detection in scale pyramid* + +To detect keypoints invariant to scale, a scale pyramid is built by filtering the input image with the Gaussian filter with a different scale. The scale pyramid generated by a Gaussian filter with different parameters is referred to as an octave. The difference between the Gaussian space is derived and keypoints are detected in each different space in each of the octaves. + +###### *ii) Gradient computation for keypoints* + +To ensure rotational invariance, gradients of each keypoint are derived by comparing the pixel values in the window centred by the keypoint. + +###### *iii) Feature description of keypoints* + +Gradients in the window centred by the keypoint are gathered together to form a gradient vector, which serves as the feature of the keypoint. + +###### SURF + +SURF is partly inspired by the SIFT descriptor. The standard version of SURF is several times faster than SIFT. To detect interest points, SURF uses an integer approximation of the determinant of Hessian blob detector, which can be computed with three integer operations using a precomputed integral image. Its feature descriptor is based on the sum of the Haar wavelet response around the point of interest. It can also be computed with the aid of the integral image. + +##### HOG + +HOG first counts occurrences of gradient orientation in localized portions of an image. The second step is orientation binning by creating the cell histograms. Each pixel within the cell casts a weighted vote for an orientation-based histogram channel based on the values found in the gradient computation. To account for changes in illumination and contrast, the gradient strengths must be locally normalized, which requires grouping the cells together into larger, spatially connected blocks. The HOG descriptor is then the concatenated vector of the components of the normalized cell histograms from all of the block regions. + +##### BRIEF + +BRIEF converts image patches into a binary feature vector so that together they can represent an object. In BRIEF, each key point is described by a feature vector which is a 128–512 bit string. BRIEF deals with the image at the pixel level so it is very noise-sensitive. By pre-smoothing the patch, this sensitivity can be reduced, thus increasing the stability and repeatability of the descriptors. BRIEF uses a Gaussian kernel for smoothing images. BRIEF is very fast both to build and match. BRIEF outperforms other fast descriptors such as SURF and SIFT in terms of speed and in terms of recognition rate in many cases. + +##### 7.1.1.5 Motion features + +Motion features indicate the change in the position of an object relative to its surroundings or a change in the surroundings relative to an object. Motion features are detected from video sequences where a temporal change of contents occurs. + +Typically, motion features are represented in the format of a motion vector indicating the direction and distance between the starting position and the ending position of an object or an image area when it moves in a video for a time slot. + +#### 7.1.2 Audio features + +audio features include time domain, spectral domain and cepstral domain features. + +##### Time domain features + +The windowing technique is employed and the long non-stationary signal is analysed as short chunks of the quasi-stationary signal. Windowing can be seen as multiplying a signal with a window function that is zero everywhere except in the region of interest. The resultant windowed signal is the subset of the original signal which is passed through the window; for the rest of the time the signal is zero. + +##### Frequency domain features + +To analyse a signal in terms of frequency, the time domain signal is converted into a frequency domain signal using Fourier transform or auto-regression analysis. Frequency domain analysis is a tool of utmost importance in audio signal processing. + +##### Cepstral domain features + +A cepstrum is obtained by taking the inverse Fourier transform of the logarithm of the spectrum of the signal. There is a complex, power, phase and real cepstrum. Among all of these, the power cepstrum is the one most relevant to speech signal processing. The analysis of the cepstrum is called cepstrum analysis, quefrency analysis (equivalent to frequency analysis in spectrum domain) or liftering (equivalent to the filtering in spectrum domain). The cepstrum features are mainly used in pitch detection, speech recognition and speech enhancement. + +### 7.2 Latent domain features + +Deep neural networks (DNN) have been widely used for intelligent analysis, where the DNN may be allocated on the mobile or client side. With the benefit of cloud computing, resources and energy at the front end can be significantly reduced, which is referred to as collaborative intelligence. Instead of images or videos, deep features in the latent domain are extracted from the raw data, compressed by occupying fewer bits than the compression of raw data occupies and then transported to the cloud server for further analysis, as shown in Figure 7-1. + +![Diagram illustrating collaborative intelligence realized by client and cloud. On the left, a tree icon is processed by a small neural network icon. Below it, icons of a smartphone, tablet, and smartwatch represent client devices. An arrow points from the tree to a latent feature vector [2.41, 0.87] and [7, 92, 0.87]. From this vector, an arrow points to a larger neural network icon and a server rack icon, representing the cloud side. A radio tower icon is shown between the client and cloud. The diagram is labeled F.748.21(22) at the bottom right.](af6be343f0c0a8f155f965dcf337b8af_img.jpg) + +F.748.21(22) + +Diagram illustrating collaborative intelligence realized by client and cloud. On the left, a tree icon is processed by a small neural network icon. Below it, icons of a smartphone, tablet, and smartwatch represent client devices. An arrow points from the tree to a latent feature vector [2.41, 0.87] and [7, 92, 0.87]. From this vector, an arrow points to a larger neural network icon and a server rack icon, representing the cloud side. A radio tower icon is shown between the client and cloud. The diagram is labeled F.748.21(22) at the bottom right. + +**Figure 7-1 – Collaborative intelligence realized by client and cloud** + +The latent domain features are the intermediate representation output by the convolutional layers or fully connected layers in the neural network. They can be different when extracted from different layers of the neural network. From the shallow layer to the deep layer, the information of deep features changes from the pixel domain to the semantic domain, as shown in Figure 7-2. The data format of the latent domain features is often tensors with multiple dimensions, typically three dimensions including weight, height and channel. Figure 7-2 shows the visualization of the deep features from different layers. The feature maps in the same layer but different channels are shown in the same row. + +The latent domain feature is related to the high-level semantic information of the content instead of describing the exact texture, colour or shape of the content. + +![Visualization of latent domain features from different layers for an input image of a bear.](2236272d3b3db6e6363337f5a8db72f6_img.jpg) + +The figure shows an input image of a bear on the left. Four arrows point from this image to four rows of feature maps, labeled Layer1, Layer2, Layer3, and Layer4. Each row contains five feature maps labeled feature\_map\_0, feature\_map\_1, feature\_map\_2, feature\_map\_3, and feature\_map\_4. The feature maps show different visualizations of the bear image, representing latent domain features extracted from different layers of a neural network. The visualizations become more abstract and less detailed from Layer1 to Layer4. A small label F.748.21(22) is present at the bottom right of the feature maps. + +Visualization of latent domain features from different layers for an input image of a bear. + +**Figure 7-2 – Visualization of the latent domain features from different layers** + +### 7.3 Semantic feature + +Semantic features are highly related to specific AI tasks. Commonly used semantic features can be classified into regions, types and semantic labels. + +#### 7.3.1 Region features + +##### Segmentation mask feature + +Image segmentation is the process of partitioning a digital image into multiple image segments, and its goal is to simplify the representation of an image into something more meaningful and easier to analyse. Image segmentation is typically used to locate objects and boundaries in images. + +The segmentation mask feature is represented by a mask format which assigns a label to every pixel in an image such that pixels with the same label share certain characteristics. + +##### Region of interest feature + +A region of interest (RoI) is a proposed region from an image, where the region attracts human or machine interest. While watching a video or image, people's attention is more probably concentrated on interesting objects (such as persons, dogs or cats), moving objects or events, leading them to ignore the other contents in the same video or image. Machine learning has been widely used to extract the RoI from the video or image to highlight the important area of the video or image. + +Typically, a salient mask is generated to indicate the RoI, which assigns different saliency values for each pixel to indicate the importance of the pixel, e.g., the pixel region containing a pedestrian or car is assigned a high saliency value while the region containing the sky or water is assigned a low saliency value. + +##### Bounding box feature + +A bounding box is an imaginary rectangle that serves as a point of reference for object detection and creates a collision box for that object. Data annotators draw these rectangles over images, outlining the object of interest within each image by defining its X and Y coordinates. This makes it easier for + +machine learning algorithms to find what they are looking for and to determine collision paths and it conserves valuable computing resources. Bounding boxes are one of the most popular image annotation techniques in deep learning. Compared with other image processing methods, this method can reduce costs and increase annotation efficiency. + +The bounding box is represented in the format of a data structure including the position of the up-left corner of the bounding box and the size of the rectangular box. + +#### **7.3.2 Semantic labels** + +##### **Class labels** + +The classical problem in computer vision, image processing and machine vision is that of determining whether or not image data contains some specific object, feature or activity. Different varieties of the recognition problem are described: 1) Object recognition (also called object classification) is when one or several pre-specified or learned objects or object classes can be recognized, usually together with their 2D positions in the image or 3D poses in the scene. 2) Identification is when an individual instance of an object is recognized. Examples include identification of a specific person's face or fingerprint, identification of handwritten digits or identification of a specific vehicle. 3) Detection is when the image data are scanned for a specific condition. Detection based on relatively simple and fast computations is sometimes used for finding smaller regions of interesting image data which can be further analysed by more computationally demanding techniques to produce a correct interpretation. + +The format of the class label is either an index indicating the extract class of a pixel or region or a vector indicating the probability of each class that a pixel or region belongs to. + +##### **Motion labels** + +Properties of moving objects in videos include position, moving speed, moving direction and trace. + +##### **Vehicle attribute labels** + +Similar to persons, cars have their own attributes such as licence plate, size, position, shape, vehicle type, primary colour and so on. + +## **8 Service requirements of feature-based distributed intelligent systems** + +### **8.1 General requirement** + +Feature-based distributed intelligent systems (FDISs) are designed to support feature-based intelligent applications. There are some different categories in FDISs. + +### **8.2 Data acquisition** + +Original input data of feature-based distributed intelligent systems are recommended to be acquired using specific front-end equipment based on scenarios. For instance, image data in industrial applications is recommended to be captured by industrial cameras with high sensitivity and low noise. It is recommended to back up a copy of the original data in case of accidental damage in the front-end. Moreover, the original data is recommended to have sufficient resolution, especially for image and audio data. + +### **8.3 Pre-processing** + +The pre-processing procedure is recommended to pre-process the original data (image, video, or audio) into particular formats, in order to extract features such as textures for the given AI tasks. + +### 8.4 Data representation + +FDISs are recommended to adapt feature extraction and feature representation scheme to establish an expression of the original data with its spatial-temporal characteristics and AI tasks to support it. + +### 8.5 Data transmission + +FDISs are recommended to implement the transformation of data from its original form to another, and meanwhile implementing its inverse process, i.e., expressing the original data in a lossy or lossless manner by using fewer data. + +### 8.6 Feature analysis + +FDISs are recommended to perform the AI task base on features with respect to the original data with comparable performance. + +FDISs are recommended to satisfy the performance requirements of the applications. The recommended performance of AI tasks can be found in Appendix II. + +### 8.7 Management + +FDISs are recommended to implement fundamental functions such as user management, authority management and system logging. + +## 9 Framework of feature-based intelligent systems + +FDISs are composed of eight functional domains: application domain, development domain, algorithm domain, service domain, data domain, device domain, management domain and security domain, as illustrated in Figure 9-1. + +![Figure 9-1: Framework of feature-based distributed intelligent systems. The diagram shows eight functional domains arranged in a grid. The central part consists of four rows of two domains each. The top row contains the Application domain (Machine applications, Human applications). The second row contains the Algorithm development domain (Model acceleration tool, Training, Evaluating, Model compress tool) and the Service domain (Other services, Vision services, Audition services). The third row contains the Data domain (Feature decoding, Data storage, Data preparation) and the Algorithm domain (Algorithm for video, Algorithm for audio, Algorithm for other). The bottom row contains the Front-end equipment domain (Data pre-processing, Feature extraction, Feature compress and encoding, Video sensing, Audio sensing, Other sensing). On the left, the Security domain is shown with four sub-components: Data security, Application security, Model security, and Privacy. On the right, the Management domain is shown with four sub-components: Service management, Algorithm management, and Front-end equipment management. A label F.748.21(22) is at the bottom right.](3f1987804d7d78bc3b3bc560e974280a_img.jpg) + +F.748.21(22) + +Figure 9-1: Framework of feature-based distributed intelligent systems. The diagram shows eight functional domains arranged in a grid. The central part consists of four rows of two domains each. The top row contains the Application domain (Machine applications, Human applications). The second row contains the Algorithm development domain (Model acceleration tool, Training, Evaluating, Model compress tool) and the Service domain (Other services, Vision services, Audition services). The third row contains the Data domain (Feature decoding, Data storage, Data preparation) and the Algorithm domain (Algorithm for video, Algorithm for audio, Algorithm for other). The bottom row contains the Front-end equipment domain (Data pre-processing, Feature extraction, Feature compress and encoding, Video sensing, Audio sensing, Other sensing). On the left, the Security domain is shown with four sub-components: Data security, Application security, Model security, and Privacy. On the right, the Management domain is shown with four sub-components: Service management, Algorithm management, and Front-end equipment management. A label F.748.21(22) is at the bottom right. + +Figure 9-1 – Framework of feature-based distributed intelligent systems + +### 9.1 Front-end equipment domain + +The front-end equipment domain consists of data acquisition capabilities for generating different categories of feature to the data domain for further processing. For effective transmission and usage, + +the data should be pre-processed and compressed. The front-end equipment domain provides data pre-processing, feature extraction, feature compression and encoding, video sensing, audio sensing, and so on. + +FDISs are recommended to support capturing different kinds of multimedia, such as video, audio and point cloud, by using sensors. + +FDISs are recommended to generate different a category of feature by choosing a different algorithm. + +FDISs are recommended to use a lower volume of data to represent the original data or feature in a lossy or lossless manner. + +FDISs are recommended to support updating feature extraction and coding algorithm. + +FDISs are recommended to support the pre-processing of acquired data, such as de-noising and image enhancement. + +FDISs are recommended to guarantee the performance of AI tasks such as average precision or accuracy. + +### 9.2 Data domain + +In this domain, the transmitted data is decoded and transformed from a compressed bitstream into a reconstructed feature or reconstructed original data. It could be the inverse process of the data encoding procedure or the transcoding. The data domain provides decoding, data storage and data preparation. + +FDISs are recommended to decode the transmitted data stream for future processing. + +FDISs are recommended to support parallel scheduling and intelligent flow control to ensure that massive bitstreams can be processed timely. + +FDISs are recommended to support storage of unstructured data and structured feature data after analysis, and provide management and control capabilities for supported feature-based tasks. + +FDISs are recommended to support data decoding, data augmentation, data quality checking, data cleaning and data annotation. + +FDISs are recommended to be able to handle noises in the sample data. + +### 9.3 Algorithm domain + +Feature-based intelligent applications are based on the analysis of multimedia data. Different applications and services need to establish different analysis models and match different algorithms. The algorithm domain implements the model to perform AI tasks. The algorithm domain provides various algorithms for video, audio, cloud point, and so on, and supports common deep learning neural network algorithm models such as residual neural networks (ResNet) and You Only Look Once (YOLO), and supports third-party algorithm libraries to meet different business needs. + +FDISs are recommended to support multialgorithm frameworks, such as VSSNet, ResNet and YOLO. + +FDISs are recommended to support various major tasks of image/video analysis, e.g., object detection and natural language processing (e.g., speech recognition). + +FDISs are recommended to support plug-in third party algorithms. + +### 9.4 Service domain + +The service domain provides intelligent services based on algorithms such as the object classification service, quality verification service and guide service. FDISs customize a batch of analysis task templates for feature-based intelligent application scenarios, such as object classification service + +templates and object detection service templates. When customers use it, they only need to set the template parameters to create an analysis task, which will be analysed and processed by the FDIS. FDISs are recommended to provide video analysis, image analysis and audio analysis capabilities. FDISs are recommended to provide cloud point analysis. + +### **9.5 Application domain** + +The application domain consists of human applications and machine applications. + +After the feature processing, the features have been restored and usually passed through a certain deep learning model, in order to generate AI results within an acceptable performance loss for machine applications. Another pathway is to prepare the restored data/feature for human vision tasks. The metrics for AI task performance include precision, recall, accuracy, WER and SER, while the metrics regarding human sense include SSIM, PSNR and PESQ. + +FDISs are recommended to support human applications and/or machine applications in given scenarios. + +FDISs are recommended to meet the performance of the tasks supporting human applications and/or machine applications. + +### **9.6 Development domain** + +In this domain, a set tools suite will be provided for feature-based model development such as model acceleration, model compression and model algorithm visualization including labelling, training, modelling and evaluation. + +FDISs are recommended to support labelling the training dataset. + +FDISs are recommended to support optimizing the model. + +FDISs are recommended to support model compression. + +FDISs are recommended to support model acceleration. + +FDISs are recommended to support the visualization of model development. + +FDISs are recommended to support the model compiling. + +FDISs are recommended to support the model performance evaluation. + +### **9.7 Management domain** + +The management domain provides front-end equipment management, algorithm management, service management, data management and application management. + +FDISs are recommended to manage and log the status and capabilities of the front-end equipment, algorithms, services, data and applications. + +#### **9.7.1 Front-end equipment management** + +FDISs are recommended to implement fundamental functions such as user management, authorization management, system logging and security guarantees. + +FDISs are recommended to support remote control, maintenance, update, configuration and diagnosis of front-end equipment. + +FDISs are recommended to support programmable control on front-end equipment. + +FDISs are recommended to support configuration on front-end equipment sampling information, such as frequency and resolution. + +#### **9.7.2 Algorithm management** + +FDISs are recommended to support the management of different algorithms, such as algorithm scheduling, invoke, licence and lifecycle management. + +#### **9.7.3 Service management** + +FDISs are recommended to support analysis task management, data source management and storage management. + +### **9.8 Security domain** + +FIDSs are recommended to achieve acceptable performance on both the front-end and the server-end, while guaranteeing data integrity, data access authority, personally identifiable information (PII), security and effectiveness. As a result, the systems are recommended to be designed with an endogenous safety scheme, together with management. Moreover, user access control is usually implemented through a registration mechanism in these systems. + +#### **9.8.1 Data security** + +FDISs are recommended to guarantee data security, application security, platform security, access security and equipment security. + +FDISs are recommended to guarantee data integrity and data security during the transmission between different modules within the system, which includes data acquisition, storage, analysis, processing, transmission and display. + +FDISs are recommended to support measures, such as access control, encryption and token to achieve data confidentiality. + +FDISs are recommended to support measures, such as user access controls, version controls and verification mechanisms, to achieve data integrity. + +FDISs are recommended to support measures, such as backup, recovery and firewall to achieve data availability. + +FDISs are recommended to prevent malicious data being used during the training procedure. + +#### **9.8.2 Model security** + +FDISs are recommended to guarantee the security of the deep model in case of unexpected attacks. To prevent attackers from acquiring user PII by repeatedly querying the same model, model access verification is recommended. + +FDISs are recommended to ensure the model encryption by methods such as model distillation. The system administrator is recommended to ensure the broad coverage of the training data to enhance model robustness. + +#### **9.8.3 Application security** + +With respect to hardware and software, it is recommended to ensure any bug in applications, models, platforms and chipsets can be repaired in a prompt response. + +FDISs are recommended to handle unexpected events correctly. + +FDISs are recommended to check authentication and authorization. + +FDISs are recommended to support countermeasures, such as encryption and decryption programs, anti-virus programs, spyware detection and deletion programs, to achieve application security. + +#### **9.8.4 PII protection** + +For output data, the identity of the individual is required to be removed or masked from the reconstructed data so it cannot be detected. + +Data acquisition equipment such as video cameras and sensors is recommended to follow PII protection rules. + +## Appendix I + +## Use cases of feature-based intelligent applications + +(This appendix does not form an integral part of this Recommendation.) + +This appendix shows examples. They are not requirements. + +### I.1 Smart city + +With the rise of IoT, many tasks that were usually done by workers or police officers can now be automated by an intelligent system, e.g., detection of illegal parking, detection of littering and traffic diversion. The above tasks often utilize intelligent systems to perform a list of sub-tasks such as object detection and object tracking. Besides, there is a high degree of interconnectivity between different node sensors and devices. It is important for these devices to communicate with each other to optimize and efficiently solve tasks. Therefore, the efficiency of data transmission can be achieved by applying the proposed feature-based framework. + +### I.2 ITS + +In intelligent transport systems (ITSs), cars may need to communicate between each other and other sensors in order to perform different tasks such as autonomous driving, lane detection and pedestrian detection. Sensors in the infrastructure may communicate features towards different vehicles, which then use these features to perform intelligent tasks. + +### I.3 Content rating + +Due to the generation of a huge quantity of video/image content, various forms of media including conventional broadcasting and personal broadcasting are overflowing. Protecting certain groups of people (e.g., people under 18) from inappropriate content is becoming a big issue. Meanwhile, the traditional manual review is time-consuming and labour-intensive. Thus, smart systems are being developed for film rating, inappropriate action detection and auto-mosaic. Intelligent tasks such as action recognition and object detection are performed during the process. + +### I.4 Smart manufacturing + +With the rapid development of intelligent industry, the degree of automation production has been enhanced, making working environments which are unsuitable for manual work possible, and large-scale, continuous production a reality. Production efficiency and accuracy are greatly improved. Besides, interoperability is recommended for devices to post-process the features to perform multiple tasks to ease the job for humans: malfunction detection, machine component detection, fire detection, hazardous zone monitoring, and so on. + +### I.5 Smart agriculture + +The rapid growth of machine vision ensures a high degree of automation in the agriculture industry. Various sensors and cameras are deployed to monitor the condition of crops or to track livestock. The feature-based framework can be applied to integrate the intelligent system for various tasks. + +## Appendix II + +## Performance of intelligent tasks + +(This appendix does not form an integral part of this Recommendation.) + +This appendix shows examples. They are not requirements. + +### II.1 Smart city + +#### II.1.1 Object detection + +In a feature-based distributed intelligent system, detection accuracy is recommended to achieve 85% and above, and processing speed differs according to specific algorithms. The algorithm processing time is recommended to be less than 200 ms/frame. The maximum framerate of the FDIS is recommended to be more than 25 fps. The delay is recommended to be within 2 s. The bitrate for 1080 P video is recommended to be less than 6 Mbit/s. + +#### II.1.2 Image classification + +In a feature-based distributed intelligent system, the classification accuracy is recommended to achieve 85% and above. The maximum framerate of the FDIS is recommended to be more than 25 fps. The delay is recommended to be within 2 s. The bitrate for 1080 P video is recommended to be less than 6 Mbit/s. + +#### II.1.3 Object tracking + +In a feature-based distributed intelligent system, the tracking accuracy is recommended to achieve 85% and above. The maximum framerate of the FDIS is recommended to be more than 25 fps. The delay is recommended to be within 2 s. The bitrate for 1080 P video is recommended to be less than 6 Mbit/s. + +#### II.1.4 Action recognition + +In a feature-based distributed intelligent system, the recognition accuracy is recommended to achieve 80% and above. The maximum framerate of a FDIS is recommended to be more than 25 fps. The delay is recommended to be within 2 s. The bitrate for 1080 P video is recommended to be less than 6 Mbit/s. + +### II.2 Smart manufacturing + +In a feature-based distributed intelligent system, the detection accuracy is recommended to achieve 85% and above. The real-time task is recommended to be supported. + +There are four main scenarios in smart manufacturing, namely, real-time control and application based on smart cameras, non-real-time application based on smart cameras, real-time application based on non-smart cameras, and application schemes based on industry Internet protocol cameras (IPCs). + +The first scenario transfers a small amount of data, usually less than 100 bytes, through TCP/IP or user datagram protocol (UDP). The second scenario is time delay-insensitive, such as non-real-time backup, monitoring commands and documents. Such data is usually transferred through TCP/IP or file transfer protocol (FTP), and the camera usually uploads a video file of 5 MB. The third scenario is the mainstream application scheme nowadays, which contains GigE Vision and USB. The fourth scenario uses ITU-T H.264 to perform encoding. Streams are transmitted through the Real Time Streaming Protocol RTSP or Real Time Transport Protocol (RTP). As a comparison, the typical bitrate of 1080p@30fps video is 4 M, the typical bitrate of 2K@30fps is around 6 M, and when it comes to 4K@30fps, the bitrate increases to around 16 M. + +As to the processing time consumed, in most industrial applications, time constraints differ from scenarios. For instance, in mobile phone surface lesion detection, the acceptable vision processing time is 500 ms ~ 1 s. In SPI/AIO (Solder Paste Inspection/Automatic Organic Inspection), the acceptable vision processing time is 200~300 ms. In solder joint inspection, as the cycle time is significantly longer, the acceptable vision processing time can reach 1~15 s. + +## Appendix III + +## Relation between tasks and features + +(This appendix does not form an integral part of this Recommendation.) + +In a FDIS, the features extracted and transmitted will be selected and configured according to the combination of AI tasks supporting intelligent applications. The mapping of the AI task and features supporting the given AI task is shown in Table III.1. + +**Table III.1 – Relation between AI tasks and features** + +| AI task | Feature | +|---------------------|-----------------------------------------------------| +| Object detection | Bounding box, edge, colour, corner... | +| Classification | Class labels, latent domain features, etc. | +| Search | Feature descriptions (SIFT, SURF, HOG, BRIEF, etc.) | +| Object segmentation | Region features (RoI, segmentation mask), etc. | +| Object tracking | Bounding box, motion features | + +## Bibliography + +- [b-ITU-T K.142] Recommendation ITU-T K.142 (2019), *Lightning protection and earthing of video surveillance systems.* +- [b-ISO/IEC 29100] ISO/IEC 29100:2011, *Information technology – Security techniques – Privacy framework.* + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.22-202309-I_PDF-E/35a7554182eb055209552843f341a1ae_img.jpg b/marked/F/T-REC-F.748.22-202309-I_PDF-E/35a7554182eb055209552843f341a1ae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..2afd5c1c6542079a333524e01925208950e57563 --- /dev/null +++ b/marked/F/T-REC-F.748.22-202309-I_PDF-E/35a7554182eb055209552843f341a1ae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0bbc235ce06a0ee8cd7cc4772b54a943c33e2569ff09688ab11a419c9389879d +size 4440 diff --git a/marked/F/T-REC-F.748.22-202309-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.748.22-202309-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..985ce4f109b67d2faa0bfb4095195a8c8daeb31f --- /dev/null +++ b/marked/F/T-REC-F.748.22-202309-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:df1fd24c75458ea11a4825ba5d66828f9be3dd9f00ac276be6c1bcb17fce8e9f +size 7269 diff --git a/marked/F/T-REC-F.748.22-202309-I_PDF-E/5860ad6bd2a2dd8d1ab12864b8f90f37_img.jpg b/marked/F/T-REC-F.748.22-202309-I_PDF-E/5860ad6bd2a2dd8d1ab12864b8f90f37_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..dda76611044b883614f5d340187d220830e02b81 --- /dev/null +++ b/marked/F/T-REC-F.748.22-202309-I_PDF-E/5860ad6bd2a2dd8d1ab12864b8f90f37_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d4ec3202ed4be71cec55a01cd74f728656cc190d9518316cc576b1f3c18c1daf +size 2717 diff --git a/marked/F/T-REC-F.748.22-202309-I_PDF-E/acfc53eca625d62b38aa2563efa95c3e_img.jpg b/marked/F/T-REC-F.748.22-202309-I_PDF-E/acfc53eca625d62b38aa2563efa95c3e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..4aae1c7f24a1620bd97bb63cc64aa329642e07d3 --- /dev/null +++ b/marked/F/T-REC-F.748.22-202309-I_PDF-E/acfc53eca625d62b38aa2563efa95c3e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9b2006acf97677c526fc71b0d088ec025da7828a11359698afac17f0739b5f31 +size 175207 diff --git a/marked/F/T-REC-F.748.23-202404-I_PDF-E/raw.md b/marked/F/T-REC-F.748.23-202404-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..4238b7f283ac4f6a76317950dbc45ecc3754df67 --- /dev/null +++ b/marked/F/T-REC-F.748.23-202404-I_PDF-E/raw.md @@ -0,0 +1,630 @@ + + +# Recommendation **ITU-T F.748.23 (04/2024)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +## **Requirements and framework for intelligent crowdsensing multimedia interaction based on deep learning** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +## Recommendation ITU-T F.748.23 + +# Requirements and framework for intelligent crowdsensing multimedia interaction based on deep learning + +## Summary + +Artificial intelligence (AI) techniques can greatly improve the efficiency and effectiveness of crowdsensing tasks execution and enable intelligent multimedia interaction in crowdsensing. Recommendation ITU-T F.748.23 outlines specific scenarios for crowdsensing multimedia interaction that leverage AI techniques, and subsequently defines the corresponding requirements and framework in detail. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.23 | 2024-04-26 | 16 | 11.1002/1000/15608 | + +## Keywords + +AI, crowdsensing, framework, multimedia data, requirements. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2024 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|----------------------------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Overview of intelligent crowdsensing multimedia interaction based on AI ..... | 2 | +| 7 Scenarios and requirements for AI-based intelligent crowdsensing data interaction... | 4 | +| 7.1 Scenarios and requirements for crowdsensing task assignment..... | 5 | +| 7.2 Scenarios and requirements for efficient aggregation of multimedia data..... | 9 | +| 7.3 Scenarios and requirements for crowdsensing data quality optimization ..... | 11 | +| 7.4 Requirements for multimedia crowdsensing data privacy enhancement ..... | 14 | +| 8 Framework of crowdsensing multimedia interaction based on AI..... | 15 | +| 8.1 Centralized crowdsensing platform layer ..... | 15 | +| 8.2 Crowdsensing resource layer..... | 17 | +| 8.3 AI capabilities..... | 17 | +| Bibliography..... | 19 | + + + +# Requirements and framework for intelligent crowdsensing multimedia interaction based on deep learning + +# 1 Scope + +This Recommendation specifies the requirements and framework for intelligent crowdsensing multimedia interaction based on artificial intelligence (AI). + +The scope of this Recommendation is proposed to include the following aspects: + +- General overview of intelligent crowdsensing multimedia interaction base on AI. +- Scenarios and requirements for crowdsensing multimedia interaction based on AI. +- Framework of crowdsensing multimedia interaction based on AI. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T F.748.21] Recommendation ITU-T F.748.21 (2022), *Requirements and framework for feature-based distributed intelligent systems*. + +[ITU-T Y.4217] Recommendation ITU-T Y.4217 (2022), *Service requirements and capability framework for Internet of things-related crowdsourced systems*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 crowdsensing** [b-ITU-T Y.4205]: The practice of employing crowdsourcing in collecting sensory data. + +**3.1.2 crowdsourcing participant** [b-ITU-T Y.4205]: The person contributing to a crowdsourced system by providing and potentially benefitting from ideas, content, or access to services, hardware or other system resources. In the general case, the participant is a member of a large, open and potentially undefined group of people. + +**3.1.3 crowdsourcing resource** [b-ITU-T Y.4205]: The device, hardware module, service or other system resource that a crowdsourced system has been given access to by crowdsourcing participants. + +**3.1.4 crowdsourcer** [b-ITU-T Y.4205]: The initiator of a crowdsourcing task. It can refer to a single person, a group of people, a corporation, an organization, etc. + +**3.1.5 crowdsourced systems** [b-ITU-T Y.4205]: Systems that employ crowdsourcing in order to augment their constituent infrastructure and the set of provided services or collected information. + +**3.1.6 artificial intelligence (AI)** [b-ETSI GR ENI 004]: Computerized system that uses cognition to understand information and solve problems. + +**3.1.7 machine learning** [b-ITU-T Y.3172]: Processes that enable computational systems to understand data and gain knowledge from it without necessarily being explicitly programmed. + +## **3.2 Terms defined in this Recommendation** + +None. + +# **4 Abbreviations and acronyms** + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-----|-------------------------------| +| AI | Artificial Intelligence | +| DL | Deep Learning | +| GPS | Global Positioning System | +| IoT | Internet of Things | +| ML | Machine Learning | +| NLP | Natural Language Processing | +| OCR | Optical Character Recognition | +| RNN | Recurrent Neural Network | +| UAV | Unmanned Aerial Vehicle | +| UE | User Equipment | + +# **5 Conventions** + +In this Recommendation: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement needs not be present to claim conformance. + +# **6 Overview of intelligent crowdsensing multimedia interaction based on AI** + +With the development and popularization of the intelligent mobile devices, wearable devices and Internet of things (IoT) and its built-in sensors (such as accelerometer, gyroscope, camera, compass, global positioning system (GPS), microphone, connected car communication module, etc.), intelligent crowdsensing has gradually developed into a new perception mode. More and more industries require the use of intelligent crowdsensing technology to perceive spatial-temporal information, extract knowledge, and understand multimodal data. During the execution of crowdsensing tasks, crowdsensing technology can leverage the advantages of large scale participants' widespread distribution, flexible mobility, and real-time connectivity. However, the following issues still need to be optimized by artificial intelligence (AI) technologies such as machine learning (ML) or deep learning (DL) when implementing intelligent crowdsensing: + +Firstly, in crowdsensing systems, there are usually multiple parallel tasks that need to be assigned. Due to the heterogeneity, complexity, and uncertainty in both the crowdsensing nodes and crowdsourcing participants, it is necessary to introduce AI schemes and jointly optimize the assignment of multi tasks based on the characteristics of varied sensing nodes and participants. This can enhance collaboration among multiple sensing nodes and better integrate comprehensive perception of multimedia data. + +Secondly, when executing the crowdsensing tasks, the multimedia data interaction will consume the network resources and incur costs for crowdsourcing participants. In order to save network resources and reduce data transmission costs, it is necessary to carry out efficient aggregation of multimedia data by extracting multimedia key features, filter redundant information through AI techniques on crowdsourcing resources, and transmit key features and parameters instead of all the media content to the crowdsensing system for further processing. + +Thirdly, the perceived multimedia data may have the characteristics of multimodality, low quality, information redundancy, and large data volume. If manual screening is used to extract multimedia data information, the crowdsensing tasks executing efficiency will be very low. Therefore, in order to improve the efficiency of crowdsensing tasks, it is necessary to adopt AI techniques and design adaptive frameworks for evaluating the quality of multimedia data, selecting high-quality data, extracting multimedia data features, and efficiently analysing multimedia data. + +Lastly, privacy and security is a crucial concern in crowdsensing, as the perception of multimedia data may involve information such as location, image or voice. AI techniques such as data encryption, anonymization and secure transmission can be used to ensure the security and privacy of crowdsensing tasks. Moreover, intelligent mechanisms can be developed to detect abnormal or malicious behaviours, protecting the reliability and effectiveness of data perception, processing, and transmission. + +In conclusion, AI schemes can significantly enhance the efficiency and effectiveness of crowdsensing tasks executing, with benefits in task assignment, multimedia data processing, data transmission efficiency, and privacy and security protection. By establishing requirements and a framework for crowdsensing multimedia interactions based on AI, it can offer low-cost, intelligent capabilities for various crowdsensing scenarios and achieve effective crowdsensing results. An overview of intelligent crowdsensing multimedia interaction based on AI is shown in Figure 6-1. + +![Figure 6-1: Overview of intelligent crowdsensing multimedia interaction based on AI. The diagram is organized into three layers: Application layer, Network layer, and Sensing layer. The Application layer includes scenarios like Smart city, Smart warehouse, Smart factory, and Smart transformation. The Crowdsensing task execution with AI section shows Task assignment (Task 1, Task 2 -> AI technology -> Sub-task 1, 2, 3 -> The crowd -> Crowdsourced resources), Efficient data aggregation (Effective raw data -> AI technology -> Effective data and redundant data -> Crowdsourced resources), and Data quality optimization (Multimodal raw data -> AI technology -> Optimized data). The Network layer shows various communication methods like cellular, satellite, cloud, Bluetooth, and WiFi. The Sensing layer shows multimedia sensing data from wearable and mobile devices, Internet and APPs, and IoT devices, all contributed by 'The crowd'.](a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg) + +The diagram illustrates the architecture of intelligent crowdsensing multimedia interaction based on AI, organized into three layers: + +- Application layer:** Includes scenarios such as Smart city, Smart warehouse, Smart factory, and Smart transformation. +- Crowdsensing task execution with AI:** + - Task assignment:** Task 1 and Task 2 are processed by AI technology to generate Sub-task 1, Sub-task 2, and Sub-task 3, which are then executed by The crowd using Crowdsourced resources. + - Efficient data aggregation:** Effective raw data from Crowdsourced resources is processed by AI technology to generate Effective data and redundant data. + - Data quality optimization:** Multimodal raw data is processed by AI technology to generate Optimized data. +- Network layer:** Shows various communication methods including cellular, satellite, cloud, Bluetooth, and WiFi. +- Sensing layer:** Shows multimedia sensing data from wearable and mobile devices, Internet and APPs, and IoT devices, all contributed by The crowd. + +Figure 6-1: Overview of intelligent crowdsensing multimedia interaction based on AI. The diagram is organized into three layers: Application layer, Network layer, and Sensing layer. The Application layer includes scenarios like Smart city, Smart warehouse, Smart factory, and Smart transformation. The Crowdsensing task execution with AI section shows Task assignment (Task 1, Task 2 -> AI technology -> Sub-task 1, 2, 3 -> The crowd -> Crowdsourced resources), Efficient data aggregation (Effective raw data -> AI technology -> Effective data and redundant data -> Crowdsourced resources), and Data quality optimization (Multimodal raw data -> AI technology -> Optimized data). The Network layer shows various communication methods like cellular, satellite, cloud, Bluetooth, and WiFi. The Sensing layer shows multimedia sensing data from wearable and mobile devices, Internet and APPs, and IoT devices, all contributed by 'The crowd'. + +F.748.23(24) + +Figure 6-1 – Overview of intelligent crowdsensing multimedia interaction based on AI + +# 7 Scenarios and requirements for AI-based intelligent crowdsensing data interaction + +This clause describes how AI schemes foster the four scenarios within crowdsensing data interaction: task assignment, efficient data aggregation, multimedia data optimization, and privacy and security protection. Figure 7-1 shows the relationship between these four scenarios. The requirements for each scenario are also further classified. + +![Figure 7-1: The relationship between crowdsensing data interaction scenarios. The diagram shows four layers of interaction with upward arrows indicating data flow. The layers are: Application layer (Crowdsensing applications), Cloud layer (Crowdsensing data quality optimization based on AI), Edge layer and sensing layer (Efficient data aggregation based on AI), and Sensing layer (Crowdsensing task assignment based on AI). A vertical blue box labeled 'Data security protection' spans the bottom three layers.](cfef993dcc8fb513de79eb1f93cf26ae_img.jpg) + +The diagram illustrates the relationship between crowdsensing data interaction scenarios across four layers, with data flow indicated by upward arrows: + +- Application layer:** Crowdsensing applications +- Cloud layer:** Crowdsensing data quality optimization based on AI. This layer includes three sub-components: Multimodal data relationship analysis, Multimedia data reliability evaluation, and Data quality optimization. +- Edge layer and sensing layer:** Efficient data aggregation based on AI. This layer includes two sub-components: Data pre-processing and pre-optimization, and Data transmission and aggregation. +- Sensing layer:** Crowdsensing task assignment based on AI. This layer includes three sub-components: Task decomposition and allocation, Sensing node Recommendation, and Participants relevancy evaluation. + +A vertical blue box labeled **Data security protection** spans the bottom three layers (Sensing, Edge, and Cloud). + +F.748.23(24) + +Figure 7-1: The relationship between crowdsensing data interaction scenarios. The diagram shows four layers of interaction with upward arrows indicating data flow. The layers are: Application layer (Crowdsensing applications), Cloud layer (Crowdsensing data quality optimization based on AI), Edge layer and sensing layer (Efficient data aggregation based on AI), and Sensing layer (Crowdsensing task assignment based on AI). A vertical blue box labeled 'Data security protection' spans the bottom three layers. + +**Figure 7-1 – The relationship between crowdsensing data interaction scenarios** + +## 7.1 Scenarios and requirements for crowdsensing task assignment + +Crowdsensing tasks assignment is mainly to select the appropriate crowdsourcing participants and match related sensing nodes to complete the crowdsensing tasks, while meeting certain constraint conditions to achieve the optimal allocation of resources. The traditional crowdsensing task assignment approach primarily relies on random, isolated, or roughly allocated assignments based on manual experience. However, with the inclusion of heterogeneous sensing nodes, the uncertainty surrounding participants, and the increased complexity of crowdsensing tasks, it is imperative to employ advanced AI schemes. By applying AI schemes to crowdsensing task assignment, the crowdsensing nodes and crowdsourcing participants can form an effective collaboration to complete the crowdsensing task. + +### 7.1.1 Scenarios for crowdsensing task assignment based on AI + +AI schemes play a vital role in discovering the capabilities of heterogeneous sensing nodes, evaluating the correlation among crowdsourcing participants, and considering incentive mechanisms for various task execution qualities. In the context of smart warehouse scenarios, one of the crowdsensing tasks is to efficiently arrange the goods, especially when dealing with irregularly shaped items to achieve the most space-efficient arrangement within a sorting container. The goods sorting container consists of a box body equipped with multiple sensors and mechanical rotating shafts and these components are connected with the crowdsensing platform. The heterogeneous sensors and rotating shafts on the sorting container are required to work together to conduct efficient arranging of goods. An essential challenge in this scenario is the dynamic assignment of the tasks, taking into account the unique capabilities and characteristics of each sensor involved, AI techniques can be leveraged on crowdsensing tasks assignment: + +**Task decomposition and allocation:** AI algorithms play a vital role in improving the efficiency of crowdsensing tasks by breaking them down into smaller, more manageable sub-tasks and optimize the allocation of tasks or sub-tasks by considering various factors such as task features, capabilities of sensing nodes, workload balance, and time constraints. For the sorting container scenarios, a goods arrangement task is recommended to be decomposed into sub-tasks. For sub-task 1, the sorting container together with heterogeneous sensors obtains the sensing data (such as weight, shape, density, position) of each placed item sequentially and transfers the sensing data to the crowdsensing platform. For sub-task 2, the crowdsensing platform utilizes the collected sensing data of each placed item and uses the trained neural network AI inference model to estimate the initial rotation parameters + +of the sorting container. These rotation parameters are then transmitted to the mechanical shafts. The multi modal data such as the sensing data of the goods placed in sequence, the number of goods, the volume of the sorting container, and other related factors can be utilized to train the neural network AI inference model. This decomposition allows for a better understanding of the dependencies and requirements associated with each task, facilitating more efficient assignment and resource allocation. Furthermore, AI technology can continuously learn and improve its arrangement strategies over time. According to the initial rotation parameters received from the crowdsensing platform, the mechanical axis starts rotating and drives the sorting container to rotate in multiple directions, achieving efficient arrangement of goods. By analysing historical data and real-time feedback, the rotation parameters are continuously optimized, causing the container to rotate continuously and prompting the goods to constantly change position, shift space, and exchange placement order within the container. Ultimately, this process leads to finding the placement scheme that maximizes the utilization of the container's space. This adaptive and self-improving nature of AI enables warehouses to continuously enhance their efficiency. + +**Sensing nodes recommendation:** AI algorithms are capable of analysing historical data and leveraging experience to suggest suitable sensing nodes based on the individual characteristics and requirements of the task. For the sorting container scenarios, the sensing nodes are recommended to include multiple mechanical shafts located at the bottom corners of the sorting container, heterogeneous sensors such as infrared cameras, gravity sensors, and gyroscopes positioned within the container. These mechanical shafts and sorting container sensors are equipped with communication modules to facilitate communication with the crowdsensing platform individually. By taking into account factors such as computation capability, network communication availability and location, AI can recommend the most appropriate sensors for the goods arrangement task. This helps in optimizing task allocation and ensures that each sensing node is assigned tasks that align with their capabilities and resources. + +### 7.1.2 Requirements for crowdsensing task assignment based on AI + +- It is required for a crowdsensing platform to quantize a crowdsourcer's request into a computable task according to certain rules, and match one or more task execution sensing nodes. Both deep learning methodology or a combinatorial optimization methodology are accepted. + +#### 7.1.2.1 To conduct the crowdsensing task assignment based on deep learning methodology + +- It is required to define the crowdsensing task feature and attribute information. The crowdsensing task feature can be defined as a tuple {Feature 1, Feature 2, Feature 3..., Feature n, etc.}. + +The task features are recommended to include: + +Feature 1: Task content. Task content refers to a variety of perceptual operations, such as perceiving noise, taking a photo, taking a video, etc.; + +Feature 2: Description. Description refers to the requirements and parameters of the perception task, such as the recording duration, the pixels of the photo, the video format, etc.; + +Feature 3: Task spatial type. The task spatial type includes static space task, dynamic space task. + +Feature 4: Location. Location refers to the task execution area, such as the centre coordinates and radius of the sensing tasks location; + +Feature 5: Time. Time refers to the perceived timeliness of task execution. + +Feature 6: Assurance level. According to the target of the crowdsensing task, the task assurance level is recommended to be divided into the following categories: time-sensitive tasks, high-reliability tasks, emergency service tasks, normal tasks, etc. + +Feature 7: Sensing node type. The sensing node type includes: single sensing node types and multiple sensing nodes types. + +Feature 8: Other factors, etc. + +- It is required to apply multiple devices as sensing nodes to realize crowdsensing tasks. The sensing node device types include: mobile devices, wearable devices, IoT devices and sensors, gateway, edge computing server and so on. +- It is required to collect suitable sensing nodes to execute the tasks in order to achieve balanced coverage, optimal task allocation and effective tasks execution. +- It is recommended to define the sensing node features, which represent the characteristics or attributes of each sensing node that can be used for clustering and selection purpose. +- The specific features are recommended to depend on the feature of the sensing task and the type of multimedia data being collected. +- The feature of the sensing node is recommended to be represented as a tuple {Feature 1, Feature 2, Feature 3, ... , Feature n, etc.}. + +The sensing node features are recommended to include: + +Feature 1: Location distribution + +Feature 2: Movement characteristics such as for example the movement speed, the movement distance, the movement duration of an unmanned aerial vehicle (UAV) + +Feature 3: Multimedia acquisition capability + +Feature 4: Sensing node power consumption and remaining capacity + +Feature 5: Calculation accuracy + +Feature 6: Network statuses (e.g., the network throughput, the network signal strengthens, the network connectivity) + +Feature 7: Sensor sensitivity + +Feature 8: Other factors, etc. + +- Once the sensing node features are defined, it is recommended to apply AI clustering algorithms to group similar sensing nodes together. + +The clustering algorithms is to analyse the feature values of each sensing node and group them into clusters based on their similarity. This process organizes the sensing nodes into distinct groups according to their shared characteristics, employing methods such as k-means, hierarchical clustering, or density-based clustering. + +- After clustering the sensing nodes, it is required to collect the sensing nodes from clusters that will execute specific crowdsensing tasks. The criteria for task node collection are required to vary depending on the specific requirements of the crowdsensing task. +- It is recommended that each crowdsensing task contains a diverse set of sensing nodes from different clusters, enabling a well-rounded coverage of the sensing space and avoiding of over-representing a single cluster. + +- It is recommended to employ AI algorithms to construct the task sensing node assignment model. + +This model facilitates the identification of sensing nodes responsible for executing crowdsensing tasks by using the task feature, the sensing node feature and the cluster result. Various AI algorithms such as the logistic regression, decision trees, random forests, support vector machines (SVM) or neural networks can be employed for this purpose. + +#### 7.1.2.2 To conduct the crowdsensing task assignment based on combinatorial optimization algorithm + +- It is required to formulate the task assignment as a computational optimization problem based on the objective function, constraints, and chosen optimization algorithm. +- It is required to define the objective function, which quantifies the quality or effectiveness of the sensing node collection solution. The objective function can be one or multiple optimization objectives. + - It is required that the objectives function is capable to guide the optimization algorithm in finding the best sensing node collection solution. +- It is required to identify the constraint conditions during the sensing node collection process. +- The constraint conditions are recommended to include: spatial constraint, resource availability constraints (e.g., memory, bandwidth), time duration constraint, incentive budget constraint and accuracy constraint. +- It is required to choose an appropriate optimization algorithm that can handle the defined objective function and constraints. The goal of the optimization algorithm is recommended to include but not be limited to: + - Maximize the multimedia data quality; + - Minimize the task incentive costs; + - Minimize the task completion duration; + - Minimize the network resource consumption; + - Maximize the task completion rate, etc. +- It is required that the optimization algorithms are varied depending on the crowdsensing task complexity, the size of the data, the specific constraints and the objective functions that have been defined. +- For time-sensitive crowdsensing tasks, it is recommended to ensure a prompt completion rate and meet high timeliness requirements for task execution. Therefore, it is recommended to define specific locations for sensing nodes to complete crowdsensing tasks, rather than randomly assign locations for sensing nodes to move around. +- In situations where sensing node resources are limited, it is recommended that the optimization objectives of task assignment maximize the task completion rate while minimizing the total distance moved to complete the tasks, thereby reducing the time required for crowdsensing task completion. For example, the minimum-cost maximum flow model is capable to solve this problem. +- In situations where sensing nodes resources are abundant, it is recommended select suitable sensing nodes from a large pool to complete the crowdsensing tasks. +- The optimization objectives of task assignment is recommended to minimize the incentive cost while also minimizing the total distance moved to complete the tasks in order to reduce the participants' burden. For example, the linear weighting method or constraint-based method is capable to solve this problem. +- It is required to filter and classify the tasks, and then combine or decompose different categories of tasks before releasing the crowdsensing tasks. +- For tasks of the same type, it is recommended to apply clustering AI algorithms for tasks classification, and then combine multiple tasks into one task in order to enhance the crowdsourcing resources efficiency or avoid repeated contributions of sensing nodes. +- For tasks of different categories, it is recommended to apply classification AI algorithms to match different sensing nodes or participants for different task categories to improve the quality of task execution. + +- For complex tasks, it is recommended to decompose and divide the complex task into multiple sub-tasks. +- It is required that the task assignment also includes the pending tasks and related requirements for each pending task. +- It is recommended to allocate and apply unmanned aerial vehicle (UAV) and satellite remote sensing nodes as the crowdsourced resources for emergency service tasks, such as floods, earthquakes and fire monitoring. +- It is recommended to allocate incentives for crowdsensing participants with AI models to reduce biases and make fair and data-driven decisions. These factors are recommended to be taken into consideration: crowdsensing task completion times, the difficulty of the task, the level of expertise required, the participant's sensing nodes capabilities. +- It is required to take participants adjustments and replacements as necessary to ensure the successful completion of tasks. +- To address the challenge of unknown data sparsity in a new crowdsensing platform, it is recommended to select a subset of participants who have consistently provided valuable contributions to submit crowdsensing data from their respective locations. By leveraging spatiotemporal AI models, it is capable to infer information about non-perceived areas, effectively mitigating the issue of data sparsity and enhancing the accuracy of the crowdsensing data. For example, association recurrent neural network (RNN) models is optional to be applied to solve the crowdsensing data sparsity issue. + +## 7.2 Scenarios and requirements for efficient aggregation of multimedia data + +The multimedia raw data often contains a significant amount of redundant and low-quality information. Uploading all these massive data to a centralized crowdsensing platform would consume a substantial amount of communication network resources, participants' time, and sensing nodes hardware expenses. Therefore, before transmitting the data to the centralized crowdsensing platform, it is essential to optimize the multimedia data aggregation through the application of AI schemes in order to reduce the communication costs, save participants' time, and minimize the hardware expenses associated with sensing resources. + +### 7.2.1 Scenarios for efficient aggregation of multimedia data based on AI + +**Reduced communication costs:** AI-based schemes can facilitate the extraction of relevant and valuable information from multimedia data, allowing for data compression and prioritization. This reduces the amount of data that needs to be transmitted, leading to significant savings in communication costs. Additionally, AI algorithms can identify and filter out high-quality data, and redundant or irrelevant data, further optimizing the network bandwidth utilization. + +**Save crowdsourced resources:** AI-based multimedia data aggregation can leverage distributed computing architectures, such as edge computing and cloud computing. By offloading data processing to edge computing platforms or cloud platforms, the computational load on sensing resources can be reduced, potentially minimizing the required hardware capabilities and associated expenses. + +### 7.2.2 Requirements for efficient aggregation of multimedia data based on AI + +#### 7.2.2.1 To reduce the communication costs + +- It is recommended to conduct raw data quality analysis on the sensing node side before uploading the raw data to the centralized crowdsensing platform. +- It is recommended to evaluate the timeliness, integrity, and accuracy of the raw data with AI methods. + +The timeliness of the data refers to how quickly it is captured and transmitted from the sensing nodes. AI algorithms can be used to analyse the timestamps associated with the data to identify any delays or inconsistencies. This analysis helps ensure that the data is up-to-date and relevant to the crowdsensing task. + +The data integrity refers to the completeness and correctness of the information. For example, AI techniques can be used to verify the integrity of the raw data by examining factors such as data formatting, completeness of fields, and internal consistency. Anomalies or discrepancies in the data can be identified and flagged for further investigation or corrective actions. + +The accuracy of the raw data refers to how well it reflects the true state of the observed phenomenon. For example, AI algorithms can be used to perform data validation and verification, comparing the captured data against expected patterns or reference data sets. Statistical analysis, ML models, or pattern recognition techniques can be employed to identify potential errors, outliers, or inconsistencies in the raw data. + +- Based on the raw data quality evaluation result, it is recommended to filter out the high-quality data on the sensing node side and only transfer these high-quality data to the centralized crowdsensing platform in order to reduce data transmission costs. +- It is recommended to set different evaluation criteria when using AI schemes to filter high-quality data from different crowdsensing tasks. + +For example, for urban 3D modelling scenarios, the diversity of images is an important indicator to evaluate the data quality. However, for single point of failure reporting scenarios, the similarity of pictures needs to be used as an indicator to evaluate the data quality. + +- It is recommended not to transfer the redundant data to the centralized crowdsensing system by applying AI analysis on data redundancy. +- When conducting AI analysis on data redundancy analysis, it is required to consider the semantic feature information within the raw data. Commonly used semantic features can be classified into regions, types and semantic labels [ITU-T F.748.21]. + +For example, in an earthquake rescue crowdsensing task, participants and sensing cameras are required to upload images of the scene, the characteristics such as time, place, angle, and distance need to be used to describe the image data redundancy definition. If the entities within them are the same but the photo shooting time, shooting angle, and shooting distances are different, the semantic features of these images are not redundant. This is because different shooting angles can provide additional and valuable insights or a clear view of obstacles for the rescue work. + +- It is recommended to support the participants for choosing an appropriate time and network access to upload crowdsourcing data. The collected data could be stored locally in user equipment (UE) rather than uploaded to the crowdsensing platform immediately. Participants can freely choose the appropriate time and network for data transmission (such as home broadband network, public network, etc.) [ITU-T Y.4217]. + +NOTE – This Recommendation is not applicable to time-sensitive tasks and emergency service tasks. + +- It is recommended to apply an AI scheme for image compression and video compression on the sensing node side to enhance the multimedia data coding efficiency and reduce the data redundancy. +- It is recommended to extract the key feature of the multimedia crowdsensing data on the sensing node side and submit the key character information to centralized crowdsourcing platform rather than the whole raw data. +- For crowdsensing tasks that accept text as expected crowdsensing data, it is recommended to apply optical character recognition (OCR) to scan and convert the image into text on participants' crowdsourced resource. + +- It is recommended to recognize the content and entities within the image with AI models, and convert the recognized contents entities into text descriptions by applying natural language processing (NLP) on participants' crowdsourced resource. Then the converted texts are sent to the centralized crowdsensing platform instead of the images. + +#### 7.2.2.2 To save the crowdsourced resources + +- For crowdsensing tasks without specific reporting period requirements, it is recommended that the crowdsensing data reporting period should be as long as possible, to save the power consumption of the crowdsourced resource. +- For crowdsensing tasks that need to carry out AI training or inference on participants' crowdsourced resource, it is recommended to implement light-weight AI models schemes and thin the model to reduce the computation costs and speed up the computation speed. + +The light-weight AI models implementation examples: neural network pruning, sparse connection, neural network weight quantization, neural network distillation, hyper-parameter optimization, reducing the channels number or the feature map size of the neural network. + +- It is recommended to provide edge AI and fog computing for crowdsourced resource in order to reduce latency for the lightweight AI models implemented on resource-constrained devices. +- In order to reduce the amount of data collection and lower task costs for unperceived areas, predictive AI models are recommended to be applied on the edge computing side to estimate or infer the data in the unsensed areas. + +The goal is to analyse the available data patterns and relationships to make informed predictions about the unperceived areas. For example, given the population density and base station traffic in a specific area, the intelligence crowdsensing task is to determine the population density in the unperceived area. It can compare the base station traffic data in the specific area with the base station traffic data in the unperceived area, and then analyse and predict the population density in the unperceived area. + +- It is recommended to enhance the accuracy of data predictions for the unperceived areas by continuously learning and improving over time as more raw data becomes available. +- It is required to support parallel submission of multimedia crowdsensing data during voice calls or video calls in order to enhance the crowdsensing task efficiency. +- If the crowdsourced resource is incapable of performing AI training or inference, it is recommended to transport the raw data to the edge computing platform for AI training or inference. + +## 7.3 Scenarios and requirements for crowdsensing data quality optimization + +Due to the diversity of crowdsourcing equipment, environmental impacts, uncertainty of crowdsourcing participants' behaviour and other factors, the quality of collected raw data usually deviates. On the other hand, since the crowdsourcing platform collects massive multi modal data from large number of crowdsourcing participants, the workload of data analysis is heavy. In order to save the cost of manual analysis and improve the efficiency of data processing, it is essential to process the multi modal multimedia data with AI models to enhance the data quality, evaluate the data reliability and intelligently analyse the data relationship according to related crowdsensing tasks. + +### 7.3.1 Scenarios for crowdsensing data quality optimization based on AI + +**Data relationship intelligent analysis:** The crowdsensing multimedia data encompasses a wide range of information including diverse participants, environmental factors, Internet sources, and physical sensing node entities. These carriers contain multiple dimensions and modalities of heterogeneous content, such as text, images, audio and video, as well as environmental perception + +information. To effectively handle this heterogeneous data, AI schemes are needed for multi modal data representation and heterogeneous data coupled relationships analysis. + +**Data reliability evaluation:** Some participants might provide noisy or erroneous data. Additionally, redundant data collected by multiple participants for the same crowdsensing task can lead to inefficiencies and duplicate processing. Thus, it is essential to evaluate the data reliability for correct decision-making of crowdsensing tasks and apply AI reasoning models to obtains real and reliable multimedia data. + +**Data quality enhancement:** The quality of the data collected by sensing nodes varies significantly. Improving the multimedia data quality before or after transferring to a crowdsensing platform by building an AI model will boost the overall performance of the crowdsensing tasks. + +### 7.3.2 Requirements for crowdsensing data quality optimization based on AI + +#### 7.3.2.1 For data relationship intelligent analysis + +- When multiple modalities of data coexist, it is required to apply different AI models to extract data features from multi-source heterogeneous data. +- For text data, it is required to extract text features through AI models such as NLP methods. +- For image data, it is required to detect the image features through computer vision algorithm or machine learning methods. +- For graph-based representations that capture the relationships and interactions between different entities and data points, it is required to apply graph neural networks models to analyse the complex coupling relationships among the heterogeneous data elements. +- When extracting the data features, it is required to standardize and normalize the data of each modality to ensure consistent formats and units. +- It is required to align multi modal data features in terms of time, space, granularity, and semantics, and to ensure that multi modal data features can complement and support each other. +- It is required to ensure that the data features from different modalities are aligned in time. For example, this can be achieved through techniques such as timestamp interpolation, timestamp alignment algorithms, and synchronization. +- It is required to ensure that the spatial information from different modalities is aligned. For example, this can be achieved through techniques such as GPS or spatial interpolation methods. +- It is recommended to ensure that the granularity feature from different modalities is aligned within the same time or spatial range. +- It is recommended to ensure the semantic information alignment between different modalities. For example, this can be achieved through NLP techniques such as word vector representation, semantic similarity calculation, entity recognition, and relation extraction. +- After feature extraction is completed for each modality, the sensing data from different modalities exhibit varying correlation relationships and causal interactions in terms of patterns, structures, and distributions. To enhance the collaborative perception of multi-modal data, it is recommended to utilize multi modal data coupling models that integrate and associate features from diverse modalities. + +For example, when dealing with a sudden traffic accident at location A, it is not possible to determine the impact of the accident on other locations solely based on the text description reported by the user. Therefore, it is recommended to incorporate features from multi-modal data sources. This can include image data reported by users, noise information from intelligent streetlights, and location information from the sensors from UAV. By leveraging + +this comprehensive set of multi modal data, it can make accurate predictions regarding the impact of the traffic accident at location A located five kilometres away. + +The multi modal data coupling model refers to the integration or combination of multi modal data to solve a specific problem or achieve a specific goal. It describes the interdependence between modal data, explores relationships within different modal data, minimizes data mismatches, enhances the correlation between modalities, and complements each other's data features. + +- When utilizing the multi modal data coupling mode, it is recommended to capture both the explicit and implicit interaction relationship between multi modal data. This facilitates a deeper understanding of the interplay between diverse data sources. + +In the context of multi-modal data coupling for crowdsensing tasks, explicit interaction refers to the explicit correlations and interactions between different modalities. For example, visual data and audio data are collected by the crowdsensing platform, explicit interactions can be captured by examining how certain visual features correspond to specific audio features, or how changes in one modality affect the other modality. These interactions can be quantified and analysed using techniques such as feature fusion, where features from multiple modalities are combined to extract meaningful patterns and relationships. + +In the context of multi-modal data coupling for crowdsensing tasks, implicit interactions refers to the interplay between modalities that may not have obvious connections on the surface. For example, implicit interactions could refer to the influence of ambient noise on the accuracy of visual sensor data. While there might not be a clear, direct correlation between the noise level and visual observations, implicit interactions can reveal that certain noise patterns impact the reliability of visual data in specific contexts. + +#### **7.3.2.2 For crowdsensing multimedia data reliability evaluation** + +- It is recommended to utilize AI schemes to aggregate valuable information from the raw data and filter out irrelevant or redundant data. +- It is required to assess the consistency of data across different modalities and dimensions. For instance, if images captured by participants contradict the accompanying location data, it could be an indication of unreliable or manipulated data. +- It is required to compare crowdsensing data with data from authorized or trusted sources to ensure accuracy and reliability. +- When comparing environmental sensing data, it is recommended to cross-reference with data from official environmental monitoring stations. Any discrepancies between the two sources can raise concerns regarding the reliability of the crowdsensing data. + +#### **7.3.2.3 For crowdsensing multimedia data quality enhancement** + +- It is required that the crowdsensing platform enhances the raw data quality by AI models. For example, AI models can generate high-quality images or videos by reducing noise, enhancing details, improving colour balance, or increasing resolution. By employing AI models for image and video enhancement, the visual sensing accuracy of a crowdsensing task can be significantly improved. +- It is recommended to apply AI models to correct errors or reconstruct missing data in multimedia data. For example, machine learning models can be trained to detect and correct errors or fill in missing segments in sensor data or video frames. This approach ensures the accuracy, fidelity and completeness of the collected data. +- It is recommended to apply AI schemes such as NLP models to extract meaningful information from voice or video calls in real time. + +For example, AI models can analyse conversations and detect keywords or phrases related to specific crowdsensing tasks. This contextual understanding helps in associating the data being uploaded with the relevant crowdsensing tasks, ensuring accurate tagging and categorization. + +- It is recommended to apply AI models on historical crowdsensing data for predictive analysis. For instance, based on previous patterns and trends, AI algorithms can estimate the expected data volume, identify potential anomalies or outliers, or even provide real-time feedback and suggestions to participants for optimizing their data collection efforts. +- It is required to apply AI schemes to assess the quality of multimedia data. For example, AI models can analyse the clarity, resolution, or overall quality of images or videos uploaded by participants. This allows for automatic identification of data with potential issues or lower quality, which can be flagged for further examination or refinement. +- If the crowdsensing tasks are incapable to be completed due to the lack of useful information, it is recommended to restart the crowdsensing task and calls for new data contributions. +- It is recommended to apply AI models to simulate and generate an expected content based on existing samples. For example, AI can generate the expected key components of entities within images, and then these generated image examples will be distributed to participants to help participants submit accurate image sensing data. +- If the crowdsensing tasks are incapable of being completed due to image low quality, it is recommended to apply AI models to enhance and reconstruct the image quality. Image quality enhancements include the following cases: image defogging, image contrast enhancement, image definition enhancement, removing unnecessary obstructions in the image and other image optimization methods. +- For crowdsensing tasks that involve long-term participation from multiple individuals, the collected images will vary across different time periods. It is recommended to employ time series prediction AI models to monitor and forecast changes in image details. +- For time-sensitive tasks where it is challenging to accumulate a sufficient amount of high-quality data in a short period, it is required to initially utilize coarse-grained crowdsensing data to complete the crowdsensing tasks. +- As the crowdsensing data continues to accumulate, it is recommended to progressively select high quality sensing data and apply reinforcement learning AI models and transfer learning models for data retraining in the later stages of the crowdsensing tasks. +- For crowdsensing tasks focused on monitoring entity details, basic image classification AI models are insufficient for distinguishing the subtle changes in entity details. It is recommended to employ the image segmentation AI models to get a comprehensive understanding of the information within the images. + +## **7.4 Requirements for multimedia crowdsensing data privacy enhancement** + +- It is recommended to add noise or perturbation to the raw data to prevent individual identification while preserving the utility of the sensed data. For instance, differential privacy measures are recommended to be used. +- It is recommended to apply AI techniques for anonymizing and de-identifying multimedia data to remove personally identifiable information from the uploaded content. AI models can be employed to automatically detect and blur faces, remove metadata, or encrypt sensitive information within the data. + +- When conducting crowdsensing tasks that involve highly confidential raw data, it is recommended to apply collaborative AI training models without sharing raw data among participants. +For instance, federated learning or secure multi-party computation is recommended while maintaining the confidentiality and privacy of the sensitive data provided by individual contributors. +- It is recommended to utilize encryption protocols and secure communication channels to ensure the privacy of multimedia data during transmission. +- It is recommended to apply AI schemes by defining fine-grained access policies based on roles, privileges, and specific user attributes. This ensures that only authorized individuals or entities can access and analyse the data. +- It is recommended to develop user-friendly interfaces that offer clear information about the data collection process and empower users to customize their privacy preferences. +- It is recommended to implement secure data storage practices, including encryption of stored data, regular backups, and secure data deletion when no longer needed. +- It is recommended to leverage AI in monitoring data storage systems for vulnerabilities and identifying anomalous access or usage patterns. +- It is recommended to conduct periodic privacy audits and assessments of the crowdsensing system to identify potential privacy risks and vulnerabilities. + +# 8 Framework of crowdsensing multimedia interaction based on AI + +Figure 8-1 illustrates the framework of crowdsensing multimedia interaction based on AI, which is detailed in clauses 8.1, 8.2 and 8.3. + +![Figure 8-1: The framework of intelligent crowdsensing multimedia interaction based on AI. The diagram shows two main layers: the Centralized crowdsensing platform layer and the Crowdsensing resource layer, connected by network connections. The platform layer includes units for task solution pool, task assignment, sensing data storage, data processing, sensing nodes management, incentive evaluation, data security management, and task result evaluation. The resource layer includes units for task execution, raw data pre-processing, data flow management, and participants PII management. Both layers are supported by AI capabilities, AI model services, and computing resources.](79e1709a7317ead45379cbb8ff3ba802_img.jpg) + +The diagram illustrates the framework of intelligent crowdsensing multimedia interaction based on AI. It consists of two main layers connected by network connections (indicated by lightning bolts and a double-headed arrow). The top layer is the **Centralized crowdsensing platform layer**, which contains eight units arranged in a 2x4 grid: Task solution pool, Task assignment, Sensing data storage, Data processing, Sensing nodes management, Incentive evaluation, Data security management, and Task result evaluation. The bottom layer is the **Crowdsensing resource layer**, which contains four units: Task execution, Raw data pre-processing, Data flow management, and Participants PII management. To the right of the platform layer, there is a box labeled **AI capabilities** containing **AI model services**. To the right of the resource layer, there is a box labeled **Computing resources**. Arrows indicate that both layers interact with these AI and computing resources. The diagram is labeled F.748.23(24) in the bottom right corner. + +Figure 8-1: The framework of intelligent crowdsensing multimedia interaction based on AI. The diagram shows two main layers: the Centralized crowdsensing platform layer and the Crowdsensing resource layer, connected by network connections. The platform layer includes units for task solution pool, task assignment, sensing data storage, data processing, sensing nodes management, incentive evaluation, data security management, and task result evaluation. The resource layer includes units for task execution, raw data pre-processing, data flow management, and participants PII management. Both layers are supported by AI capabilities, AI model services, and computing resources. + +Figure 8-1 – The framework of intelligent crowdsensing multimedia interaction based on AI + +## 8.1 Centralized crowdsensing platform layer + +### 8.1.1 Task solution pool unit + +The task solution pool unit conducts a comprehensive analysis of crowdsensing task requirements and categories, enabling the task assignment unit to accurately determine the most suitable solution, participants, and resources required to accomplish each task. + +This unit effectively manages a diverse range of task solutions, including online or offline task assignment methods, homogeneous or heterogeneous sensing nodes, and task distribution within a single region or across multiple regions. By incorporating these versatile solutions, the pool + +significantly enhances flexibility, efficiency, and scalability in managing and utilizing crowd-generated solutions. + +### **8.1.2 Task assignment unit** + +In line with the requirements identified in clause 7.1, the task assignment unit calls the AI capabilities to dynamically and intelligently allocate tasks to specific participants and sensing nodes based on multiple factors. + +The unit leverages AI capabilities for task decomposition, sub-task combination, task distribution, and task scheduling. + +The unit leverages AI capabilities and collaborates with the task solution pool units to assess suitable sensing node recommendations for varied tasks. + +### **8.1.3 Sensing data storage unit** + +The sensing data storage unit offers robust data storage, synchronization, and backup capabilities, enabling reliable and secure storage of multimedia data generated by crowdsensing participants or sensing nodes. It is capable of quick retrieval functionality, data security measures, scalability, and flexible integration. + +### **8.1.4 Data processing unit** + +In line with the requirements defined in clause 7.3 and clause 7.4, the data processing unit calls the AI capabilities to execute multimedia data quality enhancement and multimedia content analysis on the centralized crowdsensing platform layer. + +The unit conducts data reliability evaluation. It evaluates the reliability of the collected data, enhancing its quality and reliability by eliminating noise, outliers, and irrelevant data points through the process of data cleaning. + +The unit performs data feature engineering and multi-modal data coupling by leveraging AI capabilities. This allows it to integrate and consolidate multimedia data from multiple sources or solution providers, enabling the derivation of more accurate and comprehensive information. + +The unit exploits AI capabilities to enhance the quality of multimedia data, thereby improving the overall performance of crowdsensing tasks. + +The unit prioritizes data privacy protection and takes measures to safeguard the privacy and confidentiality of the crowd sensed data. + +### **8.1.5 Sensing nodes management unit** + +The sensing nodes management unit is responsible for managing the operations status of sensing nodes used in the crowdsensing platform. + +One of the key capabilities of the sensing nodes management unit is that it is capable of registering and provisioning new crowdsensing nodes onto the crowdsensing platform and of managing the crowdsensing nodes profiling, including gathering information such as sensing devices specifications, computation capabilities, supported protocols, and available features. + +The sensing nodes management unit also tracks device status and monitors important operation parameters such as battery life, connectivity, computation capability and functionality to ensure the optimal sensing node recommendation for varied crowdsensing tasks. + +### **8.1.6 Incentive evaluation unit** + +In line with the requirements defined in clause 7.1, the incentive evaluation unit matches proper incentives for participants and promotes fair incentive allocation. The unit calls the AI capabilities to allocate incentives that match the participants' performance and engagement. + +### **8.1.7 Result evaluation unit** + +The result evaluation unit calls the AI capability to evaluate the overall result of task execution. This unit provides the feedback for AI capabilities to help improve the training models. + +### **8.1.8 Data privacy management unit** + +In line with the requirements identified in clause 7.4, the data privacy management unit implements comprehensive security measures to protect the collected data. Access control mechanisms are implemented to regulate data access and ensure that only authorized individuals can retrieve or modify the stored data. + +## **8.2 Crowdsensing resource layer** + +### **8.2.1 Task execution unit** + +The task execution unit is capable of executing various types of tasks assigned by the crowdsensing platform. These tasks include data collection from different sensors, such as GPS, accelerometer, microphone, or camera. + +The unit collects sensing data from the sensors available on the crowdsensing node. It is capable of interfacing with different sensors and of retrieving the relevant sensing data, such as location coordinates, environmental parameters, noise levels, image, or video. + +### **8.2.2 Raw data pre-processing unit** + +In line with the requirements identified in clause 7.2, the raw data pre-processing unit calls AI capabilities to carry out data pre-processing on the crowdsensing resource layer. + +This unit calls the AI capabilities to perform redundant data filtering, data aggregation, and transformation operations to extract valuable insights or features from the raw data on the crowdsensing resource layer. + +This unit allows for data compression and prioritization to reduce the network transmission costs. + +### **8.2.3 Data flow management unit** + +In line with the requirements identified in clause 7.2, the data flow management unit collaborates with the device capability management unit to collect network connection statuses and device statuses of sensing nodes. It then calls AI capabilities to evaluate data transmission consumption, including energy consumption, network costs, transmission delays, and other relevant factors. By utilizing this analysis, this unit is capable of optimizing data transmission processes, improving resource utilization, and enhancing the overall efficiency and performance of the crowdsensing task. + +This unit is capable of transporting the raw data to the edge computing platform for AI training or inference if the participants' crowdsourced resource is incapable of performing AI computation. + +### **8.2.4 Participants privacy management unit** + +In line with the requirements identified in clause 7.4, this unit implements comprehensive security measures to protect the participants' data before transmitting the raw data to the centralized crowdsensing platform. It employs encryption techniques to safeguard sensitive information, ensuring that privacy and confidentiality are maintained. + +## **8.3 AI capabilities** + +### **8.3.1 AI model service unit** + +The AI model service unit provides AI training, AI interference and AI algorithms development for the crowdsensing platform layer and crowdsensing resource layer. By optimizing AI models, fine-tuning AI algorithms, and providing efficient AI model computation services, it facilitates real-time + +data processing, analysis, and decision-making for the crowdsensing tasks. Continuous monitoring and performance evaluation further enhance AI model effectiveness and reliability, enabling the unit to meet the evolving needs of crowdsensing tasks. + +### **8.3.2 Computing resources unit** + +The computing resource unit furnishes essential infrastructure, encompassing hardware such as CPUs and GPUs, alongside storage facilities, vital for processing crowdsensing data and executing AI models. Additionally, it provisions edge AI and fog computing resources tailored for lightweight AI models deployed within crowdsensing resources layer. Ensuring scalability, the unit dynamically adjusts computing resources in response to workload fluctuations, thereby furnishing the necessary flexibility to manage diverse data volumes and processing needs effectively. + +# Bibliography + +- [b-ITU-T Y.3172] Recommendation ITU-T Y.3172 (2019), *Architectural framework for machine learning in future networks including IMT-2020*. +- [b-ITU-T Y.4205] Recommendation ITU-T Y.4205 (2019), *Requirements and reference model of IoT-related crowdsourced systems*. +- [b-ETSI GR ENI 004] ETSI GR ENI 004 (2018), *Experiential Networked Intelligence (ENI); Terminology for Main Concepts in ENI*. + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.24-202404-I_PDF-E/raw.md b/marked/F/T-REC-F.748.24-202404-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..21d96007bcb95358795ee77be2d7b0fa8b5e8839 --- /dev/null +++ b/marked/F/T-REC-F.748.24-202404-I_PDF-E/raw.md @@ -0,0 +1,859 @@ + + +# Recommendation + +## **ITU-T F.748.24 (04/2024)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +## --- **Trusted contribution evaluation framework on federated machine learning services** + +![ITU logo](84a1d09fb489061482111515543b60dc_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.24 + +## Trusted contribution evaluation framework on federated machine learning services + +## Summary + +Federated machine learning (FML) is an emerging distributed framework that enables collaborative machine learning (ML) and model construction across distributed and decentralized datasets. FML service has distinctive features, such as location of data in the calculation, and data availability without visibility. It allows participants to jointly train ML models without sharing raw data, which can technically break data isolation and promote cooperation among data owners. + +FML service involves multiple participants who usually perform different contributions to ML model training tasks due to their many impact factors. An effective and trusted contribution evaluation mechanism for FML service is essential to increase participation of the parties involved and can promote sustainable development of FML services. + +Recommendation ITU-T F.748.24 introduces a trusted contribution evaluation service for FML service that combines and takes advantage of FML and distributed ledger technology functionalities, and provides relevant concepts, characteristics, requirements and use cases, and specifies a relevant reference framework and common capabilities. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.24 | 2024-04-15 | 16 | 11.1002/1000/15609 | + +## Keywords + +Contribution degree, evaluation service federated machine learning. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2024 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|------|----------------------------------------------------------------------------------------------|----| +| 1 | Scope ..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 2 | +| 4 | Abbreviations and acronyms ..... | 2 | +| 5 | Conventions ..... | 2 | +| 6 | Overview of a contribution evaluation service for federated machine learning services ..... | 2 | +| 7 | Characteristics and requirements of a contribution evaluation service for FML services ..... | 4 | +| 7.1 | Characteristic of contribution evaluation service for FML services ..... | 4 | +| 7.2 | Common requirements of a contribution evaluation service for FML services ..... | 5 | +| 8 | Reference framework of a contribution evaluation service for FML services ..... | 5 | +| 8.1 | FML service agent..... | 6 | +| 8.2 | Data collection-functional component ..... | 6 | +| 8.3 | Contribution calculator-functional component..... | 6 | +| 8.4 | ML model management-functional component ..... | 6 | +| 8.5 | DLT agent-functional component ..... | 7 | +| 8.6 | Task management-functional component..... | 7 | +| 9 | Common capabilities of a contribution evaluation service for FML services..... | 7 | +| 9.1 | Data storage of FML services..... | 7 | +| 9.2 | Identity registration of FML services ..... | 7 | +| 9.3 | FML model-training task registration of FML services ..... | 8 | +| 9.4 | FML model-training task execution (one-time computation) ..... | 10 | +| 9.5 | FML model-training task execution (multiple repetitive computation) ..... | 11 | +| 9.6 | Evaluation of FML model-training task contribution ..... | 13 | +| 10 | General functional components of coordinators of and participants in FML training..... | 14 | +| 10.1 | FML training coordinator ..... | 14 | +| 10.2 | FML training participant ..... | 16 | +| 11 | Security consideration ..... | 17 | +| | Appendix I – Use cases of a contribution evaluation service for FML services ..... | 18 | +| | Bibliography..... | 19 | + + + +## Trusted contribution evaluation framework on federated machine learning services + +# 1 Scope + +This Recommendation provides a trusted contribution evaluation service (CES) for federated machine learning (FML) services, and specifies relevant requirements, a reference framework and common capabilities. + +This Recommendation includes: + +- the concept, characteristics and requirements of a CES for FML services; +- a reference framework and common capabilities of a CES for FML services. + +Use cases of trusted a CES for FML services. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 distributed ledger** [b-ITU-T X.1400]: A type of ledger that is shared, replicated, and synchronized in a distributed and decentralized manner. + +**3.1.2 distributed ledger technology (DLT)** [b-ITU-T X.1400]: Technology that enables the operation and use of distributed ledgers. + +**3.1.3 distributed ledger technology system (DLT system)** [b-ITU-T X.1400]: A system that implements a distributed ledger. + +**3.1.4 federated machine learning (FML)** [b-IEEE 3652.1]: Federated machine learning (FML) is a framework or system that enables multiple participants to collaboratively build and use machine learning models without disclosing the raw and private data owned by the participants while achieving good performance. + +**3.1.5 ledger** [b-ITU-T X.1400]: Information store that keeps final and definitive (immutable) records of transactions. + +**3.1.6 service** [b-ITU-T Y.2091]: A set of functions and facilities offered to a user by a provider. + +**3.1.7 trust** [b-ITU-T Y.3052]: The measurable belief and/or confidence which represents accumulated value from history and the expecting value for the future. + +NOTE – Trust is quantitatively and/or qualitatively calculated and measured. Trust is used to evaluate values of entities, value-chains among multiple stakeholders and human behaviours, including decision-making. + +**3.1.8 trusted system** [b-ISO/TR 15801]: Information technology system with the capability of managing electronically stored information in a trustworthy manner. + +**3.1.9 trustworthy** [b-ISO/TR 15801]: Ability to demonstrate authenticity, integrity and availability of electronically stored information over time. + +## 3.2 Terms defined in this Recommendation + +None. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-----|-------------------------------------| +| CC | Contribution Calculator | +| CES | Contribution Evaluation Service | +| DC | Data Collection | +| DLT | Distributed Ledger Technology | +| FC | Functional Component | +| FML | Federated Machine Learning | +| ML | Machine Learning | +| MMM | ML Model Management | +| PII | Personally Identifiable Information | +| TM | Task Management | + +# 5 Conventions + +This Recommendation uses the following conventions: + +- The phrase "is required to" indicates a requirement that must be strictly followed and from which no deviation is permitted if conformance with this Recommendation is to be claimed. +- The phrase "is recommended" indicates a requirement that is recommended, but which is not absolutely required to claim conformance with this Recommendation. +- The phrase "can optionally" indicates an optional requirement that is permissible, without implying any sense of being recommended. This term is not intended to imply that the vendor's implementation must provide the option, and the feature can be optionally enabled by the network operator or service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with this Recommendation. + +# 6 Overview of a contribution evaluation service for federated machine learning services + +FML is an emerging distributed framework that enables collaborative ML and model construction across distributed and decentralized datasets. FML service has distinctive features, such as location of data in the calculation, and data availability without visibility. It allows participants to jointly train ML models without sharing raw data directly, which can technically break data isolation and promote cooperation among data owners. + +FML service involves multiple participants who usually perform different contributions to ML model-training tasks due to many impact factors of the participants, such as quantity, quality and + +types of datasets, and capabilities of computing and communication. An effective and trusted contribution evaluation mechanism for FML service is essential to increase participation of the parties involved and can promote the sustainable development of FML services. + +In the traditional coordination paradigm of FML services, depicted in Figure 6-1-a, FML training coordinators exchange information directly with each FML training participants for ML model training and utilization, in which the exchanged information usually includes modules, parameters and dataset indexes. FML training coordinators can store all exchanged information and final ML models, but FML training participants only can store part of the exchanged information generated by themselves individually. In this paradigm, for an ML model-training task, usually only FML training coordinators have full information to calculate the contribution degree of FML training participants. + +Distributed ledger technologies (DLTs) have inherent advantages to leverage data management and sharing, such as peer to peer communication, decentralization, data immutability, openness, crowd consensus, and support and automation of smart contracts. DLT systems can be used as trustworthy shared storage for exchanging data for ML model-training tasks, and also for algorithms and modules to evaluate contributions. In this case, FML training coordinators and FML training participants can calculate contribution degrees for ML model-training tasks independently and consistently. + +The trusted CES takes advantages of DLT and enhances contribution evaluation capability on FML services. Figure 6-1-b depicts the relevant enhanced coordination paradigm of FML services for contribution evaluation. In this paradigm, FML training coordinators and FML training participants exchange information for ML model training and utilization through CES and DLT systems. The CES manages ML model-training tasks, and acts as a bridge for FML training coordinators and FML training participants to exchange information for ML model training and utilization. The CES also stores relevant information in DLT systems. DLT systems guarantee the reliability, integrity, consistency, availability and non-tampering of relevant information. The CES, supported by DLT systems, provides a trusted evaluation service to evaluate contribution degree on FML services for each participant. + +![Figure 6-1: Overview of a trusted contribution evaluation service for FML services. (a) Traditional coordination paradigm: FML training participants and FML training coordinators exchange information directly. (b) Enhanced coordination paradigm: FML training participants, FML training coordinators, and a Contribution evaluation service (CES) on FML services all interact with a central Trustworthy shared storage (such as DLT systems).](7a0db9703b68b3d06cdaeefc084c0006_img.jpg) + +The diagram illustrates two coordination paradigms for FML services. Part (a), titled 'Traditional coordination paradigm of FML services', shows a group of 'FML training participants' (represented by three overlapping dashed boxes) and 'FML training coordinators' (represented by a single dashed box) connected by a double-headed arrow. Below the coordinators, a list indicates the information exchanged: '- ML models', '- Information for ML model training and utilization ...'. Part (b), titled 'Enhanced coordination paradigm of FML services for contribution evaluation', shows the same 'FML training participants' and 'FML training coordinators' boxes, but they are now connected to a central 'Contribution evaluation service (CES) on FML services' box. This CES box is further connected to a 'Trustworthy shared storages (such as DLT systems)' box at the bottom. All three components (participants, coordinators, and CES) have double-headed arrows connecting them to the shared storage. A small label 'F.748.24(24)' is present in the bottom right corner of part (b). + +Figure 6-1: Overview of a trusted contribution evaluation service for FML services. (a) Traditional coordination paradigm: FML training participants and FML training coordinators exchange information directly. (b) Enhanced coordination paradigm: FML training participants, FML training coordinators, and a Contribution evaluation service (CES) on FML services all interact with a central Trustworthy shared storage (such as DLT systems). + +**Figure 6-1 – Overview of a trusted contribution evaluation service for FML services** + +In addition, FML training coordinators can store modules for ML model training and modules for evaluating contributions in DLT systems. FML training participants can store indexes of datasets in DLT systems. Furthermore, FML training coordinators and FML training participants can initiate ML model-training tasks through CES and DLT systems. + +# **7 Characteristics and requirements of a contribution evaluation service for FML services** + +## **7.1 Characteristic of contribution evaluation service for FML services** + +### **7.1.1 Full-lifecycle data collection and management for FML services** + +A CES can collect and manage process data for FML services over their full lifecycle, including data concerning, at least: + +- creation of ML training tasks; +- management of ML model-training tasks; +- transmission between FML training participants and FML training coordinators; +- ML model training and utilization. + +### **7.1.2 Trustworthy storage for FML services** + +A CES adopts DLT system(s) as trustworthy storage for exchanged data for ML model-training tasks, and also for algorithms and modules to evaluate contributions. DLT system(s) guarantee(s) the reliability, integrity, consistency, availability and non-tampering of data. + +NOTE – A DLT system acts as trustworthy shared storage of CES, but CES is independent of DLT systems. DLT itself lies outside the scope of this Recommendation. + +### **7.1.3 Pluggable modules for contribution degree evaluation** + +A CES provides pluggable modules to evaluate contribution degree. Pluggable modules are usually transparent and stored in DLT systems. FML training participants and coordinators can use and validate the pluggable modules independently. + +NOTE – The model for calculating the contribution degree is jointly formulated and provided by the FML training participants and coordinators, and the CES. Calculation of contribution degree lies outside the scope of this Recommendation. + +#### **7.1.4 Independent contribution degree evaluation** + +An FML service involves multiple participants and allows them to jointly training ML models without sharing raw data directly. A CES provides a decentralized contribution degree evaluation approach to ensure that each participant can calculate their contribution without revealing data. + +#### **7.1.5 Consistency in contribution degree calculation** + +When any participant in and coordinator of FML training uses the same contribution calculation approach provided by the CES to calculate the contribution degree for a FML training task, the result is same. + +The results obtained by different contribution calculation approaches are consistent. For example, if the contribution degree of FML training participant A is more than that of B, which is calculated by one approach, the contribution degree is consistent with that obtained by another method. + +FML training participants and coordinators are able to evaluate the trustworthiness of the results obtained by the contribution calculation approaches provided by a CES. + +#### **7.1.6 Security and personally identifiable information** + +A CES supports data security and personally identifiable information (PII) protection for data collection (DC) and data storage. A CES exposes the data and capabilities according to its security and PII protection policies. + +### **7.2 Common requirements of a contribution evaluation service for FML services** + +### **7.2.1 Requirements for data storage** + +- A CES is required to support participants in and coordinators of FML training by storing information about FML model-training tasks (such as time, data source, data quantity and data quality) in DLT systems after each round of aggregation of an FML model-training task. +- A CES is required to support participants in and coordinators of FML training by storage of the status of FML model-training tasks (such as model update time, model parameter changes and model result) in DLT systems after each round of aggregation of an FML model-training task. +- A CES is required to support FML training coordinators by storage of operational data of FML model-training tasks in DLT systems after each round of aggregation of an FML model-training task. +- A CES is required to support participants in and coordinators of FML training by retrieval of data from trustworthy shared storages (such as DLT system(s)). + +NOTE – DLT systems can be used as trustworthy storage for exchanged data for FML model-training tasks, and also for algorithms and modules to evaluate contributions. + +### **7.2.2 Requirements for collecting data to calculate contribution degree** + +- A CES is required to support FML training participants and coordinators by collecting data for FML model-training tasks (such as time, data source, data quantity and data quality) from relevant DLT systems to evaluate contribution degree. +- A CES is required to support FML training participants and coordinators by collection of the model update status of FML training tasks (such as model update time, model parameter changes and model result) from relevant DLT systems to evaluate contribution degree. + +### **7.2.3 Requirements for rewarding participants and coordinators** + +- A CES is recommended to build mechanisms to allow FML training participants and coordinators by obtaining and validating their contribution degree for FML services. +- A CES can optionally incentivize FML training participants by providing feedback on their respective contribution to the overall FML training task. + +### **7.2.4 Requirements for ML model management to calculate contribution degree** + +- A CES is required to provide an ML model management (MMM) mechanism to ensure ML model security. +- A CES is required to provide pluggable modules for FML training participants and coordinators to evaluate contribution degree. +- A CES is required to provide contribution degree assessment mechanisms to ensure that contribution calculation results are trustworthy. + +### **7.2.5 Requirements for security protection** + +- A CES is required to provide a security mechanism when up- or downloading data to or from DLT systems. +- A CES is required to provide a PII protection mechanism when up- or downloading data to or from DLT systems. + +# **8 Reference framework of a contribution evaluation service for FML services** + +A CES consists of six main functional components (FCs), including FML service agents, DC, contribution calculator (CC), model management, DLT agents and task management. Figure 8-1 shows a reference framework of a trusted CES for FML services. + +![Figure 8-1: Reference framework of a contribution evaluation service for FML services. The diagram shows a layered architecture. At the top, 'FML services' are represented by nested dashed boxes containing 'FML training participants', 'Model parameters', 'Data usages', and 'FML training coordinators'. These components are connected to a central 'CES' (Contribution Evaluation Service) block via double-headed arrows. The 'CES' block contains 'FML service agents', 'Data collection', 'Contribution calculator', 'Model management', and 'DLT agents', which are all connected to a 'Task management' component. The 'CES' block is also connected to 'Trustworthy shared storages (such as DLT systems)' at the bottom via double-headed arrows. The label 'F.748.24(24)' is at the bottom right.](d4af765160d04ecef538e5066006dc77_img.jpg) + +Figure 8-1: Reference framework of a contribution evaluation service for FML services. The diagram shows a layered architecture. At the top, 'FML services' are represented by nested dashed boxes containing 'FML training participants', 'Model parameters', 'Data usages', and 'FML training coordinators'. These components are connected to a central 'CES' (Contribution Evaluation Service) block via double-headed arrows. The 'CES' block contains 'FML service agents', 'Data collection', 'Contribution calculator', 'Model management', and 'DLT agents', which are all connected to a 'Task management' component. The 'CES' block is also connected to 'Trustworthy shared storages (such as DLT systems)' at the bottom via double-headed arrows. The label 'F.748.24(24)' is at the bottom right. + +**Figure 8-1 – Reference framework of a contribution evaluation service for FML services** + +## 8.1 FML service agent + +FML service agents interact with external FML services. FML training participants and FML training coordinators connect to FML service agents to exchange information about data usage and model parameters. + +FML service agents expose functional interfaces for FML training participants and FML training coordinators to interact with each other. + +## 8.2 Data collection-functional component + +A DC-functional component (DC-FC) supports participants in and coordinators of FML training. A DC-FC collects and manages information related to data usage, data exchange, ML model training, model management, etc. + +## 8.3 Contribution calculator-functional component + +A CC-functional component (CC-FC) is used to calculate the contribution degree. A CC-FC supports participants in and coordinators of FML training by calculating the contribution degree based on the FML training status, data usage, data quality, data quantity, model-training parameters, etc. + +## 8.4 ML model management-functional component + +An MMM-functional component (MMM-FC) provides trustworthy full-lifecycle management of ML model-training tasks. An MMM-FC includes at least: + +- initiation or cessation of ML training; +- MMM; +- ML result management; +- management of contribution degree result for FML services. + +## 8.5 DLT agent-functional component + +A DLT agent of a CES interacts with DLT systems, which provides capabilities related to trustworthy storage, traceability and their management. + +## 8.6 Task management-functional component + +A task management-functional component (TM-FC) provides management mechanism and services policies for FML model-training tasks, including security authentication mechanisms. + +# 9 Common capabilities of a contribution evaluation service for FML services + +This clause provides common capabilities and procedures of a trusted CES on FML services, according to the requirements specified in clause 7 and reference framework specified in clause 8. + +## 9.1 Data storage of FML services + +See Figure 9-1. FML training coordinators and FML training participants for FML service can store data directly in a DLT system (steps 1 and 4) or store data through a CES (steps 2 and 5). A CES can process data sent by FML services, including checking validity and authority (steps 3 and 6). Data after checking and processing is usually stored in the DLT system (step 7). + +FML training participants and FML training coordinators attach identity information when storing data. If information about an FML training task is stored, its identity is also required. If the relevant data is stored in the DLT system through a CES, the CES attaches its identity-related information to the data stored in the DLT system. Attaching identity is to facilitate further searching, extraction and processing of relevant data. + +![Sequence diagram showing data storage procedures in an FML system. Lifelines: FML training participants, FML training coordinators, FML service agent, Task management, DLT agent, and DLT system. The diagram shows two main flows: one where participants store data directly in the DLT system (steps 1, 4, 7) and another where they store data via the CES (steps 2, 3, 5, 6, 7).](e9d825d87c5f85c8dba0664eace96ef4_img.jpg) + +``` +sequenceDiagram + participant FMLTP as FML training participants + participant FMLTC as FML training coordinators + subgraph CES + FSA as FML service agent + TM as Task management + DA as DLT agent + end + LS as DLT system + + Note left of FMLTP: FML services system + Note right of DA: CES + + FMLTP->>LS: 1. Store data to the DLT system + FMLTP->>FMLTC: 2. Store data to the CES + FMLTC->>FSA: 3. Process request + FSA->>TM + TM->>DA + DA->>LS: 7. Data storage + FMLTC->>LS: 4. Store data to the DLT system + FMLTC->>FMLTP: 5. Store data to the CES + FMLTP->>FSA: 6. Process request + FSA->>TM + TM->>DA + DA->>LS: 7. Data storage +``` + +F.748.24(24) + +Sequence diagram showing data storage procedures in an FML system. Lifelines: FML training participants, FML training coordinators, FML service agent, Task management, DLT agent, and DLT system. The diagram shows two main flows: one where participants store data directly in the DLT system (steps 1, 4, 7) and another where they store data via the CES (steps 2, 3, 5, 6, 7). + +Figure 9-1 – Reference procedures of data storage in FML system + +## 9.2 Identity registration of FML services + +FML training coordinators and FML training participants firstly register in a CES before they can organize or participate in FML training tasks. During registration, FML training coordinators and FML training participants need to provide user identity, public key and certificates. FML training coordinators and FML training participants store their private key corresponding to their public key. A CES uses its own private key to sign certificates, store signature information in the corresponding certificates and feeds back to the FML training coordinators and FML training participants. The CES stores registration information about FML training coordinators and FML training participants, and related information can also be stored in the DLT system. See Figure 9-2. + +![Sequence diagram showing identity registration in the FML system. The diagram is divided into two main lifelines: 'FML services system' and 'CES'. The 'FML services system' contains 'FML training participants' and 'FML training coordinators'. The 'CES' contains 'FML service agent', 'Task management', 'DLT agent', and 'DLT system'. The sequence of messages is: 1. Request to register device (from participants and coordinators to service agent); 2. Process request (from service agent to task management); 3. Check strategy (self-call on task management); 4. Store registration information (from task management to DLT system via DLT agent); 5. Process request (from task management back to service agent); 6. Feedback registration request (from service agent to participants and coordinators). A label F.748.24(24) is at the bottom right.](4801720824e4b5e2361a5564f91cfb70_img.jpg) + +``` + +sequenceDiagram + participant P as FML training participants + participant C as FML training coordinators + participant SA as FML service agent + participant TM as Task management + participant DA as DLT agent + participant DS as DLT system + + Note left of P: FML services system + Note right of DS: CES + + P->>SA: 1. Request to register device + C->>SA: 1. Request to register device + SA->>TM: 2. Process request + TM->>TM: 3. Check strategy + TM->>DA: 4. Store registration information + DA->>DS: + TM->>SA: 5. Process request + SA->>C: 6. Feedback registration request + SA->>P: 6. Feedback registration request + +``` + +F.748.24(24) + +Sequence diagram showing identity registration in the FML system. The diagram is divided into two main lifelines: 'FML services system' and 'CES'. The 'FML services system' contains 'FML training participants' and 'FML training coordinators'. The 'CES' contains 'FML service agent', 'Task management', 'DLT agent', and 'DLT system'. The sequence of messages is: 1. Request to register device (from participants and coordinators to service agent); 2. Process request (from service agent to task management); 3. Check strategy (self-call on task management); 4. Store registration information (from task management to DLT system via DLT agent); 5. Process request (from task management back to service agent); 6. Feedback registration request (from service agent to participants and coordinators). A label F.748.24(24) is at the bottom right. + +**Figure 9-2 – Reference procedures of identity registration in FML system** + +The relevant steps mainly include the following. + +Step 1. One or more FML training coordinator and one, usually more, FML training participant register themselves with a CES. + +Step 2. The FML services agent of the CES receives and processes relevant requests, such as conducting permission and other aspects of detection. + +Step 3. The TM-FC of the CES detects related request information. If it passes the check, the TM-FC uses the private key of the CES to sign the certificates of the multiple FML training participants and FML training coordinators. + +Step 4. If it passes the inspection, the TM-FC of the CES stores relevant data in the DLT system through the DLT agent. + +In the process of checking the FML training coordinators and the FML training participants, the CES can obtain the parameter information required for inspection from the DLT platform. + +Steps 5 and 6. The CES feeds registration information back to the FML training coordinators and the FML training participants. If the registration is successful, the FML training coordinators and FML training participants can obtain their own certificates signed by the CES, as well as obtaining its certificate and public key. If the registration fails, the FML training coordinators and FML training participants can obtain registration failure information, including its causes. + +## 9.3 FML model-training task registration of FML services + +FML model-training tasks are registered by FML training coordinators. Before the registration of the FML training task, the FML training coordinators and the FML training participants (can be multiple) negotiate the FML model training task, including the selection of the FML model, and relevant number and data set information. + +When FML model-training task negotiation is completed, whether successful or not, the FML training coordinators and the FML training participants store the relevant information in the DLT system. Related operations can be stored either via a CES or directly in the DLT system. + +When the FML model-training task negotiation is successful, FML training coordinators register the negotiated FML training task information in the CES, and obtain the unique identity of FML service. + +After successful negotiation and registration of the FML model-training task, FML training coordinators distribute the FML model-training task information, including the identity of the FML model-training task, to FML training participants. + +When the FML model-training task is registered successfully, the FML model-training task is executed between the FML training coordinators and the FML training participants. See Figure 9-3. + +![Sequence diagram showing the reference procedures of FML model-training task registration in the FML system. The diagram is divided into two main lifelines: FML services system and CES. The FML services system includes FML training participants and FML training coordinators. The CES includes FML service agent, Task management, DLT agent, and DLT system. The sequence of messages is: 1. Negotiate FML tasks (between participants and coordinators), 2. Request register FML tasks (coordinators to service agent), 3. Process request (service agent to task management), 4. Check strategy (task management self-call), 5. Store FML tasks information (task management to DLT system), 6. Process request (task management to service agent), 7. Feedback registration information (service agent to coordinators), 8. Inform FML tasks (coordinators to participants), 9. Receive and prepare FML tasks (participants to coordinators), 10. Feedback reception and prepare FML tasks (coordinators to service agent), 11. Stores FML task registration information (service agent to DLT system).](724c7777b608e53be38b12b6fb3c43bc_img.jpg) + +``` + +sequenceDiagram + participant FML participants as FML training participants + participant FML coordinators as FML training coordinators + participant FML service agent as FML service agent + participant Task management as Task management + participant DLT agent as DLT agent + participant DLT system as DLT system + + Note left of FML participants: FML services system + Note right of DLT system: CES + + FML participants->>FML coordinators: 1. Negotiate FML tasks + FML coordinators->>FML service agent: 2. Request register FML tasks + FML service agent->>Task management: 3. Process request + Task management->>Task management: 4. Check strategy + Task management->>DLT system: 5. Store FML tasks information + Task management->>FML service agent: 6. Process request + FML service agent->>FML coordinators: 7. Feedback registration information + FML coordinators->>FML participants: 8. Inform FML tasks + FML participants->>FML coordinators: 9. Receive and prepare FML tasks + FML coordinators->>FML service agent: 10. Feedback reception and prepare FML tasks + FML service agent->>DLT system: 11. Stores FML task registration information + +``` + +F.748.24(24) + +Sequence diagram showing the reference procedures of FML model-training task registration in the FML system. The diagram is divided into two main lifelines: FML services system and CES. The FML services system includes FML training participants and FML training coordinators. The CES includes FML service agent, Task management, DLT agent, and DLT system. The sequence of messages is: 1. Negotiate FML tasks (between participants and coordinators), 2. Request register FML tasks (coordinators to service agent), 3. Process request (service agent to task management), 4. Check strategy (task management self-call), 5. Store FML tasks information (task management to DLT system), 6. Process request (task management to service agent), 7. Feedback registration information (service agent to coordinators), 8. Inform FML tasks (coordinators to participants), 9. Receive and prepare FML tasks (participants to coordinators), 10. Feedback reception and prepare FML tasks (coordinators to service agent), 11. Stores FML task registration information (service agent to DLT system). + +**Figure 9-3 – Reference procedures of FML model-training task registration in FML system** + +In order to protect data privacy, each participant in the FML model-training task encrypts data by using the public key of a CES or the encryption algorithm or key agreed in advance when storing data. Thus, only the private key owner of the CES or the owner of the previously agreed decryption algorithm or key can decrypt and use the relevant data. At the same time, the CES stores information about each participant in the FML model-training task, so that when calculating contributions to the FML model-training task, only participants are allowed to query and calculate the contribution degree through task identity. + +In the registration of an FML model-training task, if the registration is successful, the CES generates its unique identity, and stores the relevant identity and the information about the FML model-training task itself or in the DLT system. At the same time, participants in the FML model-training task and the CES negotiate the encryption methods and parameters of the subsequent data. For relevant encryption methods, data can be encrypted directly using the public key of the CES or other pre-set encryption or decryption methods and random encryption or decryption keys can be used. + +The relevant steps mainly include the following. + +Step 1. The FML training coordinators and the FML training participants negotiate the FML model-training task. + +Step 2. If the negotiation is successful, FML training coordinators send the negotiated FML model-training task description information to the CES to register the FML model-training task information. + +Steps 3, 4. The TM-FC of the CES checks the service policy that is expected to be specified, for e.g., whether the FML training coordinators and the FML training participants are registered, and whether the FML model-training task description information is complete. + +Steps 5, 6, 7. The CES stores processing results; in addition, relevant processing results can be stored in the DLT system. At the same time, registration processing results are fed back to FML training coordinators. + +Step 8. FML training coordinators sends feedback to each relevant FML training participant. + +Steps 9, 10, 11. Each relevant FML training participant sends information to FML training coordinators about whether it accepts and intends to participate in the FML model-training task, and may also send it to the CES to store the relevant information. + +## 9.4 FML model-training task execution (one-time computation) + +After successful negotiation and registration of the FML model-training task, FML training coordinators notify each relevant FML training participant to perform the FML training task. Relevant process and result data can be recorded in the DLT system through the CES. The stored data can be encrypted using either the public key of the CES or a pre-agreed encryption algorithm and key before being written to the DLT system. + +FML training participants count information such as the amount of computation spent to execute the FML model-training task and the amount of data accessed to its own data, and upload it to the DLT system through the CES. See Figure 9-4. + +![Sequence diagram showing the reference procedures of FML model-training task execution (one-time computation). The diagram is divided into two main lifelines: 'FML services system' and 'CES'. The 'FML services system' includes 'FML training participants' and 'FML training coordinators'. The 'CES' includes 'FML service agent', 'Data collection', 'Task management', 'DLT agent', and 'DLT system'. The sequence of messages is as follows: 1. Start a (one-time) FML task (from participants to coordinators); 2. Record task information (from coordinators to service agent); 3. Review and record task information (from service agent to DLT agent); 4. Store information (from DLT agent to DLT system); 5. Perform FML tasks (self-call on participants); 6. Feedback the result of FML task execution (from participants to coordinators); 7. Record task information (from participants to service agent); 8. Review and record task information (from service agent to DLT agent); 9. Store information (from DLT agent to DLT system); 10. Record task information (from coordinators to service agent); 11. Review and record task information (from service agent to DLT agent); 12. Store information (from DLT agent to DLT system). Dashed boxes group steps 3-4 and 8-9. The diagram is labeled F.748.24(24) at the bottom right.](cb4cfa42ce34febde7bdb882f3fc3094_img.jpg) + +Sequence diagram showing the reference procedures of FML model-training task execution (one-time computation). The diagram is divided into two main lifelines: 'FML services system' and 'CES'. The 'FML services system' includes 'FML training participants' and 'FML training coordinators'. The 'CES' includes 'FML service agent', 'Data collection', 'Task management', 'DLT agent', and 'DLT system'. The sequence of messages is as follows: 1. Start a (one-time) FML task (from participants to coordinators); 2. Record task information (from coordinators to service agent); 3. Review and record task information (from service agent to DLT agent); 4. Store information (from DLT agent to DLT system); 5. Perform FML tasks (self-call on participants); 6. Feedback the result of FML task execution (from participants to coordinators); 7. Record task information (from participants to service agent); 8. Review and record task information (from service agent to DLT agent); 9. Store information (from DLT agent to DLT system); 10. Record task information (from coordinators to service agent); 11. Review and record task information (from service agent to DLT agent); 12. Store information (from DLT agent to DLT system). Dashed boxes group steps 3-4 and 8-9. The diagram is labeled F.748.24(24) at the bottom right. + +**Figure 9-4 – Reference procedures of FML model-training task execution (one-time computation)** + +The relevant steps mainly include the following. + +Step 1. FML training coordinators notify each FML training participant to perform the one-time FML model-training task after negotiation and registration. The notification contains the identity, model-training modules and parameters of the FML model-training task. + +The notification content may contain the method and key of encryption for the returned data, which is used by FML training participants to return the data after calculation. + +The notification content may include information about other FML training participants, such as task identity and description. + +On its dispatch, FML training coordinators can encrypt notification content by using the public key of FML training participants or other encryption and decryption mechanisms negotiated in advance. When FML training participants receive the information, they can decrypt the notification content by using their private key (or encryption and decryption mechanism negotiated in advance). + +Other mechanisms can also be used to protect communication security and privacy between coordinators of and participants in FML training. + +Steps 2, 3, and 4. FML training coordinators store status information about FML model-training task execution and initiate execution in the DLT system through the CES. + +When a FML training coordinator stores information, it can use its private key pair to store information signatures. The signature method can be generation of a digest of key information (or all information) stored in the information, and then encryption of the generated digest using its own private key. Signature information is stored together with the stored information in the DLT system via the CES. + +FML training coordinators can simultaneously encrypt their own stored information and its signature information using the public key of the CES. In this way, only the private key of the CES can be owned to view the relevant information, to ensure data security and privacy of FML training coordinators. This encryption process can be handled by FML training coordinators themselves or via the CES. FML training coordinators may also encrypt relevant data using other previously agreed encryption algorithms and keys. + +Step 5. Relevant FML training participants receive and analyse the FML model-training task from the FML training coordinators, and then executes the FML model-training task. + +The FML model-training task requires prior negotiation. + +If the FML model-training task information is encrypted, FML training participants decrypt the FML model-training task information according to the agreed rules. + +The relevant FML training participants inspect and certify the FML model-training task. If certified, the FML model-training task is performed. + +During the inspection, certification and implementation process, the relevant parties calculate the relevant information recorded by the participants for uploading to storage and contribution evaluation. + +Step 6. The relevant FML training participants feed the processing results back to the multiple FML training coordinators. If accepted and executed, feedback FML results; if the command is not accepted or the command fails to be executed, the command sends related instructions. + +Steps 7, 8, 9. Relevant FML training participants upload the information related to the inspection, authentication, implementation and feedback operations to the DLT system via the CES. + +Similarly to coordinators, relevant FML training participants calculate a digest of the uploaded data information and sign it. At the same time, the uploaded relevant information can be encrypted using the CES public key (or other previously agreed algorithms or keys). + +After the completion of the FML in steps 10, 11 and 12, the FML training coordinator writes the execution of the FML service into the DLT system through the CES. These steps are similar to 2, 3 and 4. + +## **9.5 FML model-training task execution (multiple repetitive computation)** + +The FML model-training task with multiple repetitive calculations refers to the need for FML training participants to perform repetitive calculations, and need to cooperate with other participants in the + +calculation process (such as exchange of intermediate parameters, etc.). FML training participants can communicate and cooperate with each other through FML training coordinators. + +In the process of multiple repetitive FML model-training tasks, FML training participants perform the calculation multiple times, and during the calculation process, FML training coordinators exchange parameters with other participants, such as intermediate calculation results. At the same time, to successfully execute the FML model-training task, FML training coordinators can also actively send a task evaluation notification to the FML training participants to check and judge the execution effect. + +The termination of multiple repeatable FML model-training tasks may be based on pre-set computation termination conditions or the decision of whether and when to terminate can be made by FML training coordinators. In multiple repeatable FML model-training tasks, each FML training participant performs calculations independently, based on its data, as well as parameters transmitted by the FML training coordinators, and each FML training participant cannot directly transfer data to other participants. + +In the process of repeated FML model-training tasks, the FML training coordinators and the FML training participants continue to write information that can be used for contribution evaluation to the DLT system via the CES as required. Related write procedures and processing mechanisms are similar to FML model-training task execution (one-time computation task) (see clause 9.4). The difference between the two is that in multiple repetitive FML model-training tasks, the FML training coordinators and the FML training participants need to attach the FML model training task identity clearly when writing information to facilitate the subsequent contribution evaluation calculation. See Figure 9-5. + +![Sequence diagram of FML model-training task execution for multiple repetitive computations.](9b9d2abd741ed4bafe7f78f89961c663_img.jpg) + +``` + +sequenceDiagram + participant P as FML training participants + participant C as FML training coordinators + participant SA as FML service agent + participant DC as Data collection + participant TM as Task management + participant DA as DLT agent + participant DLT as DLT system + + Note over P, C: FML services system + Note over SA, DA: CES + + C->>P: 1. Inform start FML task (Continuous type) + P->>C: 2. Start execution + + rect rgb(240, 240, 240) + Note right of P: The continuous execution of FML + P->>P: 3. Execute and update parameters + C->>SA: C1. Record task information + SA->>DA: A. Review and record task information + DA->>DLT: B. Store information + P->>C: 4. Inform evaluation + C->>P: 5. Feedback evaluation + P->>SA: C2. Record execution information + SA->>DA: C. Review and record task information + DA->>DLT: D. Store information + end + + P->>C: 6. End execution + +``` + +The diagram is a sequence diagram showing interactions between two main systems: **FML services system** and **CES**. The **FML services system** contains two lifelines: **FML training participants** and **FML training coordinators**. The **CES** contains four lifelines: **FML service agent**, **Data collection**, **Task management**, and **DLT agent**. The **DLT system** is shown as a separate entity on the far right. + +The sequence of messages is as follows: + +- 1. Inform start FML task (Continuous type)**: FML training coordinators send this message to FML training participants. +- 2. Start execution**: FML training participants send this message to FML training coordinators. +- 3. Execute and update parameters**: A self-loop message on the FML training participants lifeline. This step is part of "The continuous execution of FML", indicated by a dashed box. +- 4. Inform evaluation**: FML training participants send this message to FML training coordinators. +- 5. Feedback evaluation**: FML training coordinators send this message to FML training participants. +- 6. End execution**: FML training participants send this message to FML training coordinators. + +Concurrent with the execution steps, the following information recording occurs: + +- C1. Record task information**: FML training coordinators send this message to the **FML service agent**. +- A. Review and record task information**: The **FML service agent** sends this message to the **DLT agent**. +- B. Store information**: The **DLT agent** sends this message to the **DLT system**. +- C2. Record execution information**: FML training participants send this message to the **FML service agent**. +- C. Review and record task information**: The **FML service agent** sends this message to the **DLT agent**. +- D. Store information**: The **DLT agent** sends this message to the **DLT system**. + +Labels **C1**, **A**, **B**, **C**, and **D** are placed within dashed boxes to group related actions. The diagram also includes a footer reference "F.748.24(24)". + +Sequence diagram of FML model-training task execution for multiple repetitive computations. + +**Figure 9-5 – Reference procedures of FML model-training task execution (multiple repetitive computations)** + +The relevant steps mainly include the following. + +Step 1. FML training coordinators notify relevant FML training participants to perform the FML model-training task. This step is similar to step 1 in Figure 9-4. + +Step 2. Each FML training participant checks and judges whether to accept the FML model-training task, and informs FML training coordinators whether to accept and execute the FML model-training task. + +Steps 3, 4 and 5. If FML training participants accept the FML model-training task, it shall be continuously executed, and during the execution process, the parameters of the execution intermediate process shall be continuously updated to FML training coordinators; meanwhile, the evaluation task sent by FML training coordinators shall be accepted, processed and fed back. + +An evaluation task is sent by the FML training collaborative to evaluate the intermediate result performed by FML training participants to facilitate their performance of the computing processes assigned to them. + +Step 6. After processing the FML model-training task, FML training participants notify FML training coordinators of the completion and final result of their execution. + +Steps A and B. FML training coordinators store execution information of the FML task via the CES. These steps are similar to steps 3 and 4, and steps 11 and 12 in Figure 9-4. + +Steps C and D. FML training participants store execution information of the FML model-training task via the CES. These steps are similar to steps 8 and 9 in Figure 9-4. + +In the process of repetitive FML model-training task execution, coordinators of and participants in FML training continuously store their own status via the CES, so as to facilitate subsequent contribution evaluation. + +## **9.6 Evaluation of FML model-training task contribution** + +Any party to the FML model-training task may apply for and seek the contribution of the parties (or agreed parties) to the FML model-training task. + +The contribution calculation module refers to the parties involved in the FML model-training task that are used to calculate the contribution of each party in the FML model-training task. Contribution computing modules are usually stored in an executable program (e.g., smart contracts). The contribution calculation modules can be stored in the CES or DLT system. + +The contribution calculation is based on relevant data stored in the DLT system. The DLT system ensures: the security and consistency of the data stored by participants in the FML training task; and the privacy protection of the relevant data via the CES. At the same time, the reliability of the FML model-training task contribution is checked by correlation of the calculated values and those in the actual data set. + +The FML training coordinators and the FML training participants can negotiate the contribution calculation approach (steps 1 and 2), and store the negotiated contribution calculation approach in the DLT system via the CES (steps 2.1 and 2.2). The parties involved in the FML service system may choose the contribution calculation approach determined in advance and stored in the DLT system. See Figure 9-6. + +![Sequence diagram showing the reference procedures of FML model-training task contribution evaluation. The diagram is divided into two main lifelines: 'FML services system' and 'CES'. The 'FML services system' includes 'FML training participants' and 'FML training coordinators'. The 'CES' includes 'FML service agent', 'Contribution calculator', 'Data collection', 'Task management', 'DLT agent', and 'DLT system'. The sequence of messages is: 1. Negotiate contribution calculation model (Participants to Coordinators); 2. Negotiate contribution calculation model (Coordinators to Participants); 2.1. Store contribution calculation model (Coordinators to DLT agent); 2.2. Store contribution calculation model (DLT agent to DLT system); 3. Request calculate contribution degree (Participants to Service agent); 3. Request calculate contribution degree (Coordinators to Service agent); C1. Process request (Service agent to Calculator); C2. Check request (Calculator to Data collection); C3. Check contribution calculation model and data (Data collection to Task management); B1. Read information (Task management to DLT system); C4. Calculate contribution degree (Calculator to DLT system); B1. Read information (DLT system to Task management); C5. Feedback degree (Calculator to Service agent); 4. Feedback contribution information (Service agent to Coordinators); 4. Feedback contribution information (Coordinators to Participants).](b6671cfafda3820aafe9a24fa7a4d8c7_img.jpg) + +Sequence diagram showing the reference procedures of FML model-training task contribution evaluation. The diagram is divided into two main lifelines: 'FML services system' and 'CES'. The 'FML services system' includes 'FML training participants' and 'FML training coordinators'. The 'CES' includes 'FML service agent', 'Contribution calculator', 'Data collection', 'Task management', 'DLT agent', and 'DLT system'. The sequence of messages is: 1. Negotiate contribution calculation model (Participants to Coordinators); 2. Negotiate contribution calculation model (Coordinators to Participants); 2.1. Store contribution calculation model (Coordinators to DLT agent); 2.2. Store contribution calculation model (DLT agent to DLT system); 3. Request calculate contribution degree (Participants to Service agent); 3. Request calculate contribution degree (Coordinators to Service agent); C1. Process request (Service agent to Calculator); C2. Check request (Calculator to Data collection); C3. Check contribution calculation model and data (Data collection to Task management); B1. Read information (Task management to DLT system); C4. Calculate contribution degree (Calculator to DLT system); B1. Read information (DLT system to Task management); C5. Feedback degree (Calculator to Service agent); 4. Feedback contribution information (Service agent to Coordinators); 4. Feedback contribution information (Coordinators to Participants). + +**Figure 9-6 – Reference procedures of FML model-training task contribution evaluation** + +Contribution can be calculated in a variety of ways, e.g., through the effective contribution ratio, including total computation, actual computation, effective computation and data validity. + +Data validity is determined by comparing the correlation between the actual calculated data and the data set for the FML model-training task. When the FML model-training task is executed, the data set on which it depends can be agreed. The relevant judgement approach is involved in the previous description. + +The following is a simple calculation approach for contribution degree evaluation as an example. + +$\text{Contribution} = \text{Effective contribution} / \text{Total effective contribution}$ + +$\text{Total effective contribution} = \text{the sum of the effective contribution of all relevant parties}$ + +$\text{Effective contribution} = \text{volume of calculated data} \times \text{number of calculations} + \text{number of updated and accepted parameters}$ + +When calculating contribution, the CES needs to check the relevant identity information of the FML training participants, the FML training coordinators, the task information of the FML training task and the relevant log information. Relevant logs and other information can usually be stored in the DLT system and stored encrypted using the aforementioned encryption method. + +# 10 General functional components of coordinators of and participants in FML training + +## 10.1 FML training coordinator + +The main FCs of FML training coordinators include device registration, model management, contribution management, data management, task coordination, log update, task management, participants device synchronization and log synchronization. + +- The device registration FC of FML training coordinators is used to deal with the registration operation and related information management of FML training coordinators. The FML + +collaborative device registers itself with CES using the device registration FC, and can search and verify the registration information related to the registered FML training participants. + +- The model management FC of the FML training coordinators is used to manage the models of FML model-training tasks. +- The contribution management FC of the FML training coordinators is used to negotiate the contribution calculation modules of the FML model-training task and request the contribution calculation of the FML model-training task. +- The data management FC of the FML training coordinators is used to manage the index information of the target computing data set of the FML model-training task. Relevant index information, excluding the original data, contains at least the identifier of the data, a description and classification of the data information, and digest information. The relevant index information needs to be sufficient for the FML training coordinators to measure the FML model-training task and to identify a data item. At the same time, during the execution of FML service, the FML model-training task execution module will use the same algorithm to generate and compare summaries to make statistics and evaluate whether the pre-set data set is used. For a FML training task that is computed once, there is no need to count and evaluate whether to use a pre-set data set when executing a FML model-training task. +- The task coordination FC of the FML training coordinators is mainly used to cooperate with the FML training participants according to the selected FML training task model and related parameters. For FML training tasks with repetitive computation, the task coordinators FC provides continuous services and provides coordinated transmission of relevant intermediate result parameters and related information between FML training participants. +- The log update FC of the FML training coordinators is mainly used to generate log information during the execution of the FML model-training task, including data access, information synchronization, etc., according to the execution of the task cooperation FC. +- The TM-FC of FML training coordinators is mainly used to manage FML model-training tasks and related information. +- The log synchronization FC of the FML training coordinators is mainly used to write the process log of the FML service to CES. + +The synchronization of the participants of the FML training coordinators is mainly used to cooperate with each FML training coordinators, including the negotiation of FML training tasks, synchronization of FML model-training results or parameters of the intermediate process. + +![Figure 10-1: Main functional components of FML training coordinators and FML training participants. The diagram shows two main groups: 'FML training coordinators' and 'FML training participants', both connected to 'CES and trustworthy shared storages'. The coordinators group includes Device registration, Model management, Contribution management, Data management, Task coordination, Log update, Task management, Participant synchronization, and Log synchronization. The participants group includes Device registration, Model loading, Task management, Data management, Task execution, Parameter update, Participant synchronization, Contribution management, Log processing, and Log synchronization. Arrows indicate the flow and interaction between these components.](e69b9188aa2c14ec6b21c83f711fef65_img.jpg) + +``` + +graph TD + subgraph Coordinators [FML training coordinators] + DR1[Device registration] + MM[Model management] + CM[Contribution management] + DM1[Data management] + TC[Task coordination] + LU[Log update] + TM1[Task management] + PS1[Participant synchronization] + LS1[Log synchronization] + + DR1 --> DM1 + MM --> TC + CM --> LU + DM1 <--> TC + TC <--> LU + TM1 --> PS1 + PS1 --> LS1 + end + + subgraph Participants [FML training participants] + DR2[Device registration] + ML[Model loading] + TM2[Task management] + DM2[Data management] + TE[Task execution] + PU[Parameter update] + PS2[Participant synchronization] + CM2[Contribution management] + LP[Log processing] + LS2[Log synchronization] + + DR2 --> DM2 + ML --> TE + TM2 --> PS2 + DM2 --> TE + TE <--> PU + PU <--> PS2 + TE --> LP + LP --> LS2 + end + + CES[CES and trustworthy shared storages] + + PS1 --> CES + LS1 --> CES + PS2 --> CES + LS2 --> CES + CES --> CM + CES --> LU + CES --> PS1 + CES --> LS1 + CES --> PS2 + CES --> LS2 + CES --> LP + CES --> LS2 + +``` + +Figure 10-1: Main functional components of FML training coordinators and FML training participants. The diagram shows two main groups: 'FML training coordinators' and 'FML training participants', both connected to 'CES and trustworthy shared storages'. The coordinators group includes Device registration, Model management, Contribution management, Data management, Task coordination, Log update, Task management, Participant synchronization, and Log synchronization. The participants group includes Device registration, Model loading, Task management, Data management, Task execution, Parameter update, Participant synchronization, Contribution management, Log processing, and Log synchronization. Arrows indicate the flow and interaction between these components. + +**Figure 10-1 – Main functional components of FML training coordinators and FML training participants** + +## 10.2 FML training participant + +The main FCs of FML training participant mainly include device registration, model loading, task management, data management, task execution, parameter update, participant device synchronization, contribution management, log processing and log synchronization, etc. + +- The device registration FC of FML training participants is used to register itself and find the registration information of other participants in and coordinators of FML training. +- The model-loading FC is used to obtain the FML calculation model. The FML training model is a set of executable modules, which can come from FML training participants themselves or from FML training coordinators, etc. FML training participants can perform the FML model-training task by executing computational modules. +- The TM-FC of the FML training participants is used to negotiate and obtain the FML model-training task and the associated computing model code. +- The data management FC of FML training participants is used to provide computational data to the task execution FC when performing the FML task. If the data set for the FML model-training task is agreed upon in advance, the data management FC is responsible for checking and providing the data set. +- The task execution FC of FML training participants is used to execute applicable modules of the loaded FML training model, and relies on the data set provided by the data management FC. The final or intermediate result is synchronized with FML training coordinators through the parameter update FC and the coordinators synchronization FC. At the same time, the logs of intermediate process and data usage used in the processing process are sent to the CES through the log-processing module and log synchronization FC. The task execution FC can review the calculated data set to check whether the previously agreed data set is used, and at the same time generate logs of the relevant results to send to the CES. + +- The contribution management FC of FML training participants is used to negotiate the contribution evaluation model with FML training coordinators and to cooperate with CES to calculate the contribution evaluation information of the FML training task. +- The log processing FC and log synchronization FC of FML training participants are used to send the log information to the CES. + +The FCs of the CES can be deployed in coordinators of and participants in FML training and perform the functionalities of the CES. + +# 11 Security consideration + +FML model-training tasks can be based on the agreed raw data sets of the parties. The raw data sets themselves do not need to be shared, just the index information of the raw data sets (such as data identifier and digest). At the same time, parts of the FML model-training tasks are executed according to the same executable program, so as to ensure the certainty of the calculation. The execution module to train ML models of the FML model-training task can be executed in the secure container of the coordinators of and participants in FML training, and the execution modules can check the raw data set during the execution process, so as to further ensure the reliability of the calculation. + +The FCs for task coordination and contribution management of the FML training participants, the FCs for task execution and contribution management of the FML training participants, and the DC-FC and CC-FC of the CES may all run in a secure execution environment. The secure execution environment may be provided by software (e.g., a secure container or secure virtual machine) or a hardware environment (e.g., a secure chip). + +The data set information, task executable FC, and log information of FML model-training tasks can be safely stored in the DLT system, which can not only provide security for data and modules, but also protect the privacy of the data, and make multi-party secure computation more traceable and trusted. + +FML training participants and coordinators may come from different companies, countries and regions, and the CES should provide mutual authorization and authentication mechanisms to ensure communication security. An FML service involves multiple parties and should provide a security mechanism to ensure data security and reliability for data communication and computation. + +# Appendix I + +## Use cases of a contribution evaluation service for FML services + +(This appendix does not form an integral part of this Recommendation.) + +In the construction of smart cities, various government departments and enterprises have various kinds of massive data. At the time of publication, some socially important development projects, such as systems for credit information and transportation planning need to be completed jointly with the data of government and enterprises. Considering the confidentiality of governmental and commercial data, the project can only be completed by using FML services to jointly learn between government and enterprises. + +For example, in order to build a safe public traffic system, public security departments, traffic management departments and automobile enterprises gather data to train image recognition and traffic prediction models by FML services. Public security departments, traffic management departments and automobile enterprises act as FML training participants, and regulatory departments act as FML training coordinators. A CES manages FML training tasks and acts as a bridge between coordinators of and participants in FML training to exchange training data. In addition, a CES stores the exchanged training data in a DLT system. A CES also manages itself and stores data in DLT systems. DLT systems guarantee the reliability, integrity, consistency, availability and immutability of data. As a result, a CES, supported by DLT system(s), provides a trusted contribution degree evaluation service for each participant. See Figure I.1. + +![Diagram illustrating the architecture of a safe public traffic system using a Contribution Evaluation Service (CES) and a Distributed Ledger Technology (DLT) system.](7e1c9b51e067a48cd0fcc9748d8bd8d8_img.jpg) + +The diagram illustrates the architecture of a safe public traffic system. At the top, a box contains 'Data' (represented by a database icon) and a 'Computing unit' (represented by a whale icon), which are connected by a double-headed arrow. This box is labeled 'Public security departments'. Below it, a central 'Coordinator' (represented by three circles) is connected to three 'CES' (Contribution Evaluation Service) boxes via dashed arrows. The 'Public security departments' box is connected to the central 'CES' box by a double-headed arrow. The 'Coordinator' is connected to the central 'CES' box by a double-headed arrow. The central 'CES' box is connected to the 'Traffic management departments' and 'Automobile enterprises' boxes by dashed arrows. The 'Traffic management departments' and 'Automobile enterprises' boxes each contain 'Data' and a 'Computing unit' connected by a double-headed arrow. The 'Traffic management departments' box is connected to the 'Automobile enterprises' box by a double-headed arrow. Both the 'Traffic management departments' and 'Automobile enterprises' boxes are connected to a 'DLT system' (represented by a long horizontal bar) by double-headed arrows. The 'DLT system' is also connected to the 'Public security departments' box by a double-headed arrow. The 'Traffic management departments' and 'Automobile enterprises' boxes are connected to the 'DLT system' by double-headed arrows. The 'Public security departments' box is connected to the 'DLT system' by a double-headed arrow. The 'Public security departments' box is connected to the 'Traffic management departments' and 'Automobile enterprises' boxes by arrows labeled 'Encryption'. + +F.748.24(24) + +Diagram illustrating the architecture of a safe public traffic system using a Contribution Evaluation Service (CES) and a Distributed Ledger Technology (DLT) system. + +Figure I.1 – Using a CES to build a safe public traffic system + +## Bibliography + +- [b-ITU-T X.1400] Recommendation ITU-T X.1400 (2020), *Terms and definitions for distributed ledger technology*. +- [b-ITU-T Y.2091] Recommendation ITU-T Y.2091 (2011), *Terms and definitions for next generation networks*. +- [b-ITU-T Y.3052] Recommendation ITU-T Y.3052 (2017), *Overview of trust provisioning for information and communication technology infrastructures and services*. +- [b-IEEE 3652.1] IEEE 3652.1-2020, *IEEE Guide for architectural framework and application of federated machine learning*. +- [b-ISO/TR 15801] Technical Report ISO/TR 15801:2017, *Document management – Electronically stored information – Recommendations for trustworthiness and reliability*. + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.25-202309-I_PDF-E/16c1175b5f05a4b55e6d396fc51b15b3_img.jpg b/marked/F/T-REC-F.748.25-202309-I_PDF-E/16c1175b5f05a4b55e6d396fc51b15b3_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..15ca8be331dfe05ea3846e738432718dcac6e710 --- /dev/null +++ b/marked/F/T-REC-F.748.25-202309-I_PDF-E/16c1175b5f05a4b55e6d396fc51b15b3_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f82432042f62fd7bc6571a996e8510f014c9f7da4a2238849a5277ee7f95ccad +size 47525 diff --git a/marked/F/T-REC-F.748.25-202309-I_PDF-E/2ba086df3506f81bae3a9b53725dcfea_img.jpg b/marked/F/T-REC-F.748.25-202309-I_PDF-E/2ba086df3506f81bae3a9b53725dcfea_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..72f30988a44cb4ebe95d1ff42528a3ea1ab63cdc --- /dev/null +++ b/marked/F/T-REC-F.748.25-202309-I_PDF-E/2ba086df3506f81bae3a9b53725dcfea_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:42fea834d7dcdbd8607015846c661cfea4c79e8fdbd43a59868d3bc5f43a132d +size 41447 diff --git a/marked/F/T-REC-F.748.25-202309-I_PDF-E/5cab96b2d23174c25919840ecd50aa48_img.jpg b/marked/F/T-REC-F.748.25-202309-I_PDF-E/5cab96b2d23174c25919840ecd50aa48_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..cf60fe3063f7c00ae7923236e7f48109f7277a79 --- /dev/null +++ b/marked/F/T-REC-F.748.25-202309-I_PDF-E/5cab96b2d23174c25919840ecd50aa48_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7ed76955fc741b5f89991b5d3dd26b160d1fba558d6813208699b7d93c384687 +size 53554 diff --git a/marked/F/T-REC-F.748.25-202309-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg b/marked/F/T-REC-F.748.25-202309-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..e02c92ed295c41329fe654547ec6c6b3098a286b --- /dev/null +++ b/marked/F/T-REC-F.748.25-202309-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6872f17b7c25b3399b72338b1891044801cb6d4c7403864338a3b7e16c064916 +size 44047 diff --git a/marked/F/T-REC-F.748.25-202309-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.748.25-202309-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..25f8fbfaef317942ed3102c74b40a7d42f80608b --- /dev/null +++ b/marked/F/T-REC-F.748.25-202309-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:83cab7dcabfe939b223d97242d8523e461968d21f7c5432693c30a64f2d95521 +size 7050 diff --git a/marked/F/T-REC-F.748.25-202309-I_PDF-E/8fbdfc3d17fb1dae7b2d8f5a287fa9fc_img.jpg b/marked/F/T-REC-F.748.25-202309-I_PDF-E/8fbdfc3d17fb1dae7b2d8f5a287fa9fc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..8190b60f11051f8c5d108c930c858aad4697fa57 --- /dev/null +++ b/marked/F/T-REC-F.748.25-202309-I_PDF-E/8fbdfc3d17fb1dae7b2d8f5a287fa9fc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:50cbed63792293a2e0cf8c12a57b8b484ed0f557363c31307ab9c0bdbd9277df +size 49459 diff --git a/marked/F/T-REC-F.748.25-202309-I_PDF-E/9c6461e1e94afae4dec455e69a2ce152_img.jpg b/marked/F/T-REC-F.748.25-202309-I_PDF-E/9c6461e1e94afae4dec455e69a2ce152_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..36f093ff5f56f656bdd75c24aa028ef6b302dfd2 --- /dev/null +++ b/marked/F/T-REC-F.748.25-202309-I_PDF-E/9c6461e1e94afae4dec455e69a2ce152_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4208767b09005fdd91e49647e7fcf5568453f8f7b173542dbc3e1e5715c5556b +size 58311 diff --git a/marked/F/T-REC-F.748.25-202309-I_PDF-E/a24e89a6fe9bb70c83f8bf5202baba95_img.jpg b/marked/F/T-REC-F.748.25-202309-I_PDF-E/a24e89a6fe9bb70c83f8bf5202baba95_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..5585ba0c82ad61d7b940a4bb6c877ac691ff73e4 --- /dev/null +++ b/marked/F/T-REC-F.748.25-202309-I_PDF-E/a24e89a6fe9bb70c83f8bf5202baba95_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1d84f99c2c349562c11bd24ea9db9dfb17d27a571b611017d14c3cd384572016 +size 40237 diff --git a/marked/F/T-REC-F.748.25-202309-I_PDF-E/bd0b93e7a46ede276d0a3b79ac487bd9_img.jpg b/marked/F/T-REC-F.748.25-202309-I_PDF-E/bd0b93e7a46ede276d0a3b79ac487bd9_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a1c0072520253c74eb7a49c71f83d4be2aa8fd65 --- /dev/null +++ b/marked/F/T-REC-F.748.25-202309-I_PDF-E/bd0b93e7a46ede276d0a3b79ac487bd9_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f5f2bc6c256b810095b10a657e843fcd28cc69d3cbcebb76070bc2cc4e76ad7a +size 50199 diff --git a/marked/F/T-REC-F.748.25-202309-I_PDF-E/cfda9df1319e04207eb28bcefd1dab7b_img.jpg b/marked/F/T-REC-F.748.25-202309-I_PDF-E/cfda9df1319e04207eb28bcefd1dab7b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a4fd0ec722d58e88d7c576af1f5f73c900eddd51 --- /dev/null +++ b/marked/F/T-REC-F.748.25-202309-I_PDF-E/cfda9df1319e04207eb28bcefd1dab7b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5bf922a888244d801e08aa8326ee8ab74a87ef6fb2ccb2e85a9b12c18c4542b3 +size 73680 diff --git a/marked/F/T-REC-F.748.26-202402-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.748.26-202402-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c1629394c02f697f60ae954ad36bf274fa8082cb --- /dev/null +++ b/marked/F/T-REC-F.748.26-202402-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3763acd7a0ed56aa30f3f90468c1fb0cb54573eb9045bca3eddf8976b4f0f41d +size 7219 diff --git a/marked/F/T-REC-F.748.26-202402-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.748.26-202402-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..26b343ad25126457cdcdf66b1074a609b8edf766 --- /dev/null +++ b/marked/F/T-REC-F.748.26-202402-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:70953a4be3577ee231719897b4d91d8f42b8730c7dca66141dffd9ea873071fd +size 89304 diff --git a/marked/F/T-REC-F.748.26-202402-I_PDF-E/raw.md b/marked/F/T-REC-F.748.26-202402-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..c9b0734bed0eefedae7810aca71a1247c6af1c95 --- /dev/null +++ b/marked/F/T-REC-F.748.26-202402-I_PDF-E/raw.md @@ -0,0 +1,537 @@ + + +# Recommendation **ITU-T F.748.26 (02/2024)** + +SERIES F: Non-telephone telecommunication services +Multimedia services + +--- + +## **Technical specification for artificial intelligence cloud platforms: Performance evaluation** + +![ITU logo](390120de4fe440c42fea8154fcaad334_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with white grid lines and the letters 'ITU' in a bold, blue, sans-serif font. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.26 + +# Technical specification for artificial intelligence cloud platforms: Performance evaluation + +## Summary + +Recommendation ITU-T F.748.26 provides a comprehensive performance evaluation framework for artificial intelligence (AI) cloud platforms. Recommendation ITU-T F.748.26 gives an overview of the evaluation framework, configuration specification, workloads, metrics, requirements on evaluation results and evaluation suggestions. Recommendation ITU-T F.748.26 can be a unified guideline for developers, users, third party test agencies and researchers to analyse and access the performance of AI cloud platforms. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.26 | 2024-02-13 | 16 | 11.1002/1000/15847 | + +## Keywords + +Artificial intelligence, benchmark, metrics, performance, workload. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2024 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-----------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Overview of AI cloud platform performance evaluation framework..... | 2 | +| 6.1 Evaluation object ..... | 2 | +| 6.2 Evaluation principle..... | 2 | +| 6.3 Workflow of performance evaluation framework..... | 2 | +| 7 Configuration specification ..... | 3 | +| 7.1 Computing resource cluster configuration ..... | 3 | +| 7.2 Node configuration..... | 3 | +| 7.3 Software configuration ..... | 4 | +| 7.4 Physical environment configuration..... | 4 | +| 8 Evaluation workloads and metrics ..... | 4 | +| 8.1 Operation level ..... | 4 | +| 8.2 Model level..... | 6 | +| 8.3 Platform level ..... | 7 | +| 9 Requirements on evaluation results ..... | 8 | +| 9.1 Benchmark report ..... | 8 | +| 9.2 Benchmark materials ..... | 8 | +| Appendix I – Evaluation suggestions..... | 9 | +| I.1 Evaluation program ..... | 9 | +| I.2 Dataset preparation ..... | 9 | +| I.3 Evaluation workloads ..... | 9 | +| Bibliography..... | 10 | + + + +# Recommendation ITU-T F.748.26 + +## Technical specification for artificial intelligence cloud platforms: Performance evaluation + +# 1 Scope + +This Recommendation provides a comprehensive performance evaluation framework for artificial intelligence (AI) cloud platforms. + +In particular, this Recommendation includes: + +- an overview of the performance evaluation framework; +- a configuration specification for an AI cloud platform for performance evaluation; +- workloads and metrics for AI cloud platform performance evaluation; and +- requirements on evaluation results. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T F.748.11] Recommendation ITU-T F.748.11 (2020), *Metrics and evaluation methods for a deep neural network processor benchmark*. + +[ITU-T F.748.17] Recommendation ITU-T F.748.17 (2022), *Technical specification for artificial intelligence cloud platform – Artificial intelligence model development*. + +[ITU-T F.748.18] Recommendation ITU-T F.748.18 (2022), *Metric and evaluation methods for AI-enabled multimedia application computing power benchmark*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 artificial intelligence** [b-ISO/IEC 2382]: Branch of computer science devoted to developing data processing systems that perform functions normally associated with human intelligence, such as reasoning, learning, and self-improvement. + +**3.1.2 benchmark** [ITU-T F.748.11]: Benchmark is an evaluation method with a long-term application in the entire computer field. Example: As computer architecture advanced, it became more difficult to compare the performance of various computer systems simply by looking at their specifications. Therefore, tests were developed that allowed comparison of different architectures (i.e., providing benchmarks). + +**3.1.3 model** [b-ITU-T F.748.16]: An output created by an algorithm running on training data that can generate an inference or prediction, based on input data. + +## 3.2 Terms defined in this Recommendation + +None. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|------------------------------------| +| AI | Artificial Intelligence | +| CPU | Central Processing Unit | +| FP | Floating Point | +| GPU | Graphics Processing Unit | +| IO | Input/Output | +| IOPS | Input/output Operations Per Second | + +# 5 Conventions + +The following conventions are used in this Recommendation: + +- The phrase "**is required**" indicates a requirement that must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The phrase "**is recommended**" indicates a requirement that is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. + +# 6 Overview of AI cloud platform performance evaluation framework + +## 6.1 Evaluation object + +The evaluation object is the AI cloud platform. It manages the full lifecycle of AI model development and deployment. It usually consists of hardware resources and software stacks. It is recommended that heterogeneous AI computing power, mainstream open-source deep learning framework and multiple development environment for developers to build AI models efficiently be supported. For requirements for an AI cloud platform, see [ITU-T F.748.17]. + +## 6.2 Evaluation principle + +- Reproducibility: The environment configuration, workloads, evaluation metrics and the implementation of the benchmark is required to be clearly defined and avoid ambiguity. Hence, the developers of the AI cloud platform or the testing agencies can reproduce the evaluation on demand. +- Scalability: The evaluation is recommended to be able to test the ability of the platform to handle heavy workloads and a large number of users. +- Practicality: The evaluation is recommended to be applicable to mainstream AI cloud platforms. Moreover, it is recommended to provide results practical for the detection of performance bottlenecks and optimization of computing efficiency, resource usage and power consumption. + +## 6.3 Workflow of performance evaluation framework + +The workflow of the performance evaluation framework for an AI cloud platform consists of five parts as shown in Figure 6-1. The evaluation object is the AI cloud platform with specified configuration. The input workloads are classified by level as operation, mode and platform. Each workload is associated with specific metrics. Moreover, the output results include an evaluation report and evaluation material. + +![Figure 6-1: A recommended workflow for a performance evaluation framework. The diagram shows the flow from Input workload to Evaluation object (AI cloud platform) to Output results, influenced by Metrics and Configuration.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +The diagram illustrates a performance evaluation workflow. On the left, the 'Input workload' is categorized into three levels: + + +- Operation-level:** Includes 'Computing operations', 'Network operations', and 'IO operations'. +- Operation-level:** Includes 'Model training', 'Algorithm development', 'Model inference', and 'Model deployment'. +- Platform-level:** Includes 'Multi-task execution'. + + Arrows from these workload categories point to a central box labeled 'Evaluation object' containing 'AI cloud platform'. Above this box is a 'Metrics' table, and below it is a 'Configuration' table. + + The 'Metrics' table contains: + + +| | | +|-------------------|------------| +| Executing time | Latency | +| Speedup ratio | Throughput | +| Convergence time | Bandwidth | +| Energy efficiency | ... | + + The 'Configuration' table contains: + + +| | | | +|----------------------|--------|----------| +| Cluster | Device | Software | +| Physical environment | | | + + Arrows from both the 'Metrics' and 'Configuration' tables point to the 'Evaluation object' box. An arrow from the 'Evaluation object' box points to the 'Output results' on the right, which include 'Evaluation report' and 'Evaluation material'. + + A small label 'F.748.26(24)' is located in the bottom right corner of the diagram area. + +Figure 6-1: A recommended workflow for a performance evaluation framework. The diagram shows the flow from Input workload to Evaluation object (AI cloud platform) to Output results, influenced by Metrics and Configuration. + +**Figure 6-1 – A recommended workflow for a performance evaluation framework** + +# 7 Configuration specification + +The performance of an AI cloud platform is affected by a large number of factors, such as hardware, libraries, schedulers, deep learning frameworks, etc. For performance evaluation, the configuration about AI cloud platform is recommended to be clearly specified. + +## 7.1 Computing resource cluster configuration + +An AI cloud platform is deployed on a computing resource cluster. The architecture and composition of a computing resource cluster are crucial to the performance of the platform. Therefore, the following information about a computing resource cluster is recommended to be specified. + +- Nodes: A computing resource cluster comprises various types of node, which include those for computing, storage and management. The number of each type of node is required to be specified. +- Topology: Topology determines how nodes are connected to each other and how data is transmitted in a computing resource cluster. The topology of the computing resource cluster is recommended to be specified. +- Network: The network refers to the actual communication infrastructure between the nodes. The information such as the network protocol and the bandwidth are recommended to be specified. + +## 7.2 Node configuration + +Each node in a computing resource cluster comprises heterogeneous devices with different computational capabilities. The following information is recommended to be specified. + +- Computing devices: A node may comprise heterogeneous computing devices, which include but are not limited to a central processing unit (CPU), graphics processing unit (GPU), neural network processing unit, field programmable gate array and application specific integrated circuit. The number of computing devices (all types) is recommended to be specified. +- Computing power: The parameters that contribute to the computing power of each device are recommended to be specified. For example, a CPU device is specified by the following indicators: the micro-architecture; the number of cores; the number of threads; CPU clock speed; cache; word size; manufacturing process of CPU; etc. A GPU may include additional indicators such as GPU memory capacity and the number or type of streaming processors. + +- Memory: The memory contributes to the amount of data the node can process at once. The type, the bandwidth, speed and the capacity of memory are recommended to be specified. +- Disk: The disk contributes to the storage capacity of the node. The disk type, bandwidth, speed and storage capacity are recommended to be specified. +- System bus: The system bus determines the speed of inter-device transmission of data, address and control instructions. The generation, speed and bandwidth are recommended to be specified. +- The network adapter: The network adapter determines the speed of intra-device data transmission. The bandwidth, interface and network protocol are recommended to be specified. + +## 7.3 Software configuration + +Each node in a computing resource cluster has its own software stack. The following information is recommended to be specified. + +- Operating system: The operating system is software that manages the hardware and software resources of the node. The type, version and release number of the operating system are recommended to be specified. +- Hardware driver: The hardware driver is software that allows the operating system to access hardware devices. The version of the hardware driver is recommended to be specified. +- Runtime environment: The runtime environment is software that provides a running environment for tasks. Common runtime environments used in computing resource clusters include but are not limited to the following: bare metal server; virtual machine; and container. The type of runtime environment is recommended to be specified. + +## 7.4 Physical environment configuration + +A computing resource cluster is designed to operate within a specific physical environment to ensure its optimal performance and longevity. The following information is recommended to be specified. + +- Temperature: The computing resource cluster is required to run within a certain range of temperature to prevent overheating and damage to hardware devices. The temperature is recommended to be specified. +- Humidity: The computing resource cluster is required to run within a certain range of humidity to prevent corrosion and damage to electronic components. The humidity is recommended to be specified. +- Power supply: The computing resource cluster is required to be supported by a certain level of electrical power. The voltage and the power of the power adapter are recommended to be specified. + +# 8 Evaluation workloads and metrics + +The performance of an AI cloud platform can be indicated by the execution time and the resource consumption of a number of computing tasks. The tasks are classified into three levels: operation; model; and platform. + +## 8.1 Operation level + +In the context of an AI cloud platform, the notion of operations refers to atomic building blocks of complex AI tasks. They are mathematical calculations or data transformations that are used to process input data and generate output data. The performance of an AI computing task is influenced by the performance of each operation that comprises the entire task. The basic operations are classified into those for: computing; network; and input/output (IO). + +### 8.1.1 Computing operations + +#### 8.1.1.1 Workloads for computing operations + +The recommended computing operations used for evaluation include but are not limited to general matrix multiply, element-wise matrix calculation, convolution, pooling, rectified linear units, batch normalization, residual addition and self-attention. + +#### 8.1.1.2 Metrics for computing operations + +Execution time ( $T_{co}$ ): the time it takes for an operation to be executed from the start time point $T_{co}^s$ to the end time point $T_{co}^e$ . + +$$T_{co} = T_{co}^e - T_{co}^s \quad (8-1)$$ + +Throughput ( $TP_{co}$ ): the number of operations $N_{co}$ that can be executed per time $T_u$ . + +$$TP_{co} = \frac{N_{co}}{T_u} \quad (8-2)$$ + +Energy efficiency ( $EE_{co}$ ): The ratio of computing power to electric power consumption. The computing power ( $CP_{co}$ ) is measured in floating point of operations per second or operations per second. The electric power ( $EP_{co}$ ) consumption can be measured with the power meter during the execution time [ITU-T F.748.18]. + +$$EE_{co} = \frac{CP_{co}}{EP_{co}} \quad (8-3)$$ + +### 8.1.2 Network operations + +#### 8.1.2.1 Workloads for network operations + +Network operations are used to evaluate communication efficiency. The typical workloads include but are not limited to scatter, all-reduce and all-to-all. + +#### 8.1.2.2 Metrics for network operations + +Execution time ( $T_{no}$ ): The time it takes for an operation to be executed from the start time point $T_{no}^s$ to the end time point $T_{no}^e$ . + +$$T_{no} = T_{no}^e - T_{no}^s \quad (8-4)$$ + +Bandwidth ( $B$ ): The transmission speed of data and model parameters between devices, measured in tera-, giga-, mega or kilobytes per second. For each data transmission between two nodes, record the volume of transmitted data ( $D_{no}$ ) and the time duration $T_{no}$ of transmission. + +$$B = \frac{D_{no}}{T_{no}} \quad (8-5)$$ + +### 8.1.3 IO operations + +#### 8.1.3.1 Workloads for IO operations + +The IO operations used for evaluation include but are not limited to parallel IO and data replication. + +#### 8.1.3.2 Metrics for IO operations + +IOPS: the number of input/output operations per second that can be processed. + +Response time ( $T_{io}$ ): The time it takes for the disk to respond to an IO request. Record the start time point $T_{io}^s$ of receiving an IO request, and the end time point $T_{io}^e$ of the disk completing the response. + +$$T_{io} = T_{io}^e - T_{io}^s \quad (8-6)$$ + +## 8.2 Model level + +Based on the lifecycle of AI development, the model level tasks for performance evaluation include: model training; algorithm development; model inference; and model deployment. + +### 8.2.1 Workloads for model level benchmark + +The model level tasks are recommended to be representative AI applications, such as image classification, machine translation or recommendation. The recommended workloads include but are not limited to the following: residual network; single shot detector; bidirectional encoder representations from transformers; deep learning recommendation model; mask region-based convolutional neural network; and transformer. + +For recommended workloads and test information, see Table 10-1 of [ITU-T F.748.11]. Further, the following workload-related information is recommended to be specified. + +- Model definition: The model structure, model optimizer, loss function, initializer, etc. +- Dataset: The dataset, data enhancement approaches and data arrival modes for inference and deployment. +- Computing precision: The computing precision, which can be INT8 (where INT is integer), FP16, FP32, FP64 (where FP is floating point), mixed, etc. +- Hyperparameters: The batch size, learning rate, momentum, etc. +- Accuracy: The target accuracy for each workload. +- Development tools: The AI framework, third party libraries, integrated development environment, etc. +- Computing resource: The type and number of computing devices. + +### 8.2.2 Metrics for model training task + +Training time ( $T_{\text{mt}}$ ): From the start time point $T_{\text{mt}}^{\text{s}}$ of loading and training data until the time point $T_{\text{mt}}^{\text{e}}$ that the training process ends. + +$$T_{\text{mt}} = T_{\text{mt}}^{\text{e}} - T_{\text{mt}}^{\text{s}} \quad (8-7)$$ + +Convergence time $T_{\text{ct}}$ : From the time point of starting the first iteration $T_{\text{ct}}^{\text{s}}$ until the time point $T_{\text{ct}}^{\text{e}}$ of reaching the target accuracy. + +$$T_{\text{ct}} = T_{\text{ct}}^{\text{e}} - T_{\text{ct}}^{\text{s}} \quad (8-8)$$ + +Speedup ratio for distributed training $SR_p$ : The ratio between the computing time $T_{\text{mt}-1}$ needed by one processor and the computing time $T_{\text{mt}-p}$ needed by $p$ processors. + +$$SR_p = \frac{T_{\text{mt}-1}}{T_{\text{mt}-p}} \quad (8-9)$$ + +Throughput: The amount of data that is processed per time, for the calculation method, see Equation (8-2). + +Energy efficiency: For the calculation method, see Equation (8-3). + +### 8.2.3 Metrics for algorithm development task + +Development environment launch time $T_{\text{ad}}$ : From the time point $T_{\text{ad}}^{\text{s}}$ of calling for launch a development environment until the time point $T_{\text{ad}}^{\text{e}}$ where the development environment gets ready. + +$$T_{\text{ad}} = T_{\text{ad}}^{\text{e}} - T_{\text{ad}}^{\text{s}} \quad (8-10)$$ + +Throughput: The amount of data that is processed per time, for the calculation method, see Equation (8-2). + +Energy efficiency: For the calculation method, see Equation (8-3). + +### 8.2.4 Metrics for model inference task + +Inference latency ( $T_{\text{mi}}$ ): from the time point $T_{\text{mi}}^{\text{s}}$ of start sending the first sample until the time point $T_{\text{mi}}^{\text{e}}$ of receiving the inference result of the last sample. + +$$T_{\text{mi}} = T_{\text{mi}}^{\text{e}} - T_{\text{mi}}^{\text{s}} \quad (8-11)$$ + +Throughput: The amount of data that is processed per time, for the calculation method, see Equation (8-2). + +Energy efficiency: For the calculation method, see Equation (8-3). + +### 8.2.5 Metrics for model deployment task + +Response latency ( $T_{\text{md}}$ ): The response time duration consumed by the complete inference request for a sample. Record the start time point $T_{\text{md}}^{\text{s}}$ of inference request for a sample, and the time point $T_{\text{md}}^{\text{e}}$ of receiving the inference result of the sample. + +$$T_{\text{md}} = T_{\text{md}}^{\text{e}} - T_{\text{md}}^{\text{s}} \quad (8-12)$$ + +Speedup ratio for batched data ( $SR^{\text{b}}$ ): The ratio between the computing time $\sum_{i=1}^p T_{\text{md}}^i$ needed by processing $p$ requests in order and the computing time $T_{\text{bmd}}^p$ needed by processing a batch of $p$ requests. + +$$SR^{\text{b}} = \frac{\sum_{i=1}^p T_{\text{md}}^i}{T_{\text{bmd}}^p} \quad (8-13)$$ + +Request throughput: The amount of inference requests processed per time, for the calculation method, see Equation (8-2). + +Energy efficiency: For the calculation method, see Equation (8-3). + +## 8.3 Platform level + +AI cloud platforms can support multiple tasks simultaneously; these tasks can be homogeneous or heterogeneous. The efficiency of processing multi-tasks is an important indicator of the AI cloud platform performance. The workloads for platform level tasks are a combination of model level tasks as specified in clause 8.2.1. + +### 8.3.1 Platform level task information + +In addition to model level task information, the following information is recommended to be specified. + +- Type of tasks: The type of tasks that are being processed simultaneously by the platform. +- Number of tasks ( $n$ ): The number of each type task that are being processed simultaneously by the platform. +- Order of task requesting: The order in which tasks are requested from the platform. + +### 8.3.2 Metrics for platform level performance + +It is recommended to evaluate the platform level performance according to the following metrics: + +In this clause, it is assumed that a queue of $n$ tasks is requested by users. The request time, the start execution time and the end point of execution time of each task are denoted by $T_{\text{r}}^i$ , $T_{\text{s}}^i$ and $T_{\text{e}}^i$ ( $i = 1, 2, \dots, n$ ), respectively, where $i$ is the task number. + +Total execution time ( $T_{\text{tt}}$ ): The time duration to complete all tasks. + +$$T_{\text{tt}} = \max(T_{\text{e}}^i) - \min(T_{\text{r}}^i) \quad (8-14)$$ + +Maximum waiting time ( $T_{\text{max-wt}}$ ): The maximum waiting time for tasks to be executed. + +$$T_{\max\text{-wt}} = \max(T_s^i - T_r^i) \quad (8-15)$$ + +Average waiting time ( $T_{\text{avg-wt}}$ ): The average waiting time of all tasks. + +$$T_{\text{avg-wt}} = \frac{\sum_{i=1}^n (T_s^i - T_r^i)}{n} \quad (8-16)$$ + +Energy efficiency: For the calculation method, see Equation (8-3). + +# 9 Requirements on evaluation results + +The evaluation results are recommended for formulation into a benchmark report for the test agency and for maintenance of necessary benchmark materials. + +## 9.1 Benchmark report + +It is recommended that all test information and results be arranged in a benchmark report. A benchmark report comprises the following components. + +- Metadata information: It is recommended that metadata information of the test be recorded, including the test agency, the test time and the test object. +- Configuration information: It is recommended that the test information specified in clause 7 be included. +- Workloads descriptions: It is recommended that the descriptions of workloads specified in clause 8 be included. +- Metrics: It is recommended that the metrics specified in clause 8 be included. + +## 9.2 Benchmark materials + +It is recommended that all benchmark materials be submitted to the test agency. The benchmark materials include but are not limited to the following. + +- Source code: It is recommended that all source code implemented for the benchmark be included. +- Logs: It is recommended that the test logs, which include but are not limited to the timestamp, the number of runs, the number of epochs, the number of queries or samples processed, the accuracy, the throughput or other necessary information, be included. +- Model file: It is recommended that the model file that records the weights and structure information be included. +- Metrics calculation: It is recommended that the scripts that calculate and record the metrics be included. + +# Appendix I + +## Evaluation suggestions + +(This appendix does not form an integral part of this Recommendation.) + +The performance of the AI cloud platform is recommended to be evaluated by implementing a certain benchmark. The choice, implementation and execution of the benchmark follow the principles of the performance evaluation specified in clause 6. In this appendix, some suggestions for benchmark implementation are provided. + +## I.1 Evaluation program + +For an evaluation program, recommendations follow: + +- implementation of interfaces to collect performance information; +- implementation of functions to compute performance metrics; +- implementation of functions to record testing logs; +- maintaining the immutability of the test program during the test. + +## I.2 Dataset preparation + +For data preparation, recommendations follow: + +- application of necessary data pre-processing procedures; +- utilization of a high-performance distributed file system for data storage; +- maintenance of the integrity and orderliness of the dataset and the immutability of the dataset samples; +- avoidance of unnecessary observation of training or validation dataset by extracting features from the samples. + +### I.3 Evaluation workloads + +For evaluation workloads, recommendations follow: + +- avoidance of the application of additional optimization approaches, such as weight deletion, model pruning, knowledge distillation and re-training in model level task; +- maintenance of the model architecture and hyperparameters unchanged during the test; +- Avoidance of the execution of any process that differs from the test program during the test. + +# Bibliography + +- [b-ITU-T F.748.16] Recommendation ITU-T F.748.16 (2022), *Requirements for applications and services in smart manufacturing based on machine vision.* +- [b-ISO/IEC 2382] International Standard ISO/IEC 2382:2015, *Information technology – Vocabulary.* + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.27-202401-I_PDF-E/21ad58fee90f2be50708ff541d225507_img.jpg b/marked/F/T-REC-F.748.27-202401-I_PDF-E/21ad58fee90f2be50708ff541d225507_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..cbd5708a9c69de3f333c39ffc781c47481680de6 --- /dev/null +++ b/marked/F/T-REC-F.748.27-202401-I_PDF-E/21ad58fee90f2be50708ff541d225507_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c648928ffc9c43cac4468e17d757d9afdde8157166d8c843a94296f68b060795 +size 24926 diff --git a/marked/F/T-REC-F.748.27-202401-I_PDF-E/29f586959675cafdf81cf934954908eb_img.jpg b/marked/F/T-REC-F.748.27-202401-I_PDF-E/29f586959675cafdf81cf934954908eb_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c7eb31579b5196c733eef5895d68f0820e91e236 --- /dev/null +++ b/marked/F/T-REC-F.748.27-202401-I_PDF-E/29f586959675cafdf81cf934954908eb_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8ec703c0b1e04dbe22012df1de1bb85ade06cfd4a6bfd283de9b7008f2d9cce1 +size 24366 diff --git a/marked/F/T-REC-F.748.27-202401-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.748.27-202401-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c1629394c02f697f60ae954ad36bf274fa8082cb --- /dev/null +++ b/marked/F/T-REC-F.748.27-202401-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3763acd7a0ed56aa30f3f90468c1fb0cb54573eb9045bca3eddf8976b4f0f41d +size 7219 diff --git a/marked/F/T-REC-F.748.27-202401-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg b/marked/F/T-REC-F.748.27-202401-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..b2c7f3364dbd06597baf7b2b90144d20f396352d --- /dev/null +++ b/marked/F/T-REC-F.748.27-202401-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:108512c8650987abd3e919a66fdd19f16b6e9041d6dcc96b11d6e3a149b4a843 +size 89287 diff --git a/marked/F/T-REC-F.748.27-202401-I_PDF-E/dfe556fea00682b09a59427aaf72051c_img.jpg b/marked/F/T-REC-F.748.27-202401-I_PDF-E/dfe556fea00682b09a59427aaf72051c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d3a8269cefa9b7c77f27fffca99c76c77c4b9e01 --- /dev/null +++ b/marked/F/T-REC-F.748.27-202401-I_PDF-E/dfe556fea00682b09a59427aaf72051c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:352b4ef0f6670b87dc31dd298f96d1d0928dc53e34960e62fcdca05b8a25c313 +size 44123 diff --git a/marked/F/T-REC-F.748.28-202406-I_PDF-E/raw.md b/marked/F/T-REC-F.748.28-202406-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..db54c347b602f4f14d81521015b57ead0e7c3a06 --- /dev/null +++ b/marked/F/T-REC-F.748.28-202406-I_PDF-E/raw.md @@ -0,0 +1,552 @@ + + +# Recommendation + +# **ITU-T F.748.28 (06/2024)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +## --- **Requirements and functional architecture of digital twin platforms for supporting multimedia services** + +![ITU logo](84a1d09fb489061482111515543b60dc_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.28 + +# Requirements and functional architecture of digital twin platforms for supporting multimedia services + +## Summary + +Multimedia services are becoming increasingly useful for education, video-based marketing, digital museum, e-health, etc., providing an enriched user experience in e-meetings, distance training and product demonstrations. + +Digital twin realizes the interconnection, intercommunication and interoperability between the physical world and the digital world, constructs the description, diagnosis, prediction and decision-making of the physical world in the virtual world. + +At present, the digital twin system in different multimedia services has common requirements and functions, and it is necessary to develop a digital twin platform to realize the services for upper multimedia applications. The digital twin platform provides general services, twin services, interactive services for multimedia applications such as the fields of education, video-based marketing, digital museum, E-health, etc. + +Recommendation ITU-T F.748.28 specifies the requirements and functional architecture of the digital twin platform for supporting multimedia services. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.28 | 2024-06-13 | 16 | 11.1002/1000/15910 | + +## Keywords + +Digital twin, digital twin platform, functional architecture, multimedia services, requirements. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2024 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|---------------------------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 1 | +| 5 Conventions ..... | 2 | +| 6 Overview of the digital twin platform for supporting multimedia services ..... | 2 | +| 7 Requirements of digital twin platform..... | 3 | +| 7.1 Hardware requirements ..... | 3 | +| 7.2 Network requirements ..... | 3 | +| 7.3 Storage requirements ..... | 4 | +| 7.4 Scalability requirements ..... | 4 | +| 7.5 Precision requirements ..... | 5 | +| 7.6 Compatibility requirements ..... | 5 | +| 7.7 Security requirements ..... | 5 | +| 8 Functional architecture of the digital twin platform ..... | 6 | +| 8.1 General functions..... | 6 | +| 8.2 Twin services..... | 7 | +| 8.3 Interactive functions ..... | 10 | +| Appendix I – Use cases of the digital twin platform for supporting multimedia services..... | 12 | +| I.1 Use case of digital twin for education ..... | 12 | +| I.2 Use case of digital twin for video-based marketing ..... | 12 | +| I.3 Use case of digital twin for digital museum..... | 12 | +| I.4 Use case of digital twin for e-health..... | 13 | +| Bibliography..... | 14 | + + + +# Recommendation ITU-T F.748.28 + +## Requirements and functional architecture of digital twin platforms for supporting multimedia services + +## 1 Scope + +This Recommendation specifies the requirements and framework of digital twin platform for supporting multimedia services. + +The scope of this Recommendation includes: + +- Overview of digital twin platform for supporting multimedia services; +- Requirements for digital twin platform; +- Functional architecture of digital twin platform. + +NOTE – Use cases of the digital twin platform for supporting multimedia services are provided in Appendix I. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T F.700] Recommendation ITU-T F.700 (2000), *Framework Recommendation for multimedia services*. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following term defined elsewhere: + +**3.1.1 digital twin** [b-ITU-T Y.4600]: A digital representation of an object of interest. + +NOTE – A digital twin may require different capabilities (e.g., synchronization, real-time support) according to the specific domain of application. + +### 3.2 Terms defined in this Recommendation + +None. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|-----------------------------------| +| 2D | 2-Dimension | +| 3D | 3-Dimension | +| AI | Artificial Intelligence | +| API | Application Programming Interface | +| CPUs | Central Processing Units | + +| | | +|---------|-----------------------------------------------------------------------| +| CSV | Comma-Separated Values | +| DTP | Digital Twin Platform | +| GeoJSON | Geographic JavaScript Object Notation | +| GIS | Geographic Information System | +| GPUs | Graphics Processing Units | +| HTML | Hyper Text Markup Language | +| HTTP | Hyper Text Transfer Protocol | +| ICT | Information and Communications Technology | +| IoT | Internet of Things | +| JSON | JavaScript Object Notation | +| MQTT | Message Queuing Telemetry Transport | +| OBJ | Object File Format | +| OPC-UA | Object Linking and Embedding for Process Control Unified Architecture | +| PaaS | Platform as a Service | +| STEP | Standard for the Exchange of Product Model Data | +| STL | Stereolithography | +| WebRTC | Web Real-Time Communications | +| XML | extensible Markup Language | + +## 5 Conventions + +In this Recommendation: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. + +## 6 Overview of the digital twin platform for supporting multimedia services + +A digital twin is regarded as a virtual representation that serves as the real-time digital counterpart of a physical object or process. Digital twins in different multimedia applications have common requirements in data processing, resource management, operation and maintenance, and need a common digital twin platform to provide unified interfaces and services. + +Based on the reference model of [ITU-T F.700], the structure of the digital twin system is shown in Figure 1. The digital twin system includes a perception layer, an infrastructure layer, a digital twin platform layer and an application layer, which correspond to media component level, communication task level, service level and application level of multimedia services reference model respectively. The perception layer realizes real-time and dynamic data collection. The infrastructure layer includes network, computing and storage, providing operational support for upper-layer platforms and applications. The digital twin platform is in the third layer. It is the carrier of data and applications, and the basic support platform for industry digital twins. It provides rich information services and development interfaces for related applications to support the construction and operation of applications in various industries. The application layer provides interfaces for digital twin applications in multimedia services. + +![Figure 1 – Overview of the digital twin system. The diagram shows four layers: Application layer (Education, Video-based marketing, Digital museum, E-health), Digital twin platform layer (General services, Twin services, Interactive services), Infrastructure layer (Network, Compute, Storage), and Perception layer. Arrows indicate bidirectional flow between adjacent layers.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +``` + +graph TD + subgraph Application_layer [Application layer] + A1[Education] + A2[Video-based marketing] + A3[Digital museum] + A4[E-health] + end + subgraph Platform_layer [Digital twin platform layer] + P1[General services] + P2[Twin services] + P3[Interactive services] + end + subgraph Infrastructure_layer [Infrastructure layer] + I1[Network] + I2[Compute] + I3[Storage] + end + subgraph Perception_layer [Perception layer] + end + Application_layer <--> Platform_layer + Platform_layer <--> Infrastructure_layer + Infrastructure_layer <--> Perception_layer + +``` + +Figure 1 – Overview of the digital twin system. The diagram shows four layers: Application layer (Education, Video-based marketing, Digital museum, E-health), Digital twin platform layer (General services, Twin services, Interactive services), Infrastructure layer (Network, Compute, Storage), and Perception layer. Arrows indicate bidirectional flow between adjacent layers. + +F.748.28(24) + +**Figure 1 – Overview of the digital twin system** + +The digital twin platform collects and analyses multimedia data. It is driven by artificial intelligence (AI) and integrates information and communication technologies (ICTs) to provide digital twin functions for various interactive multimedia services. + +The platform raises the needs for running environments that can facilitate multimedia services without major modifications in the platform as a service (PaaS) domain + +## 7 Requirements of digital twin platform + +### 7.1 Hardware requirements + +The hardware is required to meet the requirements of the digital twin system for computing performance, rendering performance and storage performance. + +- **Computing power:** It is required to support special hardware such as high-performance central processing units (CPUs), graphics processing units (GPUs), to provide higher computing performance and parallel processing capabilities, meeting the operational requirements of processing large amounts of data and large-scale models and algorithms in digital twin systems. +- **Rendering ability:** It is required to support high-performance graphics cards to provide computing power and large capacity graphics memory, accelerate image processing and visualization operations of digital twins. +- **Storage capacity:** It is required to support high-performance storage devices, such as high-speed disk arrays and solid-state drive, to store data and related calculation results. + +### 7.2 Network requirements + +A digital twin platform needs to address network-related issues such as bandwidth, latency, reliability and security for multimedia services as follows: + +- **Bandwidth:** The bandwidth requirement for a digital twin platform for multimedia services will depend on the type of content being streamed and the number of users. For example, a platform that streams high-definition video will require more bandwidth than a platform that streams audio only. + +- **Latency:** A digital twin platform for multimedia services will require to have low latency in order to provide a smooth and responsive user experience. +- **Reliability:** A digital twin platform for multimedia services is recommended to be reliable in order to ensure that users can access the content they need when they need it. The platform is recommended to be able to withstand spikes in traffic and recover from outages quickly. +- **Security:** A digital twin platform for multimedia services is recommended to be secure in order to protect user data. The platform is recommended to use strong encryption and have robust security measures in place to prevent unauthorized access. This includes encryption, secure transmission protocols and other network security mechanisms to protect the digital twin system from potential network attacks and data leakage risks. +- **Error detection and correction mechanism:** It is required to support error detection and correction mechanism and have redundancy and failover capability. +- **IoT protocols:** It is required to support appropriate security mechanisms to prevent unauthorized access or interception, including encryption, secure transmission protocols, and other network security mechanisms, to protect digital twin systems from potential network attacks and digital leakage risks. + +### 7.3 Storage requirements + +The digital twin platform is required to have corresponding storage functions, supporting the storage of large amounts of data. Its storage requirements include data storage and 3-dimension (3D) model storage. + +- **Storage capacity:** It is required to support sufficient storage capacity to store a large amount of data, such as databases, data lake, etc. +- **Multiple types:** It is required to support the storage of multiple types of digital twin system data, including image and video data, text data, model data, metadata, etc. +- **Backup and recovery:** It is required to support data backup and recovery capabilities, have redundant backup and recovery strategies for data storage, operational procedures for data backup and recovery management and prevent the impact of data loss on the digital twin system. +- **File formats:** It is required to support storing in digital twin file formats such as extensible markup language (XML), standard for the exchange of product model data (STEP), stereolithography (STL), object file format (OBJ), etc. These files typically store the structure, attributes, constraints and activation conditions of the model. +- **Storage types:** It is recommended to support storing model files locally or in cloud storage for quick access and modification. +- **Version control:** It is recommended to support version control storage for digital twin models, which can track and manage model versions to resolve model conflicts and errors, and have appropriate version control and management mechanisms. + +### 7.4 Scalability requirements + +The scalability of the digital twin platform requires the platform to effectively adapt to and cope with the growing demand and scale, including processing large amounts of data, supporting more users and dealing with more complex computing tasks. + +- **Distributed deployment architecture:** It is recommended to support a distributed deployment architecture to ensure platform scalability, support the addition of service nodes and support the operation of servers within nodes as service hosting requirements increase to improve platform hosting capabilities. + +- **Programming:** It is recommended to support scalable programming languages and frameworks to automatically adjust the functionality and performance of platform as application scenarios increase or decrease. +- **Computing capabilities:** It is required to support scalable computing capabilities, with the ability to scale horizontally as the demand for computing resources increases, to support simulation of larger or more complex modules. +- **Application programming interface (API) and data exchange mechanism:** It is required to have an appropriate API and data exchange mechanism to support interoperability with other systems or platforms. + +### 7.5 Precision requirements + +The digital twin platform has a high degree of precision to ensure accurate modelling simulation and decision-making. + +- It is recommended to support a distributed deployment architecture to ensure platform scalability, support the addition of service nodes, and support the operation of servers within nodes as service hosting requirements increase to improve platform hosting capabilities. +- It is required to support fine model resolution to capture the relevant physical phenomena and characteristics of the modelling system. +- It is required to acquire and process up-to-date data in a timely manner and perform real-time simulations and predictions. +- It is required to support high precision of data acquisition and sensors to ensure that the digital twin model accurately reflects the state and behaviour of the real system. + +### 7.6 Compatibility requirements + +The digital twin platform needs to be compatible enough to ensure that data and model integration, as well as applicability to different operating systems, support platform collaboration. + +- **Data source compatibility:** It is required to support compatibility with a variety of data sources. It needs to be compatible with different manufacturers, different types of equipment and sensor access. +- **Data compatibility:** It is required to have strong data compatibility and seamless access to multiple data standards and support the integration of multiple types of data into a three-dimensional visualization space for a high degree of fusion and mining analysis. +- **Model compatibility:** It is required to support different types of mathematical models and algorithms for simulation, prediction and optimization. +- **Communication protocol compatibility:** It is required to support various communication protocols to exchange data and communicate with other systems and devices. Including hypertext transfer protocol (HTTP), message queuing telemetry transport (MQTT), object linking and embedding for process control unified architecture (OPC-UA), etc. + +### 7.7 Security requirements + +Since an effective digital twin system can only be built based on trusted data, to increase the user's awareness of the security of the digital twin platform and to prevent users from introducing security risks in the process of using the services, the security service requirements of the digital twin platform include data security, cyber security and management security requirements. + +- **Data security:** It is required to have the ability to ensure data security of the digital twin platform. Data security protection includes data privacy protection, data integrity and confidentiality protection. The confidentiality, integrity and reliability of sensitive data + +requires that it be maintained throughout its lifecycle, including creation, storage, use, sharing, archiving and destruction. + +- **Heterogeneous methods:** It is required have the ability to use different security protection methods for data security protection based on different levels of sensitive data. +- **Network security:** It is required to have the ability to ensure the network security of the digital twin platform and avoid all possible attacks. It is required to have the ability to detect and defend against all kinds of attacks, support the monitoring of network attacks on key network nodes and important servers, have a defence plan after the network has been maliciously attacked, support regular security detection of attacks and issue an alert in the event of a serious intrusion. + +It is required to have the ability to secure the management of the digital twin platform. Supports the management of the digital twin platform, including authentication and authorization of digital twin platform (DTP) platform users and authorization of operations such as create-read-update-delete for lifecycle data/models/applications. + +## 8 Functional architecture of the digital twin platform + +The framework of digital twins is shown in Figure 2, which includes general functions, twin functions and interactive functions. + +![Figure 2 – Functional architecture of the digital twin platform. The diagram is a hierarchical block diagram titled 'Digital twin platform'. It is divided into three main columns: 'General functions', 'Twin functions', and 'Interactive functions'. 'General functions' include 'Platform security' (User authentication, Log audit, Access control, Vulnerability scanning) and 'Maintenance' (Fault recovery, Auto save, Performance analysis). 'Twin functions' are organized into four sub-blocks: 'Visualization' (Video fusion, Monitoring, Dashboard, 3D visualization), 'Resource management' (Data management, Model management, Multimedia data processing, Collaborative management, Video data, Audio data, Text data, Device management), 'Intelligent service' (Simulation, Intelligent control, Intelligent decision, Alert), and 'Deployment' (Modelling, 3D rendering, Model training, Version control, Model development). 'Interactive functions' include 'Multimedia input' (Multimedia data input), 'Interactive processing', and 'Multimedia output' (Audio and video transmission, 3D visualization interaction). A legend at the bottom indicates that dashed boxes represent 'The required capabilities' and solid boxes represent 'The recommended capabilities'.](7f17c430b9598e4d748a8041457810b3_img.jpg) + +**Digital twin platform** + +| General functions | Twin functions | Interactive functions | +|----------------------------------------------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Platform security
User authentication
Log audit
Access control
Vulnerability scanning | Visualization
Video fusion, Monitoring, Dashboard, 3D visualization
Resource management
Data management, Model management, Multimedia data processing, Collaborative management, Video data, Audio data, Text data, Device management
Intelligent service
Simulation, Intelligent control, Intelligent decision, Alert
Deployment
Modelling, 3D rendering, Model training, Version control, Model development | Multimedia input
Multimedia data input
Interactive processing
Multimedia output
Audio and video transmission, 3D visualization interaction | +| Maintenance
Fault recovery
Auto save
Performance analysis | | | + +  The required capabilities +   The recommended capabilities + +F.748.28(24) + +Figure 2 – Functional architecture of the digital twin platform. The diagram is a hierarchical block diagram titled 'Digital twin platform'. It is divided into three main columns: 'General functions', 'Twin functions', and 'Interactive functions'. 'General functions' include 'Platform security' (User authentication, Log audit, Access control, Vulnerability scanning) and 'Maintenance' (Fault recovery, Auto save, Performance analysis). 'Twin functions' are organized into four sub-blocks: 'Visualization' (Video fusion, Monitoring, Dashboard, 3D visualization), 'Resource management' (Data management, Model management, Multimedia data processing, Collaborative management, Video data, Audio data, Text data, Device management), 'Intelligent service' (Simulation, Intelligent control, Intelligent decision, Alert), and 'Deployment' (Modelling, 3D rendering, Model training, Version control, Model development). 'Interactive functions' include 'Multimedia input' (Multimedia data input), 'Interactive processing', and 'Multimedia output' (Audio and video transmission, 3D visualization interaction). A legend at the bottom indicates that dashed boxes represent 'The required capabilities' and solid boxes represent 'The recommended capabilities'. + +**Figure 2 – Functional architecture of the digital twin platform** + +### 8.1 General functions + +#### 8.1.1 Platform security + +The digital twin platform supports security measures to ensure the confidentiality, integrity and availability of data and systems, including security measures such as data encryption, authentication, log management and auditing, access control and vulnerability scanning. + +- **Data encryption:** The digital twin platform provides security measures for data confidentiality protection, implements appropriate measures such as encryption technology and data leakage prevention technology, ensures data confidentiality, and prevents unauthorized disclosure of sensitive data during static or transmission processes. +- **Key management:** The digital twin platform provides services for key management, such as generation, distribution, sharing, reallocation and revocation of encryption keys for data confidentiality and authentication. +- **User authentication:** The digital twin platform provides services for identity authentication, which ensures that only authorized users have access to sensitive data and systems and that the identity repository is regularly reviewed and updated to ensure validity (i.e., to ensure that identities are still needed for access). +- **Log audit:** The digital twin platform provides logging and auditing services to realize the log management function for the operation and management of the digital twin system, and the system logs mainly include operation logs, running logs and security logs to detect the operation and access activities of the platform and detect and respond to abnormal behaviours in a timely manner. +- **Access control:** The digital twin platform provides access control services that use verified entity identities or entity capabilities to determine and enforce entity access privileges, deny unauthorized or improper access attempts, and control the time, path and physical location of attempted access. +- **Vulnerability scanning:** The digital twin platform provides vulnerability scanning services to identify potential security vulnerabilities and threats, and ensures that vulnerability scanning is conducted regularly to ensure system security. + +#### 8.1.2 Maintenance + +The digital twin platform provides multiple functions for maintenance needs, including fault recovery, automatic saving, performance analysis and PaaS service management. + +- **Fault recovery:** The digital twin platform provides services for state detection and automatic fault recovery, supporting real-time monitoring of the state of physical assets or systems, alarm actions in the event of failure and intelligent recovery. +- **Auto save:** The digital twin platform provides automatic storage services to ensure the automatic storage of relevant data and information, reduce the risk of data loss or damage during maintenance and ensure the security and reliability of data. +- **Performance analysis:** The digital twin platform includes system monitoring and provides performance analysis services, which monitor relevant performance data in the platform, such as network latency and response time. It measures and analyses relevant data, and implements optimization and adjustment based on the analysis results. +- **PaaS service management:** The digital twin platform supports the management of the full life cycle process of PaaS services, requiring regular monitoring of asset management, account passwords, compliance configurations, as well as the ability to analyse the performance of PaaS services, to achieve visual management of various types of information. + +### 8.2 Twin services + +#### 8.2.1 Visualization + +The digital twin platform has a range of visualization services for the visual representation of information and data, including video fusion, real-time visualization and 3D visualization, to ensure interaction and exploration with digital representations of physical assets or systems. + +- **Visualizing models:** According to application requirements, it supports visualizing models that display static traffic objects, dynamic traffic objects and surrounding environments. +- **Real-time visualization:** It supports configuring the parameters of the model to achieve real-time visualization. +- **Video fusion:** It supports video fusion and achieves the integration of real-time video data with physical assets or system digital twin models. This integration can provide valuable information on the performance and status of modelling objects, and supports maintenance and monitoring. +- **Dashboard:** It supports dashboard functionality, allowing users to view and interact with data and perspectives of physical assets or system digital twin models. The dashboard is required to provide a high-level overview of asset or system status and performance. +- **3D visualization:** it supports 3D visualization and creates highly detailed 3D models of physical assets for simulating various scenarios and testing different configurations or designs. + +#### 8.2.2 Deployment + +By providing powerful deployment services, the digital twin platform can effectively manage the deployment and operation of digital twin instances. The deployment requirements of the digital twin platform include digital twin modelling, simulation training, version control and deployment automation. + +- **Modelling:** It supports the modelling of physical entities in the real world in computers, integrates multi-source heterogeneous data, and models static objects, dynamic objects and the motion of objects. Furthermore, it supports the modelling of real-world scenes in computers, utilizing 2-dimension (2D) or 3D reconstruction techniques to maximize the reproduction of the surrounding environment. +- **Version control:** It supports platform version control mechanisms for managing different versions of digital twin configurations and software artefacts to support rollback and tracking of changes made to digital twin instances. +- **Model training:** It supports the construction of simulation deductive models for training and optimizing physical systems or processes. +- **Deployment automation:** It provides automation capabilities for deploying digital twin instances, ensuring consistency and reliability across different environments. + +#### 8.2.3 Intelligent functions + +The intelligent functions of the digital twin platform consist of simulations, intelligent controls, intelligent decisions and alerts. It utilizes advanced analytics and AI technologies to provide predictive, prescriptive and autonomous capabilities for physical assets or systems. These intelligent services assist in decision making and improve the overall efficiency and effectiveness of the physical asset or system. It supports the implementation of digital twins for real-time simulation and analysis, as well as diagnosis and prediction through artificial intelligence models and mechanism models. + +- **Simulation:** It has the ability to support multi-scenario simulation calculations and exercises based on real-time data and models. +- **Intelligent control:** It supports for autonomous control, which enables physical assets or systems to make autonomous decisions and adjustments based on real-time data and insights through artificial intelligence techniques such as reinforcement learning. +- **Intelligent decision:** It provides multimodal fusion analysis capabilities in simulation, emulation, prediction and other scenarios to realize intelligent information analysis and intelligent decision making. + +- **Optimization algorithms:** It supports the use of optimization algorithms that intelligently recommend the best policy based on the evaluation of the digital twin's optimization policy to help determine the most effective and efficient way to manage physical assets or system operations. +- **Alerts:** It supports intelligent alerts for monitoring and notification through automated means to help monitor the health status of the platform. + +#### 8.2.4 Resource management + +The resource management of the digital twin platform includes data management, model management, collaboration management and device management to enable the allocation and optimization of resources needed to create, maintain and use digital twin models of physical assets or systems. + +- **Data management:** It supports data management and facilitates the management of the entire lifecycle process of data storage, modelling and instantiation, and includes classification management of multimedia data, video data, audio data and text data. +- **Model management:** It supports model management and provides services for the creation, storage and update of various digital twin model instances, as well as model combination and application association management. + +Specifically, for the dynamic management of 3D models, a model processing system is required to meet the following requirements: + +- a) The model processing system includes a task server and dynamic loading module. +- b) The task server is used to obtain task information to be processed, including model identifier, model deployment information and task type (one of deletion type, addition type and modification type) of the 3D model, and send the task information to the dynamic loading module. +- c) After receiving the task information to be processed, the dynamic loading module needs to fetch the task type. If the task type is addition type or modification type, it obtains the mesh information in the model file corresponding to the model identifier. The model file corresponding to each model identifier is constructed for the 3D model indexed by the model identifier. The 3D model is processed based on the model deployment information and the mesh information when the 3D rendering engine runs the scene file of the virtual 3D scene. +- d) When the task type is deletion type, the dynamic loading module needs to delete the mesh body corresponding to the model identifier in the task information to be processed in the virtual 3D scene. Whereas the mesh body corresponding to each model identifier is generated based on the model file corresponding to the model identifier. +- e) When the task type is addition type, the dynamic loading module needs to determine the deployment position and orientation of the 3D model in the virtual 3D scene based on the model deployment information, and generate the 3D model in the virtual 3D scene at this deployment position and orientation based on the mesh information. +- f) When the task type is modification type, the dynamic loading module needs to first perform the deletion process specified in item d), and then perform the addition process of specified in item e) to complete the modification of the 3D model. +- g) The deployment position and orientation need to be represented by the coordinate system information of 3D model coordinate system. The mesh information in each model file includes the position and orientation information of each mesh body in the 3D model corresponding to the model file, which is the 3D model coordinate system of the 3D model. + +- **Collaborative management:** It supports collaborative management and multi-party collaboration on digital twin platforms through cloud platforms. +- **Device management:** It supports device management, including hardware and software resources required for digital twin systems. For hardware resources, it is necessary to optimize their configuration and performance, monitor their usage and maintenance. For software resources, such as simulation tools and data analysis software, it is necessary to optimize their configuration and performance to ensure compatibility and interoperability. + +### 8.3 Interactive functions + +#### 8.3.1 Overview + +Interactive functions are mainly used to provide real-time interaction capability for digital twin applications. Figure 3 shows a typical interaction process in which the multimedia input service receives the 3D model interaction requests of a digital twin application submitted by the user from the client. Interactive processing establishes audio and video transmission channels based on web real-time communications (WebRTC). Based on the interaction data, generating the audio and video stream by the 3D model rendering engine, transmitting it to the client by the audio and video transmission channels to complete 3D visualization interaction. + +![Figure 3 – Capabilities of interactive services. The diagram illustrates a flow from input data to interactive processing and finally to output services. On the left, input data (Image data, Video data, Audio data, Sensor data, ...) are shown. Arrows point from this data to a central processing block. This block contains three sub-processes: 'Respond to request', 'Processing interactive data', and 'Calling the rendering engine'. To the right of these are two additional components: 'Data transmission channel' and 'Obtain configuration information'. Arrows from the central processing block point to a final output block on the right. This output block contains three services: 'Offline and real-time rendering', 'Audio and video streaming transmission', and 'Visualization interaction'. A small label 'F.748.28(24)' is at the bottom right of the diagram.](16152cf1d84aea10848758f51a91ff6a_img.jpg) + +Figure 3 – Capabilities of interactive services. The diagram illustrates a flow from input data to interactive processing and finally to output services. On the left, input data (Image data, Video data, Audio data, Sensor data, ...) are shown. Arrows point from this data to a central processing block. This block contains three sub-processes: 'Respond to request', 'Processing interactive data', and 'Calling the rendering engine'. To the right of these are two additional components: 'Data transmission channel' and 'Obtain configuration information'. Arrows from the central processing block point to a final output block on the right. This output block contains three services: 'Offline and real-time rendering', 'Audio and video streaming transmission', and 'Visualization interaction'. A small label 'F.748.28(24)' is at the bottom right of the diagram. + +Figure 3 – Capabilities of interactive services + +#### 8.3.2 Multimedia input + +The multimedia input service of the digital twin platform is crucial for obtaining various types of multimedia data and integrating them into the digital twin system. The input data are required to meet the following requirements: + +- **Multi-source heterogeneous data:** It supports the collection of multi-source heterogeneous data, including data collected by various sensors, such as keyboard input, touch screen input, camera input and other media input methods. These data may belong to different institutions and third parties to meet the requirements of digital twin platform data collection. +- **Multiple data types:** It supports input of multiple data types, including image data, audio data, video data, sensor data and industrial equipment data, and supports the integration of those data into the digital twin platform. +- **Multiple data formats:** It supports input in multiple data formats, including JavaScript object notation (JSON), XML, hypertext markup language (HTML), comma-separated values (CSV), etc., while also supporting the conversion of data to 3D data formats such as 3D tiles, geographic JavaScript object notation (GeoJSON) and other 3D modelling software formats. It supports 3D model software and material format import, such as 3dmax, maya, etc. + +#### 8.3.3 Interactive processing + +Interactive processing completes to respond to requests, process the interaction data, call the 3D rendering engine and obtain user configuration information to establish data transmission channels. + +- **3D model interaction:** It responds to the 3D model interaction request of digital twin application, acquires user interaction data and parses it. +- **User interaction:** It reconstructs the user interaction data to match the data format required by the 3D model rendering engine. +- **Interactive component:** It calls the 3D model rendering engine to analyse the reconstructed interaction data and calls the encapsulated interactive component to perform interaction functions of digital twin application to obtain the rendering stream. +- **User configuration:** It obtains user configuration information from interactive data and establishes audio and video transmission channels. +- **Multiple channels:** It establishes multiple audio and video transmission channels corresponding to multiple clients for different interaction requests. + +#### 8.3.4 Multimedia output + +Multimedia output enables the 3D visualization interaction and audio and video transmission. It receives the 3D model interaction request submitted by the user from the client, runs the 3D rendering engine based on the interaction data, generates the audio and video stream after 3D visualization interaction and transmits it to the user for visualization. + +- **Offline and real-time rendering:** It supports offline and real-time rendering to achieve the rendering of twin scenes, to make timely decisions based on the data generated by the digital twin platform. The platform supports multiple renderers +- **Audio and video streaming transmission:** It transmits the rendered audio and video stream for visualization. +- **Visualization interaction:** It supports a digital twin visualization rendering engine, 2D/3D geographic information system (GIS) rendering, environmental scene rendering and 3D model rendering. + +## Appendix I + +### Use cases of the digital twin platform for supporting multimedia services + +(This appendix does not form an integral part of this Recommendation.) + +The digital twin platform can fuse and calculate the data of perception, simulation and control, making the algorithm faster and more powerful, and ensuring the speed and accuracy of analysis, deduction and decision-making. At present, it has been widely used in fields of education, video-based marketing, digital museum and e-health. + +### I.1 Use case of digital twin for education + +As shown in Figure I.1, the digital twin platform provides multimedia services for education applications, such as distance training, visual classroom, campus visualization, campus security, experimental simulation, etc. The digital twin platform provides 3D visualization, resource management, platform security and maintenance, deployment, intelligent services, etc. + +![Diagram of education use case](8592a32c2fdf17c1e562f0ba6b7e8e1a_img.jpg) + +The diagram illustrates the architecture for the education use case. It consists of two main layers. The top layer, labeled 'Education applications', contains five individual boxes: 'Distance training', 'Visual classroom', 'Campus visualization', 'Campus security', and 'Experimental simulation'. The bottom layer is a single, wide grey box labeled 'Digital twin platform'. Three vertical double-headed arrows connect the 'Digital twin platform' box to the 'Education applications' layer, indicating bidirectional data flow and service interaction. The reference code 'F.748.28(24)' is located in the bottom right corner of the diagram. + +Diagram of education use case + +Figure I.1 – Use case of digital twin for education + +### I.2 Use case of digital twin for video-based marketing + +As shown in Figure I.2, the digital twin platform provides multimedia services for video-based marketing applications, such as commodity exhibition, remote shopping, shopping behaviour analysis, customer experience personalization, etc. The digital twin platform provides 3D visualization, resource management, platform security and maintenance, deployment, intelligent services, etc. + +![Diagram of video-based marketing use case](124c6108c63173818afb8ed49521e22d_img.jpg) + +The diagram illustrates the architecture for the video-based marketing use case. It features two primary layers. The upper layer, titled 'Marketing applications', includes four boxes: 'Commodity exhibition', 'Remote shopping', 'Shopping behaviour analysis', and 'User experience personalization'. The lower layer is a grey box labeled 'Digital twin platform'. Three vertical double-headed arrows connect the 'Digital twin platform' to the 'Marketing applications' layer, representing bidirectional communication. The reference code 'F.748.28(24)' is positioned in the bottom right corner of the diagram. + +Diagram of video-based marketing use case + +Figure I.2 – Use case of digital twin for video-based marketing + +### I.3 Use case of digital twin for digital museum + +As shown in Figure I.3, the digital twin platform provides multimedia services for digital museum applications, such as online exhibition, remote visit, cultural relics visualization, recurrence of historical events and museum security. The digital twin platform provides 3D visualization, resource management, platform security and maintenance, deployment, intelligent services, etc. + +![Diagram of digital twin use case for digital museum. A top box labeled 'Digital museum applications' contains five sub-boxes: 'Online exhibition', 'Remote visit', 'Cultural relics visualization', 'Recurrence of historical events', and 'Museum security'. Below this is a large grey box labeled 'Digital twin platform'. Three double-headed vertical arrows connect the top box to the bottom box. The text 'F.748.28(24)' is at the bottom right.](9c6461e1e94afae4dec455e69a2ce152_img.jpg) + +Diagram of digital twin use case for digital museum. A top box labeled 'Digital museum applications' contains five sub-boxes: 'Online exhibition', 'Remote visit', 'Cultural relics visualization', 'Recurrence of historical events', and 'Museum security'. Below this is a large grey box labeled 'Digital twin platform'. Three double-headed vertical arrows connect the top box to the bottom box. The text 'F.748.28(24)' is at the bottom right. + +**Figure I.3 – Use case of digital twin for digital museum** + +### **I.4 Use case of digital twin for e-health** + +As shown in Figure I.4, the digital twin platform provides multimedia services for e-health applications, such as remote surgery, patient status monitoring, medical equipment management, online outpatient and simulation teaching, etc. The digital twin platform provides 3D visualization, resource management, platform security and maintenance, deployment, intelligent services, etc. + +![Diagram of digital twin use case for e-health. A top box labeled 'E-health applications' contains five sub-boxes: 'Remote surgery', 'Patient status monitoring', 'Medical equipment management', 'Online outpatient', and 'Simulation teaching'. Below this is a large grey box labeled 'Digital twin platform'. Three double-headed vertical arrows connect the top box to the bottom box. The text 'F.748.28(24)' is at the bottom right.](10781f43062bf3e9601a1e086710556c_img.jpg) + +Diagram of digital twin use case for e-health. A top box labeled 'E-health applications' contains five sub-boxes: 'Remote surgery', 'Patient status monitoring', 'Medical equipment management', 'Online outpatient', and 'Simulation teaching'. Below this is a large grey box labeled 'Digital twin platform'. Three double-headed vertical arrows connect the top box to the bottom box. The text 'F.748.28(24)' is at the bottom right. + +**Figure I.4 – Use case of digital twin for e-health** + +## Bibliography + +[b-ITU-T Y.4600] Recommendation ITU-T Y.4600 (2022), *Requirements and capabilities of a digital twin system for smart cities*. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.29-202406-I_PDF-E/raw.md b/marked/F/T-REC-F.748.29-202406-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..6a6e2950071f40fbd7f1be27a37ebff1ee081121 --- /dev/null +++ b/marked/F/T-REC-F.748.29-202406-I_PDF-E/raw.md @@ -0,0 +1,522 @@ + + +# Recommendation **ITU-T F.748.29 (06/2024)** + +SERIES F: Non-telephone telecommunication services +Multimedia services + +--- + +# **Framework and requirements of computer audition-based machinery fault diagnosis systems** + +![ITU logo](390120de4fe440c42fea8154fcaad334_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## TU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.29 + +# Framework and requirements of computer audition-based machinery fault diagnosis systems + +## Summary + +Building intelligent machinery fault diagnosis systems is essential for preventing unexpected breakdowns of machinery, thus enhancing production efficiency and ensuring safety during operations. The integration of computer audition into machinery fault diagnosis systems can detect subtle faults imperceptible to vision through non-destructive ways. Recommendation ITU-T F.748.29 specifies the basic concept of computer audition and identifies the framework and requirements of computer audition-based machinery fault diagnosis systems. + +## History\* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.29 | 2024-06-13 | 16 | 11.1002/1000/15911 | + +## Keywords + +Computer audition, framework, machinery fault diagnosis, requirements. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2024 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-----------------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Overview ..... | 2 | +| 6.1 Basic concept of computer audition (CA) ..... | 2 | +| 6.2 Introduction of CA based machinery fault diagnosis systems (MFDS)..... | 2 | +| 7 Framework of computer audition-based machinery fault diagnosis systems..... | 3 | +| 7.1 Overview ..... | 3 | +| 7.2 Application layer ..... | 4 | +| 7.3 Service support and application support layer..... | 4 | +| 7.4 Network layer ..... | 5 | +| 7.5 Device layer..... | 5 | +| 7.6 Management capabilities ..... | 5 | +| 7.7 Security capabilities..... | 6 | +| 8 Requirements for computer audition-based machinery fault diagnosis systems..... | 6 | +| 8.1 High-level requirements ..... | 6 | +| 8.2 Requirements for the service support and application support layer ..... | 8 | +| 8.3 Requirements for the network layer ..... | 10 | +| 8.4 Requirements for the device layer..... | 10 | +| Bibliography..... | 12 | + + + +# Framework and requirements of computer audition-based machinery fault diagnosis systems + +## 1 Scope + +This Recommendation specifies framework and requirements of computer audition-based machinery fault diagnosis systems. The scope of this Recommendation includes: + +- Overview. +- Framework of computer audition-based machinery fault diagnosis systems. +- Requirements for computer audition-based machinery fault diagnosis systems. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T Y.4000] Recommendation ITU-T Y.4000 (2012), *Overview of the Internet of things*. + +# 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 algorithm** [b-ISO/IEC 11557]: Set of rules for transforming the logical representation of data. + +**3.1.2 application** [b-ITU-T Y.101]: A structured set of capabilities, which provide value-added functionality supported by one or more services. + +**3.1.3 machine learning** [b-ISO/IEC 22989]: Process of optimizing model parameters through computational techniques, such that the model's behaviour reflects the data or experience. + +**3.1.4 system** [b-ISO/IEC 24028]: Combination of interacting elements organized to achieve one or more stated purposes. + +NOTE – A system is sometimes considered as a product or as the services it provides. + +**3.1.5 training** [b-ISO/IEC 24028]: Process to establish or to improve the parameters of a machine learning model based on a machine learning algorithm by using training data. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 computer audition**: An interdisciplinary scientific field that deals with the complex problem of understanding and analysing sound by computer. + +**3.2.2 machinery fault diagnosis**: A procedure that involves monitoring mechanical equipment through sensors, extracting relevant features from the data, assessing current health state of the equipment and predicting the remaining useful life and trend of the fault. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|----------------------------------------| +| API | Application Programming Interface | +| CA | Computer Audition | +| CNN | Convolutional Neural Network | +| IMT | International Mobile Telecommunication | +| IoT | Internet of Things | +| LSTM | Long Short-Term Memory | +| MFCC | Mel-Frequency Cepstral Coefficients | +| MFDS | Machinery Fault Diagnosis Systems | +| QoS | Quality of Service | +| RNN | Recurrent Neural Network | +| SFFT | Sparse Fast Fourier Transform | +| TDNN | Time-Delay Neural Network | +| TLS | Transport Layer Security | +| TLCP | Transport Layer Cryptography Protocol | + +## 5 Conventions + +In this Recommendation: + +- The keywords "**is required**" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The keywords "**is recommended**" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement needs not be present to claim conformance. + +# 6 Overview + +### 6.1 Basic concept of computer audition (CA) + +Computer audition is an emerging interdisciplinary subject inspired by the human hearing system, covering science, engineering, medicine, art and other disciplines. Its goal is to enable computers to perceive the objective world close to or even beyond the limits of human hearing, namely including audible sound and inaudible sound (ultrasound and infra-sound), based on cutting-edge acoustic signal processing and machine learning/deep learning technologies. + +## 6.2 Introduction of CA based machinery fault diagnosis systems (MFDS) + +Machinery fault diagnosis is a scientific technology to ensure the safe operation of mechanical equipment by monitoring the signals of running machinery or detecting static equipment engineering structures, diagnosing the status of equipment and predicting the operational trends. Common fault diagnosis methods can be grouped into three main categories: model-based diagnosis, knowledge-based diagnosis and data-based diagnosis. Model-based diagnosis allows for the identification of possible root causes of system misbehaviour by comparing actual system performance against predefined models, inferring faults when discrepancies are observed. Knowledge-based diagnosis leverages rule-based expert systems, focusing on the application of specialized domain knowledge and expertise in a structured format for diagnostic reasoning. Lastly, data-based diagnosis utilizes + +various techniques such as statistical analysis, signal processing and artificial intelligence to enable a data-driven approach to fault identification and analysis. + +With the vigorous development of industrial Internet and Internet of things (IoT), the growing amount of multi-source sensors for mechanical equipment and monitoring interaction data, driven by artificial intelligence and big data technology, the era of intelligent machinery fault diagnosis is coming. Intelligent fault diagnosis emulates the human thought process for inference. It captures, transfers, and processes diagnostic information, allowing for an intelligent determination of the operating condition and faults of the monitored machine. Intelligent diagnosis reduces the over-reliance of traditional fault diagnosis methods on the experience of diagnostic experts and professional technicians to a certain extent, improves the decision-making efficiency of the systems and reduces the production and operation costs of enterprises. + +The CA based MFDS integrates the computer audition technology, audio structuralizing, deep learning technology, edge computing, security management and other technologies to provide manufacturers with a series of machinery fault diagnosis capabilities including data acquisition, feature extraction, fault detection and data management. + +# **7 Framework of computer audition-based machinery fault diagnosis systems** + +### **7.1 Overview** + +Figure 1 shows the end-to-end layered framework of MFDS based on the IoT reference model [ITU-T Y.4000]. It is composed of four layers as well as management capabilities and security capabilities that are associated with the four layers. + +The four layers are as follows: + +- application layer; +- service support and application support layer; +- network layer; +- device layer. + +![Figure 1 – Framework of CA based MFDS. The diagram shows a four-layer architecture: Application layer (Systems level: Wind turbine, Gas turbine, Transformer, High-speed train; Component level: Bearing, Gear wheel, Rotor, Belt), Service support and application support layer (Algorithm/model library, Data management, Algorithm/model development, Federated prototype learning), Network layer (Networking and transport, Wireless network access: Wireless LAN, IMT-Advanced/IMT-2020, NB-IoT; Wired network access: Optical fibre, EIA-485), and Device layer (Vibration sensor, Acoustic sensor, Acoustic emission sensor). Management capability and Security capability are shown as vertical bars on the left and right sides respectively. F.748.29(24) is noted at the bottom right.](a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg) + +F.748.29(24) + +Figure 1 – Framework of CA based MFDS. The diagram shows a four-layer architecture: Application layer (Systems level: Wind turbine, Gas turbine, Transformer, High-speed train; Component level: Bearing, Gear wheel, Rotor, Belt), Service support and application support layer (Algorithm/model library, Data management, Algorithm/model development, Federated prototype learning), Network layer (Networking and transport, Wireless network access: Wireless LAN, IMT-Advanced/IMT-2020, NB-IoT; Wired network access: Optical fibre, EIA-485), and Device layer (Vibration sensor, Acoustic sensor, Acoustic emission sensor). Management capability and Security capability are shown as vertical bars on the left and right sides respectively. F.748.29(24) is noted at the bottom right. + +**Figure 1 – Framework of CA based MFDS** + +### 7.2 Application layer + +The application layer provides a range of fault diagnosis applications, which can be categorized into the system level diagnosis and component level diagnosis. For example, the three basic rotating components (bearing, gear wheel and rotor) play an important role in industrial rotating and transport machinery applications. + +Evaluating the health status of the entire assembled system holds more significance compared to assessing individual components. This is primarily because the fault diagnosis in systems is essentially a multiple-input and multiple-output inverse problem. For example, typical faults in rotating machinery include rotor bow, rigid coupling misalignment, transverse crack and axial asymmetry. + +## 7.3 Service support and application support layer + +The service support and application support layer consist of algorithm/model library, algorithm/model development, data management and federated prototype learning. + +The algorithm/model library is pre-loaded with a variety of generic models as well as specific fault recognition models, i.e., sound recognition models and vibration recognition models. This can help enterprises reduce the time and cost of developing and configuring models, thereby lowering the difficulty of implementation. + +#### Abnormal data recognition model + +Collected data contain a mix of normal and abnormal data. Recognizing abnormal data is an essential step that helps in maintaining the quality of the dataset. The abnormal data recognition model consists of a generator, a simulated data recognition module and an anomaly data recognition module. The model training process is as follows: First, use the generator to imitate the normal and abnormal + +sample data from the collected data, producing simulated data. Next, construct two types of training sample data pairs: the first consisting of simulated data and abnormal sample data, and the second comprising simulated data, normal sample data and abnormal sample data. Then, conduct the adversarial training of the generator and the simulated data recognition module using the first type of training pairs, and conduct the adversarial training of the generator and the anomaly data recognition module using the second type of training pairs. These two adversarial training processes are alternately. Upon completion of the training, a pre-trained abnormal data recognition model is acquired. + +#### **Fault recognition model** + +The fault recognition model comprises a primary network module, a threshold prediction layer, a classification layer, and at least one secondary network module. The primary and secondary network modules consist of one or more of the following deep neural network modules: recurrent neural network (RNN), long short-term memory (LSTM), time-delay neural network (TDNN) and Transformer. The threshold prediction layer consists of the global average pooling layer and the softmax layer. In this architecture, the primary network module is connected to the threshold prediction layer and the classification layer respectively. At least one secondary network module is connected between the primary network module and the classification layer. Multiple secondary network modules, if present, are sequentially linked and each connects to the classification layer. During training, the fault recognition model adopts cross-entropy as its loss function. + +The algorithm/model development provides professional and systematic development tools and environments to developers. These tools enhance the efficiency of model optimization and evaluation, making it easier for developers to handle technical challenges and focus more on creating solutions and solving problems. + +The data management provides capabilities for data processing, data storage, data computing and data integration. + +The federated prototype learning functionality enables multiple developers to build a common, robust machine learning model without sharing data, thus addressing critical issues such as data privacy and data security. + +### **7.4 Network layer** + +The network layer includes wireless access network (e.g., international mobile telecommunication-2000 (IMT-2020), wireless LAN and NB-IoT), wired access (e.g., optical fibre) and transport network. + +### **7.5 Device layer** + +The device layer comprises vibration sensors, acoustic sensors and acoustic emission sensors. They complete the sound and vibration signal acquisition of machinery. + +### **7.6 Management capabilities** + +#### **7.6.1 User management** + +UM-01: It is required to authenticate user identities. When the number of failed authentication attempts exceeds a preset threshold, immediate security measures should be activated to address potential security risks. + +UM-02: It is required to protect authentication information in the system, preventing any form of unauthorized disclosure. + +UM-03: It is required to make predefined access control rules, while also allowing system administrators the flexibility to make necessary modifications. Before modifying the rules, users should re-authenticate their identities. + +UM-04: It is required to allow authorized users to access applications, system data and other resources according to predefined rules, while preventing access by unauthorized users and applications. + +#### **7.6.2 Service management** + +The requirements for service management are as follows: + +SM-01: It is recommended to provide mechanical fault detection algorithm/model to third parties through an application programming interface (API) and/or other methods. + +SM-02: It is recommended to support calling mechanical fault detection algorithm/model from third parties through an API and/or other methods. + +### **7.7 Security capabilities** + +#### **7.7.1 Sensor security** + +DS-01: It is recommended to support IP54 rating [b-IEC 60529]. + +DS-02: The sensor should have a unique device serial number that is recognizable, tamper-resistant and non-erasable. + +DS-03: It is recommended to implement alert mechanisms for unauthorized operation of sensors. + +#### **7.7.2 Network and transport security** + +NTS-01: It is required to support data transmission integrity verification mechanisms. + +NTS-02: It is required to have mechanisms to manage communication delays and interruptions. + +NTS-03: It is required to support dynamic configuration of security protection policies, preventing malicious scanning and attacks on gateway device ports by external applications. + +NTS-04: It is recommended to support transport layer security (TLS) and/or transport layer cryptography protocol (TLCP) transport protocols. + +# **8 Requirements for computer audition-based machinery fault diagnosis systems** + +### **8.1 High-level requirements** + +Figure 2 shows the fundamental process of CA based MFDS, including data acquisition, data processing, feature extraction and fault diagnosis. + +![Flowchart of the fundamental process of CA based MFDS. The process starts with 'Data acquisition', followed by 'Data pre-processing', and 'Feature extraction'. A decision diamond 'Fault diagnosis' follows. If 'No', it loops back to 'Feature extraction'. If 'Yes', it branches into five outcomes: 'Development trend', 'Fault type', 'Fault severity', 'Fault position', and 'Counter-measure'. The code F.748.29(24) is shown at the bottom right.](af7916c89a458fdab6c3f443217388ae_img.jpg) + +``` + +graph TD + A[Data acquisition] --> B[Data pre-processing] + B --> C[Feature extraction] + C --> D{Fault diagnosis} + D -- No --> C + D -- Yes --> E[Development trend] + D -- Yes --> F[Fault type] + D -- Yes --> G[Fault severity] + D -- Yes --> H[Fault position] + D -- Yes --> I[Counter-measure] + +``` + +F.748.29(24) + +Flowchart of the fundamental process of CA based MFDS. The process starts with 'Data acquisition', followed by 'Data pre-processing', and 'Feature extraction'. A decision diamond 'Fault diagnosis' follows. If 'No', it loops back to 'Feature extraction'. If 'Yes', it branches into five outcomes: 'Development trend', 'Fault type', 'Fault severity', 'Fault position', and 'Counter-measure'. The code F.748.29(24) is shown at the bottom right. + +**Figure 2 – Fundamental process of CA based MFDS** + +#### 8.1.1 Data acquisition + +While collecting mechanical sound signals, certain machines may encounter significant ambient noise, resulting in heavily mixed sound signals. Therefore, when designing the data acquisition component, various factors must be considered. To capture precise information from the machinery or equipment, the data acquisition requirements are outlined as follows: + +DA-01: It is required to support acquiring physical field signals, i.e., sound signals and vibration signals, generated by running machinery using high sensitivity sensors. + +DA-02: It is recommended to support multi-channel sensor array acquisition, which is conducive to sound source location and background noise elimination. + +DA-03: It is recommended to support the joint acquisition of vibration signal and sound signal. + +DA-04: It is recommended to support detecting static machinery structures by acoustic emission sensor and ultrasonic radar. + +#### 8.1.2 Data pre-processing + +Data pre-processing techniques will be adopted according to the needs of a given system. The requirements for data pre-processing are as follows. + +DP-01: It is required to support abnormal data recognition to prevent them from causing interference in the subsequent analysis process. + +DP-02: It is required to support noise reduction that does not affect accurate feature extraction to facilitate audio feature extraction and classification. + +DP-03: It is recommended to support signal-to-noise ratio enhancement, techniques including singular value decomposition filtering, wavelet filtering, empirical mode decomposition filtering, etc. + +DP-04: It is recommended to support channel conversion, resampling, decompression or pre-emphasis. + +DP-05: It is recommended to examine data consistency, identify and rectify anomalous data, and eliminate redundant data to optimize the subsequent feature extraction and analysis. + +#### 8.1.3 Feature extraction + +Feature extraction is the name for methods that select and combine variables into features, effectively reducing the amount of data that must be processed, while still accurately and completely describing the original data set. Because of the complexity of vibration and sound signals in fault diagnosis, mathematical transformations, signal processing and pattern recognition techniques are widely used to extract essential features from these signals. The requirements for feature extraction are outlined as follows. + +FE-01: It is required to support segmenting sound signals into frames according to predefined rules, extract information from these frames at different time points, and derive multiple feature values for various time instances. + +FE-02: It is required to support extracting features in frequency domain. Commonly used frequency domain analysis methods include refined spectrum analysis, power spectrum analysis and cepstrum analysis. + +FE-03: It is recommended to support the further selection of features that are more conducive to classification from the feature set. + +FE-04: It is recommended to support feature fusion, which involves extracting and constructing corresponding features from heterogeneous data, and then combining them into a feature set that integrates various features. This approach enables the exploitation of complementary advantages of multiple features, resulting in more robust and accurate recognition outcomes. + +#### 8.1.4 Fault diagnosis + +The principal function of the MFDS is to discern the operational status of machinery as being normal or abnormal. Furthermore, its advanced functionalities should enable the identification of fault types, fault severity and fault position. It should also predict the potential development of these faults and provide reasonable countermeasures. The requirements for fault diagnosis are outlined as follows. + +FD-01: In a machinery fault diagnosis model, the primary network module analyses feature values at the current moment and obtains results, the classification layer then classify these results to obtain the sound type of the machinery to be identified at the current moment. Subsequently, the threshold prediction layer processes the results to obtain the current moment's confidence level, thereby determining the number of frames skipped or the number of secondary network modules to be activated at the next moment. + +FD-02: The fault detection results are recommended to be delivered with low latency. The rapid delivery of fault detection outcomes can enable timely actions to address potential issues. + +FD-03: Providing incremental learning capabilities is recommended, as this enables the classifier to expand its own knowledge and thereby enhance the accuracy of fault diagnosis. + +FD-04: Implement predictive maintenance is recommended to enhance the operational efficiency of machinery. This approach involves analysing historical data to forecast potential issues and devise proactive maintenance plans to prevent sudden machinery downtime. + +### 8.2 Requirements for the service support and application support layer + +#### 8.2.1 Algorithm/model library + +The requirements for algorithm/model library are as follows: + +AML-01: It is required to support sound recognition models that can process the current moment's features to identify the sound type and determine the running status of the machinery based on this sound type. + +AML-02: It is required to support vibration recognition models that can analyse vibration features to determine vibration types and determine the running status of the machinery. + +AML-03: It is recommended to support frequency domain conversion pre-processing algorithm, such as sparse fast Fourier transform (SFFT), Mel-frequency cepstral coefficients (MFCC) and filter banks. + +AML-04: It is recommended to support generic neural network units, such as convolutional neural network (CNN), RNN, Transformer and LSTM. + +AML-05: It is recommended to support simple frequency domain feature extraction method, such as point-wise CNN, CNN, RNN, Transformer and LSTM. + +AML-06: It is recommended to support feature extraction models, such as models that extract d-vector, I-vector and x-vector. + +AML-07: It is recommended to support real-time, low-power consumption classification models, such as dynamic neural networks and auto-encoder neural networks. + +AML-08: It is recommended to support frequency domain feature recovery algorithm after the auto-encoder for better interpretation, such as point-wise CNN, CNN, RNN, transformer and LSTM. + +AML-09: It is recommended to support plug-in third party algorithms/models. + +#### **8.2.2 Algorithm/model development** + +The requirements for the algorithm/model development domain are as follows: + +AMD-01: It is recommended to support data labelling, model training and fine-tuning. + +AMD-02: It is recommended to support model optimization functions, such as model compression and inference acceleration. + +AMD-03: It is recommended to support model compiling. + +AMD-04: It is recommended to support model performance evaluation. + +AMD-05: It is recommended to support deep learning frameworks. + +AMD-06: It is recommended to provide a customized development environment, such as supporting the installation of necessary dependencies. + +AMD-07: It is recommended to support model visualization capabilities, such as the visualization of model structures, network weights, etc. + +#### **8.2.3 Data management** + +The requirements for data management are as follows: + +DM-01: It is required to store critical data, including the historical running sound/vibration signals from machinery, along with corresponding anomaly sound/vibration type labels and anomaly confidence level labels. These data are crucial for training the machinery fault recognition model, i.e., sound recognition model and vibration recognition model. + +DM-02: It is required to support both structured and unstructured storage, such as relational databases, analytical databases, object storage, cache databases, etc. + +DM-03: It is required to support centralized storage as well as distributed storage. + +DM-04: It is recommended to support hierarchical data storage. Data with real-time or high access demands should be stored in a real-time database, while data without these demands should be stored in a long-term preservation database. + +DM-05: It is recommended to provide capabilities for batch computing, stream computing, real-time computing and query computing. + +DM-06: It is required to support mechanisms for data confidentiality and data integrity. + +#### 8.2.4 Federated prototype learning + +Due to the extreme scarcity of industrial negative sample data, it is not feasible to rely on the data from a single enterprise for the training of an intelligent machinery fault diagnosis model. Therefore, multiple enterprises must provide both positive and negative sample data. Furthermore, each enterprise has its own privacy requirements regarding their data. Therefore, the sharing and aggregation of data must be accomplished through a privacy computing framework instead of direct sharing. In other words, the gradient must be aggregated and shared through this framework. As a result, MFDS is required to support the federated prototype learning model. The specific processing flow is as follows: + +- 1) Collect industrial data from each manufacturer, and the industrial data is divided into training dataset and test dataset. +- 2) Construct the federated network model, initialize model parameters, input the training set data to the federated network model for model training, and verify the model through the test dataset to form the final federated network model. +- 3) The final federated network model is distributed to the corresponding edge server of each manufacturer for update, so as to carry out quality inspection on the industrial data to be verified by each manufacturer and obtain the quality inspection result. +- 4) Input training dataset into the federated network model for feature extraction, and the feature vector is extracted for each data in the training set and the average value is calculated to obtain a plurality of classification prototypes. Then, extract feature vectors from the data except training set to build up verification set. +- 5) The distance between the verification set and each classification prototype is calculated, and the model parameters are adjusted by a predetermined loss function until the model converges to obtain the final federated network model. As the change value of the loss function is less than the predetermined change threshold, the model converges. + +### 8.3 Requirements for the network layer + +The requirements for the network layer are as follows. + +NL-01: It is recommended to support periodic transmission and event-driven transmission to deliver data to the service support and application support layer. + +NL-02: It is recommended to support quality of service (QoS) and reliable transmission to meet service requirements. + +NL-03: It is recommended to support the capability of acquiring sensors' location. + +### 8.4 Requirements for the device layer + +The device layer accommodates various capabilities outlined in [ITU-T Y.4000] and the following requirements to acquire sensing data and deliver it to the network layer: + +DL-01: It is required to support non-destructive detection of machinery, which is fundamental in ensuring that the machinery is evaluated accurately without causing any damage or alterations to its original state. + +DL-02: It is recommended to support non-contact detection methods for machinery to avoid unnecessary interference or damage to the industrial facilities. + +DL-03: It is required to support data caching to prevent data loss during network disconnection, allowing deferred uploading of cached data upon network restoration, thereby ensuring data integrity. + +DL-04: It is recommended to support generic data pre-processing and computing capabilities, such as background noise elimination. + +DL-05: It is recommended to support automatic deployment capabilities, which allow for rapid configuration and initiation of applications. + +DL-06: It is recommended to support dynamic adjustment and configuration of sensors based on their location. + +## Bibliography + +- [b-ITU-T Y.101] Recommendation ITU-T Y.101 (2000), *Global Information Infrastructure terminology: Terms and definitions.* +- [b-IEC 60529] IEC 60529:1989/AMD2:2013, *Degree of protection provided by enclosures (IP Code).* +- [b-ISO/IEC 11557] ISO/IEC 11557:1992, *Information technology – 3,81 mm wide magnetic tape cartridge for information interchange – Helical scan recording – DDS-DC format using 60 m and 90 m length tapes.* +- [b-ISO/IEC 22989] ISO/IEC 22989:2022, *Information technology – Artificial intelligence – Artificial intelligence concepts and terminology.* +- [b-ISO/IEC 24028] ISO/IEC TR 24028:2020, *Information technology – Artificial intelligence – Overview of trustworthiness in artificial intelligence.* + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.31-202410-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/F/T-REC-F.748.31-202410-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ac295b1b0c7f14b1c5a33ef6682b1b017c812477 --- /dev/null +++ b/marked/F/T-REC-F.748.31-202410-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37aed4fc114a39843438175326e0b7e1d09a9d6075723d82fea32040fd560ee4 +size 7192 diff --git a/marked/F/T-REC-F.748.31-202410-I_PDF-E/e1a0d046fbe7f28f5e93a47091851747_img.jpg b/marked/F/T-REC-F.748.31-202410-I_PDF-E/e1a0d046fbe7f28f5e93a47091851747_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..0a9643022316534b1f4417cf05966452f27bbe15 --- /dev/null +++ b/marked/F/T-REC-F.748.31-202410-I_PDF-E/e1a0d046fbe7f28f5e93a47091851747_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:847b39b682cf58ad11f6b36a98184d4b28f6e47dbb7af0c83afbe97318aac0ca +size 64206 diff --git a/marked/F/T-REC-F.748.31-202410-I_PDF-E/raw.md b/marked/F/T-REC-F.748.31-202410-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..3b6e19d18739b73c720add622d935547de55df46 --- /dev/null +++ b/marked/F/T-REC-F.748.31-202410-I_PDF-E/raw.md @@ -0,0 +1,436 @@ + + +# Recommendation**ITU-T F.748.31 (10/2024)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +### **Technical requirements and evaluation methods of 3D digital human system based on smart mobile devices** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS Non-telephone telecommunication services + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.31 + +# Technical requirements and evaluation methods of 3D digital human system based on smart mobile devices + +## Summary + +Recommendation ITU-T F.748.31 develops the overall architecture, functional requirements and performance requirements of 3-dimensional (3D) digital human system based on smart mobile devices. The evaluation methods of some unique requirements for smart mobile devices application are also specified. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.31 | 2024-10-29 | 16 | 11.1002/1000/16186 | + +## Keywords + +3D, digital human, functional requirement, performance requirement, smart mobile device. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2025 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|---------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References ..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation ..... | 1 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Overview ..... | 2 | +| 7 Functional requirements ..... | 3 | +| 7.1 Image ..... | 3 | +| 7.2 Speech ..... | 4 | +| 7.3 Animation ..... | 4 | +| 7.4 Interaction ..... | 5 | +| 7.5 Multimodal input ..... | 5 | +| 7.6 Multimodal output ..... | 6 | +| 8 Performance requirements and evaluation methods ..... | 6 | +| 8.1 Memory usage ..... | 6 | +| 8.2 Power consumption ..... | 6 | +| 8.3 Screen fluency ..... | 8 | + + + +# Recommendation ITU-T F.748.31 + +# Technical requirements and evaluation methods of 3D digital human system based on smart mobile devices + +## 1 Scope + +This Recommendation develops the overall architecture, functional requirements and performance requirements of 3-dimensional (3D) digital human system based on smart mobile devices (SMD). + +The scope of this Recommendation includes: + +- Overall architecture of 3D digital human system based on SMD; +- Functional requirements of 3D digital human system based on SMD; +- Performance requirements of 3D digital human system based on SMD; +- Evaluation methods of some unique requirements for smart mobile devices application. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T F.748.14] Recommendation ITU-T F.748.14 (2022), *Requirements and evaluation methods of non-interactive 2D real-person digital human application systems*. + +[ITU-T F.748.15] Recommendation ITU-T F.748.15 (2022), *Framework and metrics for digital human application systems*. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following term defined elsewhere: + +**3.1.1 digital human** [ITU-T F.748.15]: A computer application that integrates the technologies of computer graphics, computer vision, intelligent speech and natural language processing. It can be used for digital content generation and human-computer interaction to help improve content production efficiency and user experience. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following term: + +**3.2.1 3D digital human**: A digital human who has a three-dimensional image based on digital technology. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-----|---------------------| +| 3D | 3-Dimensional | +| FPS | Frames Per Second | +| SMD | Smart Mobile Device | + +## 5 Conventions + +This Recommendation uses the following conventions: + +- The phrase "is required to" indicates a requirement that must be strictly followed and from which no deviation is permitted if conformity to this Recommendation is to be claimed. +- The phrase "is recommended" indicates a requirement that is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformity. + +## 6 Overview + +The overall architecture of the 3-dimensional (3D) digital human system based on smart mobile devices consists of six major parts including image, speech, animation, interactive processing, multimodal input, and multimodal output, which is shown in Figure 1. The first four parts are closely related to digital human characters, the last two parts are the attributes that digital humans need to have for presentation or interaction with smart mobile devices (SMDs). + +![Figure 1 – Framework for 3D digital human system based on SMD. The diagram shows a flow from Image to Speech and Animation, which then lead to Multimodal output (SMD). The SMD interacts with the User. The User provides input to Multimodal input (SMD and peripherals), which then leads to Interactive processing. Interactive processing consists of Perception and recognition, which leads to Understanding and decision making. Understanding and decision making then leads back to Speech and Animation. The entire system is labeled '3D digital human system based on SMD'.](e1a0d046fbe7f28f5e93a47091851747_img.jpg) + +The diagram illustrates the framework for a 3D digital human system based on a Smart Mobile Device (SMD). It is divided into two main sections: the internal system components and the external user interaction. The internal components are grouped within a dashed box labeled "3D digital human system based on SMD". At the top is a box for "Image". Below it, "Speech" and "Animation" are shown in separate boxes. Arrows indicate a flow from "Image" down to "Speech" and "Animation". From "Speech" and "Animation", an arrow points right to a box labeled "Multimodal output (SMD)". Below the "Speech" and "Animation" boxes is a dashed box labeled "Interactive processing". Inside this box, "Perception and recognition" and "Understanding and decision making" are shown in separate boxes, with an arrow pointing from "Perception and recognition" to "Understanding and decision making". An arrow points up from "Understanding and decision making" to "Speech" and "Animation". To the right of the "Interactive processing" box is another box labeled "Multimodal input (SMD and peripherals)". An arrow points from this box to "Perception and recognition". Outside the dashed box, on the far right, is a vertical box labeled "User". Arrows point from the "Multimodal output (SMD)" box to the "User" box, and from the "User" box to the "Multimodal input (SMD and peripherals)" box. The text "F.748.31(24)" is located at the bottom right of the diagram. + +Figure 1 – Framework for 3D digital human system based on SMD. The diagram shows a flow from Image to Speech and Animation, which then lead to Multimodal output (SMD). The SMD interacts with the User. The User provides input to Multimodal input (SMD and peripherals), which then leads to Interactive processing. Interactive processing consists of Perception and recognition, which leads to Understanding and decision making. Understanding and decision making then leads back to Speech and Animation. The entire system is labeled '3D digital human system based on SMD'. + +**Figure 1 – Framework for 3D digital human system based on SMD** + +## **7 Functional requirements** + +### **7.1 Image** + +#### **(1) Character characteristic** + +The normative description on character characteristic about 3D digital human system is recommended to include the following information: + +- Identity setting; +- Appearance style such as cartoon, realistic or ultra-realistic. + +#### **(2) Quality** + +The quality of 3D digital human system is required to be free from the following defects: + +- Serious deformation; +- Serious interpenetration; +- Points that are not welded; +- Broken surface; +- Other damage. + +#### **(3) Fineness** + +Fineness is the detail degree of 3D digital human system, which is recommended to include the following information: + +- The number of model surfaces; +- The modelling of facial details, e.g., eyeballs, lenses, eyelashes, internal structure of oral cavity and hair; +- The wiring structure of the character face; +- The resolution of character mapping. + +#### **(4) Image comfort** + +Image comfort is the degree of 3D digital human system to make users physiologically comfortable and is always described by a subjective score on a Likert scale from 1 worst to 5 best as shown in Table 1 of [ITU-T F.748.15]. + +#### **(5) Expression richness** + +The higher the expression richness, the stronger the sense of reality for digital humans. + +The expression richness of 3D digital human system can be evaluated by subjective score on a Likert scale from 1 worst to 5 best according to the following three aspects: + +- Natural speech expression: refers to how much natural and smooth the mouth shape and pronunciation movement of a digital human, and whether the movements of opening and closing eyes are natural when the digital human speaks; +- Basic emotional expression: refers to whether the digital human can simulate the basic emotions of real people, such as happiness, anger, fear, sadness and surprise; +- Detail micro-expression: refers to whether a digital human can imitate the micro-expression of a real person to achieve a more delicate emotional expression. + +**Table 1 – Evaluation degrees for expression richness** + +| Evaluation degrees | 5 | 4 | 3 | 2 | 1 | +|----------------------------|----------------|---------------------|--------------------|--------------------|------------------| +| Natural speech expression | Quite natural | Relatively natural | Basically natural | Not quite natural | Quite unnatural | +| Basic emotional expression | Quite accurate | Relatively accurate | Basically accurate | Not quite accurate | Quite inaccurate | +| Detail micro-expression | Quite similar | Relatively similar | Basically similar | Not quite similar | Quite dissimilar | + +#### **(6) Compliance** + +Compliance refers to whether the digital human system complies with legal and other regulations. Any one of the following situations is considered non-compliant: + +- Use unauthorized images; +- Use modified images without authorization; +- Use images that contain bad leads; +- Use images with exposure of sensitive body parts; +- Use images that cause serious perception discomfort, such as blood, nausea, etc. + +### **7.2 Speech** + +#### **(1) Pronunciation accuracy** + +Pronunciation accuracy concerns the accuracy during text-to-speech, which could be evaluated by the pronunciation word correct rate and the pronunciation sentence correct rate, see clause 7.3.1 of [ITU-T F.748.15]. The pronunciation accuracy of a 3D digital human system is required to meet the needs of users and to be free from defects such as missing sounds, excessive pronunciation, phoneme errors and intonation errors. + +#### **(2) Rhythm accuracy** + +Rhythm accuracy concerns the accuracy during speech synthesis which includes sentence pause, pitch, length, volume, accent position and focus position, which correspond to natural speech rules such as focus pronunciation and exclamatory sentence intonation. The rhythm accuracy of a 3D digital human system is required to meet the needs of users and for Chinese rhythm, it could be evaluated by the gold-standard for word segment issued by the Association for Computational Linguistics (ACL). + +#### **(3) Speech comfort** + +Speech comfort is the degree of 3D digital human system to make users physiologically comfortable and is always described by subjective score on a Likert scale from 1 worst to 5 best as shown in Table 2 of [ITU-T F.748.15]. + +### **7.3 Animation** + +#### **(1) Movement fit** + +Movement fit relates to the degree of fit between the movement of digital humans and the current context. The fit reflection of 3D digital human system can be evaluated by movement types shown in Table 3 of [ITU-T F.748.15]. + +#### (2) Movement comfort + +Movement comfort is the degree of 3D digital human system to make users physiologically comfortable and is always described by subjective score on a Likert scale from 1 worst to 5 best as shown in Table 4 of [ITU-T F.748.15]. + +### 7.4 Interaction + +The interaction experience of 3D digital human system can be evaluated by subjective score on a Likert scale from 1 worst to 5 best according to the following three aspects: + +- Interaction accuracy: refers to whether the digital human system can accurately interact with users in different usage scenarios, such as accurately ask and answer questions, accurately recognize interruptions and accurately help the user complete the task according to the context; +- Style consistency: refers to whether the digital human system maintains a consistent style and conforms to the set character characteristic during the interaction process; +- Risk identification: refers to whether the digital human system can respond to sensitive words appropriately, and will not make expressions, actions and voices that do not comply with laws, humanistic ethics and other related regulations. + +**Table 2 – Evaluation degrees for interaction expression** + +| Evaluation degrees | 5 | 4 | 3 | 2 | 1 | +|----------------------|------------------|-----------------------|----------------------|----------------------|--------------------| +| Interaction accuracy | Quite accurate | Relatively accurate | Basically accurate | Not quite accurate | Quite inaccurate | +| Style consistency | Quite consistent | Relatively consistent | Basically consistent | Not quite consistent | Quite inconsistent | +| Risk identification | Yes | – | – | – | No | + +### 7.5 Multimodal input + +The multimodal input refers to the interaction input modes which 3D digital human system can support on mobile terminals, mainly including text, voice and touch, and can be evaluated by subjective score on a Likert scale from 0 worst to 2 best as shown in Table 3. + +**Table 3 – Evaluation degrees for multimodal input** + +| Evaluation dimension | Description | 2 | 1 | 0 | +|----------------------|-------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------|--------------------------|-----------------------------------------| +| Text | Support text input through the system keyboard | Support text input | – | Not support text input | +| Voice | Support voice input through local or external microphones | Support both local and external microphones input | Only support local input | Not support voice input | +| Touch | Operations with digital humans such as single-click, double-click, long-press, single-finger slide and two-finger slide | Support touch operations with human | – | Not support touch operations with human | + +### 7.6 Multimodal output + +#### (1) Real time factor + +Access the real time factor in the broadcast system, i.e., the ratio of video synthesis time to output video duration referred to [ITU-T F.748.14] and [ITU-T F.748.15]. Video resolution is inversely proportional to the time of synthesis. The lower the resolution is, the shorter the time of synthesis required. + +#### (2) Picture accuracy + +Access the accuracy of the picture when the system generates a video with a fixed number of frames refer to [ITU-T F.748.14] and [ITU-T F.748.15]. If errors such as frame skipping and jamming occur, the picture is regarded as inaccurate. + +#### (3) Audio and video matching degree + +Assess the matching degree of audio and video when the system generates video with a fixed duration (in seconds) refer to [ITU-T F.748.14] and [ITU-T F.748.15]. In the case of redundant or missing mouth shape, the dislocation such as audio advance and delay is regarded as an audio and video mismatch. + +#### (4) Multimodal output mode + +The output types or modes supported by 3D digital human system including but are not limited to SMDs such as smartphones, tablets and head-mounted displays. + +## 8 Performance requirements and evaluation methods + +### 8.1 Memory usage + +Requirement: This item refers to the memory usage of 3D digital human system when it is running on a smart mobile device and can be evaluated from 1 worst to 3 best according to Table 4 which is by memory usage at the same expressiveness. + +**Table 4 – Evaluation degrees for memory usage** + +| Evaluation dimension | 3 | 2 | 1 | +|----------------------|------------------|---------------------|-------------------| +| Memory usage | Low memory usage | Middle memory usage | High memory usage | + +Evaluation steps: + +- Exit all background applications of the smart mobile device, start the memory monitoring tool and select the application under test; +- Start the 3D digital human system and execute the test scene for a certain time; +- Record the memory usage. + +NOTE – For digital human composite applications, the memory usage of non-digital human should be deducted. + +### 8.2 Power consumption + +#### 8.2.1 Bottom current increment + +Requirement: After running the 3D digital human system and killing the process, the increase of the bottom current is recommended to be smaller than a specified value, for example 1mA, otherwise it is considered unqualified. + +Evaluation methods: + +- Connect the smart mobile device to an ammeter; +- Exit all background applications of the SMD, turn off the screen and wait for a few minutes; +- Use the software tool of the ammeter to start recording the current for a certain time to find the minimum value $a$ of the current; +- Start the 3D digital human system on the SMD and execute the test scene; +- Exit the 3D digital human system under test, turn off the screen and wait for the system to calm down completely; +- Using the software tool of the ammeter, record the current for a same certain time to find the minimum value $b$ of the current; +- Calculate the value of $b-a$ . + +NOTE – For digital human composite applications, the current occupation of non-digital human should be deducted. + +#### 8.2.2 Average current increment + +Requirement: After running the 3-dimensional (3D) digital human system and killing the process, the increase of the average current is recommended to be smaller than a specified value, otherwise it is considered unqualified. + +Evaluation steps: + +- Connect the smart mobile device to an ammeter; +- Exit all background applications of the SMD, turn off the screen and wait for a few minutes; +- Use the software tool of the ammeter to start recording the current for a certain time to find the average value $c$ of the current; +- Start the 3D digital human system on the SMD and execute the test scene; +- Exit the 3D digital human system under test, turn off the screen and wait for the system to calm down completely; +- Using the software tool of the ammeter, record the current for a same certain time to find the average value $d$ of the current; +- Calculate the value of $d-c$ . + +NOTE – For digital human composite applications, the current occupation of non-digital human should be deducted. + +#### 8.2.3 Working current + +Requirement: During the execution of the test scene of 3D digital human system, the average current of SMD at maximum volume and brightness is recommended to be smaller than a specified value, for example 900 mA, otherwise it is considered unqualified. + +Evaluation methods: + +- Connect the smart mobile device (SMD) to an ammeter; +- Exit all background applications on the SMD, turn off the screen and wait for the system to calm down completely; +- Adjust the volume and brightness of the SMD to the maximum value; +- Start the 3D digital human system on the SMD and execute the test scene; +- Using the software tool of the ammeter, record the current of the test scene, find the average value $e$ of the current. + +NOTE – For digital human composite applications, the current occupation of non-digital human should be deducted. + +### 8.3 Screen fluency + +Requirement: This item refers to the screen fluency of 3D digital human system when it is running on a smart mobile device and can be evaluated from 1 worst to 3 best according to Table 5 which is by frame-drop-ratio. + +**Table 5 – Evaluation degrees for screen fluency** + +| Evaluation degree | 3 | 2 | 1 | +|--------------------------|----------------------|----------------------|---------------------| +| Frame-drop-ratio | Little dropped frame | Middle dropped frame | Large dropped frame | + +Evaluation steps: + +- Exit all background applications on the SMD, start the fluency testing tool; +- Start the 3D digital human system (the frame number $f$ could be set as 60 frames per second (FPS) or 30 FPS), and execute the test scene for a certain time; +- Calculate the proportion of frame rates below $f$ within the test time. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.32-202410-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/F/T-REC-F.748.32-202410-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c1629394c02f697f60ae954ad36bf274fa8082cb --- /dev/null +++ b/marked/F/T-REC-F.748.32-202410-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3763acd7a0ed56aa30f3f90468c1fb0cb54573eb9045bca3eddf8976b4f0f41d +size 7219 diff --git a/marked/F/T-REC-F.748.32-202410-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg b/marked/F/T-REC-F.748.32-202410-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..46659a2e0fc7a4bf167254ef065fd3c430b9f660 --- /dev/null +++ b/marked/F/T-REC-F.748.32-202410-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e289912e3dc7db2cbe43c096f85d23d8eafc5abc6694bb734a32719d528571a5 +size 78468 diff --git a/marked/F/T-REC-F.748.33-202410-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg b/marked/F/T-REC-F.748.33-202410-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..25f8fbfaef317942ed3102c74b40a7d42f80608b --- /dev/null +++ b/marked/F/T-REC-F.748.33-202410-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:83cab7dcabfe939b223d97242d8523e461968d21f7c5432693c30a64f2d95521 +size 7050 diff --git a/marked/F/T-REC-F.748.33-202410-I_PDF-E/dbe553cf16dd14073b89a8263a428664_img.jpg b/marked/F/T-REC-F.748.33-202410-I_PDF-E/dbe553cf16dd14073b89a8263a428664_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f3766cfcdc85b8d597f0f00dc4edc46ea2f737f4 --- /dev/null +++ b/marked/F/T-REC-F.748.33-202410-I_PDF-E/dbe553cf16dd14073b89a8263a428664_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6f576cbf45f660e0ea0322196245a17c33914739512f56218c950685df78807c +size 85636 diff --git a/marked/F/T-REC-F.748.34-202406-I_PDF-E/raw.md b/marked/F/T-REC-F.748.34-202406-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..5098dfecfbc641883c81f1f6aa9d8c6c23b9bffe --- /dev/null +++ b/marked/F/T-REC-F.748.34-202406-I_PDF-E/raw.md @@ -0,0 +1,400 @@ + + +# Recommendation **ITU-T F.748.34 (06/2024)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +## **Requirements for the construction of multimedia knowledge graph database structure based on artificial intelligence** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with white lines representing latitude and longitude, and the letters 'ITU' in a bold, blue, sans-serif font superimposed on the globe. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.34 + +# Requirements for the construction of multimedia knowledge graph database structure based on artificial intelligence + +## Summary + +Recommendation ITU-T F.748.34 specifies the requirements for effectively constructing a knowledge graph database capable of processing multimedia data using artificial intelligence technology. It specifies the framework and requirements for the construction of the multimedia knowledge graph database. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.34 | 2024-06-13 | 16 | 11.1002/1000/15913 | + +## Keywords + +Artificial intelligence, entity relationship, knowledge extraction, knowledge graph, multimedia data. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2024 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-----------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Overall framework..... | 2 | +| 7 Functional requirements ..... | 3 | +| 7.1 Data collection..... | 3 | +| 7.2 Knowledge modelling ..... | 4 | +| 7.3 Knowledge extraction..... | 4 | +| 7.4 Knowledge fusion..... | 5 | +| 7.5 Knowledge evaluation ..... | 5 | +| 7.6 Knowledge storage ..... | 6 | +| 8 Non-functional requirements ..... | 6 | +| 8.1 Scalability requirements ..... | 6 | +| 8.2 Reliability requirements ..... | 6 | +| 8.3 Compatibility requirements ..... | 6 | +| 8.4 Performance requirements ..... | 7 | +| 8.5 Security requirements ..... | 7 | +| Bibliography..... | 8 | + + + +# Requirements for the construction of multimedia knowledge graph database structure based on artificial intelligence + +# 1 Scope + +This Recommendation specifies requirements and framework for the construction of multimedia knowledge graph database structure. + +The scope of this Recommendation includes: + +- Overview; +- Framework for the construction of multimedia knowledge graph database structure; +- Requirements for the construction of multimedia knowledge graph database structure. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +# 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 attribute** [b-IEEE 2807]: Describable features. + +**3.1.2 entity** [b-IEEE 2807]: An object existed independently in the real world. + +**3.1.3 event** [b-IEEE 2807]: Changes of things or states comprised by one or more actions with the participation of one or more roles, which occurs at a specific point in time or within a specific geographical area or time period. + +**3.1.4 knowledge** [b-IEEE 2807]: Facts, opinions, actions, or skills acquired through learning, practicing, or exploring. + +**3.1.5 knowledge fusion** [b-IEEE 2807]: An activity that integrates knowledge cell(s) to form a knowledge graph with global unified knowledge identifications. + +**3.1.6 knowledge graph** [b-IEEE 2807]: Assemblies of knowledge elements and their relations described in a structured form. + +**3.1.7 knowledge modelling** [b-IEEE 2807]: An activity that constructs ontology of a knowledge graph and its formal expressions. + +**3.1.8 knowledge storage** [b-IEEE 2807]: An activity that designs the storage architecture and conducts knowledge storing, querying, maintaining, and managing by using software and hardware infrastructures. + +**3.1.9 ontology** [b-ITU-T Y.4500.12]: Formal specification of a conceptualization, i.e., defining concepts as objects with their properties and relationships versus other concepts. + +**3.1.10 relation** [b-IEEE 2807]: Connection between entities, entity types, combination of entities, or combination of entity types. + +**3.1.11 schema** [b-IEEE 2807]: Formalized expression of ontology models. + +### **3.2 Terms defined in this Recommendation** + +None. + +## **4 Abbreviations and acronyms** + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-----|-----------------------------------| +| AI | Artificial Intelligence | +| API | Application Programming Interface | +| OCR | Optical Character Recognition | +| SQL | Structured Query Language | + +# **5 Conventions** + +In this Recommendation: + +- The keywords "**is required to**" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "**is recommended**" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement needs not be present to claim conformance. +- The keywords "**can optionally**" and "**may**" indicate an optional requirement which is permissible, without implying any sense of being recommended. These terms are not intended to imply that the vendor's implementation must provide the option, and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification + +# **6 Overall framework** + +The knowledge graph is a series of various diagrams that depict the process of knowledge development and its structural relationships. It employs visualization techniques to describe knowledge resources and their carriers, mining, analysing, constructing, drawing, and displaying knowledge and its interconnections. + +While the knowledge graph primarily focuses on displaying textual information, it can also present non-textual multimedia data, such as images, audio and video. This variety of data enriches the knowledge graph's expressiveness, enabling users to learn more comprehensively about entities and their relationships through diverse information forms. + +The construction of multimedia knowledge graph database goes through stages such as data acquisition, knowledge modelling, knowledge extraction, knowledge fusion, knowledge evaluation and knowledge storage. The overall framework is shown in Figure 1. + +![Figure 1: Overall framework for construction of a multimedia knowledge graph database structure. The diagram shows six sequential stages: Data collection, Knowledge modelling, Knowledge extraction, Knowledge fusion, Knowledge evaluation, and Knowledge storage. Each stage is in a box with sub-tasks listed below it.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +| | | | | | | +|---------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------------| +| Data collection
Data acquisition

Data import | Knowledge modelling
Defining schema
Automated modelling
Hybrid indices
Real-time scheduling | Knowledge extraction
Structured data
Semi-structured text data
Unstructured multimedia data | Knowledge fusion
Entity normalization
Multi-source fusion
Real-time update | Knowledge evaluation
Rule-based evaluation
Manual evaluation
Intelligent evaluation | Knowledge storage
Structural design
Access support
Management and maintenance | +|---------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------------| + +Figure 1: Overall framework for construction of a multimedia knowledge graph database structure. The diagram shows six sequential stages: Data collection, Knowledge modelling, Knowledge extraction, Knowledge fusion, Knowledge evaluation, and Knowledge storage. Each stage is in a box with sub-tasks listed below it. + +F.748.34(24) + +**Figure 1 – Overall framework for construction of a multimedia knowledge graph database structure** + +The functions of each part are as follows: + +- a) Data collection: Identify the sources of data needed to build the knowledge graph, prepare relevant data, import it, or crawl for it using crawlers and similar tools. +- b) Knowledge modelling: Define the entity categories and entity relationship categories in the knowledge graph. +- c) Knowledge extraction: Extract knowledge information from structured data, semi-structured data and unstructured multimedia data. +- d) Knowledge fusion: Integrate knowledge through the dynamic evolution of multi-source heterogeneous semantics. This process includes conflict detection, consistency checks and knowledge reasoning to determine the accuracy of knowledge, which is then organized into the knowledge base through alignment, association and entity linking, facilitating comprehensive knowledge sharing. +- e) Knowledge evaluation: Employ methods to guarantee the quality of knowledge when new knowledge is added to the knowledge graph. +- f) Knowledge storage: Store extracted knowledge in a database. + +# 7 Functional requirements + +### 7.1 Data collection + +During data collection, it is essential to clearly identify the sources of data for constructing the knowledge graph database. Appropriate acquisition strategies should be developed for different data sources, and data should be collected accordingly. This step sets the foundation for the subsequent construction process, which involves data acquisition and import. + +The functional requirements are as follows: + +- a) It is required to clearly determine the data sources necessary for constructing the knowledge graph, such as unstructured data from literature and books, structured data from databases, and non-text data like images, audio and video. +- b) It is required to use de-identification techniques for the organization's internal data before utilization. +- c) It is required to acquire data from external partners either through application programming interfaces (APIs) or by data file transfers. +- d) It is recommended to import the collected data into databases or storage systems for further analysis. + +### 7.2 Knowledge modelling + +During knowledge modelling, business knowledge is first transformed into a graph-based representation. This transformation completes the modelling from structured data to semantic, logic-based knowledge and builds the content of the knowledge graph. + +Knowledge modelling includes defining the schema, automating the modelling process, creating hybrid indices and implementing real-time scheduling. + +The functional requirements for knowledge modelling are as follows: + +- a) It is required to support the capability to abstract business domain knowledge into the entities, relationships, attributes and constraints of a graph through the graph's schema modelling ability. +- b) It is required to support operations such as the addition, deletion, modification and querying of schema. +- c) It is required to support automated knowledge modelling services. +- d) It is required to support the capability of index configuration for querying and recall abilities for text, vectors and spatial indices. +- e) It is recommended to support knowledge construction through both top-down and bottom-up approaches. +- f) It is recommended to support the referencing of existing relationships, entity definitions, etc., from other domains. +- g) It is recommended to support multi-person online collaborative editing with real-time updates. + +### 7.3 Knowledge extraction + +### 7.3.1 Structured data + +Knowledge extraction of structured data is mainly carried out from relational databases. + +It is recommended to map the data directly into the knowledge graph by database tables, fields, Structured Query Language (SQL) queries, and views. If the field corresponds to a foreign key, replace its value with the value corresponding to the foreign key. + +### 7.3.2 Semi-structured text data + +Knowledge extraction of semi-unstructured text data is mainly oriented to open linked data, and usually the typical input is natural language or multimedia documents. + +Knowledge extraction is generally divided into entity extraction, relation extraction and attribute extraction. Through automatic or semi-automatic technology, available entities, relations and attributes are extracted to form a series of high-quality expression of facts, which lays the foundation for the construction of the upper layer. + +The requirements for knowledge extraction for semi-structured text data are as follows: + +- a) It is recommended to support entity extraction, including Time class, Entity class, and Number class, such as person, place name, organization name, date, time, currency, percentage, etc. +- b) It is recommended to support entity disambiguation to solve the situation that entities with multiple meanings can express different meanings in the same document. +- c) It is recommended to support entity unification to determine whether multiple entities belong to the same entity. +- d) It is recommended to support reference resolution, so that different references representing the same entity can be divided into an equivalent set. + +- e) It is recommended to support relation extraction, which can extract the relation between entities from the correlative corpus. + +#### 7.3.3 Unstructured multimedia data + +Knowledge extraction of unstructured multimedia data refers to extracting knowledge from multimedia data such as images, audio and video. + +For image data, it is recommended to use optical character recognition (OCR) to recognize the text information in images and extract knowledge according to the above-mentioned knowledge extraction method for text. For some applications, deep learning algorithms can be used to automatically generate text descriptions of images, and then extract knowledge from the description text. + +For audio data, it is recommended to use audio-to-text technology to convert audio to text and extract knowledge from the text. + +For video data, it is recommended to extract knowledge after obtaining text information from video images and audio data by combining the above methods for image and audio data processing. + +### 7.4 Knowledge fusion + +The functional requirements for the knowledge fusion includes entity normalization, multi-source fusion and real-time update. + +The requirements are as follows: + +- a) It is recommended that each entity have unique instances in a homogeneous environment. +- b) It is recommended to support fusion of heterogeneous multi-source entities and homogeneous entities under different graph. +- c) It is recommended to support the fusion of multi-source knowledge subgraphs by enabling subgraph reduction, graph embedding, and subgraph similarity. +- d) It is recommended to support the hierarchical inheritance ability of relation to facilitate concept management. +- e) It is recommended to support multilingual knowledge fusion. +- f) It is recommended to support the fusion of new knowledge in real time. + +### 7.5 Knowledge evaluation + +The knowledge evaluation is used to ensure the quality of knowledge newly added to the knowledge graph. It mainly includes three methods: rule-based evaluation, manual evaluation and intelligent evaluation. + +Knowledge evaluation requires the following aspects: + +- a) It is recommended to utilize rule-based evaluation in conjunction with external interface calls to assess the correctness of knowledge, or to depend on system rules for evaluating aspects such as the timeliness and coverage of knowledge. +- b) It is recommended to ensure seamless integration between manual evaluations and the knowledge annotation platform. +- c) It is recommended to support unbiased sampling of entities, properties and relationships under certain conditions, facilitating the automatic integration of outsourcing and crowdsourcing evaluations to continuously monitor the quality of knowledge. +- d) It is recommended to incorporate intelligent evaluation with algorithmic models for assessing the correctness and structural stability of knowledge. + +### 7.6 Knowledge storage + +Knowledge storage is to store the concept layer and data layer of the knowledge graph in a format that can be recognized by the computer. Knowledge storage is a storage method designed for knowledge graph, which can store all kinds of knowledge to support the effective management and calculation of large-scale graph data. + +The storage of the knowledge graph does not have to rely on a specific underlying structure, and the general practice is to adopt different underlying storage according to the needs of data and applications. It mainly includes three aspects: structural design, access support, management and maintenance. + +The functional requirements for the knowledge storage components of the knowledge graph system are as follows: + +- a) It is recommended to support scalability to adapt to the growing demands of knowledge over time. +- b) It is recommended to support querying, reading, calculating, and applying requirements to facilitate interaction with various storage and computing components. +- c) It is recommended to support security, reliability, maintenance, and management by implementing role-based access control, encryption, multi-user capabilities, high availability, and backup and restore functions. + +# 8 Non-functional requirements + +### 8.1 Scalability requirements + +The scalability requirements for the AI-based multimedia knowledge graph database structure are as follows: + +- a) It is recommended to support expansion and contraction at any time according to business needs. +- b) It is recommended to support a trusted channel management scheme with an automatic scaling mechanism, thereby achieving the purpose of simple and efficient management of trusted computing channels. + +### 8.2 Reliability requirements + +The reliability requirements for the AI-based multimedia knowledge graph database structure are as follows: + +- a) It is recommended to ensure that data can only be used as agreed upon, to avoid misuse of data. +- b) It is recommended to support a multi-data centre coordination mechanism to solve the problem of cross-data centre disaster recovery. + +### 8.3 Compatibility requirements + +The compatibility requirements for the AI-based multimedia knowledge graph database structure are as follows: + +- a) It is recommended to ensure compatibility with mainstream graph query languages and graph analysis tools. +- b) It is recommended to operate smoothly on mainstream hardware in the industry. +- c) It is recommended to support batch data import and update from multiple data sources. +- d) It is recommended to support compatibility with mainstream commercial and open-source community edition operating systems. + +### **8.4 Performance requirements** + +The performance requirements for the AI-based multimedia knowledge graph database structure are as follows: + +- a) It is recommended to support solving problems like load balancing, failover and dynamic scaling in online services through a clustering solution. +- b) It can optionally support second-level queries for data scales of billions or more. + +### **8.5 Security requirements** + +The security requirements for the AI-based knowledge graph system are as follows: + +- a) It is recommended to support encryption transmission and storage of sensitive data in the knowledge graph (such as personal information, business data, etc.), supporting access control to data through encryption and decryption methods. +- b) It is recommended to support restricting the transmission of data in the knowledge graph among specific authorized entities. +- c) It is recommended to ensure the confidentiality, integrity and availability of knowledge graph data. +- d) It is recommended to ensure the confidentiality and integrity of input data and output results of the knowledge graph, ensuring they are not obtained by unauthorized users. + +## Bibliography + +- [b-ITU-T Y.4500.12] Recommendation ITU-T Y.4500.12 (2018), *oneM2M base ontology*. +- [b-IEEE 2807] IEEE Standard 2807-2022, *IEEE Standard for Framework of Knowledge Graphs*. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.35-202406-I_PDF-E/raw.md b/marked/F/T-REC-F.748.35-202406-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..9e3132913cb46801abf57eb81b6892d4197a15ca --- /dev/null +++ b/marked/F/T-REC-F.748.35-202406-I_PDF-E/raw.md @@ -0,0 +1,692 @@ + + +# Recommendation **ITU-T F.748.35 (06/2024)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +## --- **Requirement and framework of trustworthy federated machine learning based service** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.35 + +# Requirement and framework of trustworthy federated machine learning based service + +## Summary + +Federated machine learning (FML) is an emerging distributed machine learning paradigm that enables collaborative model training, learning, utilization and construction from a large number of distributed datasets on the basis of ensuring data security and legal compliance. In FML, computing takes place where the data are, and although the data are available, neither the data computing nor the data are visible. There are some challenges for FML-based services in aspects of trust as they perform in distributed or decentralized environments. All the challenges are often brought about by a lack of trust in the multiple participants of FML-based services, usually in the processes of model training and utilization, such as data indexing, data computing, parameter exchanging, etc. + +Specific functional components are needed to enhance the trustworthiness of FML-based services, such as enhancing dataset indexing, data computing, parameter exchange and model utilization. The distributed ledger technology (DLT) system is one type of trustworthy shared data system that can be used to also store FML-based service data. The FML-based service can take advantage of the convergence between FML and those components, especially to help address trust-related challenges for FML-based services. + +Recommendation ITU-T F.748.35 provides a trustworthy FML-based service, and specifies its concept, general characteristics and requirements, and reference framework. + +## History\* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.35 | 2024-06-13 | 16 | 11.1002/1000/15914 | + +## Keywords + +Distributed ledger technology (DLT), federated machine learning (FML), framework, trustworthiness. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2024 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|--------------------------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Overview of trustworthy FML-based services ..... | 2 | +| 7 General characteristics and requirements of trustworthy FML-based services..... | 4 | +| 7.1 General characteristics of trustworthy FML-based services ..... | 4 | +| 7.2 General requirements of trustworthy FML-based services ..... | 6 | +| 8 Reference framework of trustworthy FML-based services ..... | 7 | +| 8.1 Functional components of FML coordinators ..... | 9 | +| 8.2 Functional components of FML participants ..... | 9 | +| 8.3 External entities ..... | 10 | +| 8.4 Reference points ..... | 10 | +| Appendix I – Challenges and impact factors for trustworthiness of FML-based services ..... | 11 | +| I.1 Dataset provision for FML model training..... | 11 | +| I.2 Computing for FML model training..... | 11 | +| I.3 Communication for FML model training ..... | 12 | +| I.4 Model utilization of FML model ..... | 12 | +| Appendix II – Use cases of trustworthy FML-based services ..... | 13 | +| II.1 Trustworthy FML-based services in autonomous driving environment ..... | 13 | +| Bibliography..... | 15 | + + + +# Requirement and framework of trustworthy federated machine learning based service + +# 1 Scope + +This Recommendation provides the requirements and framework of trustworthy federated machine learning (FML) based services. + +The scope includes: + +- Concept, general characteristics and requirements of trustworthy FML-based services; +- Reference framework of trustworthy FML-based services. + +Use cases of trustworthy FML-based service are provided in an appendix. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 distributed ledger** [b-ITU-T X.1400]: A type of ledger that is shared, replicated, and synchronized in a distributed and decentralized manner. + +**3.1.2 distributed ledger technology (DLT)** [b-ITU-T X.1400]: Technology that enables the operation and use of distributed ledgers. + +**3.1.3 federated machine learning (FML)** [b-IEEE P3652.1]: A framework or system that enables multiple participants to collaboratively build and use machine learning models without disclosing the raw and private data owned by the participants while achieving good performance. + +**3.1.4 federated machine learning model (FMLM)** [b-IEEE P3652.1]: The result of the model-training process of a federated machine learning system. The learned model can be used in order to make certain machine-learning inference tasks on new data, e.g., classification, recognition, prediction, and recommendation. + +**3.1.5 ledger** [b-ITU-T X.1400]: Information store that keeps final and definitive (immutable) records of transactions. + +**3.1.6 trust** [b-ITU-T Y.3052]: The measurable belief and/or confidence which represents accumulated value from history and the expecting value for the future. + +NOTE – Trust is quantitatively and/or qualitatively calculated and measured. Trust is used to evaluate values of entities, value-chains among multiple stakeholders and human behaviours, including decision making. + +**3.1.7 trustworthy** [b-ISO/TR 15801]: Ability to demonstrate authenticity, integrity and availability of electronically stored information over time. + +**3.1.8 trusted system** [b-ISO/TR 15801]: Information technology system with the capability of managing electronically stored information in a trustworthy manner. + +## **3.2 Terms defined in this Recommendation** + +This Recommendation defines the following terms: + +**3.2.1 federated machine learning coordinator:** A party who composes and manages tasks for federated machine learning (FML) model training and utilization through coordination with FML participants. + +**3.2.2 federated machine learning participant:** A party who provides datasets and computing resources to participate the activities of a federated machine learning (FML)-based service, such as data pre-processing, model training, model utilization, etc. + +**3.2.3 federated machine learning model training dataset:** A dataset to be used to train federated machine learning (FML) models. + +**3.2.4 federated machine learning model training module:** An executable program to be used to train federated machine learning (FML) models with FML model training datasets. + +# **4 Abbreviations and acronyms** + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|--------|------------------------------------------| +| DLT | Distributed Ledger Technology | +| FML | Federated Machine Learning | +| FMLM | Federated Machine Learning Model | +| FMLS | Federated Machine Learning-based Service | +| ML | Machine Learning | +| RMCT | Robust Model Compression Training | +| T-FMLS | Trustworthy FML-based Service | +| TSS | Trustworthy Shared Storage | + +# **5 Conventions** + +This Recommendation uses the following conventions: + +- The keywords "is required to" indicate a requirement that must be strictly followed and from which no deviation is permitted if conformance with this Recommendation is to be claimed. +- The keywords "is recommended" indicate a requirement that is recommended, but which is not absolutely required to claim conformance with this Recommendation. + +# **6 Overview of trustworthy FML-based services** + +Federated machine learning (FML) is an emerging distributed machine learning (ML) paradigm that enables collaborative model training, learning, utilization and construction from a large number of distributed datasets on the basis of ensuring data security and legal compliance. In FML, computing takes place where the data are, and although the data are available, neither the data computing nor the data are visible. + +In traditional FML-based services (see Figure 6-1), FML coordinators and FML participants exchange information directly. FML coordinators coordinate and manage dataset indexes, model training and utilization progress, and store and manage full data of the entire lifecycle for dataset indexes, model training and utilization. FML coordinators usually act as trust points for traditional FML-based services. There are some challenges for traditional FML-based service for it works in distributed or decentralized environments, trusted or untrusted. The main challenges are listed in Appendix I. All those challenges are often brought about by a lack of trust in the multiple distributed or decentralized participants of the traditional FML-based service. Appendix I also provides more information about impact factors for trustworthiness of FML-based services. + +![Figure 6-1: Overview of traditional FML-based service. The diagram shows two main entities: 'FML participants' and 'FML coordinators'. 'FML participants' are shown in a nested box with three sub-components: 'Provision of raw data locally and dataset indexes', 'Participation for model training and utilization locally', and 'Storage and management for part of data of the entire lifecycle for dataset indexes, model training and utilization'. 'FML coordinators' are shown in a nested box with two sub-components: 'Coordination and management for dataset indexes, model training and utilization' and 'Storage and management for full data of the entire lifecycle for dataset indexes, model training and utilization'. A double-headed arrow labeled 'Information exchange for FML' connects the two entities, with a list of items: '- Dataset indexes', '- Training modules', '- Training tasks', '- Middle parameters', '- Trained models', '- Utilization', and '- ...'. Below the 'FML coordinators' box is the text 'Trust point of FML-based service'. In the bottom right corner of the diagram area is the code 'F.748.35(24)'.](acfc53eca625d62b38aa2563efa95c3e_img.jpg) + +Figure 6-1: Overview of traditional FML-based service. The diagram shows two main entities: 'FML participants' and 'FML coordinators'. 'FML participants' are shown in a nested box with three sub-components: 'Provision of raw data locally and dataset indexes', 'Participation for model training and utilization locally', and 'Storage and management for part of data of the entire lifecycle for dataset indexes, model training and utilization'. 'FML coordinators' are shown in a nested box with two sub-components: 'Coordination and management for dataset indexes, model training and utilization' and 'Storage and management for full data of the entire lifecycle for dataset indexes, model training and utilization'. A double-headed arrow labeled 'Information exchange for FML' connects the two entities, with a list of items: '- Dataset indexes', '- Training modules', '- Training tasks', '- Middle parameters', '- Trained models', '- Utilization', and '- ...'. Below the 'FML coordinators' box is the text 'Trust point of FML-based service'. In the bottom right corner of the diagram area is the code 'F.748.35(24)'. + +**Figure 6-1 – Overview of traditional FML-based service** + +Distributed ledger technologies (DLT) have inherent advantages to leverage data management and sharing, such as peer to peer communication, decentralization, immutability, openness, crowd consensus and smart contract support and automatization. The FML-based service can benefit from the convergence between FML and DLT, especially in addressing the challenges mentioned above. + +Figure 6-2 gives an overview of a trustworthy FML-based service (T-FMLS). In this paradigm, FML coordinators and FML participants exchange information through a trustworthy shared storage (TSS), not directly with each other. The TSS is used to store and manage the information of the entire lifecycle for dataset indexes, model training and utilization for the T-FMLS. + +NOTE – T-FMLS is independent from TSS. TSS may be based on a DLT system. TSS and DLT are out of the scope of this Recommendation. + +In T-FMLS, FML participants store their training datasets (sets of raw data) locally and provide indexes of the training datasets in the TSS. The TSS keeps the data indexes to prevent tampering. An index of raw data includes descriptions of the raw data, information on accessing the raw data, and a digest of the raw data which is used to validate the relationship between the index and the raw data. The digest in an index can be encrypted by the FML participants. + +In T-FMLS, FML coordinators store FML model training modules in the TSS. The TSS keeps the FML model training modules to prevent tampering. FML model training modules are open to relevant FML participants. FML model training modules can be programmed as smart contracts of the TSS. In addition, FML participants can utilize the robust model compression training (RMCT) technique to lighten the model while ensuring the robustness of the compressed model compared to traditional model compression techniques. RMCT technique can be used to reduce the size of models in FML + +with limited communication and storage resources, and includes adversarial training, differential privacy, quantization-aware training, knowledge distillation, etc. + +![Figure 6-2 – Overview of trustworthy FML-based service. The diagram illustrates the interaction between FML participants, FML coordinators, and a Trustworthy shared storage (e.g., DLT system).](f4fdd410cdb84df81274da55721e56fb_img.jpg) + +The diagram shows three main components: **FML participants**, **FML coordinators**, and **Trustworthy shared storage (e.g., DLT system)**. + +- FML participants** (top left, dashed box) contain: + - Training dataset provision + - Secure computing sandbox (containing Training modules) + - Trained model management + - Trained model utilization + - Training data exchanging + - Contribution evaluation +- FML coordinators** (top right, dashed box) contain: + - FML model training + - FML model management + - FML model utilization + - Contribution evaluation +- Trustworthy shared storage (e.g., DLT system)** (bottom, dashed box) contains: + - Data indexes + - Training modules + - FML models + - Activity logs ... + +**Information exchanging for FML** is shown between the components: + +- Between **FML participants** and **Trustworthy shared storage**: + - Upward arrow: Dataset indexes, Training modules, Training tasks, Middle parameters, Trained models, Utilization ... + - Downward arrow: Dataset indexes, Training modules, Training tasks, Middle parameters, Trained models, Utilization ... +- Between **FML coordinators** and **Trustworthy shared storage**: + - Leftward arrow: Dataset indexes, Training modules, Training tasks, Middle parameters, Trained models, Utilization ... +- A dashed arrow labeled **Traditional information exchanging for FML** points from **FML participants** to **FML coordinators**. + +F.748.35(24) + +Figure 6-2 – Overview of trustworthy FML-based service. The diagram illustrates the interaction between FML participants, FML coordinators, and a Trustworthy shared storage (e.g., DLT system). + +**Figure 6-2 – Overview of trustworthy FML-based service** + +In T-FMLS, by using data indexes and training modules stored in TSS, FML coordinators can compose and manage FML model training tasks. In an FML model training task, data indexes and training modules are selected by the FML coordinator. During an FML model training task, the training modules are performed in a secure computing sandbox of each FML participant, and the training modules can validate the relationship of the indexes provided by FML coordinators with the raw data provided by FML participants. In the meantime, training parameters are exchanged between FML coordinators and FML participants through TSS, and the model training results and relevant activity logs can be stored in TSS as well. TSS provides trust capability to store data indexes, training modules, training result and relevant activity logs. Therefore, this paradigm of trustworthy FML-based services can enhance their trustworthiness and data security. + +# 7 General characteristics and requirements of trustworthy FML-based services + +## 7.1 General characteristics of trustworthy FML-based services + +### 7.1.1 Trustworthy full-lifecycle management for FML model training tasks + +T-FMLS provides trustworthy full-lifecycle management of FML model training tasks, including at least: + +- Creating and managing of FML model training tasks; +- Managing of FML model training progresses; +- Managing and utilizing relevant trained FML models; +- Tracing and tracking FML model training and result management. + +Relevant information of FML model training is stored in TSS, which prevents information tampering. + +### **7.1.2 Trustworthy module provision for FML model training** + +T-FMLS improves the following distinct characteristic of FML-based services: "computing is available but it is not visible". + +FML model training modules are usually transparent and stored in TSS. T-FMLS participants can check and validate the FML model training modules independently. They can also trace and track the utilization of the FML model training modules. + +Typically, each T-FMLS participant provides secure computing sandbox and performs the FML model training modules in the secure computing sandbox to guarantee security of computing environment of each FML participant. + +### **7.1.3 Trustworthy data provision and computing for FML model training** + +T-FMLS improves another distinct characteristic of the FML-based service, "data are available but not visible". Where the data are is where the computing takes place, and data are not moved out of their secure domain. + +In T-FMLS, FML participants keep raw data in their secure domains and share data indexes through TSS. The FML coordinators select and use the data indexes and FML model training module(s) to compose FML model training tasks. + +When FML participants perform an FML model training task, they provide raw data that match relevant data indexes. The FML model training module of the FML model training task can validate the relationship between the raw data and the relevant data indexes. Furthermore, more it may measure quality and quantity of the dataset to be used to train the FML model. All the information of activities of data provisioning and computing can be stored in TSS. FML participants and coordinators can then trace and track the whole progresses of the FML model training task and evaluate their relevant contributions to the FML model training and utilization. + +### **7.1.4 Trustworthy parameters exchange for FML model training** + +T-FMLS guarantees that no sensible information and no row data are leaked when exchanging training parameter for FML model training. The exchanged information is stored in TSS and can be traced and tracked. + +#### **7.1.5 Trustworthy model management for trained FML model** + +T-FMLS improves the following distinct characteristic of the FML-based service: "the trained model is available, but it is not visible". The trained FML models can be distributed to be stored by the FML participants and coordinators. T-FMLS guarantees trustworthy storage and management of the trained FML models and guarantee that the participants can follow the utilization of their trained FML models. + +#### **7.1.6 Compliance with national and regional data regulations** + +T-FMLS is compliant with national and regional data regulations when FML coordinators deploy FML model training tasks to FML participants and when FML coordinators and participants perform trained FML models. + +#### **7.1.7 Robustness and scalability for FML model training** + +T-FMLS guarantees the robustness and scalability of FML model training, including reducing negative impacts of failures from a single or several of the FML participants. + +T-FMLS guarantees RMCT, which ensures that FML models trained on multiple FML participants are robust against adversarial attacks. In T-FMLS, multiple FML participants can jointly train FML models while keeping their data decentralized and private. Adversarial attacks are a type of security threat where an attacker manipulates the input data to deceive the model into producing incorrect outputs. FML model training tasks are working in distributed or decentralized environments which + +may become vulnerable to adversarial attacks. RMCT ensures model robustness by allowing the compressed model to perform normal inference under various conditions, such as when dealing with noisy or perturbed data. + +## **7.2 General requirements of trustworthy FML-based services** + +### **7.2.1 Requirements for task management of FML model training** + +- It is required to provide TSS to allow FML participants and FML coordinators to save and retrieve information (such as data indexes, parameters, modules and activity logs), respectively and independently. +- It is required that each FML model training task of T-FMLS has a global unique identity which is used in its full lifecycle. +- It is required to allow activity logs of management of FML model training tasks in full lifecycle to be stored in the TSS, including at least: + - That of creating and managing of FML model training tasks; + - That of managing of FML model training progresses; + - That of managing and utilizing relevant trained FML models; + - That of tracing and tracking FML model training and result management. +- It is required that information of FML model training tasks is stored in the TSS. +- It is required to be compliant with national and regional data regulations when FML coordinators deploy FML model training tasks to FML participants. + +### **7.2.2 Requirements for module provision for FML model training** + +- It is required to allow FML model training modules to be stored in the TSS. +- It is required to provide a mechanism to allow FML participants to check and validate FML model training modules independently when they are used to train FML models, according to FML model training tasks. +- It is required to provide mechanism to allow each FML participant to provide a secure computing sandbox and to perform FML model training modules in the secure computing sandbox to guarantee security of the computing environment. +- It is required to provide a mechanism to allow FML model training modules being performed to store activity logs to the TSS, according to relevant FML model training tasks and the rules of FML participants. +- It is recommended to allow FML participants and FML coordinators to trace and track the utilization of the FML model training modules independently. + +### **7.2.3 Requirements for data provision and computing for FML model training** + +- It is required to provide a mechanism to allow FML participants to share data indexes that match corresponding raw data and allow FML coordinators to create FML model training tasks according to the shared data indexes. +- It is required to provide a mechanism to allow FML participants to avoid raw data to be leaked out of its original domain when performing FML model training. +- It is required to provide a mechanism to allow FML model training modules being performed to check and validate raw data provided by FML participants, according to relevant FML model training tasks. +- It is recommended to allow FML participants and FML coordinators to trace and track the utilization of the data indexes independently. + +- It is recommended to provide mechanism to allow FML participants and FML coordinators to measure the quality and quantity of the datasets to be used to train FML models independently. + +### **7.2.4 Requirements for parameter exchanging for FML model training** + +- It is required that no sensible information and no row data are leaked when exchanging training parameters for FML model training. +- It is recommended to provide a mechanism to allow FML participants and FML coordinators to trace and track the exchanged training parameters independently, if allowed. +- It is required to be compliant with national and regional data regulations when parameters are being exchanged for FML model training. + +### **7.2.5 Requirements for model management for trained FML models** + +- It is recommended to allow trained FML models to be stored in distributed storage by the FML participants and FML coordinators, according to policies of FML model training tasks. +- It is required to guarantee trustworthy storage and management for trained FML models, and to allow FML participants and FML coordinators to follow the utilization of their trained FML models. +- It is required to be compliant with national and regional data regulations when FML coordinators and participants perform trained FML models. + +### **7.2.6 Requirements for robustness and scalability for FML model training** + +- It is required to guarantee robust and scalable FML model training, including reducing negative impacts of failures from one or several participants. +- It is required to allow FML participants to perform RMCT, including: + - Adversarial training to ensure inference integrity by utilizing data perturbation; + - Differential privacy to prevent models from memorizing training data; + - Quantization-aware training to be robust against model inversion attacks; + - Distillation learning to protect local privacy. + +### **7.2.7 Requirements for data security protection** + +- It is required to provide a mechanism to protect data security, such as raw data, data indexes, FML model training tasks, trained FML models, etc. + +# **8 Reference framework of trustworthy FML-based services** + +This clause provides a reference framework of the trustworthy FML-based service (T-FMLS) according to the requirements listed in clause 7. + +![Diagram of reference framework of trustworthy FML-based services. The diagram shows three main components: FML participants, FML coordinators, and Trustworthy shared storage (TSS). FML participants are grouped into a 'Secure computing sandbox' containing four functional component (FC) blocks: dataset provision, data computing, data exchange, and management. Below the sandbox are 'Agents for storage systems' and 'Contribution evaluation'. FML coordinators include FCs for managing training tasks, tracking progress, managing performance, contribution evaluation, and robust model compression training. TSS (e.g., DLT system) contains Data indexes, Training modules, FML models, and Activity logs. Information exchange R1 is between TSS and Coordinators. Information exchange R2 is between TSS and Participants. Information exchange R3 is between Participants and Coordinators.](4801720824e4b5e2361a5564f91cfb70_img.jpg) + +The diagram illustrates the reference framework of trustworthy FML-based services, showing the interaction between FML participants, FML coordinators, and a Trustworthy shared storage (TSS) system. + +**FML participants** are organized into a **Secure computing sandbox** containing four functional component (FC) blocks: + +- FCs for training dataset provision (Enhanced dataset indexing and storing, Raw data provision, and tracing, Adversarial dataset provision and tracing, Traditional FCs for data provision ...) +- FCs for training data computing (Validating, tracking and tracing dataset, Validating, tracking and tracing modules, Tracing model compression training, Traditional FCs for data calculating ...) +- FCs for training data exchange (Validating, tracking and tracing output parameters, Traditional FC for transferring output parameters ...) +- FCs for management and utilization of trained model (Validating, tracking and tracing management and utilization of trained model, Traditional FCs for management and utilization of trained model ...) + +Below the sandbox are **Agents for storage systems** and **Contribution evaluation**. + +**FML coordinators** include: + +- FCs for managing and performing of FML training tasks and trained FML models +- Tracking and tracing of model training tasks and progress +- Tracking and tracing of utilization of trained models +- Probing and managing performance of training and utilization +- Contribution evaluation +- Probing and managing robust model compression training +- Traditional management and performance of FML training tasks and trained FML models + +**Trustworthy shared storage (e.g., DLT system)** contains: + +- Data indexes +- Training modules +- FML models +- Activity logs ... + +Information exchange for FML is indicated by R1, R2, and R3: + +- R1:** Information exchange for FML between TSS and FML coordinators. +- R2:** Information exchange for FML between TSS and FML participants. The data exchanged includes: + - Indexes of training datasets + - Modules for training FML models + - Parameters of model training + - Trained FML models + - Activity logs for FML model training and utilization +- R3:** Traditional information exchange for FML between FML participants and FML coordinators. + +F.748.35(24) + +Diagram of reference framework of trustworthy FML-based services. The diagram shows three main components: FML participants, FML coordinators, and Trustworthy shared storage (TSS). FML participants are grouped into a 'Secure computing sandbox' containing four functional component (FC) blocks: dataset provision, data computing, data exchange, and management. Below the sandbox are 'Agents for storage systems' and 'Contribution evaluation'. FML coordinators include FCs for managing training tasks, tracking progress, managing performance, contribution evaluation, and robust model compression training. TSS (e.g., DLT system) contains Data indexes, Training modules, FML models, and Activity logs. Information exchange R1 is between TSS and Coordinators. Information exchange R2 is between TSS and Participants. Information exchange R3 is between Participants and Coordinators. + +**Figure 8-1 – Diagram of reference framework of trustworthy FML-based services** + +The T-FMLS mainly consists of three parts, multiple FML coordinators and FML participants, and trustworthy shared storage (TSS). Each of them includes groups of logical functional components (FCs). + +Each FML coordinator of the T-FMLS includes groups of FCs for FML training tasks, trained FML models and contribution evaluation, including as least: + +- FCs for tracking and tracing model training tasks and progress; +- FCs for tracking and tracing utilization of trained models; +- FCs for probing and managing performance of training and utilization of FML models; +- FCs for contribution evaluation; +- FCs for probing and managing robust model compression training; +- FCs for traditional task composition and management for FML model training and utilization; +- Agents for storage systems. + +Each FML participant of the T-FMLS includes groups of FCs for FML training tasks, trained FML models and contribution evaluation, including as least: + +- FCs for training dataset provision; +- FCs for training data computing, in secure computing sandboxes; +- FCs for training parameter exchange; +- FCs for management and utilization of trained models; +- FCs of contribution evaluation; +- Agents for storage systems. + +The TSS is an entity external to the T-FMLS, with which the FML participants and coordinators store and exchange information for FML model training and trained FML models, such as data indexes, training modules, FML models and activity logs. + +An FML coordinator exposes two reference points, R1 and R3. R1 is for information exchange for FML model training and trained FML model utilization by using the TSS. R3 is a traditional reference point through which the traditional FML coordinators and FML participants exchange information for FML model training and trained FML model utilization directly. + +An FML participant exposes reference point R2. R2 is for information exchange for FML model training and trained FML model utilization by the TSS. + +NOTE – The TSS may be based on DLT systems. The TSS and reference point R3 are out of the scope of this Recommendation. + +## **8.1 Functional components of FML coordinators** + +### **8.1.1 FCs for tracking and tracing model training tasks and progress** + +An FML coordinator of a given T-FMLS supports tracking and tracing FML model training tasks and progress. + +### **8.1.2 FCs for tracking and tracing utilization of trained models** + +An FML coordinator of a given T-FMLS supports tracking and tracing the utilization of trained FML models. + +### **8.1.3 FCs for probing and managing performance of training and utilization** + +An FML coordinator of a given T-FMLS supports probing and managing the performance of FML model training and utilization. + +### **8.1.4 FCs for contribution evaluation** + +An FML coordinator of a given T-FMLS supports evaluation of its contribution to each task of the FML model training and utilization. + +### **8.1.5 FCs for probing and managing robust model compression training** + +An FML coordinator of a given T-FMLS supports probing and managing robust model compression training. + +#### **8.1.6 FCs for traditional task composition and management for FML model training and utilization** + +An FML coordinator of a given T-FMLS supports traditional task composition and management for FML model training and utilization. + +### **8.1.7 Agents for storage systems** + +Agents for storage systems of an FML coordinator of a given T-FMLS is to connect to trustworthy shared storage to support exchanging information for FML model training and utilization. + +## **8.2 Functional components of FML participants** + +### **8.2.1 FCs for training dataset provision** + +An FML participant of a given T-FMLS supports training dataset provision related functional components of traditional FML-based service. Besides, it supports enhanced functionalities for dataset indexing and storing, raw data providing and tracing, adversarial dataset provision and tracing. + +### **8.2.2 FCs for training data computing** + +An FML participant of a given T-FMLS supports training data computing related functional components of traditional FML-based service. Besides, it supports enhanced functionalities for validating, tracking and tracing dataset, and validating, tracking and tracing training modules, and tracing model compression training. Data computing is performed in secure computing sandboxes of FML participants. + +### **8.2.3 FCs for training data exchange** + +An FML participant of a given T-FMLS supports exchanging training parameters related functional components of traditional FML-based service. Besides, it supports enhanced functionalities for validating, tracking and tracing the parameters to be exchanged for FML model training. + +### **8.2.4 FCs for management and utilization of trained machine learning (ML) models** + +An FML participant of a given T-FMLS supports management and utilization of trained ML models of traditional FML-based service. Besides, it supports enhanced functionalities for validating, tracking and tracing management and utilization of trained models. + +### **8.2.5 FCs for contribution evaluation** + +An FML participant of a given T-FMLS supports evaluation of its contribution to each task of the FML model training and utilization. + +#### **8.2.6 Agents for storage systems** + +Agents for storage systems of an FML participant of a given T-FMLS is to connect to trustworthy shared storage to support exchanging information for FML model training and utilization. + +## **8.3 External entities** + +### **8.3.1 Trustworthy shared storage** + +The trustworthy shared storage (e.g., a DLT system) is an entity external to a T-FMLS, with which the FML participants and coordinators store and exchange information for FML model training and trained FML models. + +## **8.4 Reference points** + +An FML coordinator exposes reference point R1 for information exchange for FML model training and trained ML model utilization by using trustworthy shared storage. + +An FML participant exposes reference point R2 for information exchange for FML model training and trained ML model utilization by using trustworthy shared storage. + +An FML coordinator exposes traditional reference point R3 with which the traditional FML coordinator and FML participants to exchange information for FML model training and trained FML model utilization directly. + +# Appendix I + +## Challenges and impact factors for trustworthiness of FML-based services + +(This appendix does not form integrated part of this Recommendation.) + +There are some challenges for the traditional FML-based service as it works in distributed or decentralized environments, trusted or untrusted, mainly as given below: + +- In the model training phase of the traditional FML-based service: + - Computing: Because computing environments are diverse and managed by different participants, the challenges include to guarantee that computing environments are trusted, to guarantee that the training modules performed by each participant are consistent and secure, etc. + - Dataset provision: Because the dataset and the relevant data computing are available but are not visible, the challenges include how to have other participants trust the provided data. + - Data exchange: Because training parameters should be exchanged for model training, the challenges include how to guarantee that the raw data are not leaked during model training, etc. + - Model management: Relevant challenges include where to store the trained model, whether using one integrated model or using distributed model slices stored by different participants. How to guarantee that a trained model can be used if it is stored in distributed storage, etc. + - Robust of training: Relevant challenges include how to protect model training from interruption and disturbance by a part of the participants, etc. +- In the model utilization phase of traditional the FML-based service: + - Model utilization: Relevant challenges include how to identify a trained model to be used, and how the participants can obtain continuous benefits from the trained models if they had contributed to the training. + +According to the challenges for traditional FML-based services mentioned above, there are some important impact factors to the trustworthiness of FML-based services as given below. + +### I.1 Dataset provision for FML model training + +FML-based service has a distinct characteristic that is "data are available but they are not visible". That means that "computing takes place where the data are". Because the data used for FML model training are not visible, there are some impact factors for each participant to let other participants and the coordinator accept the dataset, including: + +- The quality and quantity of the dataset provided for model training are valuable; +- The actual calculated dataset for model training matches the pre-declared dataset, etc. + +### I.2 Computing for FML model training + +FML-based service has another distinct characteristic that is "computing is available, but it is not visible". Because the computing in FML model training progress is not visible, there are some impact factors as shown below (but not limited): + +- Whether the training modules performed by each FML participant are consistent; +- Whether the training modules performed are secure for executing environment; +- Whether the training modules performed are secure for the raw data; + +- Whether the training modules can validate and guarantee the dataset provided by each FML participant to match to pre-declared dataset; +- Whether the computing capability of each FML participant is valuable, etc. + +### **I.3 Communication for FML model training** + +It is necessary to exchange parameters when training FML models in FML-based services. Therefore, there are some risks related to the leaking of sensitive data when parameters for FML model training are exchanged. There is at least the following relevant impact factor: + +- How to guarantee that the sensible information of raw data of FML participants is not leaked when model training parameters are exchanged, etc. + +### **I.4 Model utilization of FML model** + +There is another distinct characteristic of FML-based service, "the trained model of FML is available, but it is not visible". Each participant contributes part of raw data provision and computing, and then usually only stores part of the trained FML model. Therefore, at least the following relevant impact factors exist: + +- Who and how is the trained model to be stored; +- Who can use the trained model, and how to can all the FML participants benefit according to their real contribution, etc. + +Because of the important impact factors for trustworthiness of FML-based services mentioned above, it is needed to provide technical mechanisms to improve trustworthiness among FML participants and coordinators when they collaborate to provide FML-based services. + +# Appendix II + +## Use cases of trustworthy FML-based services + +(This appendix does not form integrated part of this Recommendation.) + +### II.1 Trustworthy FML-based services in an autonomous driving environment + +The FML-based autonomous driving service consists of three entities: an autonomous driving terminal (FML participant), a trustworthy shared data storage (DB) system, and an autonomous driving system (FML coordinator). + +The autonomous driving terminal (FML participant) provides federated machine learning and inference functions in autonomous vehicle. The autonomous driving terminal may be a personal or a shared autonomous driving terminal. The autonomous driving terminal enables machine learning and inference function for user preference service. The autonomous driving terminal transfers the learning models to an autonomous driving system through communication networks. + +The communication network can utilize the existing edge network environment. For example, it supports wireless local area network (LAN), etc. The communication network provides a function of transmitting a result of processing an autonomous driving service between the autonomous driving terminal (FML participant) and the autonomous driving system (FML coordinator). + +The autonomous driving system (FML coordinator) may consist of edge communication network and shared data storage system, handles learning models from the autonomous terminal and learns from information of the surrounding environment. The system contains three functions, namely, autonomous service management function, user information management function and service information management function. The autonomous service management function supports an autonomous driving service from the surrounding environment of the user terminal. The user information management function is responsible for user registration and storage service related to autonomous driving information. The service information management function supports artificial intelligence learning on autonomous driving terminals with surrounding environment information to manage machine-based learning models. + +![A hierarchical diagram showing the architecture of trustworthy FML based service in an autonomous vehicle environment. At the top is a central 'FML coordinator' (Autonomous service management) connected to a 'Local shared DB system' (User and service information). Below this, two identical branches are shown. Each branch contains an 'FML coordinator' (Autonomous service management) connected to its own 'Local shared DB system' (User and service information). These coordinators are connected via a 'Network' cloud to four 'FML participant' (Autonomous driving terminal) units at the bottom. The participants are represented by icons of autonomous shuttles.](b6671cfafda3820aafe9a24fa7a4d8c7_img.jpg) + +``` + +graph TD + TopFML[FML coordinator +Autonomous service management] + TopDB[(Local shared DB system +User and service information)] + + TopFML --- TopDB + + Subgraph LeftBranch + L1FML[FML coordinator +Autonomous service management] + L1DB[(Local shared DB system +User and service information)] + L1Network((Network)) + L1P1[FML participant +Autonomous driving terminal] + L1P2[FML participant +Autonomous driving terminal] + + L1FML --- L1DB + L1FML --- L1Network + L1Network --- L1P1 + L1Network --- L1P2 + end + + Subgraph RightBranch + R1FML[FML coordinator +Autonomous service management] + R1DB[(Local shared DB system +User and service information)] + R1Network((Network)) + R1P1[FML participant +Autonomous driving terminal] + R1P2[FML participant +Autonomous driving terminal] + + R1FML --- R1DB + R1FML --- R1Network + R1Network --- R1P1 + R1Network --- R1P2 + end + + TopFML --- L1FML + TopFML --- R1FML + +``` + +A hierarchical diagram showing the architecture of trustworthy FML based service in an autonomous vehicle environment. At the top is a central 'FML coordinator' (Autonomous service management) connected to a 'Local shared DB system' (User and service information). Below this, two identical branches are shown. Each branch contains an 'FML coordinator' (Autonomous service management) connected to its own 'Local shared DB system' (User and service information). These coordinators are connected via a 'Network' cloud to four 'FML participant' (Autonomous driving terminal) units at the bottom. The participants are represented by icons of autonomous shuttles. + +F.748.35(24) + +**Figure II.1 – Trustworthy FML based service in autonomous vehicle environment** + +# Bibliography + +- [b-ITU-T X.1400] Recommendation ITU-T X.1400 (2020), *Terms and definitions for distributed ledger technology*. +- [b-ITU-T Y.3052] Recommendation ITU-T Y.3052 (2017), *Overview of trust provisioning in information and communication technology infrastructures and services*. +- [b-IEEE P3652.1] IEEE Standard P3652.1 (2020), *IEEE Guide for Architectural Framework and Application of Federated Machine Learning*. +- [b-ISO/TR 15801] ISO/TR 15801:2017, *Document management – Electronically stored information – Recommendations for trustworthiness and reliability*. + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.40-202410-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.748.40-202410-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c1629394c02f697f60ae954ad36bf274fa8082cb --- /dev/null +++ b/marked/F/T-REC-F.748.40-202410-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3763acd7a0ed56aa30f3f90468c1fb0cb54573eb9045bca3eddf8976b4f0f41d +size 7219 diff --git a/marked/F/T-REC-F.748.40-202410-I_PDF-E/562f471e8153729557e6a4ee6343c32c_img.jpg b/marked/F/T-REC-F.748.40-202410-I_PDF-E/562f471e8153729557e6a4ee6343c32c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ac40ac527bbc8c42ecb99c488b568f44770565d9 --- /dev/null +++ b/marked/F/T-REC-F.748.40-202410-I_PDF-E/562f471e8153729557e6a4ee6343c32c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f12007cf754cad4529a2503ef3f1e077a8265089618ee517bd17289e03a7f1bf +size 58376 diff --git a/marked/F/T-REC-F.748.40-202410-I_PDF-E/raw.md b/marked/F/T-REC-F.748.40-202410-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..1bd9bce68be6e96c2cf71f1f982a40ca2f18e67d --- /dev/null +++ b/marked/F/T-REC-F.748.40-202410-I_PDF-E/raw.md @@ -0,0 +1,425 @@ + + +# Recommendation **ITU-T F.748.40 (10/2024)** + +SERIES F: Non-telephone telecommunication services +Multimedia services + +--- + +## **Technical specification for artificial intelligence cloud platform: Data annotation** + +![ITU logo](390120de4fe440c42fea8154fcaad334_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.40 + +# Technical specification for artificial intelligence cloud platform: Data annotation + +## Summary + +Recommendation ITU-T F.748.40 provides technical requirements for data annotation capabilities for artificial intelligence (AI) cloud platforms. This Recommendation could support the construction and operation of such platforms by providing general user guidelines. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.40 | 2024-10-29 | 16 | 11.1002/1000/16189 | + +## Keywords + +Annotation tool, artificial intelligence, data annotation. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2025 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-------------------------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 1 | +| 5 Conventions ..... | 2 | +| 6 Technical specification for artificial intelligence cloud platform-data annotation..... | 3 | +| 6.1 Input data ..... | 3 | +| 6.2 Data annotation..... | 4 | +| 6.3 Intelligent annotation..... | 6 | +| 6.4 Data delivery ..... | 6 | +| 6.5 Production management ..... | 7 | +| Bibliography..... | 8 | + +## **Introduction** + +The importance of data annotation is gaining attention due to its role in improving the training of algorithmic models. Therefore, it is very important to evaluate the capability of data annotation. This proposal outlines the general requirements for the data annotation capabilities of AI cloud platforms, that can help industry players analyse and optimize their products, while also assisting users in building and selecting products that meet these requirements. + +# Recommendation ITU-T F.748.40 + +## Technical specification for artificial intelligence cloud platform: Data annotation + +# 1 Scope + +This Recommendation is part of the series of technical specification for artificial intelligence cloud platforms and provides functional requirements for the data annotation section of the AI cloud platform. From the perspective of the entire AI data annotation process, this Recommendation mainly focuses on the data annotation capabilities that help both users and product parties to understand the level of annotation ability. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 artificial intelligence** [b-ISO/IEC 22989]: Research and development of mechanisms and applications of AI systems. + +**3.1.2 dataset** [b-ISO/IEC 22989]: Collection of data with a shared format. + +**3.1.3 data annotation** [b-ISO/IEC 22989]: Process of attaching a set of descriptive information to data without any change to that data. + +**3.1.4 label** [b-ISO/IEC 22989]: Target variable assigned to a sample. + +## 3.2 Terms defined in this Recommendation + +None. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-----|-----------------------------------| +| 3D | Three-dimensional | +| AAC | Advanced Audio Coding | +| AI | Artificial Intelligence | +| API | Application Programming Interface | +| AVI | Audio Video Interleave | +| BMP | Bitmap | + +| | | +|--------|-----------------------------------| +| COCO | Common Objects in Context Dataset | +| CSV | Comma-Separated Values | +| FLAC | Free Lossless Audio Codec | +| FLV | Flash Video | +| GIF | Graphics Interchange Format | +| HLS | HTTP Live Streaming | +| JPG | Joint Photographic Group | +| JSON | JavaScript Object Notation | +| MKV | Matroska Video File | +| MOV | QuickTime File Format | +| MP3 | MPEG-1or MPEG-2 Audio Layer III | +| MP4 | MPEG-4 Part 14 | +| PCD | Point Cloud Data | +| PLY | Polygon File Format | +| PNG | Portable Network Graphics | +| PTS | Pro Tools Session | +| RAR | Roshal Archive | +| RTMP | Real-Time Messaging Protocol | +| RTSP | Real Time Streaming Protocol | +| SDK | Software Development Kit | +| STL | STereoLithography | +| TAR | Tape Archive | +| TAR.GZ | Tarball Gzip | +| TIFF | Tagged Image File Format | +| TXT | Text File | +| VOC | Visual Object Classes Challenge | +| WAV | Waveform Audio File Format | +| XLS | Excel Spreadsheet | +| XML | Extensible Markup Language | +| YOLO | You Only Look Once | +| ZIP | Zipped File | + +# 5 Conventions + +In this Recommendation: + +- The keywords "**is required to**" indicate a requirement that must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "**is recommended to**" indicate a requirement that is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. + +# 6 Technical specification for artificial intelligence (AI) cloud platform-data annotation + +The process of data annotation is mainly divided into data input, data annotation (manual annotation methods or intelligent annotation methods), and data delivery. There are different annotation tools for different data types. At the same time, production management guarantees the capability of the entire data annotation process. The standard organizes the callout functions involved in these steps. + +![Figure 1 – Artificial intelligence data annotation framework. The diagram shows a flow from Input data to Data annotation to Data delivery, with a Production management layer at the bottom.](562f471e8153729557e6a4ee6343c32c_img.jpg) + +The diagram illustrates the Artificial Intelligence data annotation framework. It consists of three main stages: Input data, Data annotation, and Data delivery, connected by arrows. Below these stages is a Production management layer. + +- Input data:** Contains sub-components for Data type and Input method. +- Data annotation:** Divided into two paths: + - Manual annotation:** Includes Image, Text, Audio, Video, 3D point cloud, and Multimodal. + - Intelligent annotation:** Includes Intelligent annotation method and Intelligent annotation type. +- Data delivery:** Includes Annotation quality inspection, Data delivery format, Data delivery content, and Data delivery method. +- Production management:** A horizontal bar at the bottom containing Platform management, Project management, Authority management, and Statistical analysis of data. + +F.748.40(24) + +Figure 1 – Artificial intelligence data annotation framework. The diagram shows a flow from Input data to Data annotation to Data delivery, with a Production management layer at the bottom. + +Figure 1 – Artificial intelligence data annotation framework + +## 6.1 Input data + +### 6.1.1 Data type + +- The platform is required to support the import of data, including the following categories: + - Image: portable network graphics (PNG), joint photographic group (JPG), bitmap (BMP), tagged image file format (TIFF), graphics interchange format (GIF). + - Text: text file (TXT), JavaScript object notation (JSON), extensible markup language (XML), comma-separated values (CSV). + - Audio: waveform audio file format (WAV), MPEG-1 or MPEG-2 audio layer III (MP3), advanced audio coding (AAC), free lossless audio codec (FLAC). + - Video: MPEG-4 Part 14 (MP4), audio video interleave (AVI), QuickTime file format (MOV), matroska video file (MKV). + - three-dimensional (3D) point cloud: polygon file format (PLY), STereoLithography (STL), point cloud data (PCD), pro tools session (PTS). + - Multimodal: e.g., data fused by image and 3D point clouds. +- The platform is recommended to support the import of the labelled data, and common datasets such as COCO (common objects in context dataset), VOC (visual object classes challenge), and YOLO (you only look once). + +### 6.1.2 Data input mode + +- The platform is required to support the import of data as offline methods, where pre-generated data can be accessed as single or in batches, including: + - Upload files:** files, folders. + - Archive files:** zipped file (ZIP), tape archive (TAR), tarball gzip (TAR.GZ), roshal archive (RAR). + - Data sources:** object storage, file storage, database. + +- b) The platform is required to support the import of data as online methods, which can continuously access the data generated in real-time in the form of streams, including application programming interface (API), software development kit (SDK), and other real-time video streaming protocols, such as real time streaming protocol (RTSP), http live streaming (HLS), real-time messaging protocol (RTMP), and flash video (FLV). + +## **6.2 Data annotation** + +### **6.2.1 Image data annotation** + +- a) The platform is required to provide annotation tools to support the annotation of image data, including: + - i) Annotation tools to support outline point or key point annotation on images. + - ii) Annotation tools to support line annotation on images, such as polylines, table lines. + - iii) Annotation tools to support region annotation on images, such as bounding boxes, polygons, ellipses. + - iv) Annotation tools to support 3D frames annotation on images. + - v) Annotation tools to support pixel-level segmentation of images. + - vi) Annotation tools to support attribute annotation of images, such as the license plate, brands, and categories of a car. +- b) The platform is recommended to provide aid tools to support the annotation of image data: + - i) Aid tools to improve the efficiency of image annotation, e.g., keyboard shortcuts. + - ii) Aid tools to improve the flexibility of image annotation, such as segmentation and cutting capabilities when annotating polygonal images. + - iii) Aid tools to improve the accuracy of image annotation, e.g., zoom. + +### **6.2.2 Text data annotation** + +- a) The platform is required to provide annotation tools to support the annotation of text data: + - i) Annotation tools to support the labelling of text data, e.g., attributes. + - ii) Annotation tools to support intervals annotation of text data. + - iii) Annotation tools to support relational annotation of text data. + - iv) Annotation tools to support instruction fine-tuning annotation of text data. +- b) The platform is recommended to provide aid tools to support the annotation of text data: + - i) Aid tools to improve the efficiency of text annotation, e.g., keyboard shortcuts. + - ii) Aid tools to improve the flexibility of text annotation, such as providing functions to modify attributes and global attributes. + - iii) Aid tools to improve the accuracy of text annotation, e.g., text comparison. + +#### **6.2.3 Audio data annotation** + +- a) The platform is required to provide annotation tools to support the annotation of audio data, including: + - i) Annotation tools to support the annotation of single-paragraph audio. + - ii) Annotation tools to support the annotation of multi-paragraph continuous audio. + - iii) Annotation tools to support the annotation of multi-paragraph discontinuous audio. +- b) The platform is recommended to provide aid tools to support the annotation of audio data: + - i) Aid tools to improve the efficiency of audio annotation, e.g., keyboard shortcuts. + +- ii) Aid tools to improve the flexibility of audio annotation, such as providing play, pause, fast forward, fast backward. +- iii) Aid tools to improve the accuracy of audio annotation, such as providing functions to view and manage the annotated audio. + +#### **6.2.4 Video data annotation** + +- a) The platform is required to provide annotation tools to support the annotation of video data, including: + - i) Annotation tools to support video data, such as annotating outline points, key points, polylines, table lines, bounding boxes, polygons, ellipses, and attributes. + - ii) Annotation tools to support intervals annotation of video data. +- b) The platform is recommended to provide aid tools to support the annotation of video data: + - i) Aid tools to improve the efficiency of video annotation, e.g., keyboard shortcuts. + - ii) Aid tools to improve the agility of video annotation, such as play, pause, navigate to a video frame, and move a video frame forward or backward. + - iii) Aid tools to improve the accuracy of video annotation, such as the ability to view and manage annotated videos. + +#### **6.2.5 3D point cloud data annotation** + +- a) The platform is required to provide annotation tools to support the annotation of 3D point cloud data, including: + - i) Annotation tools to support the annotation of 3D point cloud data, such as outline points, key points, polylines, table lines, bounding boxes, polygons, ellipses and attributes. + - ii) Annotation tools to support intervals of 3D point cloud data labelling in consecutive frames. +- b) The platform is recommended to provide aid tools to support the annotation of 3D point cloud data: + - i) Aid tools to improve the efficiency of 3D point cloud annotation, e.g., keyboard shortcuts. + - ii) Aid tools to improve the flexibility of 3D point cloud annotation, such as providing functions to drag, rotate and modify cubes. + - iii) Aid tools to improve the accuracy of 3D point cloud annotation, e.g., the ability to switch between different view options. + +#### **6.2.6 Multimodal data annotation** + +- a) The multimodal data refers to data consists of two or more modalities. The platform is required to provide annotation tools to support the annotation of multimodal data, including: + - i) Annotation tools to support multimodal single-frame annotation, such as outline points, key points, polylines, table lines, bounding boxes, polygons, ellipses, and relationships. + - ii) Annotation tools to support multimodal cross-frame annotation, e.g., interval annotation. + - iii) Annotation tools to support instruction fine-tuning annotation of multimodal data. +- b) The platform is recommended to provide aid tools to support the annotation of multimodal data: + - i) Aid tools to improve the efficiency of multimodal data annotation, e.g., keyboard shortcuts. + - ii) Aid tools to improve the agility of multimodal data annotation, such as provide functions to drag, rotate, and modify bounding boxes. + +- ii) Aid tools to improve the accuracy of multimodal data annotation, e.g., provides the function of mapping the relationship between different modalities of data. + +## 6.3 Intelligent annotation + +#### 6.3.1 Intelligent annotation method + +- a) Intelligent annotation is the process of automatically completing labelled data using artificial intelligence technology. The platform is required to provide intelligent annotation, including: + - i) **Pre-annotation**: refers to providing corresponding labelling models according to the task type. + - ii) **AI-assisted annotation**: refers to providing the functions that users only need to complete part of the annotation task, and the platform completes the remaining task based on the AI method e.g., active learning. + - iii) **Human-machine collaboration**: refers to the platform providing algorithms to filter a certain percentage of essential data, users annotate these data, and then train AI models. The remaining labelling task would be done by the trained AI models. + - iv) **Intelligent quality review**: refers to evaluating the quality of annotation by AI models. + - v) **Management of intelligent annotation capabilities**: refers to the management of the above intelligent annotation capabilities, such as adding, editing, deleting. + +#### 6.3.2 Intelligent annotation type + +- a) The platform is required to provide images intelligent annotation and texts intelligent annotation. +- b) The platform is recommended to provide audio, video, and 3D point cloud intelligent annotation. + +## 6.4 Data delivery + +#### 6.4.1 Annotation quality inspection + +- a) The platform is required to provide the inspection of annotation quality including: + - i) **Multi-data type**: refers to the platform providing the inspection including image, text, audio, and video data. + - ii) **Multi-method**: refers to providing at least two quality inspection methods, such as full inspection and sampling inspection. +- b) The platform is recommended to provide intelligent quality inspection, such as automatically evaluating the confidence level of data annotation by using relative algorithms. +- c) The platform is recommended to provide a quality inspection report. + +#### 6.4.2 Data delivery format + +- a) The platform is required to provide data annotation results delivery formats, including: + - i) **Label data**: JSON, XML, TXT. + - ii) **Structured label data**: CSV, Excel spreadsheet (XLS). + +#### 6.4.3 Data delivery content + +- a) The platform is required to deliver original data and labelled data. +- b) The platform is required to provide a description of the delivery data set, such as data source, data label distribution, and version management. + +- c) The platform is recommended to provide description information of the annotation process, such as abandonment rate, verification pass rate, inspection pass rate, annotation work duration, and work efficiency. + +#### **6.4.4 Data delivery method** + +- a) The platform is required to provide offline data delivery, such as ZIP, and TAR. +- b) The platform is required to provide online data delivery, such as API, and RTSP. + +## **6.5 Production management** + +### **6.5.1 Platform management** + +- a) The platform is required to provide basic operations for user management, such as user login authentication, adding, editing, and deleting. +- b) The platform is recommended to provide platform-based communication modules, such as message push, notification management, problem reporting, and email alerts. + +### **6.5.2 Project management** + +- a) The platform is required to provide the management capabilities, including: + - i) **Management of annotation tasks:** creation, query, editing, deleting. + - ii) **Management of annotation tools:** configuration of tools, label attribute. + - iii) **Management of annotation progress:** refers to providing functions that monitor the status. + - iv) **Management of priorities:** refers to providing functions that adjust the priority of tasks. + +#### **6.5.3 Authority management** + +- a) The platform is required to provide functions that manage permissions based on the role in the annotation task. + +#### **6.5.4 Statistical analysis of data** + +- a) The platform is required to provide statistical analysis of tasks and data annotators. +- b) The platform is required to provide visualization of various statistical data. +- c) The platform is recommended to provide data analysis operations, such as sorting, and filtering. + +## Bibliography + +- [b-ISO/IEC 22989] ISO/IEC 22989:2022, *Information technology – Artificial intelligence – Artificial intelligence concepts and terminology*. + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.41-202503-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.748.41-202503-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c1629394c02f697f60ae954ad36bf274fa8082cb --- /dev/null +++ b/marked/F/T-REC-F.748.41-202503-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3763acd7a0ed56aa30f3f90468c1fb0cb54573eb9045bca3eddf8976b4f0f41d +size 7219 diff --git a/marked/F/T-REC-F.748.41-202503-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.748.41-202503-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..8ad0b16fef360701c851bfde67b42d9d4b21216b --- /dev/null +++ b/marked/F/T-REC-F.748.41-202503-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1009010b875075a38ed05f6185f281b9772639d790902f5eaf74698ab4b0e47c +size 104246 diff --git a/marked/F/T-REC-F.748.41-202503-I_PDF-E/raw.md b/marked/F/T-REC-F.748.41-202503-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..f82e00e0ddc6dd3508a68a948fd64db6622ce63b --- /dev/null +++ b/marked/F/T-REC-F.748.41-202503-I_PDF-E/raw.md @@ -0,0 +1,376 @@ + + +# Recommendation **ITU-T F.748.41 (03/2025)** + +SERIES F: Non-telephone telecommunication services +Multimedia services + +--- + +## **Technical requirements and evaluation methods of artificial intelligence (AI)-based driver behaviour detection application** + +![ITU logo](390120de4fe440c42fea8154fcaad334_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.41 + +# Technical requirements and evaluation methods of artificial intelligence (AI)-based driver behaviour detection application + +## Summary + +Driver behaviour detection is one of the most important parts of a vehicle driving safety assistant system. It helps prevent safety risks caused by human beings and reduces the occurrence of serious accidents significantly. Therefore, how to realize a mature driver behaviour detection application has drawn public attention. + +The information and connections between the driver, the car and the environment need to be considered to fully understand driver behaviour in realistic driving scenarios. This project mainly focuses on the driver behaviour application using computer vision in vehicles. + +This proposal aims to provide a general framework, requirements and evaluation methods for driver behaviour detection applications and to help companies and industries boost the development of driver behaviour detection on multimedia applications. + +Recommendation ITU-T F.748.41 provides technical requirements and evaluation methods for artificial intelligence (AI)-based driver behaviour detection application. This Recommendation could support the construction of such applications by providing general user guidelines. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.41 | 2025-03-16 | 21 | 11.1002/1000/16231 | + +## Keywords + +Artificial intelligence, deep learning, driver behaviour detection, machine learning. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2025 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|---------------------------------------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Framework and functions of AI-based driver behaviour detection application ..... | 2 | +| 6.1 Framework of AI-based driver behaviour detection application..... | 2 | +| 6.2 Functions of each module..... | 3 | +| 7 Technical requirements and evaluation methods of AI-based driver behaviour detection application..... | 4 | +| 7.1 Technical requirements ..... | 4 | +| 7.2 Evaluation methods ..... | 4 | +| Bibliography..... | 8 | + +## **Introduction** + +Driver behaviour detection is one of the most important parts of a vehicle driving safety assistant system. It helps prevent safety risks caused by human beings and reduces the occurrence of serious accidents significantly. Therefore, how to realize a mature driver behaviour detection application has drawn public attention. + +The information and connections between the driver, the car and the environment need to be considered to fully understand driver behaviour in realistic driving scenarios. This project mainly focuses on the driver behaviour application using computer vision in vehicles. + +This proposal aims to provide a general framework, requirements and evaluation methods for driver behaviour detection applications and to help companies and industries boost the development of driver behaviour detection on multimedia applications. + +# Recommendation ITU-T F.748.41 + +## Technical requirements and evaluation methods of artificial intelligence (AI)-based driver behaviour detection application + +# 1 Scope + +This Recommendation focuses on the technical requirements and evaluation methods of AI-based driver behaviour detection application. The scope includes: + +- Framework for driver behaviour detection application based on AI multimedia application, including functions; +- Requirements and evaluation methods for AI-based driver behaviour detection application; +- Requirements for the data privacy protection in application. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +- [ITU-T F.749.3] Recommendation ITU-T F.749.3 (2020), *Use cases and requirements for vehicular multimedia networks.* +- [ISO/IEC 15007] ISO/IEC 15007:2020, *Road vehicles – Measurement and analysis of driver visual behaviour with respect to transport information and control systems.* +- [ISO/IEC 22989] ISO/IEC 22989:2022, *Artificial intelligence – Artificial intelligence concepts and terminology.* + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 deep learning** [b-ITU-T F.748.12]: A representation learning method, used to model high-level abstractions in data through the use of model architectures, which are composed of multiple nonlinear transformations. + +**3.1.2 input data** [ISO/IEC 22989]: A representation learning method, used to model high-level abstractions in data through the use of model architectures, which are composed of multiple nonlinear transformations. + +**3.1.3 machine learning** [b-ITU-T Y.3172]: Processes that enable computational systems to understand data and gain knowledge from it without necessarily being explicitly programmed. + +NOTE 1 – This definition is adapted from [b-ETSI GR ENI 004]. + +NOTE 2 – Supervised machine learning and unsupervised machine learning are two examples of machine learning types. + +## 3.2 Terms defined in this Recommendation + +None. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +DNN Deep Neural Network + +# 5 Conventions + +In this Recommendation: + +- The keywords "**is required to**" indicate a requirement that must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "**is recommended to**" indicate a requirement that is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. + +# 6 Framework and functions of AI-based driver behaviour detection application + +## 6.1 Framework of AI-based driver behaviour detection application + +As one of the most important parts of vehicle driving safety assistance system, driver behaviour detection application helps effectively prevent safety risks caused by human beings and reduce the occurrence of serious accidents through recognizing various driver behaviours that may carry potential risks. However, the current driver behaviour detection application lacks uniform standards for the functional modules to be included, the abnormal behaviours that can be detected, and the means of determining abnormal behaviours, which all need to be further standardized. By standardizing the basic modules and the methods of determining the abnormal behaviours of a driver behaviour detection application, it is possible to improve the accuracy of the application detection and enhance the reliability of the system in identifying potential risky behaviours, and avoid traffic accidents due to driver errors or improper behaviours. + +This clause mainly covers the overall framework for driver behaviour detection applications and the basic functions of each module, including the data input module, data preprocessing module, computation and decision module and output module. The driver behaviour detection application contains various types of AI-based multimedia tasks, including human pose estimation and object detection, which all take multimedia data as input and apply applications generated by deep learning technologies to multimedia. The driver behaviour prediction result is the comprehensive output of the end-to-end model, in which input data is multimodal. + +The driver behaviour detection application takes videos, images and audio as input and concentrates on data about the human voice and human limb, including arms, legs, trunk, head position and overall gesture. The AI-based multimedia applications such as human pose estimation, object detection and speech recognition are included in the application. The data which are not related to the AI tasks mentioned before will be blurred and anonymized. No user identification data will be collected and used. The application will not store or transmit data. + +In the data input module, the application takes multimedia data such as images, videos and audio as input. The data is cleaned and processed in the data preprocessing module. In the computation and decision module, it performs reference (usually using AI models) based on the preprocessed multimedia data and outputs the judgment results. The output module gives the result of specific behaviour types to the driver. + +![Figure 6.1 – Framework for AI-based driver behaviour detection application. The diagram shows a flow from Data input (Multimedia data: Video, Image, Audio) to Data preprocessing (Data cleaning, Data rescaling/smoothing, Data splitting/sampling, Data augmentation) to Computation and decision (AI-based multimedia application: Processed videos, Processed images, Processed audios leading to Inference functions) to Output (Driver abnormal behaviour: Drinking, Turning around, Dressing up, Making phone calls, Using control panel). A 'Driver' icon is shown on the right.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +The diagram illustrates the framework for an AI-based driver behaviour detection application. It is enclosed in a dashed box and consists of four main stages: + +- Data input:** Multimedia data (Video, Image, Audio). +- Data preprocessing:** Data cleaning, Data rescaling/smoothing, Data splitting/sampling, Data augmentation. +- Computation and decision:** AI-based multimedia application (Processed videos, Processed images, Processed audios) leading to Inference functions. +- Output:** Driver abnormal behaviour (Drinking, Turning around, Dressing up, Making phone calls, Using control panel). + +A 'Driver' icon is shown on the right side of the diagram. The reference code F.748.41(25) is located at the bottom right. + +Figure 6.1 – Framework for AI-based driver behaviour detection application. The diagram shows a flow from Data input (Multimedia data: Video, Image, Audio) to Data preprocessing (Data cleaning, Data rescaling/smoothing, Data splitting/sampling, Data augmentation) to Computation and decision (AI-based multimedia application: Processed videos, Processed images, Processed audios leading to Inference functions) to Output (Driver abnormal behaviour: Drinking, Turning around, Dressing up, Making phone calls, Using control panel). A 'Driver' icon is shown on the right. + +**Figure 6.1 – Framework for AI-based driver behaviour detection application** + +## 6.2 Functions of each module + +The AI-based driver behaviour detection application is required to include the following modules: + +- **Data input functional module:** This module is to take various types of information as input. +- **Data preprocessing functional module:** This module is to transform information and process data into formatted data that can be used directly. +- **Computation and decision functional module:** This module is to infer results and make decisions through an end-to-end model by using processed data. +- **Data output functional module:** This module is to output the type of driver behaviour. + +### 6.2.1 Data input functional module + +The data input module takes multimedia data such as real-time videos, images and audio as input. The data comes from on-board cameras and sensors, which is mainly user behaviour data that is strongly related to the application. It is required that any images of an individual be anonymized and that they do not contain sensitive personal information. + +### 6.2.2 Data preprocessing functional module + +The data preprocessing module aims to improve data quality by cleaning, conversion and other operations, to process the original data into a form that is directly used for model inference. + +The data preprocessing **functional** module includes the following functions: + +- **Data cleaning functions:** Supporting data cleaning functions, such as sorting raw data into standardized data forms (e.g., data format conversion, data type conversion), invalid sample data removal and duplicate data deletion. + +- **Data splitting and sampling functions:** Supporting functions of splitting and sampling data, such as splitting or merging structured data, upsampling (increasing data samples to enhance information) and downsampling (decreasing data samples to reduce information). +- **Data rescaling and smoothing functions:** Supporting functions of rescaling and smoothing data, such as data normalization and data standardization (e.g., Z-Score standardization). +- **Data augmentation functions:** Adding small changes to existing data or newly create synthetic data from existing data to increase the amount of data. + +### 6.2.3 Computation and decision functional module + +The computation and decision **functional** module realizes the detection of driver behaviour by performing an end-to-end inference model on multimedia data. The inference functions are generally deep neural network composed by encoder layer, concatenating embedding layer, feed forward layer and classifier layer. + +- **Processed videos:** Video data processed by the data processing module for direct use by the inference functions. +- **Processed images:** Image data processed by the data processing module for direct use by the inference functions. +- **Processed audio:** Audio data processed by the data processing module for direct use by the inference functions. +- **Inference functions:** Supports functions of extracting features (e.g., gesture, voice) from multimedia data and performing inference for classification tasks to support driver behaviour detection. + +### 6.2.4 Data output functional module + +- The data output module supports functions of output some information (e.g., send alarm message) to the driver when driver's behaviour is abnormal, such as drinking, distracted by talking, dressing up, making phone calls, turning around and using the control panel. + +# 7 Technical requirements and evaluation methods of AI-based driver behaviour detection application + +## 7.1 Technical requirements + +- **Data input:** It is required to support inputting multiple modalities of data, including audio, videos and images. +- **Data preprocessing:** It is recommended to support constructing a pipeline that contains a series performed sequentially steps of data processing to process raw input data, covering image, audio and video pipelines. +- **Computation and decision:** It is recommended to support building an end-to-end unified deep neural network (DNN) [ISO/IEC 22989] model to classify driver abnormal behaviours. +- **Data output:** It is required to support the detection of at least one kind of abnormal behaviour that could cause the driver to lose concentration on driving, such as drinking, dressing, turning around, making phone calls and using the control panel. + +## 7.2 Evaluation methods + +### Overall application performance evaluation + +- **Data collection:** Collect a diverse dataset of driving scenarios, including instances of normal and abnormal driving behaviour. + +- Data annotation: Annotate the dataset with labels indicating different types of behaviours, such as drinking, dressing, making phone calls or sleeping. +- Performance metrics: Evaluate the model's performance using various metrics, including: + - Accuracy: The proportion of correctly classified instances; + - Precision: The ratio of true positive predictions to the total number of positive predictions; + - Recall: The ratio of true positive predictions to the total number of actual positive instances; + - F1 Score: The harmonic mean of precision and recall, providing a balance between them. +- Testing on unseen data: Evaluate the trained model on the testing dataset, which contains unseen instances of driving behaviour. Calculate performance metrics to assess how well the model generalizes to new data. +- Qualitative analysis: Conduct qualitative analysis by inspecting misclassified instances and understanding the reasons behind misclassifications. This can provide insights into areas for improvement in the model or the dataset. +- Real-world testing: Deploy the model in real-world driving scenarios, either in controlled environments or as part of a pilot study. Monitor the model's performance and gather feedback from users to assess its effectiveness in detecting and mitigating driver behaviour. + +#### **Making phone calls** + +- Test purpose: Will reduce the driver's driving ability and emergency response ability due to making phone calls during driving. +- Test action: Making phone calls includes but is not limited to the following: + - 1) Answering a call with one or both hands; + - 2) Holding the mobile phone with one hand or both hands. +- Test procedures and methods: The application can detect the driver's mobile phone use in a non-contact manner and has the following functions: + - 1) The detection of making phone calls can be realized under different lighting conditions, such as day, night, dusk and dawn; with the function of several alarm level settings: + +When the vehicle is stationary, if the driver is detected using the mobile phone, the application only temporarily records the image and video information for further assessment. When the vehicle speed is lower than the alarm setting speed, if the driver's use of mobile phone behaviour is detected, a first-level alarm will be generated. When the vehicle speed is higher than the alarm setting speed, if the driver's use of mobile phone behaviour is detected, a second-level alarm will be generated. When the duration is beyond a fixed time, the alarm is upgraded to the third level. + +#### **Drinking** + +- Test purpose: Impaired vision caused by drinking during driving will reduce the driver's emergency driving ability. +- Test action: Drinking actions includes the following several types: + - 1) Picking up the cup or beverage bottle with the left hand or the right hand; + - 2) Drinking by raising the head; + - 3) Drinking by using a straw. +- Test procedures and methods: During the vehicle's driving process, the application detects the driver's drinking behaviour in a non-contact manner and generates an alarm, and has the following functions: + +- 1) Be capable of detecting drinking behaviour under different lighting conditions such as during the day, night, dusk and dawn, with the function of several alarm level settings. + +When the vehicle speed is lower than the alarm setting speed, if the driver's drinking behaviour is detected and is not confirmed as drinking, temporarily save photos and videos of the driver's behaviour for further assessment. When the vehicle speed is higher than the alarm setting speed, if the driver's drinking behaviour is detected, generate a secondary alarm and send the alarm information to the driver. When the vehicle speed is higher than the alarm setting speed and the driver is detected drinking a beverage with the obvious characteristics of an alcoholic beverage, generate a tertiary alarm. + +#### **Distracted by talking** + +- Testing purpose: During the driving process, the driver talks with passengers in the car for a long time or talks with the outside world through other communication means (e.g., in-car Bluetooth headset), which impairs the driver's attention and reduces the driving ability and the ability to deal with dangerous emergencies. +- Test action: Distracted by talking includes the following: + - 1) When talking with passengers in the car, the driver spends an excessive proportion of their time talking and listening; + - 2) Communicating with the outside world through other communication methods (e.g., in-car Bluetooth headset) for a duration exceeding the limit. +- Test procedures and methods: The application should be able to detect the driver's behaviour of making a sound for a long time, which impairs the driver's attention in a non-contact manner and generate an alarm and has the following functions: + - 1) Achieving long-term vocal behaviour detection under different lighting conditions, such as day, night, dusk and dawn, with the function of several alarm level settings. + +When the vehicle is stationary, if the driver's talking behaviour is detected for a long time, the application only temporarily records image and video information for further assessment. When the vehicle speed is lower than the alarm setting speed, if the driver's long-term talking behaviour is detected, a first-level alarm is generated. When the vehicle speed is higher than the alarm setting speed, if the driver's talking behaviour is detected for a long time, a second-level alarm is generated and alarm information is sent to the driver. + +#### **Playing with mobile phones** + +- Test purpose: The driver's driving ability and emergency response ability is reduced due to playing with mobile phones during driving. +- Test action: Playing with mobile phones actions includes the following: + - 1) Watching or swiping the mobile phone. +- Test procedures and methods: The application detects the driver's mobile phone use in a non-contact manner and has the following functions: + - 1) Mobile phone behaviour detection can be realized under different lighting conditions, such as day, night, dusk and dawn; with the function of several alarm level settings. + +When the vehicle is stationary, if the driver is detected using the mobile phone, the application only temporarily records the image and video information for further assessment. When the vehicle speed is lower than the alarm setting speed, if the driver's use of mobile phone behaviour is detected, a first-level alarm will be generated. When the vehicle speed is higher than the alarm setting speed, if the driver's use of mobile phone behaviour is detected, a second-level alarm will be generated and alarm information would be sent to the driver. When the duration exceeds a fixed duration, the alarm is upgraded to level three. The time duration requirements could be referenced in [ITU-T F.749.3] [ISO/IEC 15007]. + +#### Turning around + +- Test purpose: Drivers who frequently turn around for various reasons during driving have a high risk and relevant behaviours need to be monitored. +- Test action: Frequent turning around includes but is not limited to the following: + - 1) Taking the direction of the vehicle as a reference, turning the body exceeds a fixed angle more than specific times within a certain time; + - 2) Taking the direction of the vehicle as a reference, turning the body exceeds a fixed angle for more than a certain time. +- Test procedures and methods: The application detects the driver's turning behaviour in a non-contact manner to generate an alarm and has the following functions: + - 1) Turn detection can be realized in different light conditions, such as day, night, dusk and dawn, with the function of several alarm level settings. + +When the vehicle is stationary, if the driver's frequent turning behaviour is detected, the application only temporarily records image and video information for further assessment. When the vehicle speed is lower than the alarm setting speed, if the driver's frequent turning behaviour is detected, a first-level alarm will be generated. When the vehicle speed is higher than the alarm setting speed, if the driver's frequent turning behaviour is detected, a second-level alarm is generated and alarm information is sent to the driver. + +#### Using the control panel + +- Test purpose: The use of the control panel while driving with the driver's hands off the steering wheel can create a risk of the vehicle steering out of control and needs to be monitored. +- Test action: The use of the control panel includes the following: + - 1) Leaving the steering wheel with one hand to use the control panel; + - 2) Using the control panel with both hands off the steering wheel. +- Test procedures and methods: The application should be able to detect the use of the control panel by the driver in a non-contact manner and generate an alarm with the following functions: + - 1) The control panel can be used to detect hands off the steering wheel under different lighting conditions, such as day, night, dusk and dawn, with the function of several alarm level settings. + +When the vehicle is stationary, if the driver's hands are detected to leave the steering wheel and use the control panel, the system only temporarily records image and video information for further assessment. When the vehicle is started, if the driver is detected to leave the steering wheel with one hand using the control panel, a first-level alarm is generated. If the driver's hands are detected to leave the steering wheel, the control panel is used to generate a second-level alarm, send alarm information to the driver. When the second-level alarm exceeds the certain time, the alarm is upgraded to the third-level alarm. The time duration requirements could be referenced in [ITU-T F.749.3] [ISO/IEC 15007]. + +## Bibliography + +- [b-ITU-T F.748.12] Recommendation ITU-T F.748.12 (2021), *Deep learning software framework evaluation methodology*. +- [b-ITU-T Y.3172] Recommendation ITU-T Y.3172 (2019), *Architectural framework for machine learning in future networks including IMT-2020*. +- [b-ETSI GR ENI 004] ETSI GR ENI 004 V2.2.1 (2021), *Experiential Networked Intelligence (ENI); Terminology for Main Concepts in ENI*. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.42-202503-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg b/marked/F/T-REC-F.748.42-202503-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d4d8969f805414689ab79cfd08c333d454d482b3 --- /dev/null +++ b/marked/F/T-REC-F.748.42-202503-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:636c3839edff955a5b6208530b05fa1c983ab153d90f00216517a7a99478b988 +size 7074 diff --git a/marked/F/T-REC-F.748.42-202503-I_PDF-E/8e14350b4b669119a3bdfca7869110ca_img.jpg b/marked/F/T-REC-F.748.42-202503-I_PDF-E/8e14350b4b669119a3bdfca7869110ca_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..0b6897aed21f72cfcc86941d56ccca566993a959 --- /dev/null +++ b/marked/F/T-REC-F.748.42-202503-I_PDF-E/8e14350b4b669119a3bdfca7869110ca_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c977e595355e84863e32e2ca5885c6dbf2f35d63ac8f0cb83b5e7d113b7a8446 +size 70723 diff --git a/marked/F/T-REC-F.748.42-202503-I_PDF-E/a26e142d3df5bef41a84a9dd099d7825_img.jpg b/marked/F/T-REC-F.748.42-202503-I_PDF-E/a26e142d3df5bef41a84a9dd099d7825_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..e0addb8499554431ac4ab5dabe5f85865a36848c --- /dev/null +++ b/marked/F/T-REC-F.748.42-202503-I_PDF-E/a26e142d3df5bef41a84a9dd099d7825_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2daffba4a9b82f64fd023a8d82567729847de8c4fe1867c3e7aa4151351b3180 +size 31728 diff --git a/marked/F/T-REC-F.748.42-202503-I_PDF-E/e9d825d87c5f85c8dba0664eace96ef4_img.jpg b/marked/F/T-REC-F.748.42-202503-I_PDF-E/e9d825d87c5f85c8dba0664eace96ef4_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..92da0e07b67bf2328f25f9ea179b2ac2bda75d2d --- /dev/null +++ b/marked/F/T-REC-F.748.42-202503-I_PDF-E/e9d825d87c5f85c8dba0664eace96ef4_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:df9b4523e5c72f9a1f0f43daa73c53fe952d8bc6bd543717f10a6bba9681b8c9 +size 30959 diff --git a/marked/F/T-REC-F.748.43-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.748.43-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ac295b1b0c7f14b1c5a33ef6682b1b017c812477 --- /dev/null +++ b/marked/F/T-REC-F.748.43-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37aed4fc114a39843438175326e0b7e1d09a9d6075723d82fea32040fd560ee4 +size 7192 diff --git a/marked/F/T-REC-F.748.43-202503-I_PDF-E/d0abac95583b52a3b35f74a215567334_img.jpg b/marked/F/T-REC-F.748.43-202503-I_PDF-E/d0abac95583b52a3b35f74a215567334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a6f6a197ad4fa93bce07a0487732f7979106ad2e --- /dev/null +++ b/marked/F/T-REC-F.748.43-202503-I_PDF-E/d0abac95583b52a3b35f74a215567334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7ed10e243e7f10573e34c32b7f09cb3026b1faf609592e05f6115aaf1d14fd06 +size 101928 diff --git a/marked/F/T-REC-F.748.43-202503-I_PDF-E/raw.md b/marked/F/T-REC-F.748.43-202503-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..1447299b56e3a978def314013cac1cd1e95b6adc --- /dev/null +++ b/marked/F/T-REC-F.748.43-202503-I_PDF-E/raw.md @@ -0,0 +1,521 @@ + + +Recommendation + +**ITU-T F.748.43 (03/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +**Framework and requirements for the +foundation model platform** + +![ITU logo](84a1d09fb489061482111515543b60dc_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +# ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +## Recommendation ITU-T F.748.43 + +## Framework and requirements for the foundation model platform + +## Summary + +Recommendation ITU-T F.748.43 specifies the framework and requirements for the foundation model platform in multimedia, which aims to address issues like application barriers, resource and computational power needs, and complex processes, despite foundation models' role in content generation. It delivers artificial intelligence (AI) capabilities for multimedia applications. The formulation of this foundation model platform framework fills the vacuum of platform standards for the production and operation of foundation models in the multimedia field, helping producer enterprises guide product research and standardize their construction of foundation model platforms, and provides enterprise users with selection references for foundation model platform products. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.43 | 2025-03-16 | 21 | 11.1002/1000/16233 | + +## Keywords + +Artificial intelligence, foundation model, framework, requirements. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2025 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-----------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 1 | +| 5 Conventions ..... | 1 | +| 6 Framework..... | 2 | +| 6.1 Overall framework..... | 2 | +| 6.2 Data engineering..... | 2 | +| 6.3 Model pre-training..... | 2 | +| 6.4 Model optimization ..... | 3 | +| 6.5 Model delivery..... | 3 | +| 6.6 Model operations ..... | 3 | +| 6.7 Platform capabilities..... | 3 | +| 6.8 Monitoring..... | 4 | +| 7 Requirements ..... | 4 | +| 7.1 Data engineering..... | 4 | +| 7.2 Model pre-training..... | 5 | +| 7.3 Model optimization ..... | 5 | +| 7.4 Model delivery..... | 6 | +| 7.5 Model operations ..... | 7 | +| 7.6 Platform capabilities..... | 8 | +| 7.7 Monitoring..... | 9 | +| Bibliography..... | 10 | + + + +## Framework and requirements for the foundation model platform + +## 1 Scope + +This Recommendation specifies the framework and requirements for the foundation model platform, designed to guide producing enterprises in creating service specifications and to offer benchmarks for enterprise users in selecting foundation model platform products. It consists of seven parts: data engineering, model pre-training, model optimization, model delivery, service operations, platform capabilities and monitoring. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +# 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following term defined elsewhere: + +**3.1.1 artificial intelligence** [b-ISO/IEC 2382]: Interdisciplinary field, usually regarded as a branch of computer science, dealing with models and systems for the performance of functions generally associated with human intelligence, such as reasoning and learning. + +### 3.2 Terms defined in this Recommendation + +None. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-----|-----------------------------------| +| AI | Artificial Intelligence | +| API | Application Programming Interface | +| CPU | Central Processing Unit | +| GPU | Graphics Processing Unit | + +## 5 Conventions + +In this Recommendation: + +- The keywords "is required to" indicate a requirement that must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "is recommended to" indicate a requirement that is recommended but which is not absolutely required. Thus, this requirement needs not be present to claim conformance. + +# 6 Framework + +### 6.1 Overall framework + +Figure 6-1 illustrates the framework of the foundation model platform. Based on the pipeline of large-scale artificial intelligence (AI) model production and service, the platform can be divided into seven parts: data engineering, model pre-training, model optimization, model delivery, model operations, platform capabilities, and monitoring. + +![Figure 6-1: Framework of the foundation model platform. The diagram shows a hierarchical structure. At the top is a horizontal bar labeled 'Monitoring'. Below it are five vertical boxes representing the main pipeline stages: 'Data engineering', 'Model pre-training', 'Model optimization', 'Model delivery', and 'Model operations'. Each of these boxes contains a list of sub-components. Below these stages is another horizontal bar labeled 'Platform capabilities', which contains three sub-components: 'Plugin capability', 'Platform management', and 'Resource scheduling and management'.](d0abac95583b52a3b35f74a215567334_img.jpg) + +The diagram illustrates the framework of the foundation model platform, organized into seven main components: + +- Monitoring** (Top level) +- Data engineering**: + - Data collection + - Data processing + - Data generation + - Data annotation +- Model pre-training**: + - Distributed training + - Training completeness + - Development process stabilities +- Model optimization**: + - Prompt engineering + - Model fine-tuning + - Reinforcement learning +- Model delivery**: + - Model compression + - Model management + - Model testing + - Service deployment +- Model operations**: + - Service orchestration + - Result feedback + - Data feedback +- Platform capabilities** (Bottom level): + - Plugin capability + - Platform management + - Resource scheduling and management + +Figure 6-1: Framework of the foundation model platform. The diagram shows a hierarchical structure. At the top is a horizontal bar labeled 'Monitoring'. Below it are five vertical boxes representing the main pipeline stages: 'Data engineering', 'Model pre-training', 'Model optimization', 'Model delivery', and 'Model operations'. Each of these boxes contains a list of sub-components. Below these stages is another horizontal bar labeled 'Platform capabilities', which contains three sub-components: 'Plugin capability', 'Platform management', and 'Resource scheduling and management'. + +F.748.43(25) + +**Figure 6-1 – Framework of the foundation model platform** + +### 6.2 Data engineering + +Data engineering includes data collection, data processing, data generation, and data annotation. + +- Data collection is the process of accessing different types of data in different manners. +- Data processing is the process of transforming raw data into clean data for models through a series of operations. +- Data generation is the process of creating new data from scratch or deriving new data from existing data. It aims at providing ample and diverse datasets to enhance the performance and generalization capability of the models. +- Data annotation refers to the process of classifying and labelling data to identify and understand its characteristics. + +### 6.3 Model pre-training + +Model pre-training includes distributed training, training completeness and development process stability. + +- Distributed training is the process to enhance training efficiency. It is recommended to support various training strategies. +- Training completeness is the capability to train models across various modalities, languages, and learning methods. +- Development process stability is the capability used to ensure timely recovery from failures encountered during training. + +### **6.4 Model optimization** + +Model optimization includes prompt engineering, model fine-tuning, and reinforcement learning. + +- a) Prompt engineering refers to the technology of using optimization, design and adjustment capabilities to propose problems or tasks to foundation models. A good prompt can improve the quality of the model's inference results. +- b) Model fine-tuning involves adjusting the parameters of foundation models or training models with small datasets for a specific task. This process adapts and enhances the model's performance for particular business needs. +- c) Reinforcement learning refers to the process of optimizing a model's quality by adjusting its weights and parameters based on human feedback. + +### **6.5 Model delivery** + +Model delivery includes model compression, model management, model testing, and service deployment. + +- a) Model compression refers to the technology of reducing the model size to improve its operational efficiency and enable it to be deployed on various devices. +- b) Model management refers to the unified management of various models developed internally and by third parties, encompassing aspects such as model storage, version control, and lifecycle management. +- c) Model testing refers to evaluating the model's precision and performance by using independent testing datasets. This process aims to identify defects in the model and enable subsequent optimization and improvement. +- d) Service deployment refers to the process of transforming trained or tuned models into inference services and deploying them in the production environment. + +### **6.6 Model operations** + +Service operations include service orchestration, result feedback, and data feedback. + +- a) Service orchestration refers to the ability to schedule and manage model services, such as integrating various independent model services into one complete business process. This facilitates collaborative work across models and services, enhancing the flexibility and scalability of the services. +- b) Result feedback is the ability to manage and provide timely feedback on the inference results and user feedback on the model service. +- c) Data feedback refers to the process where results or derivative data are re-entered into the system or platform after data processing, analysis or usage. + +### **6.7 Platform capabilities** + +Platform capabilities include plug-in capability and resource scheduling and management. + +- a) Plug-in capability refers to the platform's capability to easily and flexibly add, configure, and manage various plug-ins through an extensible mechanism. This facilitates the adaptation of models to different application scenarios and enhances the capability of the foundation model. +- b) Resource scheduling and management refers to the capability of reasonably allocating, optimizing, and scheduling computing resources, storage resources, and network resources. + +### **6.8 Monitoring** + +Service monitoring refers to the capability of real-time monitoring, and analysis during the model production and service operation processes. Implementing monitoring, it enhances the ability to detect and troubleshoot issues, thereby improving service stability. + +# **7 Requirements** + +### **7.1 Data engineering** + +#### **7.1.1 Data collection** + +- a) It is required to support structured and unstructured data collection access. +- b) It is required to support encryption of sensitive data. +- c) It is required to support access to real-time data streams. +- d) It is recommended to support data quality testing, such as testing for integrity, consistency and availability. +- e) It is recommended to support data storage, management and access. +- f) It is recommended to support the periodic collection of access data. +- g) It is recommended to support the monitoring and maintenance of data collection. + +#### **7.1.2 Data processing** + +- a) It is required to support data cleaning, deduplication and standardization. +- b) It is required to support data conversion. +- c) It is recommended to support the handling of unbalanced data. +- d) It is recommended to support the basic functions of data desensitization, including built-in desensitization rules and customized desensitization rules. +- e) It is recommended to support the underlying functionality of data detoxification to enhance the fairness, security and trustworthiness of training and inference data. +- f) It is recommended to support the capability of visualization to analyse the data. +- g) It is recommended to support customized data processing. +- h) It is recommended to support automatic structured and unstructured data processing. +- i) It is recommended to support the audit function of the data desensitization and detoxification process. +- j) It is recommended to support automatic detection and correction of data in specific scenarios or tasks. +- k) It is recommended to support data version control, record the operations and changes in the process of data cleaning and screening, and facilitate the tracking of historical versions of data and comparison. + +#### **7.1.3 Data generation** + +- a) It is recommended to support enhancement capabilities for various types of data. +- b) It is recommended to support built-in data generation templates for generating commonly used data. +- c) It is recommended to support customized data generation rules to define the logic, associations and constraints of the generated data. +- d) It is recommended to support data fusion, integrating datasets from different kinds of modalities and sources to produce more accurate, reliable, complete and valid samples. + +- e) It is recommended to support automated data generation capability. +- f) It is recommended to support adaptive data generation, dynamically adjusting the data generation strategy according to the model's performance during the training process. +- g) It is recommended to support the interpretability of data generation. + +#### **7.1.4 Data annotation** + +- a) It is required to support the labelling of various collected data. +- b) It is required to support labelled data management. +- c) It is recommended to support the visualization of data labelling. +- d) It is recommended to support team labelling. +- e) It is recommended to support labelling for various types of data, such as audio, text, image, and video. +- f) It is recommended to support customized annotation rules to meet the needs of specific scenarios. +- g) It is recommended to support continuous optimization capabilities for intelligent annotation. +- h) It is recommended to support multi-level auditing and quality control. + +### **7.2 Model pre-training** + +#### **7.2.1 Distributed training** + +- a) It is recommended to support various distributed training strategies, such as operator parallelism, data parallelism, model parallelism, and pipeline parallelism. +- b) It is recommended to support at least four of the aforementioned strategies. + +#### **7.2.2 Training completeness** + +- a) It is recommended to support model training in text and image modalities. +- b) It is recommended to support learning paradigms, including supervised learning and semi-supervised learning. +- c) It is recommended to support training in various natural languages such as local languages and English. +- d) It is recommended to support training in various programming languages. +- e) It is recommended to support training in various modalities, including video and audio. +- f) It is recommended to support training in various industry scenario. + +#### **7.2.3 Development process stability** + +- a) It is recommended to support accelerator failure detection and recovery. +- b) It is recommended to support manual checkpoint recovery. +- c) It is recommended to support cluster network failure detection and recovery. +- d) It is recommended to support automatic checkpoint recovery. + +### **7.3 Model optimization** + +#### **7.3.1 Prompt engineering** + +- a) It is required to support a prompt library, which includes prompt management and visualization capabilities. +- b) It is recommended to support the capability of evaluation and continuous feedback for the prompt. + +- c) It is recommended to support customized development, management and visualization of prompt templates. +- d) It is recommended to support the capability of prompt enhancement, such as thinking chain prompting. +- e) It is recommended to support automation and optimization of the prompt to improve model response quality. +- f) It is recommended to support multi-modal prompt design, such as text, image, audio or video. + +#### **7.3.2 Model fine-tuning** + +- a) It is required to support loading different popular foundation models. +- b) It is required to support common fine-tuning algorithms. +- c) It is recommended to support the management and evaluation for fine-tuned models. +- d) It is recommended to support the visualization of the fine-tuning process. +- e) It is recommended to support the capability of automatic fine-tuning for specific tasks. + +#### **7.3.3 Reinforcement learning** + +- a) It is required to support common reinforcement learning frameworks. +- b) It is required to support the feedback mechanism, which can transform human feedback into an understandable and usable form for the model. +- c) It is recommended to support reward modelling training capabilities. +- d) It is recommended to support common reinforcement learning strategies, such as strategy gradient reinforcement learning algorithm and near-end strategy optimization algorithm. +- e) It is recommended to support the evaluation of the model after reinforcement learning. +- f) It is recommended to support the capability of automatic adjusting model weights and parameters based on human feedback. + +### **7.4 Model delivery** + +#### **7.4.1 Model compression** + +- a) It is required to support the operation of quantitative algorithms. +- b) It is recommended to support the operation of pruning algorithms. +- c) It is recommended to support the operation of distillation algorithms. +- d) It is recommended to support the capability of automatic compression for models. + +#### **7.4.2 Model management** + +- a) It is required to support model export functionality. +- b) It is required to support model storage functionality, including visual operations such as model file registration, upload, download, and deletion. +- c) It is required to support model version management, including operations such as creating new versions, rolling back to historical versions, and comparing version differences. +- d) It is recommended to support hierarchical access and usage permissions for models. +- e) It is recommended to support cross-model version dependency management to ensure that collaboration and interaction between models can proceed normally across different versions. +- f) It is recommended to support communication and collaboration capabilities between models, including data exchange, combination, nesting, and other capabilities. +- g) It is recommended to support model auditing functionality, providing visual information such as operation logs, model documentation, model metadata, and model parameters. + +#### **7.4.3 Model testing** + +- a) It is required to support the management of testing datasets. +- b) It is required to support the evaluation of model performance. +- c) It is required to support testing of the model interface. +- d) It is recommended to support the comparison of results from different models or different versions of the same model, in order to facilitate the selection of the optimal model. +- e) It is recommended to support the automatic generation of testing reports. +- f) It is recommended to support the customization of performance evaluation metrics and interface test cases. + +#### **7.4.4 Service deployment** + +- a) It is required to support deployment of modelling services via mirroring. +- b) It is required to support flexible provision of model services through application programming interface (API) interfaces. +- c) It is recommended to support the capability of visually configuring update policies. +- d) It is recommended to support monitoring of the deployment process. +- e) It is recommended to support visual orchestration and automated execution capabilities for deployment pipelines. +- f) It is recommended to support inference acceleration services. + +### **7.5 Model operations** + +#### **7.5.1 Service orchestration** + +- a) It is required to support the management of different versions of services. +- b) It is required to support the construction and provision of standardized interfaces for unified model services. +- c) It is recommended to support configuration for orchestration, such as sequential invocation, conditional judgment, and branch selection. +- d) It is recommended to support visualization and scheduling capabilities. +- e) It is recommended to support data transfer and processing between model services. +- f) It is recommended to support the ability to execute multiple services in parallel. +- g) It is recommended to support the intelligent capability of dynamic service scheduling. + +#### **7.5.2 Result feedback** + +- a) It is recommended to support storing various types of inference results. +- b) It is recommended to support accurate and complete presentation of inference results to the user, or to other systems through interfaces. +- c) It is recommended to support the collection and storage of user feedback on inference results. +- d) It is recommended to support automated analysis of inference results and user feedback. +- e) It is recommended to support tracking and logging of inference results and user feedback on issues or anomalies. +- f) It is recommended to support the intelligent capability to analyse inference results and user feedback. + +#### **7.5.3 Data feedback** + +- a) It is required to support the collection and storage of reflow data. + +- b) It is required to support the management of reflow tasks. +- c) It is required to support data reflow logging. +- d) It is recommended to support reflow monitoring and timely response based on monitored data changes. +- e) It is recommended to support periodic or triggered data reflow, such as data reflow triggered by user complaints and data drift. +- f) It is recommended to support data reflow strategy setting. +- g) It is recommended to support the traceability of data reflow, so as to trace the source and the process of data. +- h) It is recommended to support the visualization of reflow data monitoring. +- i) It is recommended to support real-time data reflow. +- j) It is recommended to support automatic exception handling mechanism, such as data conflict, data loss, reflow failure, and other exceptions. + +### **7.6 Platform capabilities** + +#### **7.6.1 Plug-in capability** + +- a) It is recommended to support plug-in registration and library management. +- b) It is recommended to support the management of plug-in access rights. +- c) It is recommended to support plug-in development. +- d) It is recommended to support plug-in operation monitoring and logging. +- e) It is recommended to support plug-in intelligent arrangement and selection. + +#### **7.6.2 Platform management** + +- a) It is recommended to support user management and visualization features, including but not limited to user registration, user login, permission control, and user classification. +- b) It is recommended to support platform log management and visualization features, including but not limited to the collection, storage, querying, and analysis of logs. +- c) It is recommended to support platform auditing and visualization features, including but not limited to recording and querying personnel, time, and behaviour of key operations performed on the platform. +- d) It is recommended to support performance metric analysis and visualization display when the platform processes tasks, including but not limited to response time, throughput, resource utilization rate, and task concurrency. +- e) It is recommended to support cost measurement and analysis, including but not limited to resource costs, labour costs, and time costs. +- f) It is recommended to support predicting future cost consumption based on historical data. +- g) It is recommended to support the platform's fault tolerance and automatic repair capabilities under certain exceptional circumstances. + +#### **7.6.3 Resource scheduling and management** + +- a) It is required to support the capability of resource allocation. +- b) It is required to support resource warning settings and automatic triggering of warning notifications. +- c) It is recommended to support configuration management of resource scheduling policies and prioritization. + +- d) It is recommended to support resource load balancing capability, automatically adjusting resource allocation according to resource usage. +- e) It is recommended to support heterogeneous resource management, such as graphics processing unit (GPU) and central processing unit (CPU) collaborative computing and scheduling. +- f) It is recommended to support elastic scaling of resource pools, automatically expanding or reducing the size of resource pools based on demand and resource utilization. +- g) It is recommended to support the ability to intelligently optimize resource scheduling strategies to improve resource utilization. + +### **7.7 Monitoring** + +- a) It is recommended to support the monitoring of model producing process. +- b) It is recommended to support the monitoring of model performance. +- c) It is recommended to support the monitoring of model service. +- d) It is recommended to support the monitoring of exceptions and events. +- e) It is recommended to support the capability to customize monitoring metrics. +- f) It is recommended to support error tolerance and auto-recovery for some scenarios. +- g) It is recommended to support intelligent monitoring and analysis. + +## Bibliography + +[b-ISO/IEC 2382] ISO/IEC 2382:2015, *Information technology – Vocabulary*. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.44-202503-I_PDF-E/07b17a620c75522d53916a11e12d1bff_img.jpg b/marked/F/T-REC-F.748.44-202503-I_PDF-E/07b17a620c75522d53916a11e12d1bff_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ef7d312f48608a9caff0751b72763ba65b3213f4 --- /dev/null +++ b/marked/F/T-REC-F.748.44-202503-I_PDF-E/07b17a620c75522d53916a11e12d1bff_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3b18cc35e4acf49877fb8987f7416b0f8bf3ac4796e34b5841e6a0dd021d9266 +size 78257 diff --git a/marked/F/T-REC-F.748.44-202503-I_PDF-E/75f0cb39f1cd165dfe4a6aa6c4d9388d_img.jpg b/marked/F/T-REC-F.748.44-202503-I_PDF-E/75f0cb39f1cd165dfe4a6aa6c4d9388d_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d21a9635d4e5f59136d14d2aa0d071d2358d798a --- /dev/null +++ b/marked/F/T-REC-F.748.44-202503-I_PDF-E/75f0cb39f1cd165dfe4a6aa6c4d9388d_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:92ec2829f42e27383608d6e21ebe40fea48340563b0c73c0573429842ed9c001 +size 61121 diff --git a/marked/F/T-REC-F.748.44-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.748.44-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ffc7ac2119eec15901d4f25fd1f6f3fa94981d1c --- /dev/null +++ b/marked/F/T-REC-F.748.44-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f2e4738329c669b6225ed42fcd4267ce117866dc503937a271986f829e5334fc +size 7020 diff --git a/marked/F/T-REC-F.748.45-202503-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.748.45-202503-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..6a574bf4c7fd169a72a97de0395e1970ee6f7092 --- /dev/null +++ b/marked/F/T-REC-F.748.45-202503-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5d863c71623a2b5a933b672e32a1463a7461603fd81d81e68543e5d90adf3b80 +size 7132 diff --git a/marked/F/T-REC-F.748.45-202503-I_PDF-E/990567efebf979be51f56d1150012c9d_img.jpg b/marked/F/T-REC-F.748.45-202503-I_PDF-E/990567efebf979be51f56d1150012c9d_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c6d21a797ed5be102812db8273f2cb034885a9bc --- /dev/null +++ b/marked/F/T-REC-F.748.45-202503-I_PDF-E/990567efebf979be51f56d1150012c9d_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:10977ab691a26e5d8336ca962f7e69c0f8d0d2493e91694aa73c303402b1382e +size 136222 diff --git a/marked/F/T-REC-F.748.46-202503-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/F/T-REC-F.748.46-202503-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c1629394c02f697f60ae954ad36bf274fa8082cb --- /dev/null +++ b/marked/F/T-REC-F.748.46-202503-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3763acd7a0ed56aa30f3f90468c1fb0cb54573eb9045bca3eddf8976b4f0f41d +size 7219 diff --git a/marked/F/T-REC-F.748.46-202503-I_PDF-E/731f533b0599c8e42a063f06e4332045_img.jpg b/marked/F/T-REC-F.748.46-202503-I_PDF-E/731f533b0599c8e42a063f06e4332045_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..417b13cb3ddc1e066664914a2c18fe82949657a6 --- /dev/null +++ b/marked/F/T-REC-F.748.46-202503-I_PDF-E/731f533b0599c8e42a063f06e4332045_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ecda550f04dd58a01c9f915185cda1669448da3b43ce5a974be0dbb0956acb92 +size 80907 diff --git a/marked/F/T-REC-F.748.46-202503-I_PDF-E/raw.md b/marked/F/T-REC-F.748.46-202503-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..2ca9eea943f9f2089e37a7293b9be265c2ca475d --- /dev/null +++ b/marked/F/T-REC-F.748.46-202503-I_PDF-E/raw.md @@ -0,0 +1,736 @@ + + +# Recommendation**ITU-T F.748.46 (03/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +### **Requirements and evaluation methods of artificial intelligence agents based on large scale pre-trained models** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +## Recommendation ITU-T F.748.46 + +# Requirements and evaluation methods of artificial intelligence agents based on large scale pre-trained models + +## Summary + +Recommendation ITU-T F.748.46 provides a framework of requirements and evaluation methods for AI agents based on pre-trained large models. It attempts to help enterprises and users understand AI agent technology and help them build and implement products and applications faster and more accurately. Furthermore, this Recommendation is intended to promote the healthy development of the industry. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.46 | 2025-03-16 | 21 | 11.1002/1000/16236 | + +## Keywords + +AI agent, evaluating methods, technical requirements. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2025 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-----------------------------------------------------------------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 1 | +| 5 Conventions ..... | 2 | +| 6 Framework of requirement and evaluation methods for artificial intelligence agents based on large scale pre-trained model ..... | 2 | +| 7 Requirements and evaluation methods of artificial intelligence agents ..... | 3 | +| 7.1 Perception and cognition ..... | 3 | +| 7.2 Planning capability ..... | 8 | +| 7.3 Memory capability..... | 10 | +| 7.4 Execution capability ..... | 11 | +| Appendix I – Formulae for common indicators..... | 14 | +| Bibliography..... | 16 | + + + +# Recommendation ITU-T F.748.46 + +# Requirements and evaluation methods of artificial intelligence agents based on large scale pre-trained models + +# 1 Scope + +This Recommendation mainly focuses on AI agents based on large scale pre-trained models. It attempts to establish technical requirements and evaluation methods for intelligent products and applications to measure their performance. The capabilities involved in this Recommendation include four aspects: + +- **Perception and cognition:** This comprises of recognition, comprehension, dialogue, generation, and reasoning. +- **Planning:** This includes task planning, task scheduling and task optimization. +- **Memory:** This includes short-term memory and long-term memory. +- **Action:** This includes virtual environment execution capability and real-world environment execution capability. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 AI agent** [b-ISO/IEC 22989]: Automated entity that senses and responds to its environment and takes actions to achieve its goals. + +**3.1.2 artificial intelligence** [b-ISO/IEC 2382]: An interdisciplinary field, usually regarded as a branch of computer science, dealing with models and systems for the performance of functions generally associated with human intelligence, such as reasoning and learning. + +## 3.2 Terms defined in this Recommendation + +None. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-----|-------------------------| +| AI | Artificial Intelligence | +| CV | Computer Vision | +| SCR | Sentence Correct Rate | + +SER Sentence Error Rate + +WCR Word Correct Rate + +WER Word Error Rate + +# 5 Conventions + +In this Recommendation: + +- The keywords "**is required to**" indicate a requirement that must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "**is recommend to**" indicate a requirement that is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. +- The keywords "**optional**" indicate an optional requirement that is permissible, without implying any sense of being recommended. These terms are not intended to imply that the vendor's implementation must provide the option, and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +# 6 Framework of requirement and evaluation methods for artificial intelligence agents based on large scale pre-trained models + +The AI agents based on the pre-trained large models are required to have the following four capabilities: perception and cognition, planning, memory and execution capability. In practical applications, the intelligent system is recommended to consider the interaction of the external environment and the agent groups. Based on the above, this Recommendation divides the overall technical structure into two modules, one focusing on the technical capabilities of the single AI agent and the other focusing on the interaction between the AI agent and the external elements. + +This Recommendation evaluates the capability levels of AI agents in two aspects: performance superiority and functional completeness. + +![Figure 1: Framework of requirement and evaluation methods for artificial intelligence agents based on large scale pre-trained models. The diagram shows a flow from 'Evaluation methods' to 'Technical capabilities' (Perception and cognition, Planning, Memory, Action) to 'Environment and AI agent groups'. 'Evaluation methods' include Performance tests and Functional test. 'Technical capabilities' are: Perception and cognition (Recognition, Comprehension, Conversational, Generating, Reasoning), Planning (Task planning, Task scheduling, Task optimization), Memory (Short-term, Long-term), and Action (Virtual environment execution, Real-world environment execution). Arrows show a sequential flow between capabilities and bidirectional arrows between the entire technical capabilities block and the environment block.](731f533b0599c8e42a063f06e4332045_img.jpg) + +The diagram illustrates the framework of requirement and evaluation methods for artificial intelligence agents. It is structured as follows: + +- Evaluation methods** (left side): + - Performance tests + - Functional test +- Technical capabilities** (center, divided into four columns): + - Perception and cognition**: Recognition capabilities, Comprehension capabilities, Conversational capabilities, Generating capability, Reasoning capability. + - Planning**: Task planning, Task scheduling, Task optimization. + - Memory**: Short-term memory, Long-term memory. + - Action**: Virtual environment execution capability, Real-world environment execution capability. +- Environment and AI agent groups** (top bar): The entire technical capabilities block has bidirectional arrows connecting it to this environment bar. + +Arrows indicate a sequential flow from left to right: Evaluation methods → Perception and cognition → Planning → Memory → Action. + +Figure 1: Framework of requirement and evaluation methods for artificial intelligence agents based on large scale pre-trained models. The diagram shows a flow from 'Evaluation methods' to 'Technical capabilities' (Perception and cognition, Planning, Memory, Action) to 'Environment and AI agent groups'. 'Evaluation methods' include Performance tests and Functional test. 'Technical capabilities' are: Perception and cognition (Recognition, Comprehension, Conversational, Generating, Reasoning), Planning (Task planning, Task scheduling, Task optimization), Memory (Short-term, Long-term), and Action (Virtual environment execution, Real-world environment execution). Arrows show a sequential flow between capabilities and bidirectional arrows between the entire technical capabilities block and the environment block. + +F.748.46(25) + +**Figure 1 – Framework of requirement and evaluation methods for artificial intelligence agents based on large scale pre-trained models** + +# **7 Requirements and evaluation methods of artificial intelligence agents** + +## **7.1 Perception and cognition** + +### **7.1.1 Recognition capability** + +AI agents are recommended to support a recognition capability. This clause is intended to evaluate the completeness and recognition performance of various recognition functions of AI agents. + +#### **7.1.1.1 Visual recognition** + +Requirements for the functional completeness and recognition performance of visual recognition of AI agents are as follows: + +- a) Task richness: The AI agent is recommended to ensure the richness of visual recognition tasks. The functions that may be involved, but not limited to, are posture estimation, movement recognition, gesture recognition, optical character recognition and text recognition. + +Evaluation methods for functional completeness and recognition performance of visual recognition of AI agents are as follows: + +- a) The performance of the AI agent is recommended to be evaluated under visual recognition tasks. It is recommended to select and calculate the appropriate indicators from accuracy, precision and recall. +- b) The functional completeness of the AI agent is recommended to be assessed by calculating the number of functions which are satisfied. + +#### **7.1.1.2 Text recognition** + +Requirements for the functional completeness and recognition performance of text recognition of AI agents are as follows: + +Task richness: The AI agent is recommended to ensure the richness of text recognition tasks. The functions that may be involved, but not limited to, are card recognition, bill recognition, form recognition and custom document recognition. + +Evaluation methods for the functional completeness and recognition performance of text recognition of AI agents are as follows: + +- a) The performance of the AI agent is recommended to be evaluated under text recognition tasks. It is recommended to select and calculate appropriate indicators such as the word correct rate (WCR) and sentence correct rate (SCR). +- b) The functional completeness of the AI agent is recommended to be assessed by calculating the number of functions which are satisfied. + +#### **7.1.1.3 Speech recognition** + +Requirements for the functional completeness and recognition performance of speech recognition of AI agents are as follows: + +Task richness: The AI agent is recommended to ensure the richness of speech recognition tasks supported by the AI agent. The functions that may be involved, but not limited to, are multi-language recognition, dialects and accents recognition, translation between different languages and recognition of noisy environments. + +Evaluation methods for the functional completeness and recognition performance of speech recognition of AI agents are as follows: + +- a) The performance of the AI agent is recommended to be evaluated under speech recognition tasks. It is recommended to select and calculate appropriate indicators such as the word error rate (WER) and sentence error rate (SER). +- b) The functional completeness of the AI agent is recommended to be assessed by calculating the number of functions which are satisfied. + +### **7.1.2 Comprehension capability** + +AI agents are recommended to support comprehension capability. This clause is intended to evaluate whether the intelligent agent has the ability to understand text, images and speech. + +#### **7.1.2.1 Visual comprehension** + +List the requirements and evaluate whether the AI agents are capable of image and video understanding. + +Requirements for the functional completeness and comprehension performance of image and video understanding of AI agents are as follows: + +- a) Image comprehension: The AI agent is required to ensure the capability of understanding images. +- b) Video understanding: The AI agent is recommended to ensure the capability of understanding videos. + +Evaluation methods for the functional completeness and comprehension performance of image and video understanding of AI agents are as follows: + +The accuracy rate of image understanding or video understanding is recommended to be evaluated. + +#### **7.1.2.2 Semantic analysis** + +Requirements for the functional completeness and comprehension performance of syntactic analysis of AI agents are as follows: + +- a) Lexical analysis: The AI agent is required to ensure the capability of lexical analysis. +- b) Syntactic analysis: The AI agent is required to ensure the capability of syntactic analysis. + +Evaluation methods for the functional completeness and comprehension performance of syntactic analysis of AI agents are as follows: + +The performance of lexical analysis and syntactic analysis of an AI agent is recommended to be evaluated based on accuracy, recall and F1 score. + +#### **7.1.2.3 Sentiment analysis** + +Requirements for the functional completeness and comprehension performance of semantic analysis of AI agents are as follows: + +- a) Emotional tendencies: The AI agent is required to ensure the capability to understand different emotions and emotional tendencies. +- b) Emotional granularity: The AI agent is recommended to ensure the capability of further segmentation for more abstract and ambiguous emotional tendencies based on basic tendencies. +- c) Emotion tracing: The AI agent is required to ensure the capability of sentiment understanding based on emotion tracing. +- d) Emotional suggestion: The AI agent is required to ensure the capability to put forward relevant emotional suggestions based on the results of emotional tracing. + +Evaluation methods for the functional completeness and comprehension performance of semantic analysis of AI agents are as follows: + +The performance of semantic analysis of AI agents is recommended to be evaluated based on accuracy. + +### **7.1.3 Dialogue capability** + +AI agents are recommended to support dialogue capability. This clause is intended to evaluate whether the AI agent has the ability to engage in single-round and multi-round dialogues. + +#### **7.1.3.1 Single-round dialogue** + +Requirements for the functional completeness and comprehension performance of single-round dialogue of AI agents are as follows: + +- a) Fuzzy intent recognition: The AI agent is required to ensure the capability of fuzzy intent recognition for dialogues on different topics. +- b) Precise intent recognition: The AI agent is required to ensure the capability of precise intent recognition for dialogues on different topics. + +Evaluation methods for the functional completeness and comprehension performance of single-round dialogue of AI agents are as follows: + +- a) It is recommended to use the single-round dialogue test data to test the AI agent. It is recommended to categorize recognition results into accurate recognition, fuzzy recognition and no recognition, and calculate the single-round dialogue accuracy rate. + +#### **7.1.3.2 Multi-round dialogue** + +Requirements for the functional completeness and comprehension performance of multi-round dialogue of AI agents are as follows: + +- a) Context recognition: The AI agent is required to ensure the capability of context recognition for multi-round dialogues on different topics. +- b) Task completion: The AI agent is required to ensure the capability of completing the multi-round dialogues on different topics. + +Evaluation methods for the functional completeness and comprehension performance of multi-round dialogues of AI agents are as follows: + +- a) It is recommended to use the multi-round dialogue test data to test the AI agent, and it is recommended to calculate the task completion rate of the multi-round conversations. +- b) It is recommended to calculate the context recognition accuracy of AI agents on different topics. + +### **7.1.4 Generative capability** + +The AI agents are recommended to support generative capability. This clause is intended to evaluate whether the AI agent has the ability to generate text, images, audio and other multimodal material. + +#### **7.1.4.1 Text generation** + +Requirements for the functional completeness and comprehension performance of generating text of AI agents are as follows: + +- a) Grammar and spelling: The AI agent is required to ensure the quality of generated texts and try to avoid spelling and grammatical errors. +- b) Fluency and coherence: The AI agent is required to ensure the fluency and coherence between utterances of generated text. +- c) Content accuracy: The AI agent is required to ensure generated text meet the proposed requirements. The requirements that may be involved, but not limited to, are content, genres and language styles. + +- d) Content diversity: + - 1) Text topics: The AI agent is recommended to ensure text generation on different topics. The topics that may be involved, but not limited to, are news, advertisements, scripts, lyrics, poems and novels. + - 2) Languages and dialects: The AI agent is recommended to ensure text generation on different languages and dialects. The languages that may be involved, but not limited to, are Chinese, English, Japanese, Korean, French, German, Italian, Spanish, Portuguese and Russian. + +Evaluation methods for the functional completeness and comprehension performance of generating text of AI agents are as follows: + +- a) It is recommended to assess the generated texts for grammatical correctness, discourse structure rationalization and word spelling. +- b) It is recommended to assess the fluency and coherence between utterances of generated texts. +- c) It is recommended to assess whether the generated text meets the proposed requirements. +- d) It is recommended to calculate the number of different topics, languages and dialects which are supported by the AI agent. +- e) It is recommended to make the subjective assessment of the fluency, richness and logic of the generated text. + +#### **7.1.4.2 Image generation** + +Requirements for the functional completeness and comprehension performance of generating images of AI agents are as follows: + +- a) Relevance: The AI agent is recommended to ensure the relevance between the generated images and proposed requirements. +- b) Smoothness of images: The AI agent is recommended to ensure the smoothness of generated images. The items that may be involved, but not limited to, are boundary line of the objects, basic physical cognition and details of objects. +- c) Diversity: + - 1) Image themes: The AI agent is recommended to adjust to different themes for image generation. The subjects that may be involved, but not limited to, are natural landscapes, urban landscapes, portraits, animal portraits, food images, works of art and science fiction images. + - 2) Image size: The AI agent is recommended to adjust to varied sizes and resolutions for image generation. + +Evaluation methods for the functional completeness and comprehension performance of generating images of AI agents are as follows: + +- a) It is recommended to make the subjective assessment of the relevance, smoothness and diversity of the generated images. +- b) It is recommended to calculate the number of different themes, sizes and resolutions the AI agent supports for image generation. + +#### **7.1.4.3 Audio generation** + +Requirements for the functional completeness and comprehension performance of audio generation of AI agents are as follows: + +- a) Pronunciation accuracy: The AI agent is required to ensure the pronunciation accuracy when generating audio. +- b) Syllable clarity: The AI agent is required to ensure syllable clarity when generating audio. + +- c) Naturalness of speech: The AI agent is required to ensure the naturalness of speech when generating audio. + +Evaluation methods for the functional completeness and comprehension performance of audio generation of AI agents are as follows: + +- a) It is recommended to evaluate the quality of generated audio in three dimensions: pronunciation accuracy, syllable clarity and naturalness of speech. Use the following formula for calculating a composite score. + +$$S = S_N \times \alpha + S_P \times \beta + S_M \times \gamma$$ + +$S$ : the composite score of audio quality + +$S_N$ : the score of pronunciation accuracy + +$S_P$ : the score of syllable clarity + +$S_M$ : the score of speech naturalness + +$\alpha, \beta, \gamma$ : weights + +- b) It is recommended to make a subjective assessment of the relevance and smoothness of the generated audio. + +#### 7.1.4.4 Multimodal generation + +Requirements for the functional completeness and comprehension performance of multimodal generation of AI agents are as follows: + +- a) Content rationality: The AI agent is required to ensure content rationality. The generated multimodal material should conform to the laws of human cognition and the actual world, and should obey rationality through logic and common sense. +- b) Content relevance: The AI agent is required to ensure content relevance. The diverse types of data in the generated multimodal material should related to each other. + +Evaluation methods for the functional completeness and comprehension performance of multimodal generation of AI agents are as follows: + +It is recommended to make a subjective assessment of the quality of the generated audio from content rationality and content relevance. + +### 7.1.5 Reasoning capability + +The AI agents are recommended to support a reasoning capability. This clause is intended to evaluate whether the AI agent can reason and comprehend. + +#### 7.1.5.1 Co-reference resolution + +Requirements for the functional completeness and comprehension performance of co-reference resolution of AI agents are as follows: + +- a) Richness of information sources: The AI agent is required to adjust to available information sources. The sources that may be involved, but not limited to, are operational systems, media and corporate databases. + +Evaluation methods for the functional completeness and comprehension performance of co-reference resolution of AI agents are as follows: + +- a) The test dataset should be collected from different information sources. +- b) It is recommended to calculate the accuracy of co-reference reasoning resolution for the AI agent system based on the dataset. + +#### **7.1.5.2 Knowledge reasoning** + +Requirements for the functional completeness and comprehension performance of knowledge reasoning of AI agents are as follows: + +- a) **Richness of reasoning themes:** The AI agent is required to adjust to different reasoning themes. The themes that may be involved, but not limited to, are healthcare, energy, humanities, culture and social issues. +- b) **Transparency of reasoning:** The AI agent is required to ensure the transparency of reasoning. The AI agent is recommended to provide the key steps of reasoning. The methodologies that may be involved, but not limited to, are chain of thought, tree of thought and graph of thought. + +Evaluation methods for the functional completeness and comprehension performance of knowledge reasoning of AI agents are as follows: + +It is recommended to calculate the accuracy of reasoning on different themes. + +## **7.2 Planning capability** + +### **7.2.1 Task planning** + +The AI agents are recommended to support task planning. This clause is intended to evaluate the ability of AI agents to coordinate and plan tasks. + +#### **7.2.1.1 Target disassembly** + +Requirements for the functional completeness and comprehension performance of target disassembly of AI agents are as follows: + +- a) **Goal recognition and cognition:** The AI agent is required to support in-depth cognition of goals for workflows, including, but not limited to, key indicators and potential obstacles. +- b) **Goal analysis and prediction:** The AI agent is required to support analysing and predicting possible results. The outcomes include, but are not limited to, goal concepts, structure, complexity, hierarchy and inter-goal relationships. +- c) **Disassembly relevance:** The AI agent is recommended to be able to manage and understand the correlation between the disassembled subgoals when disassembling the targets. +- d) **Reasonableness of disassembly:** The AI agent is recommended to be able to refer to the feasibility, dependency and priority of the disassembled subgoals when disassembling the targets, so as to ensure that the subsequent subsets are operable. +- e) **Interpretability and visualization of disassembly:** The AI agent is recommended to support the provision of detailed explanation and visualization of the disassembly planning scheme to help users or developers understand the process and results of the scheme. + +Evaluation methods for the functional completeness and comprehension performance of target disassembly of AI agents are as follows: + +It is recommended to assess the functional completeness of the AI agent by calculating the number of functions which are satisfied. + +#### **7.2.1.2 Planning strategy** + +Requirements for the functional completeness and comprehension performance of planning strategy of AI agents are as follows: + +- a) **Planning structure:** The AI agent is recommended to support task planning according to a certain structure. This involves, but is not limited to, linear, hierarchical, parallel, tree and mesh. +- b) **Planning logic:** The AI agent is recommended to promise the rationality of the process for disassembling tasks and ensuring the clear logical relationships between subtasks. + +- c) **Planning consistency:** The AI agent is recommended to promise planning consistency and coordination when organizing plans between or within tasks. The agents are recommended to avoid inconsistencies such as duplicate tasks, dead-end tasks, conflicting tasks and invalid tasks. + +Evaluation methods for the functional completeness and comprehension performance of planning strategy of AI agents are as follows: + +- a) It is recommended to calculate the number of the planning steps given by the AI Agents to accomplish the specified task are necessary and effective. +- b) It is recommended to calculate the number of the planning steps given by the AI agents to accomplish the specified task generate the correct result after execution. + +### **7.2.2 Task scheduling** + +The AI agents are recommended to support task scheduling. This clause is intended to evaluate whether the AI agent can set the priority of tasks and whether it has a flexible scheduling mechanism. + +#### **7.2.2.1 Scheduling mechanisms** + +Requirements for the functional completeness and comprehension performance of task scheduling mechanisms of AI agents are as follows: + +- a) **Mechanism enrichment:** The AI agent is recommended to support different scheduling mechanisms. The mechanisms that are involved, but not limited to, are first-come-first-served, short job priority, rotation scheduling mechanism, priority scheduling mechanism and earliest deadline priority. +- b) **Robustness:** The AI agent is recommended to quickly adapt and re-plan the tasks under abnormal situations, such as automatic intervention and manual intervention. +- c) **Autonomous cooperation:** Assess whether the agent groups can automatically schedule different agents to work and collaborate. + +Evaluation methods for the functional completeness and comprehension performance of task scheduling mechanisms of AI agents are as follows: + +It is recommended to assess the functional completeness of the AI agent by calculating the number of functions which are satisfied. + +#### **7.2.2.2 Priority setting** + +Requirements for the functional completeness and comprehension performance of priority setting of AI agents are as follows: + +- a) **Completion rate of priority setting:** The AI agent is recommended to ensure the completion of priority setting under different requests. +- b) **Accuracy of priority setting:** The AI agent is recommended to ensure the accuracy of priority setting waves in a reasonable interval. + +Evaluation methods for the functional completeness and comprehension performance of priority setting of AI agents are as follows: + +It is recommended to calculate the accuracy rate and completion rate of task priority setting. + +#### **7.2.2.3 Organizational coordination** + +Requirements for the functional completeness and comprehension performance of organizational and coordination skills of AI agents are as follows: + +- a) **Resource coordination:** The AI agent is recommended to support resource coordination. The aspects involved, but not limited to, are time allocation, computational resource management, data access and storage, and multi-source information integration. + +- b) Task allocation: The AI agent is recommended to support flexible task allocation based on actual resources. +- c) Progress monitoring: The AI agent is recommended to support monitoring the progress of process execution and providing alarms under abnormal situations. +- d) Emergency response: The AI agent is recommended to support flexible adjustment of the task allocation strategy and emergency response in emergency situations. + +Evaluation methods for the functional completeness and comprehension performance of organizational and coordination skills of AI agents are as follows: + +It is recommended to assess the functional completeness of the AI agent by calculating the number of functions which are satisfied. + +## **7.3 Memory capability** + +### **7.3.1 Short-term memory** + +The AI agents are recommended to support short-term memory. This clause is intended to evaluate whether the AI agent has the ability to perform short-term trace retrieval and prompt word management. + +#### **7.3.1.1 Trace retrieval** + +Requirements for the functional completeness and comprehension performance of trace retrieval of AI agents are as follows: + +- a) Retrieval accuracy: The AI agent is recommended to ensure the trace retrieval is accurate and matches the original input. +- b) Retrieval completeness: The AI agent is recommended to ensure the trace retrieval is complete and contains all relevant information. +- c) Average retrieval time: The AI agent is recommended to ensure the average time to complete the retrievals varies in a reasonable range. + +Evaluation methods for the functional completeness and comprehension performance of trace retrieval of AI agents are as follows: + +- a) It is recommended to evaluate the accuracy of trace retrieval. +- b) It is recommended to evaluate whether the trace retrieval is complete. +- c) It is recommended to evaluate the average time to complete the retrievals. + +### **7.3.2 Long-term memory** + +The AI agents are recommended to support long-term memory. This clause is intended to evaluate whether the AI agent has the ability to store long-term traces and quickly retrieve them. + +#### **7.3.2.1 Memory storage** + +Requirements for the functional completeness and comprehension performance of memory storage of AI agents are as follows: + +- a) Historical dialogue rounds: The AI agent is recommended to remember historical dialogue rounds. +- b) Knowledge update: The AI agent is recommended to ensure the promptness and quality of knowledge updates. +- c) Storage capacity: The AI agent is recommended to ensure the sufficiency of storage capacity. + +Evaluation methods for the functional completeness and comprehension performance of memory storage of AI agents are as follows: + +- a) It is recommended to calculate the maximum number of historical dialogue rounds that the AI agents can support without significant degradation of performance. +- b) It is recommended to assess the frequency and quality of knowledge updates of the AI agents. +- c) It is recommended to calculate the number of long-term trace characters that can be memorized and stored. + +#### **7.3.2.2 Quick search** + +Requirements for the functional completeness and comprehension performance of quick search of AI agents are as follows: + +- a) Retrieval speed: The AI agent is recommended to ensure the retrieval speed waves in a reasonable range. +- b) Retrieval accuracy: The AI agent is recommended to ensure the retrieval accuracy waves in a reasonable range. +- c) Retrieval coverage: The AI agent is recommended to ensure the retrieval coverage is complete and abundant. + +Evaluation methods for the functional completeness and comprehension performance of quick search of AI agents are as follows: + +- a) It is recommended to calculate the time taken by the AI agents from receiving the query request to returning the first word of retrieval result. +- b) It is recommended to calculate the degree of match between the retrieval results returned by the AI agents and the user's query intents. +- c) It is recommended to assess the sources and types of information that the AI agents can retrieve. + +## **7.4 Execution capability** + +### **7.4.1 Virtual environment execution capability** + +The AI agents are recommended to support virtual environment execution capability. This clause is intended to evaluate whether the AI agent has the ability to run processes and communicate with each other. + +#### **7.4.1.1 Process operation** + +Requirements for the functional completeness and comprehension performance of process operation of AI agents are as follows: + +- a) Process understanding: The AI agent is recommended to accurately understand the structure and rules of the processes, as well as the dependency and timing relationship between different tasks. +- b) Task allocation: The AI agent is recommended to reasonably allocate tasks to different performers or resources according to the process structure and task requirements. +- c) Resource scheduling: The AI agent is recommended to be able to reasonably schedule resources to support the execution of tasks according to process requirements and resource conditions. +- d) Abnormality handling: The AI agent is recommended to be able to detect and handle abnormalities in a timely manner during process execution to ensure the normal operation of the process. +- e) Scalability: The AI agent is recommended to have enough scalability to operate normally in different processes and tasks. + +- f) Tool richness: The AI agent is recommended to be able to provide abundant tools, such as document parsing, speech recognition, database access, image recognition, and so on. + +Evaluation methods for the functional completeness and comprehension performance of process operation of AI agents are as follows: + +- a) It is recommended to assess the functional completeness of the AI agent by calculating the number of functions which are satisfied. +- b) It is recommended to calculate the number of tools that can be provided by AI agents. + +#### **7.4.1.2 Communication and interaction** + +Requirements for the functional completeness and comprehension performance of communication and interaction of AI agents are as follows: + +- a) Passive service: The AI agent is required to provide services passively with human-agent interaction. +- b) Active service: The AI agent is recommended to provide services actively. + +Evaluation methods for the functional completeness and comprehension performance of communication and interaction of AI agents are as follows: + +It is recommended to make a subjective assessment of the initiative to implement the tasks. + +### **7.4.2 Real environment execution capability** + +The AI agents are recommended to support real environment execution capability. This clause is intended to evaluate whether the AI agent has the ability to locate, navigate, collaborate dynamically and perceive the environment. + +#### **7.4.2.1 Positioning and navigation** + +Requirements for the functional completeness and comprehension performance of localization and navigation of AI agents are as follows: + +- a) Intelligence degree of positioning scheme: The AI agent is recommended to provide intelligent technologies in the positioning process. It can be based on external observation signals, sensors and computer vision. +- b) Positioning environment complexity: The AI agent is recommended to adjust to complex and dynamic positioning environments, such as open fields, city streets and narrow spaces. +- c) Single-point positioning accuracy: The AI agent is recommended to ensure single-point positioning accuracy. +- d) Trajectory error: The AI agent is recommended to ensure the trajectory offset during the positioning process. +- e) Positioning solution versatility: The AI agent is recommended to consider different environments and tasks when making the positioning solution. + +Evaluation methods for the functional completeness and comprehension performance of localization and navigation of AI agents are as follows: + +- a) It is recommended to evaluate the positioning accuracy of the AI agent and evaluate the relevant performance by comparing the number of reports of the gold standard device. +- b) It is recommended to assess the performance of the positioning solution, such as positioning accuracy and ease of operation. +- c) It is recommended to evaluate the positioning accuracy of the AI agent and evaluate the related performance by testing the positioning error. + +#### **7.4.2.2 Environmental perception** + +Requirements for the functional completeness and comprehension performance of environment perception of AI agents are as follows: + +- a) Diversity of perception parameters: The AI agent is recommended to support multiple types of sensors, such as physical sensors, temperature sensors, optical sensors, etc., and it supports multiple types of parameters' perception. +- b) Fusion perception accuracy: The AI agent is recommended to support the fusion of diverse types of perception technologies for environment perception. + +Evaluation methods for the functional completeness and comprehension performance of environment perception of AI agents are as follows: + +- a) It is recommended to assess the functional completeness of the AI agent by calculating the number of functions which are satisfied. +- b) It is recommended to assess the accuracy and richness of monitored indicators. + +## Appendix I + +## Formulae for common indicators + +(This appendix does not form an integral part of this Recommendation.) + +In this Recommendation, calculating the various technical performances of AI agents draws on several common index calculation methods. The calculations of the indicators that may be involved are shown as follows: + +- a) The equations for accuracy, precision and recall. + +$$Accuracy = \frac{TP + TN}{TP + FN + FP + FN}$$ + +$$Precision = \frac{TP}{TP + FP}$$ + +$$Recall = \frac{TP}{TP + FN}$$ + +*TP*: true positive + +*TN*: true negative + +*FP*: false positive + +*FN*: false negative + +- b) The equations for WCR and SCR. + +$$WCR = \frac{CW}{TW} \times 100\%$$ + +$$SCR = \frac{CS}{TS} \times 100\%$$ + +*WCR*: word correct rate + +*SCR*: sentence correct rate + +*CW*: the number of correct words + +*TW*: the number of total words + +*CS*: the number of correct sentences + +*TS*: the number of total sentences + +- c) The single-round dialogue accuracy rate. + +$$P_F = \frac{F_1}{F} \times 100\%$$ + +*PF*: single round dialogue accuracy rate + +*F1*: number of dialogues with accurate recognition and fuzzy recognition + +*F*: total number of single round dialogues + +- d) The following recognition rate for the multi-round conversations. + +$$P_D = \frac{D_1}{D} \times 100\%$$ + +*PD*: the following recognition rate + +*D1*: the number of rounds for multi-round dialogues which are correctly recognized + +*D*: the total number of rounds for multi-round dialogues which are tested + +- e) The single-round dialogue accuracy rate. +- f) The common formulas for accuracy rate and completion rate mentioned in this Recommendation. + +$$P_a = \frac{A_1}{A} \times 100\%$$ + +$P_a$ : accuracy rate + +$A_1$ : the number of correct samples + +$A$ : the total number of samples + +$$P_G = \frac{G_1}{G} \times 100\%$$ + +$P_G$ : completion rate + +$G_1$ : the actual number of completed samples + +$G$ : the total number of samples + +- g) The average time to complete the retrievals. + +$$P_c = \frac{C_1}{C} \times 100\%$$ + +$P_c$ : the average time + +$C_1$ : total time to complete retrievals + +$C$ : the total number of tasks + +## Bibliography + +- [b-ISO/IEC 2382] ISO/IEC 2382:2015, *Information technology – Vocabulary*. +- [b-ISO/IEC 22989] ISO/IEC 22989:2022, *Information technology – Artificial intelligence – Artificial intelligence concepts and terminology*. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.47-202503-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/F/T-REC-F.748.47-202503-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c1629394c02f697f60ae954ad36bf274fa8082cb --- /dev/null +++ b/marked/F/T-REC-F.748.47-202503-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3763acd7a0ed56aa30f3f90468c1fb0cb54573eb9045bca3eddf8976b4f0f41d +size 7219 diff --git a/marked/F/T-REC-F.748.47-202503-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.748.47-202503-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c980989257ac9f6483143e8985041d11ade2b0a5 --- /dev/null +++ b/marked/F/T-REC-F.748.47-202503-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0bd6b5ee1831f952e8578873f64db77128fed9b8685e91a519405a42dbc5f4f7 +size 84674 diff --git a/marked/F/T-REC-F.748.47-202503-I_PDF-E/raw.md b/marked/F/T-REC-F.748.47-202503-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..b360184cb14f30fd0109f52d88da110885e143d3 --- /dev/null +++ b/marked/F/T-REC-F.748.47-202503-I_PDF-E/raw.md @@ -0,0 +1,413 @@ + + +# Recommendation **ITU-T F.748.47 (03/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +## **Functional requirements for artificial intelligence development within the artificial intelligence cloud platform – Automated machine learning** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.47 + +# Functional requirements for artificial intelligence development within the artificial intelligence cloud platform – Automated machine learning + +## Summary + +Recommendation ITU-T F.748.47 focuses on the automated machine learning (AutoML) functional capabilities for artificial intelligence (AI) development within the AI cloud platform. This Recommendation covers the terminology, features and reference design of an AI cloud platform using AutoML to enable the development of business-oriented intelligent multimedia applications. + +The purpose of this Recommendation is to provide a general technical specification to facilitate AutoML techniques that could better assist in the development of AI multimedia applications. + +## History\* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.47 | 2025-03-16 | 21 | 11.1002/1000/16237 | + +## Keywords + +Artificial intelligence, automated machine learning, deep learning, machine learning. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2025 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|---------------------------------------------------------------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Functional requirements for artificial intelligence development within the AI cloud platform: Automated machine learning..... | 2 | +| 6.1 Data processing ..... | 3 | +| 6.2 Model development ..... | 4 | +| 6.3 Model deployment..... | 6 | +| 6.4 Management and maintenance ..... | 6 | +| Bibliography..... | 8 | + + + +## Functional requirements for artificial intelligence development within the artificial intelligence cloud platform – Automated machine learning + +# 1 Scope + +This Recommendation provides the common requirements for automated machine learning (AutoML) for artificial intelligence (AI) development within the AI cloud platform. It addresses the following subjects: + +- The framework of AutoML in AI cloud platforms. +- Functional requirements for data processing, model development, model deployment, and management and maintenance. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 automated machine learning** [b-ITU-T F.748.17]: The process of automating the time consuming, iterative tasks of machine learning model development. It allows developers to build machine learning models automatically with high scale, efficiency, and productivity all while sustaining model quality. + +**3.1.2 deep learning** [b-ITU-T F.748.12]: A representation learning method, used to model high-level abstractions in data through the use of model architectures, which are composed of multiple nonlinear transformations. + +**3.1.3 inference** [b-ISO/IEC 22989]: In artificial intelligence, a premise is either a fact, a rule, a model, a feature or raw data. + +**3.1.4 machine learning** [b-ITU-T Y.3172]: Processes that enable computational systems to understand data and gain knowledge from it without necessarily being explicitly programmed. + +NOTE 1 – This definition is adapted from [b-ETSI GR ENI 004]. + +NOTE 2 – Supervised machine learning and unsupervised machine learning are two examples of machine learning types. + +## 3.2 Terms defined in this Recommendation + +None. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|--------|-----------------------------------------------------------------| +| AI | Artificial Intelligence | +| API | Application Programming Interface | +| AutoML | Automated Machine Learning | +| BOHB | Bayesian Optimization and Hyperband | +| DARTS | Differentiable Architecture Search | +| ENAS | Efficient Neural Architecture Search | +| GDAS | Gradient-based Search using Differentiable Architecture Sampler | +| ICE | Individual Conditional Expectation Plot | +| Isomap | Isometric Feature Mapping | +| NMS | Non-Maximum Suppression | +| PBT | Population-based Training | +| PDP | Partial Dependence Plot | +| SHAP | Shapley Additive Explanations | +| SPOS | Single Path One-Shot | + +# 5 Conventions + +In this Recommendation: + +- The keywords "**is recommend to**" indicate a requirement that is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. + +# 6 Functional requirements for artificial intelligence development within the AI cloud platform: Automated machine learning + +AutoML enables better integration of business data, attributes, rules, and evaluation metrics to develop business-oriented AI multimedia applications. This Recommendation focuses on the functional capabilities of AutoML tools in cloud platforms. The Recommendation is organized based on the AI multimedia application lifecycle that includes data processing, model development, and model deployment. Meanwhile, the platform needs to provide supporting capabilities to assist in better achieving, understanding, and managing the resources and outcomes involved in AutoML. Figure 1 shows the framework of AutoML in the AI cloud platform. + +![Figure 1: The framework of AutoML in related AI cloud platforms. The diagram shows three main stages: Data processing, Model development, and Model deployment, all supported by a Management and maintenance layer. Data processing includes automated cleaning, preprocessing, and annotation. Model development includes feature engineering, architecture search, hyper-parameter optimization, and ensemble. Model deployment includes compression and iteration. Management and maintenance include interaction methods, resource management, visualization, and version control.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +``` + +graph TD + subgraph Data_processing [Data processing] + D1[Automated data cleaning] --> D2[Automated data preprocessing] + D2 --> D3[Automated data annotation] + end + subgraph Model_development [Model development] + M1[Automated feature engineering] --> M2[Neural architecture search] + M2 --> M3[Hyper-parameter optimization] + M3 --> M4[Automated model ensemble] + end + subgraph Model_deployment [Model deployment] + P1[Automated model compression] --> P2[Automated model iteration] + end + subgraph Management_and_maintenance [Management and maintenance] + MM1[Interaction methods] + MM2[Resource management] + MM3[Visualization] + MM4[Version control] + end + D3 --> M1 + M4 --> P1 + Management_and_maintenance --> M4 + +``` + +Figure 1: The framework of AutoML in related AI cloud platforms. The diagram shows three main stages: Data processing, Model development, and Model deployment, all supported by a Management and maintenance layer. Data processing includes automated cleaning, preprocessing, and annotation. Model development includes feature engineering, architecture search, hyper-parameter optimization, and ensemble. Model deployment includes compression and iteration. Management and maintenance include interaction methods, resource management, visualization, and version control. + +F.748.47(25) + +**Figure 1 – The framework of AutoML in related AI cloud platforms** + +## 6.1 Data processing + +### 6.1.1 Automated data cleaning + +- It is recommended to support automated filtering of data based on predefined thresholds, where thresholds include video/image width, height, and resolution. +- It is recommended to support automated data cleansing for image data, e.g., removing duplicate images, removing blank images. +- It is recommended to support automated data cleaning for text data, e.g., word segmentation, removing stopped words. +- It is recommended to support automated audio cleaning for audio data, e.g., silence removal, noise reduction. + +### 6.1.2 Automated data pre-processing + +- It is recommended to support automated data processing for imbalanced distribution data, e.g., class weighting, downsampling, oversampling. +- It is recommended to support automated data processing of time series data, e.g., smoothing, discontinuous sequence completion, missing value filling. +- It is recommended to support automated data augmentation, e.g., mirror flipping, rotation, scaling, distortion for images. + +### 6.1.3 Automated data annotation + +- It is recommended to support the conversion between different file formats of the dataset and provide an export interface. +- It is recommended to support intelligent annotation, e.g., annotation by pre-trained models, annotation by active learning, and annotation assisted by pre-built algorithms. +- It is recommended to support automated saving and visualization of annotation results. +- It is recommended to support automated bounding box, segmentation, and landmark annotation for image and video data. + +- e) It is recommended to support automated named entity recognition, sentiment labelling, and keyword extraction for text data. +- f) It is recommended to support annotation task assignment, e.g., annotation tracking, annotation version control. +- g) It is recommended to support the modification of annotation results, e.g., re-annotation. + +## **6.2 Model development** + +### **6.2.1 Automated feature engineering** + +Automated feature construction capabilities including the following requirements: + +- a) It is recommended to support automated feature construction capabilities, e.g., mathematical operations, Cartesian product, and binarization. +- b) It is recommended to support automated feature construction capability for time-series data, e.g., window shift, difference, aggregation statistic. +- c) It is recommended to support deep feature synthesis. + +Automated feature extraction capabilities including the following requirements: + +- a) It is recommended to support automated feature extraction capability, e.g., principal component analysis, independent component analysis, isometric feature mapping (Isomap), nonlinear dimensionality reduction, linear discriminant analysis, and singular value decomposition. +- b) It is recommended to support automated feature extraction through pre-trained models, e.g., through neural network pre-trained models, and relational self-coding models. + +Automated feature selection capability including the following requirements: + +- a) It is recommended to support automated feature selection, e.g., filtering, wrapping and embedding methods during the process, and support search strategies, e.g., global search, heuristic search, and random search strategies. +- b) It is recommended to support automated two-stage feature selection, with two-stage search and important feature selection for models that perform better in the first stage. + +Automated feature transformation capabilities including the following requirements: + +- a) It is recommended to support automated feature transformation, e.g., feature scaling (normalization, regularization), feature transformation (function transformation, quantile transformation). + +Automated feature evaluation including the following requirements: + +- a) It is recommended to support automated feature analysis and evaluation, e.g., evaluating feature complexity, and analysing feature importance through model-specific evaluation metrics. + +### **6.2.2 Neural architecture search** + +Search space including the following requirements: + +- a) It is recommended to support common network structure search, e.g., ResNet, Hybrid. +- b) It is recommended to support the customized loss function trained according to the task. + +Search strategy including the following requirements: + +- a) It is recommended to support basic search strategies, e.g., random search, grid search, genetic evolution search, and Bayesian optimization search. + +- b) It is recommended to support a variety of accelerated search strategies, e.g., differentiable architecture search (DARTS), gradient-based search using differentiable architecture sampler (GDAS), efficient neural architecture search (ENAS), single path one-shot (SPOS) algorithm. +- c) It is recommended to support different levels of search, e.g., the level of operators and the level of topology. +- d) It is recommended to support customized search strategies. + +Evaluation strategy including the following requirements: + +- a) It is recommended to support common accuracy evaluation metrics and their visualization, e.g., accuracy, mean square error. +- b) It is recommended to support the setting of early stopping rules, such as the number of iterations, expected loss, expected accuracy, and maximum training time. +- c) It is recommended to support the setting of the proxy evaluator. + +Optimization methods including the following requirements: + +- a) It is recommended to support the rapid application of the searched network structure to different downstream tasks, e.g., the backbone network searched on the classification task could be applied to the detection task and the segmentation task by simple configuration. +- b) It is recommended to support the search of network structure under multiple objectives or constraints, e.g., model size, number of floating-point computations, inference time. +- c) It is recommended to support the search function of data pre-processing and post-processing compatible with network structure search, e.g., data enhancement method, non-maximum suppression (NMS) threshold. + +### **6.2.3 Hyper-parameter optimization** + +- a) It is recommended to support setting search space related parameters, e.g., learning rate. +- b) It is recommended to support common search strategies and the customized search strategies, e.g., grid search, random search, hyperband, population-based training (PBT), and Bayesian optimization and hyperband (BOHB). +- c) It is recommended to support setting evaluation strategies, i.e., supporting common accuracy evaluation metrics and visualizations, e.g., accuracy, mean square error, etc. +- d) It is recommended to support early stopping algorithms, e.g., curve fitting assessor, media stop assessor. +- e) It is recommended to support the setting of early stopping rules, e.g., the maximum search duration, the maximum search duration of a single task, and the maximum search accuracy. +- f) It is recommended to support customized early stop algorithms and early stop rules, e.g., search space filtering rules. +- g) It is recommended to support hyper-parameter optimization based on data sub-sampling (e.g., random sampling, stratified sampling). + +### **6.2.4 Automated model ensemble** + +Model integration strategy including the following requirements: + +- a) It is recommended to support enabling and disabling automated model integration. +- b) It is recommended to support setting and modifying the number of models to be integrated. +- c) It is recommended to support multiple model integration strategies e.g., bagging, boosting and stacking. + +Model integration weights including the following requirements: + +- a) It is recommended to support different weight allocation methods, e.g., voting, averaging and learning. +- b) It is recommended to support automated adjustment of model integration weights based on the assessment metrics. + +## **6.3 Model deployment** + +### **6.3.1 Automated model compression** + +- a) It is recommended to support automated model compression e.g., model quantization, pruning, knowledge distillation, and compression based on parameter sharing. +- b) It is recommended to support format transformation of model files. +- c) It is recommended to support computational graph optimization capabilities, e.g., operator fusion, subgraph splitting, execution scheduling. + +### **6.3.2 Automated model iteration** + +- a) It is recommended to support the upgrading of the inference service, e.g., through grey scale release strategies. +- b) It is recommended to support data iteration based on the existing models to optimize difficult case samples. + +## **6.4 Management and maintenance** + +### **6.4.1 Interaction methods** + +- a) It is recommended to support modelling approaches, e.g., visualization tools, application programming interface (API) interfaces, and command line. +- b) It is recommended to provide documentation and example code for the API interfaces. +- c) It is recommended to equip the command-line tool with support for common automated machine learning tasks, including data preprocessing, model training, and model evaluation. +- d) It is recommended to support modelling templates for multiple parts of the whole process, e.g., modelling templates based on tasks (e.g., image, text, audio, video), scenarios (e.g., binary classification, multi-classification, regression), and business scenarios (e.g., risk control in the financial industry, broadband early warning in the communication industry). + +### **6.4.2 Resource management** + +- a) It is recommended to support resource detection and dynamic adjustment, e.g., automated resource allocation based on model complexity and sample size. +- b) It is recommended to provide resource usage monitoring and alerting functionalities. +- c) It is recommended to provide resource usage reporting and analytics functionalities, including resource consumption patterns and resource allocation strategies. + +### **6.4.3 Visualization** + +- a) It is recommended to support model explainability visualization methods, e.g., SHapley additive exPlanations (SHAP), partial dependence plot (PDP), individual conditional expectation plot (ICE). +- b) It is recommended to support the visualization of the model training process, including training progress, loss function curves, accuracy curves, and other relevant metrics. +- c) It is recommended to provide model structure visualization, including the model's hierarchical structure and parameter settings. + +- d) It is recommended to support data distribution visualization, including the distribution of data points, correlations between features, and other relevant characteristics. +- e) It is recommended to provide model prediction result visualization, including prediction probability distributions, classification boundaries, and other relevant metrics. + +### **6.4.4 Version control** + +- a) It is recommended to support version management of AI assets, e.g., datasets, models, services, and templates. +- b) It is recommended to record each change, including the timestamp and a description of the modification. +- c) It is recommended to establish a comprehensive approval workflow for version releases. +- d) It is recommended to ensure that each released version undergoes rigorous testing and validation to guarantee its stability and reliability. +- e) It is recommended to provide a version rollback feature that allows users to revert to previous versions. +- f) It is recommended to enable comparison analysis between versions to assist users in quickly identifying and resolving issues. + +## Bibliography + +- [b-ITU-T F.748.12] Recommendation ITU-T F.748.12 (2021), *Deep learning software framework evaluation methodology*. +- [b-ITU-T F.748.17] Recommendation ITU-T F.748.17 (2022), *Technical specification for artificial intelligence cloud platform – Artificial intelligence model development*. +- [b-ITU-T Y.3172] Recommendation ITU-T Y.3172 (2019), *Architectural framework for machine learning in future networks including IMT-2020*. +- [b-ETSI GR ENI 004] ETSI GR ENI 004 V3.1.1 (2023), *Experiential Networked Intelligence (ENI); Terminology*. +<> +- [b-ISO/IEC 22989] ISO/IEC 22989:2022, *Information technology – Artificial intelligence – Artificial intelligence concepts and terminology* +<> + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.48-202503-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/F/T-REC-F.748.48-202503-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c1629394c02f697f60ae954ad36bf274fa8082cb --- /dev/null +++ b/marked/F/T-REC-F.748.48-202503-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3763acd7a0ed56aa30f3f90468c1fb0cb54573eb9045bca3eddf8976b4f0f41d +size 7219 diff --git a/marked/F/T-REC-F.748.48-202503-I_PDF-E/b93cbfb52e37619e688175a6aad9edd9_img.jpg b/marked/F/T-REC-F.748.48-202503-I_PDF-E/b93cbfb52e37619e688175a6aad9edd9_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..39ba10a7f855d05d61e5963c76c591bca0bd3f3f --- /dev/null +++ b/marked/F/T-REC-F.748.48-202503-I_PDF-E/b93cbfb52e37619e688175a6aad9edd9_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7f3b512c74a77f30696c524b37e11a3c034b3d4428b1a450bdcb4b7ba174ff6a +size 98560 diff --git a/marked/F/T-REC-F.748.48-202503-I_PDF-E/raw.md b/marked/F/T-REC-F.748.48-202503-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..7ec0d9174a34cedc3579a29097b3ae33981d7754 --- /dev/null +++ b/marked/F/T-REC-F.748.48-202503-I_PDF-E/raw.md @@ -0,0 +1,489 @@ + + +# Recommendation **ITU-T F.748.48 (03/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +## **Technical specification for artificial intelligence cloud platforms – Artificial intelligence model deployment** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.48 + +# Technical specification for artificial intelligence cloud platforms – Artificial intelligence model deployment + +## Summary + +Recommendation ITU-T F.748.48 specifies a framework for the cloud-based deployment of artificial intelligence (AI) models. It covers the functional requirements of AI cloud platforms to enable the deployment of AI models. It establishes the technical specifications of the platform's supporting functional modules, core functional modules and auxiliary functional modules. + +## History\* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.48 | 2025-03-16 | 21 | 11.1002/1000/16238 | + +## Keywords + +AI, compression, model deployment, monitoring, optimization, requirement. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2025 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|------|---------------------------------------------------------------------|----| +| 1 | Scope ..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 1 | +| 4 | Abbreviations and acronyms ..... | 1 | +| 5 | Conventions ..... | 2 | +| 6 | Framework of AI model deployment ..... | 2 | +| 7 | Requirements of AI model processing ..... | 3 | +| 7.1 | Model conversion ..... | 3 | +| 7.2 | Model optimization ..... | 4 | +| 7.3 | Model evaluation..... | 4 | +| 8 | Requirements of deployment management ..... | 5 | +| 8.1 | Cloud deployment ..... | 5 | +| 8.2 | End-edge deployment..... | 6 | +| | Appendix I – An example of AI model deployment on cloud..... | 8 | +| I.1 | Model preparation ..... | 8 | +| I.2 | Model optimization ..... | 8 | +| I.3 | Model evaluation..... | 8 | +| I.4 | Cloud deployment ..... | 8 | +| I.5 | Monitoring and maintenance..... | 9 | +| | Appendix II – An example of AI model deployment on end device ..... | 10 | +| II.1 | End device selection ..... | 10 | +| II.2 | Model preparation ..... | 10 | +| II.3 | Model conversion ..... | 10 | +| II.4 | End device deployment ..... | 10 | +| | Bibliography..... | 11 | + + + +# Recommendation ITU-T F.748.48 + +# Technical specification for artificial intelligence cloud platforms – Artificial intelligence model deployment + +# 1 Scope + +This Recommendation provides the framework of artificial intelligence (AI) model deployment and specifies the functional requirements of AI cloud platforms. + +The scope of this Recommendation includes: + +- the framework of AI model deployment; +- requirements of AI model processing; +- requirements of deployment management. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[[ITU-T F.748.17](#)] Recommendation ITU-T F.748.17 (2022), *Technical specification for artificial intelligence cloud platform – Artificial intelligence model development*. + +[[ITU-T F.748.38](#)] Recommendation ITU-T F.748.38 (2024), *Technical specification for artificial intelligence cloud platform: General architecture*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following term defined elsewhere: + +**3.1.1 artificial intelligence** [b-ISO/IEC 2382]: An interdisciplinary field, usually regarded as a branch of computer science, dealing with models and systems for the performance of functions generally associated with human intelligence, such as reasoning and learning. + +## 3.2 Terms defined in this Recommendation + +None. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|-----------------------------------| +| AI | Artificial Intelligence | +| API | Application Programming Interface | +| AUC | Area Under the Curve | +| CPU | Central Processing Unit | +| HTTP | Hypertext Transfer Protocol | + +| | | +|-------|-----------------------------------------| +| MLOPS | Machine Learning Operations | +| OAuth | Open Authorization | +| ROC | Receiver Operating Characteristic Curve | +| RPC | Remote Procedure Call | +| WebUI | Web User Interface | + +# 5 Conventions + +In this Recommendation: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. +- The keywords "can optionally" indicate an optional requirement which is permissible, without implying any sense of being recommended. These terms are not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +# 6 Framework of AI model deployment + +Model deployment refers to the application of trained AI models in real-world scenarios to make predictions or decisions based on input data. It is a crucial functional module within an AI cloud platform which supports the entire AI model lifecycle [ITU-T F.748.38]. As shown in Figure 6-1, the AI model lifecycle primarily includes three phases: model development, model deployment and model operation. Prior to deployment, the trained model is required to be prepared and tested with appropriate datasets to ensure readiness for production. The functional requirements for model development are specified in [ITU-T F.748.17]. Once deployed, continuous operation is needed to ensure that the AI model maintains its performance over time. + +Typically, the model deployment module of an AI cloud platform is recommended to include two main components: model processing and deployment management. + +Model processing is intended to support the effective use of AI models across different frameworks, platforms and devices, improving their performance and accessibility. The functional modules for model processing include: + +- **Model conversion:** This module is used to convert the trained AI model into a format which is compatible with the target hardware or software platform. +- **Model optimization:** This module is used to optimize the AI model to accelerate the inference process, improving both speed and efficiency. +- **Model evaluation:** This module is used to evaluate the performance of the AI model both before and after deployment, ensuring its readiness and effectiveness in the production environment. + +Deployment management is intended to support the deployment of AI models on cloud servers, edge devices and end devices for inference. The functional modules for deployment management include: + +- **Online service:** This module is used to deploy an AI model as an online service which can be accessed by users in real time. + +- **Batch service**: This module is used to deploy an AI model as a batch service which can process large volumes of data in batches. +- **Offline service**: This module is used to deploy an AI model on local devices, ensuring access to model services in environments where network connectivity is unavailable or unreliable. +- **End-edge-cloud collaborative service**: This module is used to deploy an AI model on end or edge devices that can be managed collaboratively in the cloud. + +![Figure 6-1 – Framework of model deployment](b93cbfb52e37619e688175a6aad9edd9_img.jpg) + +``` + + graph LR + subgraph MDV [Model development] + direction TB + DM[Data management] + AD[Algorithm development] + MT[Model training] + MM[Model management] + end + + subgraph MDP [Model deployment] + direction LR + subgraph MP [Model processing] + direction TB + MC[Model conversion] + MO[Model optimization] + ME[Model evaluation] + end + subgraph DMGT [Deployment management] + direction TB + subgraph CD [Cloud deployment] + OS[Online services] + BS[Batch services] + end + subgraph EED [End-edge deployment] + OFS[Offline services] + ECCS[End-edge-cloud collaborative services] + end + end + MP --> DMGT + end + + subgraph MOP [Model operation] + direction TB + MMON[Model monitoring] + LC[Log collection] + IR[Inference results] + DR[Deployment record] + end + + MDV --> MDP + MDP --> MOP + +``` + +NOTE – The solid lines represent the modules defined in this Recommendation. The dashed lines represent upstream and downstream modules related to those defined in this Recommendation, but which are not within its scope. Module names are highlighted with blue padding and function names are listed without padding. + +**Figure 6-1 – Framework of model deployment** + +Figure 6-1 – Framework of model deployment + +### 7 Requirements of AI model processing + +#### 7.1 Model conversion + +Model conversion refers to the process of converting a model's architecture, parameters and weights into a format compatible with the target software or hardware platforms. The requirements for model conversion are as follows: + +- **Conversion algorithm management**: It is recommended to support the management of conversion algorithms, offering preset conversion algorithms and enabling the upload of custom conversion algorithms. +- **Conversion task configuration**: It is required to support the configuration of model conversion tasks to specify the source model, target format, conversion algorithm and computing resources. It is recommended to provide the information of framework compatibility and conversion limitations to enable efficient cross-framework model conversion. +- **Task management**: It is required to enable users to create, delete, update, start or terminate tasks, save or download task results, and view the status of tasks. +- **Resource monitoring**: It is recommended to support real-time monitoring of resource usage during task execution. +- **Log management**: It is recommended to support log management, providing capabilities to collect, store, download and analyse detailed logs of task execution. It is recommended to + +Rec. ITU-T F.748.48 (03/2025) 3 + +support log search functionality to search by keywords, dates or other relevant criteria to quickly locate specific entries and efficiently diagnose task-related issues. + +NOTE – Task management, resource monitoring and log management are generally applicable to all the following modules and will not be repeated in this Recommendation. + +## 7.2 Model optimization + +Model optimization refers to the process of reducing the resource consumption or improving the inference speed of an AI model. It is essential for deploying models on resource-constrained devices. The requirements for model optimization are as follows: + +- **Optimization algorithm management:** It is recommended to support the management of optimization algorithms, offering preset optimization algorithms and enabling the upload of custom optimization algorithms. +- **Optimization task configuration:** It is required to support the configuration of model optimization tasks to specify the model, optimization algorithm, optionally select an evaluation dataset, configure relevant parameters, and specify any required computing resources. It can optionally support the defining of optimization goals, including model size and inference speed, and to suggest or apply appropriate optimization techniques based on model type and deployment target to achieve these goals. +- **Optimization result visualization:** It is recommended to support the visualization of model optimization results, displaying performance metrics before and after optimization, including accuracy, inference speed and model size. It is recommended to support result comparison from different model optimization tasks to compare the effects of different optimization algorithms. Additionally, it can optionally provide explanatory tools to clarify how optimization impacts specific parts of the model and identify potential bottlenecks for further optimization. + +## 7.3 Model evaluation + +Model evaluation refers to the process of comprehensively analysing and assessing the accuracy and effectiveness of a model before or after deployment. Model evaluation is recommended to support two evaluation modes: + +- **Offline evaluation:** Offline evaluation refers to the process of evaluating a model's performance using historical data before deployment in a production environment. +- **Online evaluation:** Online evaluation refers to the process of evaluating a model's performance in real time after deployment in a production environment. + +The requirements for model evaluation are as follows: + +- **Evaluation algorithm management:** It is recommended to support the management of evaluation algorithms, offering preset evaluation algorithms and enabling the upload of custom evaluation algorithms. +- **Evaluation task configuration:** It is required to support the configuration of evaluation tasks to specify the evaluation mode (online or offline), select one or multiple models to evaluate, choose the evaluation algorithm, select the evaluation dataset and specify the required computing resources. It is recommended to manage the evaluation dataset configuration according to the selected evaluation mode. For online evaluation, it is recommended to support specifying the deployed inference service to collect the evaluation dataset and configuring the traffic ratio for data collection. For offline evaluation, it is recommended to support uploading the offline evaluation dataset. +- **Evaluation result visualization:** It is recommended to provide comprehensive visualizations of model evaluation results, displaying key metrics including accuracy, precision, recall, F1-score and ROC-AUC, with the flexibility for users to define and track custom metrics + +based on specific requirements. It is recommended to offer detailed views that allow users to compare model predictions with actual results for each data point, facilitating in-depth error analysis and performance insight. Additionally, it is recommended to support the comparison of evaluation results across different models, enabling users to effectively compare model performance and select the best performing model for their applications. + +- **Manual evaluation:** It can optionally support manual evaluation, allowing users to review and validate model predictions by inspecting and annotating specific samples or outputs. Manual evaluation can help identify subtle issues that may not be detected through automated evaluation and provide more accurate feedback for further model optimization. Additionally, it can optionally compare the results of manual evaluation with automated evaluation for a more comprehensive and precise analysis. + +# 8 Requirements of deployment management + +## 8.1 Cloud deployment + +Cloud deployment refers to deploying AI models on cloud servers, allowing users to access deployed model services through web interfaces. The requirements for cloud deployment are as follows: + +- **Services management:** It is required to enable users to create, delete, update, retrieve, start and terminate deployment services, and to view detailed information about each service. It is recommended to support resource and model configuration during deployment, allowing users to select the model version, specify resource type and size, configure inference scripts and utilize advanced features tailored to specific service needs. +- **Autoscaling:** Autoscaling is crucial for adjusting the service's resource capacity automatically in response to demand. It is recommended to support autoscaling to handle fluctuating traffic and workloads, ensuring that resources are neither over-provisioned nor under-provisioned. +- **Inference engine integration:** It can optionally support integration with high-performance inference engines, including those optimized for small models and specialized engines designed for large models, utilizing distributed inference and model parallelism to optimize performance and efficiently handle complex tasks. +- **MLOPS:** Machine learning operations (MLOPS) combines AI workflows with software development practices to improve the production readiness, repeatability and manageability of AI models. It can optionally integrate MLOPS tools to facilitate the deployment, management and monitoring of AI models in production environments. + +Additionally, cloud deployment is recommended to support deployment modes, including online services and batch services. + +### 8.1.1 Online services + +Online services refer to the process of packing an AI model as a web service, accessible via hypertext transfer protocol (HTTP), remote procedure call (RPC), web user interface (WebUI) or other interfaces. It is recommended to support the following three online inference modes: + +- **Real-time inference:** Real-time inference refers to the process of generating predictions or making decisions in real time. This type of inference is crucial for requests with interactive and low latency requirements. +- **Asynchronous inference:** Asynchronous inference refers to queuing up the incoming requests and processing them asynchronously. There will be a delay between the submission of request and the receipt of results. This type of inference is ideal for requests with large payload sizes, long processing times and near real-time latency requirements. + +- **Serverless inference:** Serverless inference refers to allocating and managing the computation resources automatically to enable users to deploy AI models without configuring or managing the underlying infrastructure and resource capacity. This type of inference is suitable for workflows that have intermittent or unpredictable traffic patterns and can tolerate cold starts. + +Additionally, for online services, it is recommended to support the following functionalities: + +- **Online debugging:** It is recommended to support online debugging to identify and resolve issues in deployed AI services in real time. It supports testing and validating the inference function of AI models in deployed services, ensuring that the results generated by the models are accurate and reliable. +- **Custom start-stop:** It is recommended to support a start-stop functionality, allowing users to schedule specific times for the inference service to start or stop, in accordance with their desired timing and requirements. This functionality automates the management of the inference service, enabling users to optimize resource utilization and reduce costs by running services only when necessary. +- **Inference results collection:** It is recommended to support inference results collection, which gathers and stores data generated during the inference phase. The collected data can be used for performance analysis, model evaluation and debugging. +- **Service authentication:** Service authentication ensures the security and data privacy of online inference services. It is recommended to support authentication methods, including username/password, Open Authorization (OAuth) and application programming interface (API) keys, with granular access control. Additionally, it is recommended to support secure token management, log auditing, revocation mechanisms and data encryption. + +### 8.1.2 Batch services + +Batch services refer to the process of performing inference on a batch of data, which is suitable for offline processing when a large amount of data is available in advance and a persistent end point is not necessary. The requirements for batch services are as follows: + +- **Batch data uploading:** It is recommended to support batch data uploading, enabling easy upload of large datasets in various formats and input methods, ensuring flexibility in data preparation for batch services. +- **Distributed data processing:** It is recommended to seamlessly feed the uploaded data into the batch services for inference, using distributed computing to efficiently handle large volumes of data and accelerate inference through batch processing. +- **Results management:** It is recommended to support results management of batch services, enabling users to view and download processed data for further analysis or integration. It is recommended to organize the results for easy access, with features for filtering, sorting and exporting the results, to enhance user experience and workflow efficiency. +- **Breakpoint resumption:** Since batch services require a significant amount of time to process, it is recommended to support resuming from a breakpoint after a service interruption. This functionality ensures that when a service is unexpectedly interrupted or encounters a failure, the service can resume processing from the point where it stopped rather than starting from the beginning. + +## 8.2 End-edge deployment + +End-edge deployment refers to deploying AI models on end or edge devices, enabling local processing and inference close to the data source. End-edge deployment is recommended to support deployment modes, including offline services and end-edge-cloud collaborative services. + +### 8.2.1 Offline services + +The offline services refer to deploying a trained model on local devices, allowing model services to be accessed in the environments where network connectivity is not available or unreliable. The requirements for offline services are as follows: + +- **Offline services deployment:** This functionality refers to packaging trained models and their related dependencies into installation packages that are easy to deploy offline. It is recommended to allow users to select the model version and deployment target environment, including hardware chip type and operating system. It is recommended to generate a complete deployment package automatically, including model files, configuration files and dependency libraries. Additionally, it can optionally support necessary encryption during the model release process to ensure the security of the model and data. +- **Offline services management:** It is required to support users in viewing and managing deployed offline services, including stopping, updating, deleting and downloading services. It can optionally support offline service version control, allowing users to switch between or roll back to different versions of the models. Additionally, it can optionally support download permission control to ensure the security of deployment package transmission and storage. + +### 8.2.2 End-edge-cloud collaborative services + +The end-edge-cloud collaborative services refer to services deployed on the end or edge devices that can be managed collaboratively in the cloud. The requirements for end-edge-cloud collaborative services are as follows: + +- **Device management:** This module provides a centralized system to manage end and edge devices used for deployment. It is required to support device registration, allowing the input of device details including device identifier, type, chip model and operating system. For newly registered or offline devices, it is recommended to support activation operations, including generating and copying activation commands, which can be executed on the local device to complete the activation. +- **Deployment packages management:** This module is used for managing the deployment package for end-edge-cloud collaborative services and distributing deployment packages to devices. It is required to support the creation of deployment packages, which assemble all the necessary components including model files, configuration files and dependency libraries. It is recommended to support deleting outdated or unnecessary packages, updating existing packages to incorporate new features or address issues, and distributing the packages to activated and connected devices. +- **Resource utilization management:** This module is designed for real-time monitoring and management of resource usage on end or edge devices, including central processing units (CPUs), memory, AI acceleration cards, storage and network bandwidth, to ensure efficient utilization and proper operation of the devices. It is recommended to provide a device agent to establish a connection between the device and the platform, allowing real-time access to the device's operating status and resource usage. It is also recommended to support viewing device resource usage, deployment package status and other relevant information. Additionally, it is recommended to support removing unnecessary deployment packages from the device to optimize performance and free up resources. + +# Appendix I + +## An example of AI model deployment on cloud + +(This appendix does not form an integral part of this Recommendation.) + +This appendix provides an example of deploying a large language model Llama3 on the cloud using AI cloud platforms to provide services like dialogue generation, text generation and code generation. + +The model is optimized and accelerated to enhance the model's inference speed and overall performance. Additionally, the model is evaluated to ensure it meets the specified requirements for performance and accuracy. + +### I.1 Model preparation + +- **Obtain model:** Obtain the pre-trained Llama3 model. +- **Import model:** Import the model into the model management module by uploading model files and the container image. + +### I.2 Model optimization + +Use the model optimization module to optimize and accelerate the model. + +- **Create optimization task:** Create a model optimization task, select the Llama3 model and optimize algorithms, such as quantization or pruning, and select the computing resources required for the task. +- **Monitor optimization task:** Monitor progress and view execution logs and resource usage via the optimization task details page during task execution. +- **Review optimization results:** Review optimization results and performance metrics, such as accuracy, inference speed and model size. +- **Save model:** Save the optimized model as a new model. + +### I.3 Model evaluation + +Use the model evaluation module to evaluate the model. + +- **Create evaluation task:** Create a model evaluation task, select the offline evaluation mode, choose the optimized Llama3 model, select the evaluation algorithm, upload the evaluation dataset and select the computing resources required for the task. +- **Monitor evaluation task:** Monitor task logs and metrics from the evaluation task details page during task execution. +- **Review evaluation results:** Review evaluation results, such as accuracy, precision and recall. +- **Compare evaluation results:** Compare and analyse the results of multiple models in a comparative view if needed. + +### I.4 Cloud deployment + +Use the cloud deployment module to deploy the optimized Llama3 model on cloud. + +- **Model deployment:** Deploy the optimized model as an online service, selecting different inference modes like real-time, asynchronous or serverless based on requirements. Choose suitable computing resources based on inference requirements, starting with smaller instances and expand it when needed. + +- **Obtain model service:** Obtain the model service after deployment, such as web APIs or other interfaces. +- **Test model service:** Use the online debugging module to test the model service, such as verifying that the service can receive text and return accurate results in the expected format. + +### **I.5 Monitoring and maintenance** + +- **Monitor:** Monitor the performance of the model service continuously with the monitoring module, tracking metrics such as latency, throughput and resource utilization. +- **Maintain:** Periodically update or retrain the model to improve performance, using the MLOPS module to manage and deploy updates seamlessly to ensure continued service optimization. + +# Appendix II + +## An example of AI model deployment on end device + +(This appendix does not form an integral part of this Recommendation.) + +This appendix provides an example of deploying an image classification model on an end device for real-time image recognition using an AI cloud platform. + +### II.1 End device selection + +Select a suitable hardware platform, such as a mobile device (phones, tablets) or embedded device (Raspberry Pi, microcontrollers), based on computing power, storage capacity and power efficiency. + +### II.2 Model preparation + +- **Obtain model:** Obtain a pre-trained image classification model suitable for the target device's computing capabilities. +- **Import model:** Import the model into the model management module by uploading model files and the container image. + +### II.3 Model conversion + +Use the model conversion module to convert the model from its original framework to the framework supported by the target device. + +- **Create conversion task:** Create a model conversion task, select the model that needs to be converted, target framework, converse script and other advanced parameters, as well as specifying the computing resources required for the task. +- **Monitor conversion task:** Monitor progress and view execution logs and resource usage via the conversion task details page. +- **Save model:** Save the converted model as a new model. + +### II.4 End device deployment + +Use the end-edge-cloud collaborative services module to deploy the converted model on an end device. + +- **Register device:** Register the end device in the device management module by uploading the device details such as device identifier, type, chip model and operating system. +- **Activate device:** Obtain the device activation command, copy it to the end device and execute it to complete activation. +- **Model deployment:** Create and deploy an end-edge-cloud collaborative package, selecting the model and deployment target environment. Then, distribute the package to the activated device. +- **Monitor deployment status:** Monitor the deployment status in the resource utilization management module, such as "in progress," "deployment failed" or "deployment successful". +- **Obtain model service:** Obtain the model service from the resource utilization management module after deployment. +- **Test model service:** Test the model service by sending images for classification and check if the results are accurate and in the expected format. + +# Bibliography + +[b-ISO/IEC 2382] ISO/IEC 2382:2015, *Information technology – Vocabulary*. + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.49-202503-I_PDF-E/0b87abe67b21a93777287649c33e755d_img.jpg b/marked/F/T-REC-F.748.49-202503-I_PDF-E/0b87abe67b21a93777287649c33e755d_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3cff4b863d54647c0ae05958edfd4936b32f05a6 --- /dev/null +++ b/marked/F/T-REC-F.748.49-202503-I_PDF-E/0b87abe67b21a93777287649c33e755d_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1a8a500f4ae8b647bc0385b0cc093da85cd32d7d2aaa3265a468bc906d2cafec +size 23572 diff --git a/marked/F/T-REC-F.748.49-202503-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.748.49-202503-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..38bdd7bc1d3581fbdc3cf6b035f618693dbe3846 --- /dev/null +++ b/marked/F/T-REC-F.748.49-202503-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:75efa8e3513a799767c04a0bb7bcd87a550f2378f5880629068581808c1c9f7f +size 7353 diff --git a/marked/F/T-REC-F.748.49-202503-I_PDF-E/d0c9888da9078a08d93712fd0e113a88_img.jpg b/marked/F/T-REC-F.748.49-202503-I_PDF-E/d0c9888da9078a08d93712fd0e113a88_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9ad7da2ba4a60c8570364de44548fa34a8ad58cd --- /dev/null +++ b/marked/F/T-REC-F.748.49-202503-I_PDF-E/d0c9888da9078a08d93712fd0e113a88_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f2229fff18f3fa48c482e6a42289107a45d11cfefbf0867fa983d486fe08ba13 +size 14608 diff --git a/marked/F/T-REC-F.748.49-202503-I_PDF-E/raw.md b/marked/F/T-REC-F.748.49-202503-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..ba2e55e38f165f85b0db488dffbace7630f47c64 --- /dev/null +++ b/marked/F/T-REC-F.748.49-202503-I_PDF-E/raw.md @@ -0,0 +1,1013 @@ + + +# Recommendation **ITU-T F.748.49 (03/2025)** + +SERIES F: Non-telephone telecommunication services +Multimedia services + +--- + +## **Architecture and protocols of multi-algorithm scheduling systems** + +![ITU logo](390120de4fe440c42fea8154fcaad334_img.jpg) + +The logo of the International Telecommunication Union (ITU), featuring the letters 'ITU' in blue inside a circle with a globe-like grid pattern. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.49 + +# Architecture and protocols of multi-algorithm scheduling systems + +## Summary + +Recommendation ITU-T F.748.49 specifies the architecture and protocols of the multi-algorithm scheduling (MAS) systems for artificial intelligence (AI)-based multimedia applications, which meet the requirements and framework of MAS systems specified in Recommendation ITU-T F.748.36. The MAS systems consist of the scheduling unit, the algorithm warehouse unit, and the computing power resource management unit. The MAS systems schedule multiple algorithms (multi-type, multi-version, or from multiple providers) across heterogeneous computing power resources based on specific tasks on-demand. This Recommendation is intended to guide the design, development, implementation and application of MAS systems. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.49 | 2025-03-16 | 21 | 11.1002/1000/16239 | + +## Keywords + +Algorithm, applications, architecture, artificial intelligence, computing power resource, interfaces, multi-algorithm scheduling, multimedia, protocols, reference points. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2025 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|------|---------------------------------------------------------------------|----| +| 1 | Scope ..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 2 | +| 4 | Abbreviations and acronyms ..... | 2 | +| 5 | Conventions ..... | 3 | +| 6 | Functional architecture and reference points..... | 3 | +| 6.1 | Functional architecture of multi-algorithm scheduling systems ..... | 3 | +| 6.2 | Reference points of multi-algorithm scheduling systems ..... | 4 | +| 7 | Overall requirements of protocols ..... | 5 | +| 7.1 | Structure of protocols ..... | 5 | +| 7.2 | REST service and HTTP method ..... | 6 | +| 7.3 | REST service resource access URI ..... | 6 | +| 7.4 | Template of protocol definition..... | 6 | +| 7.5 | Security requirements ..... | 7 | +| 8 | Interface descriptions..... | 7 | +| 8.1 | Common interfaces..... | 7 | +| 8.2 | Interfaces of reference point Am..... | 7 | +| 8.3 | Interfaces of reference point Sa..... | 8 | +| 8.4 | Interfaces of reference point Cs..... | 8 | +| 8.5 | Interface of reference point Ac..... | 9 | +| 9 | Relevant protocol definitions..... | 9 | +| 9.1 | POST /MAS/Register ..... | 9 | +| 9.2 | POST /MAS/DeRegister ..... | 9 | +| 9.3 | POST /MAS/Keepalive ..... | 10 | +| 9.4 | POST /MAS/TimeSync ..... | 10 | +| 9.5 | GET /MAS/Resource..... | 10 | +| 9.6 | POST /MAS/AlgoPackage ..... | 11 | +| 9.7 | PUT /MAS/AlgoPackage/ ..... | 11 | +| 9.8 | GET /MAS/AlgoPackage/ ..... | 11 | +| 9.9 | DELETE /MAS/AlgoPackage/ ..... | 12 | +| 9.10 | GET /MAS/AlgoPackage ..... | 12 | +| 9.11 | POST /MAS/AlgoTask..... | 12 | +| 9.12 | PUT /MAS/AlgoTask/ ..... | 13 | +| 9.13 | DELETE /MAS/AlgoTask/ ..... | 13 | +| 9.14 | GET /MAS/AlgoTask..... | 13 | +| 9.15 | POST /MAS/AlgoTaskInstance ..... | 14 | + +| | Page | +|-------------------------------------------------------|------| +| 9.16 DELETE /MAS/AlgoTaskInstance/ ..... | 14 | +| 9.17 GET /MAS/AlgoTaskInstance ..... | 14 | +| Annex A – Object descriptions ..... | 15 | +| A.1 ResponseStatus ..... | 15 | +| A.2 Register ..... | 15 | +| A.3 DeRegister ..... | 15 | +| A.4 Keepalive ..... | 15 | +| A.5 TimeSync ..... | 16 | +| A.6 Resource ..... | 16 | +| A.7 AlgoPackage ..... | 16 | +| A.8 AlgoPackageList ..... | 17 | +| A.9 AlgoTask ..... | 17 | +| A.10 AlgoTaskList ..... | 17 | +| A.11 AlgoTaskCtrl ..... | 18 | +| A.12 AlgoTaskInstance ..... | 18 | +| A.13 AlgoTaskInstanceList ..... | 18 | +| Annex B – Algorithm package format descriptions ..... | 19 | +| Bibliography ..... | 20 | + +# Recommendation ITU-T F.748.49 + +## Architecture and protocols of multi-algorithm scheduling systems + +## 1 Scope + +This Recommendation specifies the architecture and protocols of the multi-algorithm scheduling (MAS) systems for artificial intelligence-based multimedia applications, which meet the requirements and framework of MAS systems specified in [ITU-T F.748.36]. + +The scope of this Recommendation includes: + +- functional architecture and reference points; +- interface descriptions and relevant protocols definitions; and +- algorithm package format descriptions. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[[ITU-T F.748.36](#)] Recommendation ITU-T F.748.36 (2024), *Requirements and framework of multi-algorithm scheduling systems*. + +[[ITU-T H.627.3](#)] Recommendation ITU-T H.627.3 (2022), *Protocols for intelligent video surveillance systems*. + +[IETF RFC 7616] IETF RFC 7616 (2015), *HTTP Digest Access Authentication*. + +[IETF RFC 9110] IETF RFC 9110 (2022), *HTTP Semantics*. + +[IETF RFC 9112] IETF RFC 9112 (2022), *HTTP/1.1*. + +[ISO 8601-1] ISO 8601-1 (2019), *Date and time – Representations for information interchange – Part 1: Basic rules*. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 algorithm** [b-ISO/IEC 11557]: A set of rules for transforming the logical representation of data. + +**3.1.2 algorithm package** [ITU-T F.748.36]: A set of algorithmic programs and its formal description information that implement specific functions. + +NOTE – The algorithmic program is executable software composed of algorithm and runtime environment. + +**3.1.3 algorithm warehouse** [ITU-T F.748.36]: A system with unified management of algorithm packages, including import, storage, retrieval, download, etc. + +**3.1.4 artificial intelligence** [b-ISO/IEC 2382]: Interdisciplinary field, usually regarded as a branch of computer science, dealing with models and systems for the performance of functions generally associated with human intelligence, such as reasoning and learning. + +**3.1.5 computing power resource** [ITU-T F.748.36]: A resource that provides data processing and computing capabilities for various applications. + +**3.1.6 functional entity** [b-ITU-T Y.2012]: An entity that comprises an indivisible set of specific functions. Functional entities are logical concepts, while groupings of functional entities are used to describe practical, physical implementations. + +**3.1.7 multi-algorithm scheduling** [ITU-T F.748.36]: A method that schedules multiple algorithms (multi-type, multi-version, or from multiple providers) across heterogeneous computing power resources based on specific tasks on-demand. + +**3.1.8 process instance** [b-ISO/IEC 33001]: Single specific and identifiable execution of a process. + +**3.1.9 reference point** [b-ITU-T Y.2012]: A conceptual point at the conjunction of two non-overlapping functional entities that can be used to identify the type of information passing between these functional entities. + +NOTE – A reference point may correspond to one or more physical interfaces between pieces of equipment. + +**3.1.10 resource** [b-ISO/IEC 2382]: Any element of a data processing system needed to perform required operations. + +**3.1.11 scheduling** [b-ITU-T X.746]: The method of controlling the timing of the execution of a scheduled activity within or represented by a managed object. + +### 3.2 Terms defined in this Recommendation + +None. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-------|------------------------------------------------------| +| AI | Artificial Intelligence | +| AIMA | Artificial Intelligence based Multimedia Application | +| AWU | Algorithm Warehouse Unit | +| CPRMU | Computing Power Resource Management Unit | +| CPU | Central Processing Unit | +| FE | Functional Entity | +| GPU | Graphics Processing Unit | +| HTTP | Hypertext Transfer Protocol | +| HTTPS | Hypertext Transfer Protocol Secure | +| ID | Identifier | +| IP | Internet Protocol | +| JSON | JavaScript Object Notation | +| MAS | Multi-algorithm Scheduling | +| NTP | Network Time Protocol | + +| | | +|-------|-------------------------------------| +| REST | Representational State Transfer | +| RTSP | Real Time Streaming Protocol | +| SU | Scheduling Unit | +| TCP | Transmission Control Protocol | +| URI | Uniform Resource Identifier | +| UTF-8 | 8-bit Unicode Transformation Format | + +## 5 Conventions + +In this Recommendation: + +- The keywords "**is required to**" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The keywords "**is recommended**" indicate a requirement which is recommended but which is not absolutely required. Thus this requirement needs not be present to claim conformance. +- The keywords "**can optionally**" indicate an optional requirement which is permissible, without implying any sense of being recommended. These terms are not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +## 6 Functional architecture and reference points + +### 6.1 Functional architecture of multi-algorithm scheduling systems + +Figure 1 shows the functional architecture of multi-algorithm scheduling (MAS) systems. + +![Functional architecture diagram of multi-algorithm scheduling (MAS) systems. The diagram shows AIMA systems at the top, connected via reference point Am to MAS systems below. The MAS systems box contains three functional entities: SU (Scheduling Unit) at the top, AWU (Algorithm Warehouse Unit) at the bottom left, and CPRMU (Computing Power Resource Management Unit) at the bottom right. Reference points Sa, Cs, and Ac connect these entities: Sa between SU and AWU, Cs between SU and CPRMU, and Ac between AWU and CPRMU.](0b87abe67b21a93777287649c33e755d_img.jpg) + +``` + +graph TD + subgraph AIMA_systems [AIMA systems] + end + subgraph MAS_systems [MAS systems] + SU[SU] + AWU[AWU] + CPRMU[CPRMU] + SU <--> |Sa| AWU + SU <--> |Cs| CPRMU + AWU <--> |Ac| CPRMU + end + AIMA_systems <--> |Am| MAS_systems + +``` + +F.748.49(25) + +Functional architecture diagram of multi-algorithm scheduling (MAS) systems. The diagram shows AIMA systems at the top, connected via reference point Am to MAS systems below. The MAS systems box contains three functional entities: SU (Scheduling Unit) at the top, AWU (Algorithm Warehouse Unit) at the bottom left, and CPRMU (Computing Power Resource Management Unit) at the bottom right. Reference points Sa, Cs, and Ac connect these entities: Sa between SU and AWU, Cs between SU and CPRMU, and Ac between AWU and CPRMU. + +**Figure 1 – Functional architecture of multi-algorithm scheduling systems** + +NOTE – The grey rectangles in Figure 1 represent the functional entities (FEs). + +The functional architecture of MAS systems is based on the requirements and framework defined in [ITU-T F.748.36]. + +The MAS systems are composed of algorithm warehouse unit (AWU), scheduling unit (SU), and computing power resource management unit (CPRMU). + +This Recommendation defines the functional entities (FEs) AWU, SU, and CPRMU in clauses 6.1.1 to 6.1.3, and reference points Am, Sa, Cs, and Ac in clause 6.2. + +#### **6.1.1 Algorithm warehouse unit** + +The AWU is a FE within the MAS systems. + +The functionalities of the AWU are: + +- to provide algorithm package management; +- to provide algorithm package retrieval; +- to provide algorithm package storage. + +#### **6.1.2 Scheduling unit** + +The SU is a FE within the MAS systems. + +The functionalities of the SU are: + +- to provide scheduling strategies management; +- to provide scheduling tasks management; +- to provide scheduling tasks monitoring. + +#### **6.1.3 Computing power resource management unit** + +The CPRMU is a FE within the MAS systems. + +The functionalities of the CPRMU are: + +- to provide computing power resource expansion and contraction; +- to provide process instance management; +- to provide computing power resource monitoring and maintenance. + +### **6.2 Reference points of multi-algorithm scheduling systems** + +This Recommendation defines the external reference points Am and internal reference points Sa, Cs, and Ac of the MAS systems. + +#### **6.2.1 Reference point Am: AIMA systems – MAS systems** + +The reference point Am is between the artificial intelligence-based multimedia application (AIMA) systems and the MAS systems. + +It is used: + +- to send registration, de-registration, keep-alive, and time synchronization commands from AIMA to MAS; +- to send system resource status querying commands from AIMA to MAS; +- to send algorithm package addition, deletion, modification, and querying commands from AIMA to MAS; +- to send algorithm package download commands from AIMA to MAS; +- to send algorithm tasks addition, deletion, and querying commands from AIMA to MAS; +- to send algorithm tasks controlling commands from AIMA to MAS. + +It uses the hypertext transfer protocol (HTTP) [IETF RFC 9110] and [IETF RFC 9112] + JavaScript object notation (JSON) [b-JSON]. + +#### 6.2.2 Reference point Sa: AWU – SU + +The reference points Sa is between the AWU and the SU. + +It is used: + +- to send registration, de-registration, keep-alive, and time synchronization commands from AWU to SU; +- to send system resource status querying commands from SU to AWU; +- to send algorithm package querying commands from SU to AWU. + +It uses HTTP [IETF RFC 9110] and [IETF RFC 9112] + JSON. + +#### 6.2.3 Reference point Cs: SU – CPRMU + +The reference points Cs is between the SU and the CPRMU. + +It is used: + +- to send registration, de-registration, keep-alive, and time synchronization commands from CPRMU to SU; +- to send system resource status querying commands from SU to CPRMU; +- to send process instance (of algorithm tasks) addition, deletion, and querying commands from SU to CPRMU. + +It uses HTTP [IETF RFC 9110] and [IETF RFC 9112] + JSON. + +#### 6.2.4 Reference point Ac: AWU – CPRMU + +The reference points Ac is between the AWU and the CPRMU. + +It is used: + +- to send algorithm package download commands from CPRWU to AWU. + +It uses HTTP [IETF RFC 9110] and [IETF RFC 9112] + JSON. + +## 7 Overall requirements of protocols + +### 7.1 Structure of protocols + +The protocols adopt the representational state transfer (REST) style and transmit through HTTP. The content-type in the message header is required to be set as application/JSON. The message body is described in JSON format, and the character set is 8-bit unicode transformation format (UTF-8). The message response status code is required to comply with the requirements of [IETF RFC 9110]. The protocol interaction and connection method is required to support HTTP long and short connections, and the implementation mechanism is required to comply with the requirements of [IETF RFC 9110]. + +Figure 2 shows the structure of protocols. + +![A diagram showing the structure of protocols as a stack of four layers. The top layer is 'Message body protocols based on JSON', followed by 'HTTP based on REST', then 'TCP', and the bottom layer is 'IP'.](d0c9888da9078a08d93712fd0e113a88_img.jpg) + +| | +|--------------------------------------| +| Message body protocols based on JSON | +| HTTP based on REST | +| TCP | +| IP | + +A diagram showing the structure of protocols as a stack of four layers. The top layer is 'Message body protocols based on JSON', followed by 'HTTP based on REST', then 'TCP', and the bottom layer is 'IP'. + +F.748.49(25) + +**Figure 2 – The structure of protocols** + +### 7.2 REST service and HTTP method + +Table 1 shows the implementation of REST service through HTTP method verbs. + +**Table 1 – HTTP method verbs** + +| HTTP method verbs | Annotation | +|-------------------|--------------------| +| POST | Resource creation | +| GET | Resource retrieval | +| PUT | Resource update | +| DELETE | Resource deletion | + +### 7.3 REST service resource access URI + +The REST service resource access uniform resource identifier (URI) is: + +**{Protocol}://{Hostname}:{Port}/{BaseURI}{ServiceURI}[?(p1=v1)&(p2=v2)...&(pn=vn)]** + +The variable part is represented by "{ }", and the optional part is represented by "[ ]"; the protocol is required to support HTTP; Hostname refers to the Internet protocol (IP) address or domain name of the system; Port can optionally support; BaseURI refers to the name of the system, which is required to be published as "MAS" for multi-algorithm scheduling; ServiceURI refers to the interface resource request URI; ?(p1=v1)&(p2=v2)...&(pn=vn) refers to a query string, where each resource defines required or optional query string parameters that appear in key-value pairs. + +### 7.4 Template of protocol definition + +Table 2 shows the template of protocol definition. + +**Table 2 – Template of protocol definition** + +| URI | | | | +|------------|--------------|----------------------|-----------------------| +| Function | | | | +| Method | Query string | Request message body | Response message body | +| POST | | | | +| GET | | | | +| PUT | | | | +| DELETE | | | | +| Annotation | | | | + +The URI item describes the interface resource request URI; the function item describes the function of the interface; the method item lists the HTTP methods involved in the resource, including POST, GET, PUT and DELETE; the query string describes the key-value pairs of the resources; the request message body item describes the message body of the HTTP request; the response message body describes the response to an HTTP request; the annotation item provides a description of the interface. + +### 7.5 Security requirements + +Both parties involved in information exchange are required to undergo identity authentication. The authentication process is required to comply with the digest access authentication specified in [IETF RFC 7616]. The server random number is required to be reset after exceeding 3600 seconds during digest access authentication. For systems with high security requirements, it is recommended to comply with hypertext transfer protocol secure (HTTPS) [IETF RFC 9110]. + +## 8 Interface descriptions + +### 8.1 Common interfaces + +Before conducting data exchange, the system is required to register first. When registration fails, a random delay of no more than 300 seconds is required to be allowed before re-registration. Both parties are required to configure a unified "heartbeat interval" (default heartbeat interval of 90 seconds). After successful registration, the registrant is required to regularly send heartbeat messages to the registered party according to the "heartbeat interval" for keep-alive. At the same time, both parties are required to configure the "keep-alive timeout parameter" (default heartbeat timeout of 3 times). The registrant is required to perform time calibration during the interaction process and proactively de-register when it stops running. For details on parameters refer to [ITU-T H.627.3]. + +Table 3 shows the common interfaces. + +**Table 3 – The common interfaces** + +| No. | Method | URI | Function | Protocol clause | +|-----|--------|-----------------|----------------------|-----------------| +| 1 | POST | /MAS/Register | Registration | 9.1 | +| 2 | POST | /MAS/DeRegister | De-registration | 9.2 | +| 3 | POST | /MAS/Keepalive | Keep-alive | 9.3 | +| 4 | POST | /MAS/TimeSync | Time synchronization | 9.4 | + +### 8.2 Interfaces of reference point Am + +Table 4 shows the interfaces of reference point Am. + +**Table 4 – The interfaces of reference point Am** + +| No. | Method | URI | Function | Interface provider | Interface caller | Protocol clause | +|-----|--------|------------------------------|------------------------------------------------------------|--------------------|------------------|-----------------| +| 1 | GET | /MAS/Resource | System resource status querying | MAS | AIMA | 9.5 | +| 2 | POST | /MAS/AlgoPackage | Algorithm package addition | MAS | AIMA | 9.6 | +| 3 | PUT | /MAS/AlgoPackage/ | Algorithm package modification | MAS | AIMA | 9.7 | +| 4 | GET | /MAS/AlgoPackage/ | Algorithm package download | MAS | AIMA | 9.8 | +| 5 | DELETE | /MAS/AlgoPackage/ | Algorithm package deletion | MAS | AIMA | 9.9 | +| 6 | GET | /MAS/AlgoPackage | Algorithm package information querying based on conditions | MAS | AIMA | 9.10 | + +**Table 4 – The interfaces of reference point Am** + +| No. | Method | URI | Function | Interface provider | Interface caller | Protocol clause | +|-----|--------|------------------------|---------------------------------------------------------|--------------------|------------------|-----------------| +| 7 | POST | /MAS/AlgoTask | Algorithm task addition | MAS | AIMA | 9.11 | +| 8 | PUT | /MAS/AlgoTask/ | Algorithm task controlling (e.g., start and stop) | MAS | AIMA | 9.12 | +| 9 | DELETE | /MAS/AlgoTask/ | Algorithm task deletion | MAS | AIMA | 9.13 | +| 10 | GET | /MAS/AlgoTask | Algorithm task information querying based on conditions | MAS | AIMA | 9.14 | + +### 8.3 Interfaces of reference point Sa + +Table 5 shows the interfaces of reference point Sa. + +**Table 5 – The interfaces of reference point Sa** + +| No. | Method | URI | Function | Interface provider | Interface caller | Protocol clause | +|-----|--------|------------------|---------------------------------|--------------------|------------------|-----------------| +| 1 | GET | /MAS/Resource | System resource status querying | AWU | SU | 9.5 | +| 2 | GET | /MAS/AlgoPackage | Algorithm package querying | AWU | SU | 9.10 | + +### 8.4 Interfaces of reference point Cs + +Table 6 shows the interfaces of reference point Cs. + +**Table 6 – The interfaces of reference point Cs** + +| No. | Method | URI | Function | Interface provider | Interface caller | Protocol clause | +|-----|--------|------------------------------------|--------------------------------------------------------------------------|--------------------|------------------|-----------------| +| 1 | GET | /MAS/Resource | System resource status querying | CPRMU | SU | 9.5 | +| 2 | POST | /MAS/AlgoTaskInstance | Algorithm task process instance creation | CPRMU | SU | 9.15 | +| 3 | DELETE | /MAS/AlgoTaskInstance/ | Algorithm task process instance deletion | CPRMU | SU | 9.16 | +| 4 | GET | /MAS/AlgoTaskInstance | Algorithm task process instance information querying based on conditions | CPRMU | SU | 9.17 | + +### 8.5 Interface of reference point Ac + +Table 7 shows the interface of reference point Ac. + +**Table 7 – The interface of reference point Ac** + +| No. | Method | URI | Function | Interface provider | Interface caller | Protocol clause | +|-----|--------|------------------------------|----------------------------|--------------------|------------------|-----------------| +| 1 | GET | /MAS/AlgoPackage/ | Algorithm package download | AWU | CPRMU | 9.8 | + +## 9 Relevant protocol definitions + +### 9.1 POST /MAS/Register + +Table 8 shows the protocol of interface POST /MAS/Register. + +**Table 8 – Protocol of interface POST /MAS/Register** + +| URI | /MAS/Register | | | +|-------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|----------------------|-----------------------| +| Function | Registration | | | +| Method | Query string | Request message body | Response message body | +| POST | None | | | +| Annotation | 1 The definition of Register is described in clause A.2;
2 The definition of ResponseStatus is described in clause A.1;
3 Registration is required before the service caller sends a request command to the service provider; the command can be sent only after successful registration; the regularly keep-alive is required to be performed; if the keep-alive fails, the registration is required to be re-performed;
4 The registration command is required to support redirection; after successfully registering with the new address according to the redirection requirements, subsequent commands are required to be sent to the new address. | | | + +### 9.2 POST /MAS/DeRegister + +Table 9 shows the protocol of interface POST /MAS/DeRegister. + +**Table 9 – Protocol of interface POST /MAS/DeRegister** + +| URI | /MAS/DeRegister | | | +|-------------------|------------------------------------------------------------------------------------------------------------------------------|----------------------|-----------------------| +| Function | De-registration | | | +| Method | Query string | Request message body | Response message body | +| POST | None | | | +| Annotation | 1 The definition of DeRegister is described in clause A.3;
2 The definition of ResponseStatus is described in clause A.1. | | | + +### 9.3 POST /MAS/Keepalive + +Table 10 shows the protocol of interface POST /MAS/Keepalive. + +**Table 10 – Protocol of interface POST /MAS/Keepalive** + +| | | | | +|-------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-----------------------------|------------------------------| +| URI | /MAS/Keepalive | | | +| Function | Keep-alive | | | +| Method | Query string | Request message body | Response message body | +| POST | None | | | +| Annotation | 1 The definition of Keepalive is described in clause A.4;
2 The definition of ResponseStatus is described in clause A.1;
3 It is required to support regularly keep-alive after successful registration;
4 It is required to support re-connect registration after unsuccessful keep-alive. | | | + +### 9.4 POST /MAS/TimeSync + +Table 11 shows the protocol of interface POST /MAS/TimeSync. + +**Table 11 – Protocol of interface POST /MAS/TimeSync** + +| | | | | +|-------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-----------------------------|------------------------------| +| URI | /MAS/TimeSync | | | +| Function | Time synchronization | | | +| Method | Query string | Request message body | Response message body | +| POST | None | | | +| Annotation | 1 The definition of TimeSync is described in clause A.5;
2 The definition of ResponseStatus is described in clause A.1;
3 Each of the functional entities is required to keep time synchronization; if there is a time synchronization issue, this command is required to be used to calibrate the system time. | | | + +### 9.5 GET /MAS/Resource + +Table 12 shows the protocol of interface GET /MAS/Resource. + +**Table 12 – Protocol of interface GET /MAS/Resource** + +| | | | | +|-------------------|--------------------------------------------------------|-----------------------------|------------------------------| +| URI | /MAS/Resource | | | +| Function | System resource status querying | | | +| Method | Query string | Request message body | Response message body | +| GET | None | None | | +| Annotation | The definition of Resource is described in clause A.6. | | | + +### 9.6 POST /MAS/AlgoPackage + +Table 13 shows the protocol of interface POST /MAS/AlgoPackage. + +**Table 13 – Protocol of interface POST /MAS/AlgoPackage** + +| | | | | +|-------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-----------------------------|------------------------------| +| URI | /MAS/AlgoPackage | | | +| Function | Algorithm package addition | | | +| Method | Query string | Request message body | Response message body | +| POST | None | | | +| Annotation | 1 The algorithm package is required to be a Zip format file; the description of algorithm package format is shown in Annex B;
2 The definition of ResponseStatus is described in clause A.1. | | | + +### 9.7 PUT /MAS/AlgoPackage/ + +Table 14 shows the protocol of interface PUT /MAS/AlgoPackage/. + +**Table 14 – Protocol of interface PUT /MAS/AlgoPackage/** + +| | | | | +|-------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-----------------------------|------------------------------| +| URI | /MAS/AlgoPackage/ | | | +| Function | Algorithm package modification | | | +| Method | Query string | Request message body | Response message body | +| PUT | None | | | +| Annotation | 1 The algorithm package is required to be a Zip format file; the description of algorithm package format is shown in Annex B;
2 The definition of ResponseStatus is described in clause A.1. | | | + +### 9.8 GET /MAS/AlgoPackage/ + +Table 15 shows the protocol of interface GET /MAS/AlgoPackage/. + +**Table 15 – Protocol of interface GET /MAS/AlgoPackage/** + +| | | | | +|-------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------|-----------------------------|-----------------------------------| +| URI | /MAS/AlgoPackage/ | | | +| Function | Algorithm package download | | | +| Method | Query string | Request message body | Response message body | +| GET | None | None |
| +| Annotation | 1 If the download fails, return ResponseStatus; if the download succeeds, return algorithm package;
2 The definition of ResponseStatus is described in clause A.1. | | | + +### 9.9 DELETE /MAS/AlgoPackage/ + +Table 16 shows the protocol of interface DELETE /MAS/AlgoPackage/. + +**Table 16 – Protocol of interface DELETE /MAS/AlgoPackage/** + +| | | | | +|-------------------|--------------------------------------------------------------|-----------------------------|------------------------------| +| URI | /MAS/AlgoPackage/ | | | +| Function | Algorithm package deletion | | | +| Method | Query string | Request message body | Response message body | +| DELETE | None | None | | +| Annotation | The definition of ResponseStatus is described in clause A.1. | | | + +### 9.10 GET /MAS/AlgoPackage + +Table 17 shows the protocol of interface GET /MAS/AlgoPackage. + +**Table 17 – Protocol of interface GET /MAS/AlgoPackage** + +| | | | | +|-------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-----------------------------|---------------------------------------| +| URI | /MAS/AlgoPackage | | | +| Function | Algorithm package information querying based on conditions | | | +| Method | Query string | Request message body | Response message body | +| GET | PackageID=xx&PackageName=xx&AlgoType=xx&Version=xx&Vendor=xx&Begin=x
x&Count=xx | None |
| +| Annotation | 1 It can optionally support specific fields filtering through PackageID, and PackageName, etc.; the definitions are described in clause A.7; multiple values are required to be supported for each field, separated by commas;
2 It can optionally support pagination queries, starting from Begin to query the number of Count devices; if the returned AlgoPackageList array size is less than Count, it indicates the end of pagination;
3 If the query fails, return ResponseStatus; if the query succeeds, return AlgoPackageList;
4 The definition of AlgoPackageList is described in clause A.8;
5 The definition of ResponseStatus is described in clause A.1. | | | + +### 9.11 POST /MAS/AlgoTask + +Table 18 shows the protocol of interface POST /MAS/AlgoTask. + +**Table 18 – Protocol of interface POST /MAS/AlgoTask** + +| | | | | +|-------------------|----------------------------------------------------------------------------------------------------------------------------|-----------------------------|------------------------------| +| URI | /MAS/AlgoTask | | | +| Function | Algorithm task addition | | | +| Method | Query string | Request message body | Response message body | +| POST | None | | | +| Annotation | 1 The definition of AlgoTask is described in clause A.9;
2 The definition of ResponseStatus is described in clause A.1. | | | + +### 9.12 PUT /MAS/AlgoTask/ + +Table 19 shows the protocol of interface PUT /MAS/AlgoTask/. + +**Table 19 – Protocol of interface PUT /MAS/AlgoTask/** + +| | | | | +|-------------------|---------------------------------------------------------------------------------------------------------------------------------|-----------------------------|------------------------------| +| URI | /MAS/AlgoTask/ | | | +| Function | Algorithm task controlling (e.g., start and stop) | | | +| Method | Query string | Request message body | Response message body | +| PUT | None | | | +| Annotation | 1 The definition of AlgoTaskCtrl is described in clause A.11;
2 The definition of ResponseStatus is described in clause A.1. | | | + +### 9.13 DELETE /MAS/AlgoTask/ + +Table 20 shows the protocol of interface DELETE /MAS/AlgoTask/. + +**Table 20 – Protocol of interface DELETE /MAS/AlgoTask/** + +| | | | | +|-------------------|--------------------------------------------------------------|-----------------------------|------------------------------| +| URI | /MAS/AlgoTask/ | | | +| Function | Algorithm task deletion | | | +| Method | Query string | Request message body | Response message body | +| DELETE | None | None | | +| Annotation | The definition of ResponseStatus is described in clause A.1. | | | + +### 9.14 GET /MAS/AlgoTask + +Table 21 shows the protocol of interface GET /MAS/AlgoTask. + +**Table 21 – Protocol of interface GET /MAS/AlgoTask** + +| | | | | +|-------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-----------------------------|------------------------------------| +| URI | /MAS/AlgoTask | | | +| Function | Algorithm task information querying based on conditions | | | +| Method | Query string | Request message body | Response message body | +| GET | TaskID=xx&AlgoType=xx&PackageID=xx&Begin=xx&Count=xx | None |
| +| Annotation | 1 It can optionally support specific fields filtering through TaskID, and AlgoType, etc.; the definitions are described in clause A.9; multiple values are required to be supported for each field, separated by commas;
2 It can optionally support pagination queries, starting from Begin to query the number of Count devices; if the returned AlgoTaskList array size is less than Count, it indicates the end of pagination;
3 If the query fails, return ResponseStatus; if the query succeeds, return AlgoTaskList;
4 The definition of AlgoTaskList is described in clause A.10;
5 The definition of ResponseStatus is described in clause A.1. | | | + +### 9.15 POST /MAS/AlgoTaskInstance + +Table 22 shows the protocol of interface POST /MAS/AlgoTaskInstance. + +**Table 22 – Protocol of interface POST /MAS/AlgoTaskInstance** + +| | | | | +|-------------------|-------------------------------------------------------------------------------------------------------------------------------------|-----------------------------|------------------------------| +| URI | /MAS/AlgoTaskInstance | | | +| Function | Algorithm task process instance creation | | | +| Method | Query string | Request message body | Response message body | +| POST | None | | | +| Annotation | 1 The definition of AlgoTaskInstance is described in clause A.12;
2 The definition of ResponseStatus is described in clause A.1. | | | + +### 9.16 DELETE /MAS/AlgoTaskInstance/ + +Table 23 shows the protocol of interface DELETE /MAS/AlgoTaskInstance/. + +**Table 23 – Protocol of interface DELETE /MAS/AlgoTaskInstance/** + +| | | | | +|-------------------|--------------------------------------------------------------|-----------------------------|------------------------------| +| URI | /MAS/AlgoTaskInstance/ | | | +| Function | Algorithm task process instance deletion | | | +| Method | Query string | Request message body | Response message body | +| DELETE | None | None | | +| Annotation | The definition of ResponseStatus is described in clause A.1. | | | + +### 9.17 GET /MAS/AlgoTaskInstance + +Table 24 shows the protocol of interface GET /MAS/AlgoTaskInstance. + +**Table 24 – Protocol of interface GET /MAS/AlgoTaskInstance** + +| | | | | +|-------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-----------------------------|--------------------------------------------| +| URI | /MAS/AlgoTaskInstance | | | +| Function | Algorithm task process instance information querying based on conditions | | | +| Method | Query string | Request message body | Response message body | +| GET | InstanceID=xx&TaskID&PackageID
&Begin=xx&Count=xx | None |
| +| Annotation | 1 It can optionally support specific fields filtering through InstanceID, TaskID, and PackageID, etc.; the definitions are described in clause A.12; multiple values are required to be supported for each field, separated by commas;
2 It can optionally support pagination queries, starting from Begin to query the number of Count devices; if the returned AlgoInstanceList array size is less than Count, it indicates the end of pagination;
3 If the query fails, return ResponseStatus; if the query succeeds, return AlgoTaskInstanceList;
4 The definition of AlgoTaskInstanceList is described in clause A.13;
5 The definition of ResponseStatus is described in clause A.1. | | | + +## Annex A + +### Object descriptions + +(This annex forms an integral part of this Recommendation.) + +### A.1 ResponseStatus + +Table A.1 shows the object descriptions of ResponseStatus. + +**Table A.1 – Object descriptions of ResponseStatus** + +| Parameter | Data type | Mandatory | Description | +|-----------|-----------|-----------|------------------------------------------------------------------------------------| +| ErrCode | Integer | Required | Error code; 0 indicates success, others indicate failure. | +| ErrMsg | String | Optional | Error message. | +| ObjectID | String | Optional | If the command is to generate an object, return the Identifier (ID) of the object. | + +### A.2 Register + +Table A.2 shows the object descriptions of Register. + +**Table A.2 – Object descriptions of Register** + +| Parameter | Data type | Mandatory | Description | +|-----------|-----------|-----------|---------------------------------------------| +| AppID | String | Required | Provide application ID when registering. | +| AppType | String | Required | Provide application types when registering. | + +### A.3 DeRegister + +Table A.3 shows the object descriptions of DeRegister. + +**Table A.3 – Object descriptions of DeRegister** + +| Parameter | Data type | Mandatory | Description | +|-----------|-----------|-----------|---------------------------------------------| +| AppID | String | Required | Provide application ID when de-registering. | + +### A.4 Keepalive + +Table A.4 shows the object descriptions of Keepalive. + +**Table A.4 – Object descriptions of Keepalive** + +| Parameter | Data type | Mandatory | Description | +|-----------|-----------|-----------|------------------------------------------| +| AppID | String | Required | Provide application ID during Keepalive. | + +### A.5 TimeSync + +Table A.5 shows the object descriptions of TimeSync. + +**Table A.5 – Object descriptions of TimeSync** + +| Parameter | Data type | Mandatory | Description | +|-----------|-----------|-----------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------| +| SyncMode | Integer | Required | Time synchronization mode; 1 indicates network time protocol (NTP) service timing, 2 indicates manual timing. | +| NtpServer | String | Optional | When SyncMode is set to 1, it is valid and provides the IP address or domain name of the NTP timing server; multiple addresses are separated by " ". | +| DateTime | String | Optional | When SyncMode is set to 2, it is valid and the value is the set time, the format is required to comply with [ISO 8601-1]; for example: YYYY-MM-DDTHH:MM:SS.sss. | + +### A.6 Resource + +Table A.6 shows the object descriptions of Resource. + +**Table A.6 – Object descriptions of Resource** + +| Parameter | Data type | Mandatory | Description | +|--------------|-----------|-----------|------------------------------------------------------| +| TotalCpuCore | Integer | Required | Total number of central processing unit (CPU) cores. | +| UsedCpuCore | Integer | Required | Number of used CPU cores. | +| TotalMemory | Float | Optional | Total memory, unit is GB. | +| UsedMemory | Float | Optional | Used memory, unit is GB. | +| TotalGpuCard | Integer | Optional | Total number of graphics processing unit (GPU). | +| UsedGpuCard | Integer | Optional | Number of used GPU. | +| TotalDisk | Float | Optional | Total number of disk, unit is GB. | +| UsedDisk | Float | Optional | Used disk, unit is GB. | + +### A.7 AlgoPackage + +Table A.7 shows the object descriptions of AlgoPackage. + +**Table A.7 – Object descriptions of AlgoPackage** + +| Parameter | Data type | Mandatory | Description | +|-------------|-----------|-----------|---------------------------------------------------------------------| +| PackageID | String | Required | Algorithm package ID; it is not required to be set during creation. | +| PackageName | String | Required | Name of algorithm package. | +| AlgoType | String | Required | Functions of algorithm package. | +| Version | String | Required | Version of algorithm package. | +| Vendor | String | Required | Vendor of algorithm package. | + +### A.8 AlgoPackageList + +Table A.8 shows the object descriptions of AlgoPackageList. + +**Table A.8 – Object descriptions of AlgoPackageList** + +| Parameter | Data type | Mandatory | Description | +|-----------------|------------------------|-----------|----------------------------------------| +| AlgoPackageList | Array
| Required | List of algorithm package information. | + +### A.9 AlgoTask + +Table A.9 shows the object descriptions of AlgoTask. + +**Table A.9 – Object descriptions of AlgoTask** + +| Parameter | Data type | Mandatory | Description | +|-----------|------------------|-----------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| TaskID | String | Required | Algorithm task ID; it is not required to be set during creation. | +| AlgoType | String | Required | Algorithm types for algorithm task execution. | +| PackageID | String | Optional | The algorithm package used for the algorithm task; if not specified, the system will automatically select one from the algorithm package warehouse that support the required algorithm type. | +| SrcURI | String | Required | The data source for algorithm task execution, such as HTTP or real time streaming protocol (RTSP) format video or image. | +| Rule | String
(JSON) | Optional | Rule parameters for algorithm tasks. | +| ResultURI | String | Required | The method of pushing the execution results of algorithm tasks, such as the HTTP method for pushing the results. | +| Status | String | Optional | Read-only data, return task execution status when querying, including running, paused, finished, and failed. | + +### A.10 AlgoTaskList + +Table A.10 shows the object descriptions of AlgoTaskList. + +**Table A.10 – Object descriptions of AlgoTaskList** + +| Parameter | Data type | Mandatory | Description | +|--------------|---------------------|-----------|-------------------------------------| +| AlgoTaskList | Array
| Required | List of algorithm task information. | + +### A.11 AlgoTaskCtrl + +Table A.11 shows the object descriptions of AlgoTaskCtrl. + +**Table A.11 – Object descriptions of AlgoTaskCtrl** + +| Parameter | Data type | Mandatory | Description | +|-----------|-----------|-----------|------------------------------------------------| +| Cmd | String | Required | Controlling command, including start and stop. | + +### A.12 AlgoTaskInstance + +Table A.12 shows the object descriptions of AlgoTaskInstance. + +**Table A.12 – Object descriptions of AlgoTaskInstance** + +| Parameter | Data type | Mandatory | Description | +|-------------|-------------|-----------|-------------------------------------------------------------------------------------------------------------------| +| InstanceID | Integer | Required | Process instance ID; it is not required to be set during creation. | +| AlgoTask | AlgoTask | Required | Algorithm tasks. | +| AlgoPackage | AlgoPackage | Required | The algorithm packages used for process instance. | +| Status | String | Optional | Status of process instance; read-only data, return status when querying, including running, finished, and failed. | + +### A.13 AlgoTaskInstanceList + +Table A.13 shows the object descriptions of AlgoTaskInstanceList. + +**Table A.13 – Object descriptions of AlgoTaskInstanceList** + +| Parameter | Data type | Mandatory | Description | +|----------------------|-----------------------------|-----------|---------------------------| +| AlgoTaskInstanceList | Array
| Required | List of process instance. | + +## Annex B + +### Algorithm package format descriptions + +(This annex forms an integral part of this Recommendation.) + +This annex provides the format descriptions of algorithm packages. + +Table B.1 shows the format descriptions of algorithm packages as a Zip format file. + +**Table B.1 – Format descriptions of algorithm packages** + +| Directory | File | Mandatory | Description | +|-----------|-----------------|-----------|-------------------------------------------------------------------------------------------------------------| +| Meta | meta.json | Required | Describe the properties of the algorithm package in JSON format, detail descriptions is shown in Table B.2. | +| Script | start.sh | Required | Algorithmic program startup script. | +| | stop.sh | Required | Algorithmic program stop script. | +| | status.sh | Required | Algorithmic program status query script. | +| App | (Self-defining) | Optional | Storage directory of executable files and dependency algorithm libraries of algorithmic programs. | +| Conf | (Self-defining) | Optional | Storage directory of configuration parameter files. | +| AlgoLib | (Self-defining) | Optional | Storage directory of algorithm library. | +| Data | (Self-defining) | Optional | Storage directory of other data files. | + +NOTE – "Self-defining" means that manufacturers are free to define file names according to their specific requirements, as this Recommendation does not establish pre-defined naming conventions. + +Table B.2 shows the format descriptions of meta.json file. + +**Table B.2 – Format descriptions of meta.json file** + +| Parameter | Data type | Mandatory | Description | +|-----------|-----------|-----------|-------------------------------------| +| AlgoID | String | Required | Algorithm unique ID representation. | +| AlgoName | String | Required | Algorithm name. | +| AlgoFunc | String | Required | Algorithm function. | +| Version | String | Optional | Algorithm version. | +| Vendor | String | Optional | Algorithm vendor. | +| AlgDesc | String | Required | Algorithm description. | +| RunEnv | Object | Required | Runtime environment. | +| +Arch | String | Required | Hardware architecture. | +| +AIACcel | String | Required | AI accelerator types. | + +## Bibliography + +- [b-ITU-T X.746] Recommendation ITU-T X.746 (2000) | ISO/IEC 10164-15:2002, *Information technology – Open Systems Interconnection – Systems Management: Scheduling function.* +- [b-ITU-T Y.2012] Recommendation ITU-T Y.2012 (2010), *Functional requirements and architecture of next generation networks.* +- [b-ISO/IEC 2382] ISO/IEC 2382:2015, *Information technology – Vocabulary.* + +- [b-ISO/IEC 11557] ISO/IEC 11557:1992, *Information technology – 3,81 mm wide magnetic tape cartridge for information interchange – Helical scan recording – DDS-DC format using 60 m and 90 m length tapes.* + +- [b-ISO/IEC 33001] ISO/IEC 33001:2015, *Information technology – Process assessment – Concepts and terminology.* + +- [b-JSON] *JSON schema: A media type for describing JSON documents* (2022). Available [viewed on 2024.08] + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.50-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.748.50-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d4d8969f805414689ab79cfd08c333d454d482b3 --- /dev/null +++ b/marked/F/T-REC-F.748.50-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:636c3839edff955a5b6208530b05fa1c983ab153d90f00216517a7a99478b988 +size 7074 diff --git a/marked/F/T-REC-F.748.50-202503-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.748.50-202503-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a68bc2d2344635d56cdb36e969a107f9d27ba6e3 --- /dev/null +++ b/marked/F/T-REC-F.748.50-202503-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8b648c7a54cb5ac4f8ca2d5bfdf8f11ef66b9477df78959f4fdcfe0b26a1cca7 +size 86273 diff --git a/marked/F/T-REC-F.748.51-202503-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/F/T-REC-F.748.51-202503-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c1629394c02f697f60ae954ad36bf274fa8082cb --- /dev/null +++ b/marked/F/T-REC-F.748.51-202503-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3763acd7a0ed56aa30f3f90468c1fb0cb54573eb9045bca3eddf8976b4f0f41d +size 7219 diff --git a/marked/F/T-REC-F.748.51-202503-I_PDF-E/41245ac07db266bea228735b9e8c8b73_img.jpg b/marked/F/T-REC-F.748.51-202503-I_PDF-E/41245ac07db266bea228735b9e8c8b73_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..8215dd1c2160a5d02b9fc811d7c9e86973bf8bec --- /dev/null +++ b/marked/F/T-REC-F.748.51-202503-I_PDF-E/41245ac07db266bea228735b9e8c8b73_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:77706e7e8ac460dfaa3770fb70cf810643c532134acde02079044d556efdad74 +size 23232 diff --git a/marked/F/T-REC-F.748.51-202503-I_PDF-E/7c6d9bfe9c31ce872722d60b73d20df1_img.jpg b/marked/F/T-REC-F.748.51-202503-I_PDF-E/7c6d9bfe9c31ce872722d60b73d20df1_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a8850f2686f570f807067e9c4d9d62ba83791a1d --- /dev/null +++ b/marked/F/T-REC-F.748.51-202503-I_PDF-E/7c6d9bfe9c31ce872722d60b73d20df1_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:de30a9c81bbe08c52ac2c5dc5236bd6c7e36d63424c0bdbeecabf8ae142e3f50 +size 12401 diff --git a/marked/F/T-REC-F.748.51-202503-I_PDF-E/a4d009d5dd6a4d83759d6d6538188e23_img.jpg b/marked/F/T-REC-F.748.51-202503-I_PDF-E/a4d009d5dd6a4d83759d6d6538188e23_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..8a5b5611f2fe78e4de54b4210488d1a5a8abe08c --- /dev/null +++ b/marked/F/T-REC-F.748.51-202503-I_PDF-E/a4d009d5dd6a4d83759d6d6538188e23_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3ad36b3979110ae5813033a459cbd2dda990b8d06ebb2f3f65fceabd0e50b174 +size 23951 diff --git a/marked/F/T-REC-F.748.51-202503-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg b/marked/F/T-REC-F.748.51-202503-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..eeb798188d609b548c90a04b5c492ab65a7800a6 --- /dev/null +++ b/marked/F/T-REC-F.748.51-202503-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e33d36dc4e7b882bdc45c2e1ca313faaaa21719a1061235c44bb1c0c7c2ae093 +size 161452 diff --git a/marked/F/T-REC-F.748.51-202503-I_PDF-E/ac852a162572ca8a8c8478c49b571af5_img.jpg b/marked/F/T-REC-F.748.51-202503-I_PDF-E/ac852a162572ca8a8c8478c49b571af5_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3082ac69e0e04056e2d1ce20992514d0038fcdf0 --- /dev/null +++ b/marked/F/T-REC-F.748.51-202503-I_PDF-E/ac852a162572ca8a8c8478c49b571af5_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9cf6d7d61b41bb2b4827f2c38b13756dd8c42ea0cf1c09c974beee8ae05211d0 +size 31116 diff --git a/marked/F/T-REC-F.748.51-202503-I_PDF-E/e9314c83043183351ed74908e9bf2f90_img.jpg b/marked/F/T-REC-F.748.51-202503-I_PDF-E/e9314c83043183351ed74908e9bf2f90_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..6fd9e37c7375ab64d6921a98995949bd5c924f6a --- /dev/null +++ b/marked/F/T-REC-F.748.51-202503-I_PDF-E/e9314c83043183351ed74908e9bf2f90_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d3efd44cc24982307f59f16b51d7c2306113d3035ee5c1d7d3be4cd3c687f53a +size 30299 diff --git a/marked/F/T-REC-F.748.51-202503-I_PDF-E/fbfaa723129f8b8e3be651c6fe7cfab7_img.jpg b/marked/F/T-REC-F.748.51-202503-I_PDF-E/fbfaa723129f8b8e3be651c6fe7cfab7_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..5b33b3c603edda6c548398c9f1c3afeb1837d5ad --- /dev/null +++ b/marked/F/T-REC-F.748.51-202503-I_PDF-E/fbfaa723129f8b8e3be651c6fe7cfab7_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:91c97322b846330cc5a6b5b3ec52325535029cc2ca198d802f4d36f5f40d37eb +size 66657 diff --git a/marked/F/T-REC-F.748.52-202503-I_PDF-E/5860ad6bd2a2dd8d1ab12864b8f90f37_img.jpg b/marked/F/T-REC-F.748.52-202503-I_PDF-E/5860ad6bd2a2dd8d1ab12864b8f90f37_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..70c87729a7b6ec5308341031ae730a13396bcd80 --- /dev/null +++ b/marked/F/T-REC-F.748.52-202503-I_PDF-E/5860ad6bd2a2dd8d1ab12864b8f90f37_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:982a96ba422585293d7dd6c57f16d6886453d284a42b14ffc6a7e4ac16fbca47 +size 130382 diff --git a/marked/F/T-REC-F.748.52-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.748.52-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..30d4bc9edd02d463a452fa930b82ff2dfd368c24 --- /dev/null +++ b/marked/F/T-REC-F.748.52-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8bdea22ddbf01b83f16dbc58bc0c9e4b409b0baae46bf6dede68381ee2f7df83 +size 7130 diff --git a/marked/F/T-REC-F.748.52-202503-I_PDF-E/raw.md b/marked/F/T-REC-F.748.52-202503-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..1e8d2c69423322cf5fea20f355952437081406cc --- /dev/null +++ b/marked/F/T-REC-F.748.52-202503-I_PDF-E/raw.md @@ -0,0 +1,784 @@ + + +# Recommendation + +## **ITU-T F.748.52 (03/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +**Requirements and evaluation methods for +retrieval augmented generation of large scale +pre-trained models** + +![ITU logo](84a1d09fb489061482111515543b60dc_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with white lines representing latitude and longitude, and the letters 'ITU' in a bold, blue, sans-serif font superimposed on the globe. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.52 + +# Requirements and evaluation methods for retrieval augmented generation of large scale pre-trained models + +## Summary + +Recommendation ITU-T F.748.52 establishes a framework of technical requirements and evaluation methods for retrieval augmented generation based on large scale pre-trained models. This Recommendation offers users a better understanding of retrieval augmented generation technology and helps them set up retrieval augmented generation applications more quickly and suitably. This Recommendation also provides guidelines for users to examine the level of retrieval augmented generation applications' capabilities and promotes the effectiveness of applications. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.52 | 2025-03-16 | 21 | 11.1002/1000/16242 | + +## Keywords + +Retrieval augmented generation. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2025 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-------------------------------------------------------------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 1 | +| 5 Conventions ..... | 2 | +| 6 Framework of requirements and evaluation methods for retrieval augmented generation of large scale pre-trained models ..... | 2 | +| 7 Requirements and evaluation methods of dataset construction..... | 2 | +| 7.1 Data access ..... | 2 | +| 7.2 Data segmentation ..... | 3 | +| 7.3 Data indexing..... | 3 | +| 7.4 Data updates ..... | 4 | +| 7.5 Data storage ..... | 4 | +| 8 Requirements and evaluation methods of technical capabilities ..... | 4 | +| 8.1 Retrieval capability..... | 4 | +| 8.2 Generative capability..... | 6 | +| 8.3 Optimization capability ..... | 7 | +| 9 Requirements and evaluation methods of application capabilities..... | 9 | +| 9.1 Application maturity..... | 9 | +| 9.2 Application stability ..... | 11 | +| Annex A – Grading requirements for generation capability..... | 15 | +| A.1 Answers relevance grading requirements..... | 15 | +| A.2 Answers faithfulness grading requirements ..... | 15 | +| A.3 Answers truthfulness grading requirements ..... | 16 | +| A.4 Answers relevance grading requirements..... | 16 | + +## **Introduction** + +Retrieval augmented generation technology utilizes external knowledge sources to improve the accuracy and relevance of large scale pre-trained models and provides an efficient and safe solution for enterprises to apply large scale pre-trained models. Therefore, it is very important to evaluate the retrieval augmented generation technology capabilities. This proposal provides general requirements for retrieval augmented generation technology and products that can provide a specification and basis for both retrieval augmented generation technology parties and application parties. + +# Recommendation ITU-T F.748.52 + +## Requirements and evaluation methods for retrieval augmented generation of large scale pre-trained models + +# 1 Scope + +This Recommendation focuses on retrieval augmented generation technology and applications. The scope of this Recommendation includes four parts: + +- Framework of requirements and evaluation methods for retrieval augmented generation of large scale pre-trained models. +- The requirement and evaluation methods for data construction capabilities. +- The requirement and evaluation methods for technical capabilities. +- The requirement and evaluation methods for application capabilities. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +# 3 Definitions + +### 3.1 Terms defined elsewhere + +None. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following term: + +**3.2.1 retrieval augmented generation:** A technique that combines retrieval models and generative models, which utilizes information from large-scale databases or knowledge bases to improve the accuracy and reliability of generative artificial intelligence models. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-------|---------------------------------------------------| +| API | Application Programming Interface | +| HTML | Hyper Text Markup Language | +| HyDE | Hypothetical Document Embeddings | +| MAP | Mean Average Precision | +| RAG | Retrieval Augmented Generation | +| ROUGE | Recall-Oriented Understudy for Gisting Evaluation | + +| | | +|-----|----------------------------| +| SDK | Software Development Kit | +| TPS | Transactions Per Second | +| XML | Extensible Markup Language | + +# 5 Conventions + +In this Recommendation: + +- The keywords "**is required to**" indicate a requirement that must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "**is recommended to**" indicate a requirement that is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. + +# 6 Framework of requirements and evaluation methods for retrieval augmented generation of large scale pre-trained models + +The development route of retrieval augmented generation technology and products consists of large-scale databases or knowledge bases construction, information retrieval, answer generation and application ecology stages. This Recommendation broadly forms an evaluation method to evaluate the capability of retrieval augmented generation techniques for dataset construction, retrieval and generation, as well as application. Figure 1 shows the framework of requirements and evaluation methods for retrieval augmented generation of a large scale pre-trained model. + +![Figure 1: Framework of requirements and evaluation methods for retrieval augmented generation of large scale pre-trained model. The diagram is a hierarchical flowchart showing three main processes: Data preparation process, Application phase process, and Various industries usage/Wide application scenarios. The Data preparation process includes Data extraction, Text segmentation, Vectorization, and Data storage, with sub-evaluation metrics like Data access, Data segmentation, Data indexing, Data updates, and Data storage. The Application phase process includes Knowledge base retrieval, Inject prompt, and Generate answers, with sub-evaluation metrics like Retrieval mode, Retrieval optimization, Prompt optimization, Generate optimization, Retrieval quality, Data optimization, Embedding optimization, and Generate quality. The top level shows various industries usage (Finance, education, new media, automobile manufacturing, agriculture ...) and wide application scenarios (Intelligent customer service, knowledge enhancement, content creation ...).](5860ad6bd2a2dd8d1ab12864b8f90f37_img.jpg) + +| | | | | | +|-----------------------------------|---------------------------------------------------------------------------|------------------------|-------------------------|------------------| +| Various industries usage | Finance, education, new media, automobile manufacturing, agriculture ... | | | | +| Wide application scenarios | Intelligent customer service, knowledge enhancement, content creation ... | | | | +| Component management | Call method | Modal support | Scene adaptation | | +| Robustness | Throughput rate | Negative rejection | Information integration | | +| ↑ | | | | | +| Application phase process | | | | | +| Knowledge base retrieval | ⇒ | Inject prompt | ⇒ | Generate answers | +| Retrieval mode | Retrieval optimization | Prompt optimization | Generate optimization | | +| Retrieval quality | Data optimization | Embedding optimization | Generate quality | | +| ↑ | | | | | +| Data preparation process | | | | | +| Data extraction | ⇒ | Text segmentation | ⇒ | Vectorization | +| ⇒ | | | | Data storage | +| Data access | Data segmentation | Data indexing | Data updates | Data storage | + +Figure 1: Framework of requirements and evaluation methods for retrieval augmented generation of large scale pre-trained model. The diagram is a hierarchical flowchart showing three main processes: Data preparation process, Application phase process, and Various industries usage/Wide application scenarios. The Data preparation process includes Data extraction, Text segmentation, Vectorization, and Data storage, with sub-evaluation metrics like Data access, Data segmentation, Data indexing, Data updates, and Data storage. The Application phase process includes Knowledge base retrieval, Inject prompt, and Generate answers, with sub-evaluation metrics like Retrieval mode, Retrieval optimization, Prompt optimization, Generate optimization, Retrieval quality, Data optimization, Embedding optimization, and Generate quality. The top level shows various industries usage (Finance, education, new media, automobile manufacturing, agriculture ...) and wide application scenarios (Intelligent customer service, knowledge enhancement, content creation ...). + +F.748.52(25) + +**Figure 1 – Framework of requirements and evaluation methods for retrieval augmented generation of large scale pre-trained model** + +## 7 Requirements and evaluation methods of dataset construction + +### 7.1 Data access + +**Technical requirements:** The retrieval augmented generation technology is required to have the capability to handle multiple data types and support large amounts of data during the dataset construction process. + +- a) Data types: + - 1) Structured data: It is required to support the structured data. + - 2) Semi-structured data: It is required to support the semi-structured data, such as hyper text markup language (HTML), and extensible markup language (XML). + - 3) Unstructured data: It is required to support the unstructured data, such as images and video clips. +- b) Data volume: It is required to support a large amount of data access, and it is recommended to support access to Internet data. + +### **Evaluation methods** + +- a) The service provider provides a data access description document that contains information on the type of data accessed and the amount of data accessed. +- b) Evaluate whether structured data, semi-structured data and unstructured data are supported based on technical verification and material review. +- c) Evaluate the volume of data integration and whether Internet data integration is supported based on document review. + +### **7.2 Data segmentation** + +**Technical requirements:** The retrieval augmented generation technology is required to have a complete data segmentation method during the dataset construction process. + +- a) Segmentation strategies: It is required to support various segmentation strategies, such as fixed-length segmentation, recursive call segmentation, punctuation segmentation, and character count segmentation. +- b) Segmentation configuration: It is required to support various segmentation configuration methods, such as custom segmentation strategy configuration, maximum segment length setting, and maximum overlap length setting. +- c) Data types: It is required to support segmentation of multiple data types, such as text, documents, and tabular data. + +#### **Evaluation methods** + +- a) The service provider provides a data segmentation description document that contains information on segmentation strategy, segmentation configuration and data types. +- b) Based on technical validation and material review, sequentially evaluate the degree of support for segmentation strategy, segmentation configuration and data types of products related to retrieval augmented generation technology. + +### **7.3 Data indexing** + +**Technical requirements:** The retrieval augmented generation technology is required to support diverse data indexing types during the dataset construction process. + +- a) Index types: It is required to support a variety of index types, such as layer indexes, inverted indexes, list indexes, and graph indexes. + +### **Evaluation methods** + +- a) The service provider provides a data indexing description document that contains indexing types and other relevant information. +- b) Evaluate the degree of support for indexing types of products related to retrieval augmented generation technology based on technical verification and material review. + +### 7.4 Data updates + +**Technical requirements:** The retrieval augmented generation technology is required to have diverse data update methods during the dataset construction process. + +- a) Manual update: It is required to support updating information in large-scale databases or knowledge bases through manual operations. +- b) Scheduled update: It is required to support regularly updating large-scale databases or knowledge bases according to a preset schedule. +- c) Incremental update: It is required to support updating only the parts that have changed to avoid the unnecessary overhead of full updates. +- d) Real-time update: It is required to support immediately updating large-scale databases or knowledge bases upon receiving new information to maintain the timeliness of information. + +### Evaluation methods + +- a) The service provider provides a data update description document that contains information on data updating methods. +- b) Based on technical verification and material review, evaluate whether the products related to retrieval augmented generation technology support diversified updating methods. + +### 7.5 Data storage + +**Technical requirements:** The retrieval augmented generation technology is required to have multiple data storage methods during the dataset construction process to ensure data security. + +- a) Storage capability: It is required to support features enabling the distributed storage of data, such as data compression, data archiving, data backup, and data partitioning. + +### Evaluation methods + +- a) The service provider provides a data storage description document that contains information on storage strategy. +- b) Based on technical verification and material review, evaluate the degree of support for storage strategies in products related to retrieval augmented generation technology. + +## 8 Requirements and evaluation methods of technical capabilities + +### 8.1 Retrieval capability + +#### 8.1.1 Retrieval mode + +**Technical requirements:** The retrieval augmented generation technology is required to support diverse retrieval methods. + +- a) Retrieval methods: + - 1) Content-based retrieval: It is required to support the content-based retrieval, such as full-text retrieval, keyword retrieval, and entity retrieval. + - 2) Model-based retrieval: It is required to support the model-based retrieval, such as adaptive retrieval, and cross-modal retrieval. + - 3) Algorithm-based retrieval: It is required to support the algorithm-based retrieval, such as iterative retrieval, hybrid retrieval, and maximum marginal relevance retrieval. + - 4) Semantic-based retrieval: It is required to support the content-based retrieval, such as sentence pattern retrieval, and semantic retrieval. + +- b) Retrieval integration: It is required to have the capability to integrate multiple retrieval methods, namely the capability to use multiple retrieval methods at the same time for one retrieval. + +#### Evaluation methods + +- a) The service provider provides a retrieval mode description document that contains information on content-based, model-based, algorithm-based, semantic-based and other retrieval methods. +- b) Based on technical validation and material review, evaluate whether the products related to retrieval augmented generation technology support rich and diversified retrieval modes. +- c) Based on technical verification and material review, evaluate the integration capability of multiple retrieval modes. + +#### 8.1.2 Retrieval quality + +**Technical requirements:** The content retrieved by the retrieval augmented generation technique is required to have high quality results. + +#### Evaluation methods + +- a) Construct relevant datasets to verify the superiority of retrieval quality in retrieval augmented generation techniques. +- b) Call product interface or other means to obtain search results, select test metrics, and calculate the performance results according to the following formula: + - 1) Accuracy: Assessing the degree of correlation between retrieval results and user query content. The formula is as follows. + +$$P_A = \frac{A_1}{A} \times 100\% \quad (1)$$ + +$P_A$ : Accuracy. + +$A_1$ : The number of retrieval results that are relevant to the query. + +$A$ : The total number of retrieval results. + +- 2) Recall: Assessing the percentage between relevant documents retrieved and the total number of relevant documents. The formula is as follows: + +$$R_A = \frac{A_1}{A_t} \times 100\% \quad (2)$$ + +$R_A$ : Recall. + +$A_1$ : The number of retrieval results that are relevant to the query. + +$A_t$ : The total number of all relevant data in large-scale databases or knowledge bases. + +- 3) Mean average precision (MAP): assessing the quality of the ranking of the retrieval results. The formula is as follows. + +$$MAP = \frac{\sum_{q=1}^Q AP(q)}{Q} \quad (3)$$ + +$Q$ : Number of retrievals. + +$AP$ : The average accuracy of a single recall result at each digit (e.g., top 1, top 5, top 10), calculated as follows. + +$$AP = \frac{\sum_{k=1}^n (P(k) \times rel(k))}{\text{Number of related documents}} \quad (4)$$ + +K: The sorted position of the retrieved result in the queue. + +n: The first n orderings of the retrieved results, with a reference value of 5. + +P(k): Accuracy of the first k results, i.e., $P(k) = \text{number of accurate documents in the first k results/k}$ . + +rel(k): Whether the document at position k is an accurate result is the composite score normalized for the composite score for each result in the test dataset. + +### 8.2 Generative capability + +#### 8.2.1 Answer relevance + +**Technical requirements:** the content generated using the retrieval augmented generation technique is required to have a strong correlation with the user's question, and the answer relevance is graded according to Table A.1. + +#### Evaluation methods + +- a) Construct relevant datasets to validate the answers relevance of retrieval augmented generation techniques and products. +- b) Call product interfaces or other methods to generate responses, and the generated content is scored for acceptability based on Table A.1, and the final score for answer relevance is the average of the acceptability scores of multiple reviewers. + +#### 8.2.2 Answer reliability + +**Technical requirements:** The content generated using the retrieval augmented generation technique is required to have a higher degree of match with the retrieved results, and the answer reliability is graded according to Table A.2. + +#### Evaluation methods + +- a) Construct relevant datasets to validate the answer reliability of retrieval augmented generation techniques and products. +- b) Call product interfaces or other methods to generate responses, and the retrieval results will be scored for acceptability based on Table A.1, and the final score for answer reliability will be the average of the acceptability scores of multiple reviewers. + +#### 8.2.3 Answer truthfulness + +**Technical requirements:** the content generated using the retrieval augmented generation technique is required to be consistent with facts or accepted knowledge, and the answer truthfulness is graded according to Table A.3. + +#### Evaluation methods + +- a) Construct relevant datasets to validate the answers truthfulness of the retrieval augmented generation techniques and products. +- b) Call product interfaces or other methods to generate responses, and the generated content is scored for acceptability based on Table A.3, with the final score for answer truthfulness being the average of the acceptability scores of multiple reviewers. + +#### 8.2.4 Answer correctness + +**Technical requirements:** The content generated using the retrieval augmented generation technique is required to have a high degree of match with the ground truth. + +#### Evaluation methods + +- a) Constructing relevant datasets to validate the answer correctness of retrieval augmented generation techniques and products. +- b) Call product interfaces or other methods to generate responses, select test metrics, and calculate the performance results according to the following formula: + - 1) Accurate: Assessing the match of the generated content with the ground truth. The formula is as follows: + +$$\text{Accurate} = \frac{\text{Count}_{\text{exactmatch}}(c,r)}{\text{Count}(c)} \quad (5)$$ + +c: Candidate answers generated by the model. + +r: The correct answer in the label. + +$\text{Count}_{\text{exactmatch}}(c,r)$ : The number of sentences in which the model generates an exact match between the answer and the correct answer. + +$\text{Count}(c)$ : The number of answers generated by the model. + +#### 8.2.5 Context relevance + +**Technical requirements:** The content generated using the retrieval augmented generation technique is required to have a high degree of closeness in terms of logical relationships and semantic connections between contexts, and the context relevance is graded according to Table A.4. + +#### Evaluation methods + +- a) Constructing datasets to validate the context relevance of retrieval augmented generation techniques and products. +- b) Call product interfaces or other methods to generate responses, and the generated content will be scored for acceptability based on Table A.4, with the final score for context relevance being the average of the acceptability scores of multiple reviewers. + +### 8.3 Optimization capability + +#### 8.3.1 Data optimization + +**Technical requirements:** The retrieval augmented generation technology is required to combine various data optimization methods to improve corpus diversity, data integrity and information richness: + +- a) Data cleaning: It is required to support data cleaning, such as text cleaning, semantic parsing, and data deduplication. +- b) Data augmentation: It is required to support data augmentation, such as text segmentation, synonym replacement, caption augmentation, text summarization, and image-to-text. +- c) Data fusion: It is required to support data fusion, such as attribute fusion, relationship fusion, cross-modal knowledge fusion, and knowledge representation fusion. + +#### Evaluation methods + +- a) The service provider provides a data optimization description document that contains information on data cleansing methods, data augmentation methods and data fusion methods. +- b) Based on technical verification and material review, evaluate whether multiple data optimization methods are supported. + +#### 8.3.2 Embedding optimization + +**Technical requirements:** The retrieval augmented generation technique is required to combine various embedding optimization methods: + +- a) Fine-tune embedding: It is required to support fine-tuning on specific task data to further optimize the word embedding model. +- b) Dynamic representation: It is required to support dynamically adjusting the representation of words based on their meaning in context. +- c) Refreshing representation: It is required to support regular updating of embedded model representations to adapt to semantic changes in the corpus. +- d) Incremental learning: It is required to support continuously injecting new data and knowledge into the word embedding model to gradually optimize the representation and performance of the model. +- e) Adaptive learning: It is required to support automatically adjusting word embedding model parameters and representations according to model performance and environment changes to optimize model effectiveness. + +#### Evaluation methods + +- a) The service provider provides an embedding optimization description document that contains information on fine-tuning embedding and dynamic representation. +- b) Based on the technical verification and material review, evaluate whether diversified representation optimization methods are supported. + +#### 8.3.3 Retrieval optimization + +**Technical requirements:** The retrieval augmented generation technique is required to combine various retrieval optimization methods: + +- a) Pre-retrieval optimization: + - 1) Query rewrite: It is required to support query rewriting and query statement rewriting. + - 2) Query expansion: It is required to support query expansion, which expands query statements to improve retrieval coverage and recall. + - 3) Sub-queries: It is required to split the query statement to improve retrieval accuracy. + - 4) Hypothetical document embeddings (HyDE): It is recommended to support HyDE, which utilizes a large model to make hypothetical responses to questions and responses as part of the retrieval statement. +- b) Post-retrieval optimization: + - 1) Result sorting: It is required to support sorting the retrieval results according to the sorting rules to ensure that more relevant results are prioritized for display. + - 2) Content filtering: It is required to support filtering of content, and automatically filtering incompatible items and sensitive content according to user preferences, filtering conditions, or business requirements. + - 3) Abstract generation: It is required to support the extraction of abstracts, extracting the core abstract of retrieval results to help users quickly understand and efficiently browse. + - 4) Alignment of results: It is required to support the alignment of search results from multiple sources to unify the dimensions and structure, so as to facilitate intuitive comparison and in-depth understanding by users. + +#### Evaluation methods + +- a) The service provider provides a retrieval capability optimization description document that contains information on pre-retrieval optimization and post-retrieval optimization. +- b) Based on technical verification and material review, evaluate whether it supports diversified retrieval optimization methods. + +#### 8.3.4 Prompt optimization + +**Technical requirements:** The retrieval augmented generation technology is required to combine various prompt optimization methods: + +- a) Prompt template: It is required to support guiding the model to generate content through templates, such as placeholders, and fill in the blank templates. +- b) Prompt constraints: It is required to support constraints on prompt words to ensure compliance of prompt. +- c) Prompt rewriting: It is required to support modifying the prompt words to ensure accurate guidance of model generated content. +- d) Prompt completion: It is required to support the auto-completion of prompt words to make it richer and more accurate. +- e) Multi-stage prompts: It is required to support the decomposition of complex tasks into multiple stages, and use different prompts or instructions for each phase. + +#### Evaluation methods + +- a) The service provider provides a prompt optimization description document that contains prompt templates, prompt constraints and other relevant information. +- b) Based on technical verification and material review, evaluate whether to support diversified prompt optimization methods. + +#### 8.3.5 Generate optimization + +**Technical requirements:** The retrieval augmented generation technology is required to combine various generation optimization methods: + +- a) Knowledge tracking: It is required to support tracking the source of knowledge in the generated content to ensure the accuracy and credibility of the information. +- b) Hallucination detection: It is required to support hallucination detection, such as hallucination localization, and misinformation screening. +- c) Content correction: It is required to support content correction, such as content correction illusions, logical errors, and fallacies in the generated content. +- d) Feedback iteration: It is required to support continuously optimizing the generated content and optimizing of the model based on user feedback. + +#### Evaluation methods + +- a) The service provider provides a generation capability optimization description document that contains information on knowledge tracking, illusion discovery and other relevant information. +- b) Based on technical verification and material review, evaluate whether diversified generation optimization methods are supported. + +## 9 Requirements and evaluation methods of application capabilities + +### 9.1 Application maturity + +#### 9.1.1 Component management + +**Technical requirements:** The retrieval augmented generation technology and product are required to have effective component management capabilities in the application stage: + +- a) Component support: It is required to support multiple types of components, such as data loaders, text embeddings, vector storage, retrievers, and generators. + +- b) Component flexibility: + - 1) Primary flexibility: It is required to support the control of components through scripting. + - 2) Intermediate flexibility: It is required to support the control of components through front-end configuration. + - 3) Advanced flexibility: It is required to support the control of components through front-end configuration, as well as support for component sorting, searching and other functions. + +#### **Evaluation methods** + +- a) The service provider provides a component management description document that contains information on component support and component flexibility. +- b) Based on technical verification and material review, evaluate whether the product supports multiple types of components in the application process. +- c) Based on technical verification and material review, evaluate whether the product has a high degree of component flexibility. + +#### **9.1.2 Call method** + +**Technical requirements:** The retrieval augmented generation technology and product are required to have rich calling methods in the application stage: + +- a) Call method: It is required to support multiple call methods, such as application programming interface (API), software development kit (SDK), and web interface. + +#### **Evaluation methods** + +- a) The service provider provides a calling method description document that contains the calling information of API interface, command line tool and other relevant information. +- b) Based on technical verification and material review, evaluate whether the product supports multiple calling methods in the application process. + +#### **9.1.3 Modal support** + +**Technical requirements:** The retrieval augmented generation technology and product are required to support multiple input and output modalities in the application stage: + +- a) Input modal: It is required to support multiple input modalities, such as texts, audio clips, images, and videos. +- b) Output modal: It is required to support multiple output modalities, such as texts, audio clips, images, and videos. + +#### **Evaluation methods** + +- a) The service provider provides a modality-supported description document that contains information on input modalities and output modalities. +- b) Based on technical verification and material review, evaluate whether the product supports texts, audios, and other input modalities in the application process. +- c) Based on technical verification and material review, evaluate whether the product supports texts, audios, and other output modes in the application process. + +#### **9.1.4 Scene adaptation** + +**Technical requirements:** The retrieval augmented generation technology and product are required to adapt to a variety of application scenarios in the application stage: + +- a) Scenario adaptation: It is required to support diverse scenarios, such as intelligent customer service, content creation, conversational retrieval, personalized recommendations, knowledge enhancement, and event planning. + +#### Evaluation methods + +- a) The service provider provides a scene adaptation description document that contains information on the degree of scene support and scene application effect. +- b) Based on technical verification and material review, evaluate whether the product supports diversified scenarios in the process of application. + +### 9.2 Application stability + +#### 9.2.1 Robustness + +**Technical requirements:** The retrieval augmented generation technique and product are required to be robust in the application phase, such as high response accuracy for noise robustness, and high error detection and error correction rates for counterfactual robustness. + +#### Evaluation methods + +- a) Construct relevant datasets to validate the robustness of retrieval augmented generation techniques and products in the application phase. +- b) Call product interfaces or other methods to generate responses, select test metrics, and calculate the performance results according to the noise robustness and counterfactual robustness. +- c) Noise robustness: Evaluates the capability of the model to extract useful information from noisy documents. + - 1) Noise robustness: Assessing the accuracy of generated content by adding different percentages of noisy documents to external documents. The formula is as follows: + +$$P_N = \frac{F_r}{F} \times 100\% \quad (6)$$ + +$P_N$ : Accuracy. + +$F_r$ : The number of questions answered correctly. + +$F$ : Total number of questions answered. + +- d) Counterfactual robustness: Assessing the capability of the model to recognize and correct known factual errors in retrieved documents given a warning. + - 1) Error detection rate: The error detection rate of the model is computed by providing retrieved documents containing factual errors, The formula is as follows: + +$$P_{EDR} = \frac{F_r}{F} \times 100\% \quad (7)$$ + +$P_{EDR}$ : Error detection rate. + +$F_r$ : The number of factual errors corrected correctly. + +$F$ : Total number of factual errors. + +- 2) Error correction rate: the error correction rate of the model is computed by providing retrieved documents containing factual errors. The formula is as follows: + +$$P_{ECR} = \frac{F_r}{F} \times 100\% \quad (8)$$ + +$P_{ECR}$ : Error correction rate. + +$F_r$ : The number of factual errors corrected correctly. + +$F$ : Total number of factual errors. + +#### 9.2.2 Security + +**Technical requirements:** The retrieval augmented generation technology and product are required to have high security in the application stage: + +- a) Security system construction: It is required to support a comprehensive security system construction program during the application phase, such as the formulation of security policies, the establishment of security training mechanisms. +- b) Data security guarantee: It is required to support the safeguarding of data security during the application phase, such as data encryption, data backup, and data usage audit. +- c) Access rights control: It is required to support access rights control during the application phase, such as implementation of strict identity authentication, and formulation of rights management policies, adhering to individual privacy. + +#### Evaluation methods + +- a) The service provider provides a security capability description document that contains information on data security, content security, and access rights control. +- b) Based on technical verification and material review, evaluate whether the product supports data encryption, data backup and other data security measures in the application process. +- c) Based on technical verification and material review, evaluate whether the product supports content security safeguards such as input/output information detection and sensitive information filtering in the application process. +- d) Based on technical verification and material review, evaluate whether the product supports access rights management measures such as identity authentication and rights management in the application process. + +#### 9.2.3 Throughput rate + +**Technical requirements:** The retrieval augmented generation technique and product are required to have a large user query capacity. + +#### Evaluation methods + +- a) The service provider provides a throughput rate technical description document that contains the real throughput rate and other information. +- b) Based on the technical verification and material review approach, the throughput rate of the product is calculated according to the transactions per second (TPS). +- c) TPS: Assessing the user query capacity of the retrieval augmented generation system. The formula is as follows: + +$$\text{TPS} = \frac{\text{Count}_{\text{requests}}}{\text{T}} \quad (9)$$ + +$\text{Count}_{\text{requests}}$ : Total number of requests. + +$\text{R\_T}_{\text{finish}}$ : Total time. + +TPS: Transactions per second. + +#### 9.2.4 Response time + +**Technical requirements:** The retrieval augmented generation technology and product are required to have a fast response time in the application phase. + +#### Evaluation methods + +- a) The service provider provides a response time technical description document that contains the response time. + +- b) Based on the technical verification and material review, the response time of the product is calculated according to the following formula: + +$$ES\_T = R\_T_{\text{finish}} - R\_T_{\text{start}} \quad (10)$$ + +$ES\_T$ : Response time. + +$R\_T_{\text{finish}}$ : The time when the system or application returns the result (non-streaming) or the time when the system or application returns the first byte of the result (streaming). + +$R\_T_{\text{start}}$ : The start time of the user-initiated request. + +#### 9.2.5 Negative rejection + +**Technical requirements:** The retrieval augmented generation technique and product are required to be able to effectively reject answers in practical applications when the retrieved documents or information cannot meet the requirements of the question. + +#### Evaluation methods + +- a) Construct relevant datasets to validate the negative rejection capability of retrieval augmented generation techniques and products in the application phase. +- b) Call product interfaces or other methods to generate responses, select test metrics, and calculate the performance results according to the negative rejection rate. +- c) Negative rejection rate: assessing the capability of the model to refuse to answer a question: + - 1) Effective rejection rate: the proportion of questions that should be refused to be answered that are actually refused to be answered. The formula is as follows: + +$$P_{\text{ERR}} = \frac{F_r}{F} \times 100\% \quad (11)$$ + +$P_{\text{ERR}}$ : Effective rejection rate. + +$F_r$ : Number of questions actually correctly refused to be answered. + +$F$ : Number of questions that should be refused to be answered. + +- 2) Effective answer rate: the proportion of questions that should be answered normally that are actually answered normally. The formula is as follows: + +$$P_{\text{EAR}} = \frac{F_r}{F} \times 100\% \quad (12)$$ + +$P_{\text{EAR}}$ : Effective answer rate. + +$F_r$ : Number of questions actually answered correctly. + +$F$ : Number of questions that should be answered correctly. + +- 3) Negative rejection rate: + +$$P_{\text{NRR}} = \frac{P_{\text{ERR}} + P_{\text{EAR}}}{2} \quad (13)$$ + +$P_{\text{NRR}}$ : Negative rejection rate. + +$F_{\text{ERR}}$ : Effective rejection rate. + +$F_{\text{EAR}}$ : Effective answer rate. + +#### 9.2.6 Information integration + +**Technical requirement:** The retrieval augmented generation technique and product are required to integrate information and correctly answer questions when the retrieval augmented generation technology needs to integrate multiple document information in practical applications. + +#### Evaluation methods + +- a) Construct relevant datasets to validate the information integration capabilities of retrieval augmented generation techniques and products in the application phase. +- b) Call product interfaces or other methods to generate responses, select test metrics, and calculate the performance results according to the following formula: + - 1) Correct rate of information integration: Assessing the accuracy of information integration by comparing the number of correct answers with the total number of questions. The formula is as follows: + +$$P_{\text{MIR}} = \frac{F_r}{F} \times 100\% \quad (14)$$ + +$P_{\text{MIR}}$ : Correct rate of information integration. + +$F_r$ : Number of questions with exactly the right answer. + +$F$ : Total number of issues. + +## Annex A + +### Grading requirements for generation capability + +(This annex forms an integral part of this Recommendation.) + +### A.1 Answers relevance grading requirements + +**Table A.1 – Answers relevance grading requirements** + +| Level | Description | +|---------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Level 1 | The generated content has no relevance to the question, is completely off-topic, and does not provide information or answers relevant to the question. | +| Level 2 | The generated content is of low relevance to the question and only touches on the fringes of the question. | +| Level 3 | The generated content is partially relevant to the question, but does not fully answer the core points of the question. | +| Level 4 | The generated content is closely related to the question, and as a whole follows the question closely, providing valuable information to the user. | +| Level 5 | The generated content matches the question perfectly, answering the core points of the question directly without deviating from the topic, and providing accurate and useful information. | + +### A.2 Answers faithfulness grading requirements + +**Table A.2 – Answer faithfulness grading requirements** + +| Level | Description | +|---------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Level 1 | There is a keyword overlap between the generated content and the retrieved results, but there is no obvious correlation in the semantic expression. | +| Level 2 | There is a keyword overlap between the generated content and the retrieved results, and the semantic representation references the content in the retrieved results, but there is a lot of fabricated content. | +| Level 3 | The generated content is basically consistent with the retrieval results in terms of semantic representation, with a small amount of fabricated content. | +| Level 4 | The generated content is consistent with the retrieved results in terms of semantic representation, but the content is incomplete and with omissions. | +| Level 5 | The generated content is consistent with the retrieval results in terms of semantic representation, and the content is complete without omissions. | + +### A.3 Answers truthfulness grading requirements + +**Table A.3 – Answer truthfulness grading requirements** + +| Level | Description | +|---------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Level 1 | The generated content is completely inconsistent with facts or accepted knowledge, and is patently false, misleading, or fictitious. | +| Level 2 | The generated content is only partially consistent with facts or recognized knowledge, and contains significant errors, omissions, or misleading content. | +| Level 3 | The generated content is generally consistent with facts or accepted knowledge, but with minor inaccuracies in some points. | +| Level 4 | The generated content is highly consistent with facts or accepted knowledge, and have great accuracy and reliability. | +| Level 5 | The generated content is fully consistent with the facts or accepted knowledge, without errors, omissions, or misleading content. | + +### A.4 Answers relevance grading requirements + +**Table A.4 – Contextual relevance grading requirements** + +| Level | Description | +|---------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Level 1 | 1. The generated content has a large number of logical contradictions with the preceding and following text.
2. The generated content has a great deal of repetition and redundancy with the preceding and following text. | +| Level 2 | 1. The generated content has a small number of logical inconsistencies with the preceding and following text.
2. The generated content has some repetition or redundancy with the preceding and following text. | +| Level 3 | 1. The generated content has no obvious logical contradiction with the preceding and following text, but the transitions with the preceding and following texts are not smooth enough.
2. The generated content has a small amount of repetition or redundancy with the preceding and following text. | +| Level 4 | 1. The generated content has no obvious logical contradiction with the preceding and following text, and the transition with the preceding and following text is smooth.
2. The generated content has no repetition or redundancy with the preceding and following text. | + + + +# SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.53-202503-I_PDF-E/12280970bb3eea7f6fbdc08200d444c2_img.jpg b/marked/F/T-REC-F.748.53-202503-I_PDF-E/12280970bb3eea7f6fbdc08200d444c2_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..961329f495e0dcb7935e978ef5eb40fd9874d216 --- /dev/null +++ b/marked/F/T-REC-F.748.53-202503-I_PDF-E/12280970bb3eea7f6fbdc08200d444c2_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9d47174032869727d84003a2b84c484be775136f0bdc9b6a373e4d7af5c55241 +size 58027 diff --git a/marked/F/T-REC-F.748.53-202503-I_PDF-E/3f8bb1be8f73c7623f1380af78dcf8c6_img.jpg b/marked/F/T-REC-F.748.53-202503-I_PDF-E/3f8bb1be8f73c7623f1380af78dcf8c6_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..03aeb05d1b7b2c53fb712ab5226f136202bb3558 --- /dev/null +++ b/marked/F/T-REC-F.748.53-202503-I_PDF-E/3f8bb1be8f73c7623f1380af78dcf8c6_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:15bf9c5af545b707a399b1a3095b64f6f3fbf392c8181d80e11b1104818989f6 +size 24251 diff --git a/marked/F/T-REC-F.748.53-202503-I_PDF-E/573293af67ee571aef04bf725fcefbe5_img.jpg b/marked/F/T-REC-F.748.53-202503-I_PDF-E/573293af67ee571aef04bf725fcefbe5_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ad181e5b742b441aa21fe1a17ed54b08f5c2b815 --- /dev/null +++ b/marked/F/T-REC-F.748.53-202503-I_PDF-E/573293af67ee571aef04bf725fcefbe5_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7739f0d44a3b28a38f2e323541e1a0c95a56cde0916dbc4d361b3e193dcfe424 +size 58837 diff --git a/marked/F/T-REC-F.748.53-202503-I_PDF-E/5cb79a4c597e9880d65949d9416dda5a_img.jpg b/marked/F/T-REC-F.748.53-202503-I_PDF-E/5cb79a4c597e9880d65949d9416dda5a_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..0a81d9aba9635be884951e1b8bc7085e50dd3c2a --- /dev/null +++ b/marked/F/T-REC-F.748.53-202503-I_PDF-E/5cb79a4c597e9880d65949d9416dda5a_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:999a7b1a2c0ea6da3c402e68045b88f9b135c4546e702f0ebd4b4e8ccf7c7f91 +size 9900 diff --git a/marked/F/T-REC-F.748.53-202503-I_PDF-E/7d1b7ee3543f5a1a9b7ad98572d52f85_img.jpg b/marked/F/T-REC-F.748.53-202503-I_PDF-E/7d1b7ee3543f5a1a9b7ad98572d52f85_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1732c36c9e3eff06808b2c4791d7c6b1dd825e91 --- /dev/null +++ b/marked/F/T-REC-F.748.53-202503-I_PDF-E/7d1b7ee3543f5a1a9b7ad98572d52f85_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:524701ae6faf4a95657e8bec82101bdc7ccc0aef7f783ae4ba8a7edf06c0e24c +size 34971 diff --git a/marked/F/T-REC-F.748.53-202503-I_PDF-E/7f17c430b9598e4d748a8041457810b3_img.jpg b/marked/F/T-REC-F.748.53-202503-I_PDF-E/7f17c430b9598e4d748a8041457810b3_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..458f6fcc7efa36b9f648f3dc9e7768c29de25caf --- /dev/null +++ b/marked/F/T-REC-F.748.53-202503-I_PDF-E/7f17c430b9598e4d748a8041457810b3_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1637519ad128607132dcb5098dc429486a16ea2e02bd04773da4b05e13fec47d +size 17720 diff --git a/marked/F/T-REC-F.748.53-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.748.53-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..abda52cbb1047d6649f89e40c898a1c33bed7267 --- /dev/null +++ b/marked/F/T-REC-F.748.53-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6f3b7e07231c0f3e116c99c493bdda98ecd52313b6735faa6cea23815cf3162b +size 7190 diff --git a/marked/F/T-REC-F.748.53-202503-I_PDF-E/8e14350b4b669119a3bdfca7869110ca_img.jpg b/marked/F/T-REC-F.748.53-202503-I_PDF-E/8e14350b4b669119a3bdfca7869110ca_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a961ea3552a5be995261ad12a8ebcb13b541a4a8 --- /dev/null +++ b/marked/F/T-REC-F.748.53-202503-I_PDF-E/8e14350b4b669119a3bdfca7869110ca_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fa97a1287e1e80a5d0c4d86cfe8437fc7880380be9b6186059a3ecfcee3b9623 +size 38580 diff --git a/marked/F/T-REC-F.748.53-202503-I_PDF-E/906c1ba8ae95f501d4cd79eb08485307_img.jpg b/marked/F/T-REC-F.748.53-202503-I_PDF-E/906c1ba8ae95f501d4cd79eb08485307_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..90f4099c662f05ef099eb55ee268d747cadf22e0 --- /dev/null +++ b/marked/F/T-REC-F.748.53-202503-I_PDF-E/906c1ba8ae95f501d4cd79eb08485307_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:305b5e1ad048c0b87f5242ea87f86dddb3b100a27ee573445c3166f17360d49c +size 18951 diff --git a/marked/F/T-REC-F.748.53-202503-I_PDF-E/9a8f8276c7199f1dd9f3754cf60555c6_img.jpg b/marked/F/T-REC-F.748.53-202503-I_PDF-E/9a8f8276c7199f1dd9f3754cf60555c6_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..8d592049aa95f67b14aa52fbc603110a308e815f --- /dev/null +++ b/marked/F/T-REC-F.748.53-202503-I_PDF-E/9a8f8276c7199f1dd9f3754cf60555c6_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:293e5468124e2bd923f9db4a2c988ffcad6f52becd509cd7de0fa58bde310878 +size 40264 diff --git a/marked/F/T-REC-F.748.53-202503-I_PDF-E/9e1c95ee7eb045977f18f011f49520c7_img.jpg b/marked/F/T-REC-F.748.53-202503-I_PDF-E/9e1c95ee7eb045977f18f011f49520c7_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1f67f43180684bd6f2bc74acdf4df90f8b86935e --- /dev/null +++ b/marked/F/T-REC-F.748.53-202503-I_PDF-E/9e1c95ee7eb045977f18f011f49520c7_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:edf32f481f5d521c51f72f4d52cf52b11c07beb20b0de5b74753af0c6e5199b1 +size 32214 diff --git a/marked/F/T-REC-F.748.53-202503-I_PDF-E/b0a5d736df5f6fe2c0c6dd800bb0ff9a_img.jpg b/marked/F/T-REC-F.748.53-202503-I_PDF-E/b0a5d736df5f6fe2c0c6dd800bb0ff9a_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d179a1c0b699dc97c958766a47fbf5af66d4a8e8 --- /dev/null +++ b/marked/F/T-REC-F.748.53-202503-I_PDF-E/b0a5d736df5f6fe2c0c6dd800bb0ff9a_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f60b7237063ca6fdb2f62f52e3a777187d651d401f175107cc4997ff91f463d2 +size 22711 diff --git a/marked/F/T-REC-F.748.53-202503-I_PDF-E/bd0b93e7a46ede276d0a3b79ac487bd9_img.jpg b/marked/F/T-REC-F.748.53-202503-I_PDF-E/bd0b93e7a46ede276d0a3b79ac487bd9_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..345811c57c79d705c816dea66345a231b6a4dd4c --- /dev/null +++ b/marked/F/T-REC-F.748.53-202503-I_PDF-E/bd0b93e7a46ede276d0a3b79ac487bd9_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b993c8bfd328871f04fa9231e178b31028e986d3d01e2f9623997e3b3a689420 +size 12694 diff --git a/marked/F/T-REC-F.748.53-202503-I_PDF-E/c036e2540a94b31357ceb0002f0cacab_img.jpg b/marked/F/T-REC-F.748.53-202503-I_PDF-E/c036e2540a94b31357ceb0002f0cacab_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..fddcd2274e9d799a9e42b620043c412a7cf44335 --- /dev/null +++ b/marked/F/T-REC-F.748.53-202503-I_PDF-E/c036e2540a94b31357ceb0002f0cacab_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:04384562716c78fce02a75641b8a77775562a049dbd32feec5f928f47804b98b +size 16501 diff --git a/marked/F/T-REC-F.748.53-202503-I_PDF-E/d9862ac8473902e7dfe441c5e88d47d8_img.jpg b/marked/F/T-REC-F.748.53-202503-I_PDF-E/d9862ac8473902e7dfe441c5e88d47d8_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7dce225b0a712b246a43474e26ea9c085fcc3db5 --- /dev/null +++ b/marked/F/T-REC-F.748.53-202503-I_PDF-E/d9862ac8473902e7dfe441c5e88d47d8_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:48ea9e1a0ea79e4e97eb3ea023130fe562f8c808e77de3fac0b75e12999db7cb +size 31016 diff --git a/marked/F/T-REC-F.748.53-202503-I_PDF-E/e9314c83043183351ed74908e9bf2f90_img.jpg b/marked/F/T-REC-F.748.53-202503-I_PDF-E/e9314c83043183351ed74908e9bf2f90_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..e6a772b2736ab60c9f55f44f7402bf5bf3a86a90 --- /dev/null +++ b/marked/F/T-REC-F.748.53-202503-I_PDF-E/e9314c83043183351ed74908e9bf2f90_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:96448fb7ae51f32c8740b4bbbeab7fb99637c032f485072f6e1178bab7f8b5d8 +size 33651 diff --git a/marked/F/T-REC-F.748.53-202503-I_PDF-E/f449330259f8d44842bfe0829754c00f_img.jpg b/marked/F/T-REC-F.748.53-202503-I_PDF-E/f449330259f8d44842bfe0829754c00f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..69999db51b2b0a5b9c94cee1d5fae52b903479b2 --- /dev/null +++ b/marked/F/T-REC-F.748.53-202503-I_PDF-E/f449330259f8d44842bfe0829754c00f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bfee62cf4ecbc4b9eac86b00835b26c119eccec391923d37250895e3e2f853f7 +size 63502 diff --git a/marked/F/T-REC-F.748.53-202503-I_PDF-E/f5dc5aa74ea5482955a1e8e161b5c549_img.jpg b/marked/F/T-REC-F.748.53-202503-I_PDF-E/f5dc5aa74ea5482955a1e8e161b5c549_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..e2720a1fd39f54b2ff1baa4fa1105fce60e1b2a4 --- /dev/null +++ b/marked/F/T-REC-F.748.53-202503-I_PDF-E/f5dc5aa74ea5482955a1e8e161b5c549_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:94e7509fa2c553c00341f91bfc12bfeed3b30f3c5474c14090c2fbc08bf615fc +size 31010 diff --git a/marked/F/T-REC-F.748.53-202503-I_PDF-E/ffe2d3e622dd30bea4caf15951c60e42_img.jpg b/marked/F/T-REC-F.748.53-202503-I_PDF-E/ffe2d3e622dd30bea4caf15951c60e42_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..296c351eeab3d88690957bb9bb44195677597bde --- /dev/null +++ b/marked/F/T-REC-F.748.53-202503-I_PDF-E/ffe2d3e622dd30bea4caf15951c60e42_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:42ec7ccecb476765ff9f4ed43d29c0e891b84b0e7d7ff95b17cbb33b7660fcfc +size 38154 diff --git a/marked/F/T-REC-F.748.53-202503-I_PDF-E/raw.md b/marked/F/T-REC-F.748.53-202503-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..a6ccbbe3b97f64bcc14cbb71848b50867a4f07d8 --- /dev/null +++ b/marked/F/T-REC-F.748.53-202503-I_PDF-E/raw.md @@ -0,0 +1,7171 @@ + + +Recommendation + +**ITU-T F.748.53 (03/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +**Representation and compression methods of +artificial intelligence models** + +![ITU logo](84a1d09fb489061482111515543b60dc_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +# ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +## Recommendation ITU-T F.748.53 + +# Representation and compression methods of artificial intelligence models + +## Summary + +Recommendation ITU-T F.748.53 provides a set of tools for efficient multimedia artificial intelligence model representation, compression and management, including model representation, compact model representation and encapsulation format. + +## History\* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.53 | 2025-03-16 | 21 | 11.1002/1000/16243 | + +## Keywords + +AI model, artificial intelligence, machine learning, model inference, model management, model representation, model storage, neural network. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2025 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|------------------------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this recommendation ..... | 3 | +| 4 Abbreviations and acronyms ..... | 3 | +| 5 Conventions ..... | 4 | +| 6 Overview..... | 4 | +| 7 Architecture framework of representation and compression methods of AI models .. | 5 | +| 7.1 AI model representation ..... | 5 | +| 7.2 AI model compression..... | 6 | +| 7.3 AI model encapsulation ..... | 7 | +| 8 AI model representation..... | 7 | +| 8.1 Overview ..... | 7 | +| 8.2 Network construction representation..... | 8 | +| 8.3 Operator representation ..... | 14 | +| 9 AI model compression ..... | 68 | +| 9.1 Compact representation ..... | 68 | +| 9.2 Compression workflow ..... | 82 | +| 10 AI model encapsulation ..... | 183 | +| 10.1 Encapsulation representation..... | 183 | +| 10.2 Encryption of model encapsulation ..... | 188 | +| 10.3 Residual updating of model encapsulation..... | 189 | +| 10.4 Distribution workflow ..... | 190 | +| Bibliography..... | 193 | + + + +# Representation and compression methods of artificial intelligence models + +# 1 Scope + +This Recommendation states the objectives and requirements of representation and compression methods of artificial intelligence (AI) models. + +It addresses the following subjects: + +- 1) The interoperable representation format of AI models to accommodate the variety of computing architectures, algorithm development frameworks and application scenarios. +- 2) The compact representation of neural-network-based AI models via different neural network compression algorithms to enable the applicability of AI models on resource-constrained devices. +- 3) Authentication and associated information of AI models during distribution, transmission and storage. + +The Recommendation can be used to guide relevant parties to improve the efficiency of resource-constrained equipment and reduce the cost of running large-scale AI models. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[IEEE 3161.9] IEEE Std 3161.9-2023, *IEEE Standard for Protocols and Interfaces of Digital Retina Systems*. + +[ISO/IEC 15938-17] ISO/IEC 15938-17:2024, *Information technology – Multimedia content description interface – Part 17: Compression of neural networks for multimedia content description and analysis*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 basic symbol vector** [b-IEEE 2941]: Including the basic symbol vector coexisting in the entire network, and the basic symbol vector corresponding to each layer, where the basic symbol vector corresponding to each layer is obtained from the basic symbol vector of the network according to a certain agreed rule. + +**3.1.2 block circulant matrix** [b-IEEE 2941]: The block circulant matrix is a matrix in which each block is a circulant matrix. + +**3.1.3 circulant matrix** [b-IEEE 2941]: The matrix arrangement of a regular structure for the current row/column that is the same as the previous row/column cycle. + +**3.1.4 compact representation** [b-IEEE 2941]: Expenses are compressed by a variety of neural networks in a compact representation of the serialized format after the algorithm is compressed. Examples: structured matrix, weight sharing among multiple models, sparse matrix of sparse coding, weighted quantization table, decomposition matrix of low-rank decomposition, and bit representation of binary neural network. + +**3.1.5 compressed neural network representation** [ISO/IEC 15938-17]: The representation of a neural network with model parameters encoded using compression tools. + +**3.1.6 convolutional neural network (CNN)** [b-ITU-T X.1235]: A deep learning algorithm that takes in an input image, assigns importance to various aspects/objects in the image and is able to differentiate one from the other. + +**3.1.7 deep learning model** [b-ITU-T F.748.12]: A deep learning algorithm used to solve a specific task, usually refers to the computational graph structure information and parameter information used to represent the deep learning algorithm. + +**3.1.8 disturbance vector** [b-IEEE 2941]: Extend the random vector so that its dimension is the dimension of the input vector of the network layer, and the resulting symbol vector. + +**3.1.9 encapsulation representation** [b-IEEE 2941]: Expenditure interfaces such as security information and identity verification. + +**3.1.10 high precision** [b-IEEE 2941]: The high-precision data or calculations generally are based on float16, float32, float64, or higher precision floating number encoding. + +**3.1.11 interoperable representation** [b-IEEE 2941]: The definition of the grammar of the neural network, the definition of supported operators, and the definition of the weight format (integer, floating-point and fixed-point). + +**3.1.12 layer** [ISO/IEC 15938-17]: A collection of nodes operating together at a specific depth within a neural network. + +**3.1.13 low precision** [b-IEEE 2941]: The low-precision data or calculations generally are based on 1 to 8 bits integer encoding. + +**3.1.14 machine learning (ML)** [b-ITU-T Y.3172]: Processes that enable computational systems to understand data and gain knowledge from it without necessarily being explicitly programmed. + +**3.1.15 model** [b-IEEE 2941]: Corresponding to a neural network that completes single or multiple tasks. + +**3.1.16 model compression** [b-IEEE 2941]: A method to reduce the scale of the neural network model and improve the operator and transmission efficiency of the neural network model. + +**3.1.17 neural network representation** [b-IEEE 2941]: Represents the basic syntax, semantics, operators, hyperparameters, and code stream definition methods of neural networks. + +**3.1.18 offset factor** [b-IEEE 2941]: One of the inherent parameters of neural networks. + +NOTE – It is also referred to simply as offset. + +**3.1.19 pruning** [ISO/IEC 15938-17]: Reduction of parameters in (a part of) the neural network. + +**3.1.20 random vector** [b-IEEE 2941]: The basic symbol vector corresponding to a single network layer. + +**3.1.21 reference random vector** [b-IEEE 2941]: The basic symbol vector for the coexistence of the entire network. + +**3.1.22 shared bias** [b-IEEE 2941]: The bias shared for multiple tasks, where one neural network handles multiple different tasks. + +**3.1.23 shared weight** [b-IEEE 2941]: The weights shared for multiple tasks, where one neural network handles multiple different tasks. + +**3.1.24 sparsification** [ISO/IEC 15938-17]: Increase of the number of zero-valued entries of a tensor. + +**3.1.25 structured matrix** [b-IEEE 2941]: A special kind of matrix can form a complete matrix by using less data and certain arrangement rules. + +**3.1.26 task** [b-IEEE 2941]: Use neural networks to achieve certain functions. Examples: image super resolution and image noise reduction. + +**3.1.27 unique weight** [b-IEEE 2941]: A neural network is used to process multiple different tasks, and for any task, it does not share the weight part. + +**3.1.28 weight** [b-IEEE 2941]: The strength of the connection between the different units. + +NOTE – Weight is one of the inherent parameters of a neural network. + +**3.1.29 weight aggregation dimension** [b-IEEE 2941]: Shared weight tensor and special weight vector for concatenation of dimensions. + +## 3.2 Terms defined in this recommendation + +This Recommendation defines the following terms: + +**3.2.1 AI model representation:** Represents the basic syntax, semantics, operators, hyperparameters and code stream definition methods of AI Models. + +**3.2.2 quantization:** A technique to change the bit width in the original high-precision complex model from a high-bit representation to a low-bit width. + +**3.2.3 decomposing matrices:** A technique to decompose the original weight matrix (or tensor) of a pretrained model into the multiplication and/or addition of multiple small weight matrices. + +**3.2.4 multiple models:** Several models with the same or similar architectures for the same or similar tasks are compressed simultaneously. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-------|-------------------------------------------------| +| AI | Artificial Intelligence | +| BN | Batch Normalization layer | +| CABAC | Context-Based Adaptive Binary Arithmetic Coding | +| Conv | Convolutional layer | +| Dict | Dictionary (data format) | +| EMA | Exponential Moving Averages | +| FC | Fully Connected layer | +| GAN | Generative Adversarial Network | +| HOSVD | Higher-Order Singular Value Decomposition | +| INT | Integer | +| KL | Kullback-Leibler divergence | +| MMSE | Minimal Mean Square Error | +| MPEG | Moving Picture Experts Group | + +| | | +|------|--------------------------------------------| +| MSE | Mean Square Error | +| NCHW | Number of batch, Channel, Height and Width | +| NNC | Neural Network Compression | +| NS | Not Specified | +| ReLU | Rectified Linear Units | +| SVD | Singular Value Decomposition | +| UINT | Unsigned Integer | + +# 5 Conventions + +The following conventions are used in this Recommendation: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted, if conformance to this Recommendation is to be claimed. +- The keywords "is recommended for" suggests the preferred course of action for achieving the desired outcome. Following this recommendation is encouraged to improve performance, efficiency, or compliance. While not mandatory, choosing this approach is likely to produce optimal results. + +# 6 Overview + +With the development of artificial intelligence technology represented by deep learning in recent years, neural network models have been widely used in the industry. However, large-scale integration of AI models into industry faces many challenges. + +- 1) Model representation considerations: The AI model representations of different AI frameworks are different in syntax and format. The different representations make them impossible to be compatible. At the same time, the implementation of a specific conversion tool is also a very complicated job. +- 2) Model compression considerations: The depth and size of neural network models permanently increases to achieve better performance, which restricts the deployment of the models on resource-constrained devices. In this case, model compression is necessary to reduce the computational complexity and memory footprints. +- 3) Model distribution and storage considerations: The plaintext model's version, corresponding deep learning framework, and even the corresponding training dataset should be considered in the process of model update and storage. However, there is no existing mechanism and standard that can support the authentication and associated information of neural network models. Efficient and economical distribution is the key concern before deployment. + +To address the above challenges, this Recommendation clarifies the objectives of the representation and compression methods of artificial intelligence models. In addition, this Recommendation specifies a complete process of AI model representation, compression and management, which is being used in the design, distribution and storage of the model. This framework curates and provides a platform for AI researchers and developers to manage, search and protect AI models for a variety of platforms and applications. + +# 7 Architecture framework of representation and compression methods of AI models + +This Recommendation defines the AI development interface, AI model interoperable representation, coding format and model encapsulated format for efficient multimedia AI model inference, storage, distribution and management. The structure schema is shown in Figure 1. + +![Figure 1: The structure schema of representation and compression methods of artificial intelligence models. The diagram shows a hierarchical structure. At the top is a dashed box labeled 'AI model encapsulation' containing 'Encapsulation representation' and 'Distribution workflow'. Below it is another dashed box labeled 'AI model compression' containing 'Compact representation' and 'Compression workflow'. At the bottom is a solid box labeled 'AI model representation'. A small label 'F.748.53(25)' is at the bottom right.](e9314c83043183351ed74908e9bf2f90_img.jpg) + +F.748.53(25) + +Figure 1: The structure schema of representation and compression methods of artificial intelligence models. The diagram shows a hierarchical structure. At the top is a dashed box labeled 'AI model encapsulation' containing 'Encapsulation representation' and 'Distribution workflow'. Below it is another dashed box labeled 'AI model compression' containing 'Compact representation' and 'Compression workflow'. At the bottom is a solid box labeled 'AI model representation'. A small label 'F.748.53(25)' is at the bottom right. + +**Figure 1 – The structure schema of representation and compression methods of artificial intelligence models** + +This Recommendation aims to break the barriers between various AI frameworks and promote the development and application of AI models in resource-constrained end-to-end computing platforms and wearable devices. The structure schema of representation and compression methods of artificial intelligence models is as follows: + +## 7.1 AI model representation + +AI model representation is a unified format that encapsulates AI models trained across various frameworks. It provides a standardized method for characterizing these models. This format thoroughly details the models' syntax and semantics, including data structures, syntactic boundaries, semantic interpretations and the complexities of training operator definitions. As illustrated in Figure 2, the AI model representation structure primarily comprises two components: model information and computational graph representation. The latter also integrates multiple operator nodes or variable nodes, with specific definitions tailored for each node type. + +![Figure 2: The structure of AI model representation. The diagram shows a nested structure. The outermost box is labeled 'AI model representation' and contains two main parts: 'Computational graph representation' and 'Model information'. The 'Computational graph representation' box contains two side-by-side boxes: 'Operator node' (with 'Input', 'Output', and 'Attributes') and 'Variable node' (with 'Tensor/Other type' and 'Tensor size'). A small label 'F.748.53(25)' is at the bottom right.](8e14350b4b669119a3bdfca7869110ca_img.jpg) + +F.748.53(25) + +Figure 2: The structure of AI model representation. The diagram shows a nested structure. The outermost box is labeled 'AI model representation' and contains two main parts: 'Computational graph representation' and 'Model information'. The 'Computational graph representation' box contains two side-by-side boxes: 'Operator node' (with 'Input', 'Output', and 'Attributes') and 'Variable node' (with 'Tensor/Other type' and 'Tensor size'). A small label 'F.748.53(25)' is at the bottom right. + +**Figure 2 – The structure of AI model representation** + +- a) **Model information:** This part encompasses details such as the model's name, version, framework, manuscript and contributors, offering a comprehensive overview of the model. + +- b) **Computational graph representation:** Serving as an intermediate representation within an AI system, the computational graph acts as a bridge between source and target languages. This facilitates program analysis and optimization by the compiler. Each computational graph is composed of operator nodes, variable nodes, parameters and values. +- 1) **Operator node:** These nodes represent the computational operators within the network. Parameters are associated with network weights and other transient variables, while values pertain to the network's constant values. Each operator node includes its own node information, inputs and outputs. The node information specifies the operator node's name, such as 'abs', 'relu', etc. Inputs and outputs refer to the respective input and output variable nodes of the operator node, while attributes define the operator node's characteristics. + - 2) **Variable node:** Every operator node possesses its own node information and value. The variable node's node information denotes the variable's name, and each variable node is classified by its variable type, such as a dense tensor or sparse tensor. + +## 7.2 AI model compression + +The deployment of artificial intelligence (AI) models in real-world applications often encounters limitations due to hardware resource constraints. To address these challenges, model compression techniques are essential for reducing model size and computational requirements while maintaining high accuracy and performance levels. This clause delves into the compression strategies of multiple models, quantization, pruning and the use of structured matrices, offering a comprehensive guide to optimizing AI model architecture for efficiency and scalability. + +- a) **Compression workflow:** Anchored in a suite of AI model compression strategies, this workflow is engineered to curtail computational complexity and memory demands tailored to varying requirements. It integrates disparate methodologies to condense the AI model, catering to distinct operational necessities. The compression workflow outlines a systematic approach for applying compression techniques to artificial intelligence (AI) models, aiming to reduce their size and computational demand without significantly sacrificing their performance. This process involves several key steps, from initial analysis through to the final deployment of the compressed model. Each step is designed to ensure that the model not only becomes more efficient in terms of storage and execution but also maintains or even improves its accuracy and effectiveness. Below is a detailed description of the compression workflow: +- 1) **Pre-compression analysis:** Assess the model to identify optimal compression targets and strategies. + - 2) **Compression implementation:** Apply the chosen compression techniques iteratively, adjusting parameters as necessary to achieve the best trade-off between size, speed and accuracy. + - 3) **Post-compression optimization:** Fine-tune the compressed model to restore or improve performance relative to the original model. + - 4) **Evaluation and deployment:** Thoroughly evaluate the compressed model against performance benchmarks and deploy in the target environment. + +Figure 3 represents the workflow of AI model compression. + +![Flowchart of the AI model compression workflow. The steps are: Compression analysis, Compression implementation, Optimization, and Deployment. There is a feedback loop from the Deployment step back to the Compression analysis step.](7f17c430b9598e4d748a8041457810b3_img.jpg) + +``` + +graph LR + A[Compression analysis] --> B[Compression implementation] + B --> C[Optimization] + C --> D[Deployment] + D --> A + +``` + +F.748.53(25) + +Flowchart of the AI model compression workflow. The steps are: Compression analysis, Compression implementation, Optimization, and Deployment. There is a feedback loop from the Deployment step back to the Compression analysis step. + +**Figure 3 – The workflow of AI model compression** + +- b) **Compact representation:** Through the lens of model compression, this paradigm streamlines AI models into a more compact form. It necessitates a well-defined serialization schema for the compressed AI model, abstracting away from the granular specifics of individual neural network compression techniques. +- 1 **Multiple models:** The concept of compressing multiple models simultaneously leverages shared characteristics among models to achieve compression. This approach is particularly effective in scenarios where several models perform similar tasks or have overlapping features. By identifying and exploiting commonalities, such as shared weights or representations, significant reductions in total model size can be achieved without sacrificing individual model performance. + - 2 **Quantization:** Quantization reduces the precision of the model's parameters from floating-point representation to fixed-point or lower bit-width integers. This process not only decreases the model size but also reduces the computational intensity of inference operations, making it highly suitable for deployment on resource-constrained devices. + - 3 **Pruning:** Pruning involves systematically removing weights or neurons from the model that contribute minimally to its output, thus reducing the model's complexity and size. Effective pruning strategies balance model simplification with the preservation of accuracy. + - 4 **Structured matrix:** Leveraging structured matrices involves reconfiguring model weights into matrices with predefined patterns or structures, such as circulant or Toeplitz matrices. This approach reduces the number of independent parameters, leading to models that are both compact and fast to execute. + +## 7.3 AI model encapsulation + +As the number and size of models increase, there are higher requirements for model transmission efficiency, security and standardization. In this clause, model encapsulation representation, model encryption and decryption, model residual update and distribution methods will be standardized. + +- a) **Encapsulation representation:** To ensure the consistency of data and operation at different ends of the AI model, additional validation and information are added to the original representation. Thus, the system environment, input and output formats, etc. are clarified to facilitate the use on different devices. +- b) **Encryption and decryption:** AI model training consumes a lot of time and resources and is an important digital asset. In the transmission of the data, there is a risk of leakage and tampering of model information due to illegal attacks. Encryption and decryption mean become an effective method to prevent attacks. +- c) **Residual updating:** The model often updates in its usage cycle, but updating the entire model often consumes significant resources. Therefore, updating in a residual manner can effectively reduce bandwidth and improve deployment efficiency. +- d) **Distribution workflow:** The components described above are accomplished through a standardized model distribution process. Similarly, the client needs to follow certain processes for implementing decryption, updating and verification for the correct inference of the model. + +# 8 AI model representation + +## 8.1 Overview + +AI model representation defines syntax and semantics of AI models, including data structure, syntax description, semantic description and training operator definitions. + +- a) **Data structure:** Defines the data structure of the AI model structure and parameters. + +- b) Syntax description: Defines the grammar of AI model representation, using definitions ranging from coarse-grained to fine-grained. The basic grammatical description of the AI model representation is constructed from structure definition, computational graph definition, node definition, etc. to tensor definition and basic data type definition. +- c) Semantic description: Defines the operators of the AI model, each operator includes input, output, type and attribute, and all the definitions are described in detail. +- d) Training operator definition: Defines loss function and inverse operators required in the process of AI models training. + +## 8.2 Network construction representation + +This clause defines the grammar of neural network representation, using definitions ranging from coarse-grained to fine-grained. The basic grammatical description of the neural network representation is constructed from structure definition, computational graph definition, node definition, etc. to tensor definition and basic data type definition. + +Overview of the principles of use: Neural network representation grammar should be implemented in accordance with standard specifications. When completed by a specific computing system (deep learning platform and related software and hardware), the following principles shall be followed: + +- a) The definition of the necessary levels should be considered, including: + - 1) Model structure definition; + - 2) Definition of computational graph; + - 3) Node definition; + - 4) Tensor definition; + - 5) Definition of basic data types. +- b) When the computing system is implemented, the grammatical elements can be adjusted according to actual needs, including but not limited to: + - 1) Keyword (parameter) naming; + - 2) Operator naming; + - 3) Data type. + +### 8.2.1 Computational graph representation + +#### 8.2.1.1 Data structure + +##### 8.2.1.1.1 Data structure of neural network structure + +This clause defines the data structure of the neural network structure, where the encoding structure of the calculation graph and model information are shown in Figure 4. Multiple fields are compressed and stored sequentially or nested, where each field uses *tag* and *value* to describe all the information. There are *required*, *repeated* and *optional* formats for a field, where *required* indicates that the field shall exist, *repeated* indicates that the field could repeat and *optional* indicates that the field is optional. + +![Diagram showing the coding structure of neural network structure. It consists of multiple 'Field' units, each containing 'Tag' and 'Value' sub-fields. The units are grouped under a 'Neural network' label. The diagram is labeled F.748.53(25).](c036e2540a94b31357ceb0002f0cacab_img.jpg) + +The diagram illustrates the coding structure of a neural network. It shows a sequence of 'Field' units. Each 'Field' unit is a rectangle containing two sub-fields: 'Tag' and 'Value'. The units are arranged horizontally, with an ellipsis (...) between the second and third units to indicate repetition. Above each unit is a bracket labeled 'Field'. Below the entire sequence of units is a bracket labeled 'Neural network'. In the bottom right corner of the diagram, the text 'F.748.53(25)' is present. + +Diagram showing the coding structure of neural network structure. It consists of multiple 'Field' units, each containing 'Tag' and 'Value' sub-fields. The units are grouped under a 'Neural network' label. The diagram is labeled F.748.53(25). + +**Figure 4 – The coding structure of neural network structure** + +The field for different data types adopting different encoding methods is shown in Figure 5: + +![Figure 5 – Different types of encoding](bd0b93e7a46ede276d0a3b79ac487bd9_img.jpg) + +| | | +|------------|-------| +| Tag | Value | +|------------|-------| + + (1) + +| | | | +|------------|---------------|-------| +| Tag | Length | Value | +|------------|---------------|-------| + + (2) + +| | | | +|------------|-------|------------| +| Tag | Value | Tag | +|------------|-------|------------| + + (3) + + +F.748.53(25) + +Figure 5 – Different types of encoding + +**Figure 5 – Different types of encoding** + +a) *Tag* coding: + +The coding of the *Tag* adopts $field\_number \ll 3 \mid wire\_type$ , where *field\_number* is the label defined for each field and *wire\_type* is the data type of the field. Finally, the above unsigned integer is adopted variant encoding. + +b) *Int32/int64/uint32/uint64/bool/enum* encoding: + +These types use the encoding method in Figure 5(1). The *Tag* is stored in the first byte by the *Tag* encoding method. The *value* is variant-encoded and stored in the second byte. + +c) *String* encoding: + +This type uses the encoding method in Figure 5(2). The first byte encodes the *Tag*, the second byte encodes the *length* of the string in a variant, and the remaining bytes store the *value*. + +d) *Float32/Float64* encoding: + +These types use the encoding method in Figure 5(1). The first byte encodes the *tag*, and the following bytes directly copy the float memory. + +e) *Nested structure* coding: + +This type uses the encoding method in Figure 5(1). The first byte encodes the *tag*, and the *value* uses encoding serialization regardless of whether it is a unit or a tuple. + +f) *Repeated* coding: + +This type uses the coding scheme in Figure 5(3). The overall number is calculated by the number of repeated tags. The first byte encodes the tag, and the value uses encoding serialization regardless of whether it is a unit or a tuple. + +##### 8.2.1.1.2 Data structure of neural network parameters + +The parameters of a neural network are encoded and stored independently, and are composed of a series of operator parameters. The parameter coding method is shown in Figure 6. + +![Figure 6 – Encoding method of neural network parameters](9e1c95ee7eb045977f18f011f49520c7_img.jpg) + +| Tag | | +|------------|-------| +| Key | Value | +| Key | Value | +| ... | | +| Key | Value | + + Layer 1 + +| Tag | | +|------------|-------| +| Key | Value | +| Key | Value | +| ... | | +| Key | Value | + + Layer 2 + +... + +| Tag | | +|------------|-------| +| Key | Value | +| Key | Value | +| ... | | +| Key | Value | + + Layer n + +F.748.53(25) + +Figure 6 – Encoding method of neural network parameters + +**Figure 6 – Encoding method of neural network parameters** + +For each operator or each node of the neural network, the *tag* corresponds to the custom name of each layer. Each operator or node corresponds to multiple fields, representing multiple parameters. The *key* of the field corresponds to the keyword or name of the parameter, and the *value* stores the corresponding array or data. + +For the *value* in Figure 6, the encoding method for the array is shown in Figure 7. It includes data types (*dtypes*), matrix dimensions (*dim count*), the size of each dimension (*dimensions*), the stride of each dimension (*strides*), and specific data (*data*). The corresponding value could be searched by index, data type and the stride of each dimension. + +| | | | +|-------------------|---------|---| +| dtype | float32 | | +| dim count | 2 | | +| dimensions | 3 | 3 | +| strides | 12 | 4 | +| data | * | | + +F.748.53(25) + +**Figure 7 – Data structure of array** + +#### 8.2.1.2 Definition of computational graph + +The computational graph is an intermediate representation in AI system. It is a representation of a program between the source and target languages, which facilitates program analysis and optimization for the compiler. + +The definition of the computational graph is shown in Table 1. + +The *id* of the computational graph represents the unique serial number of the computational graph. The computational graph may call several sub-computational graphs: + +- a) The conditional statement, the internal operator and variable node of the loop statement in the neural network representation constitute a new operator node or a subgraph; +- b) Several operator nodes in the neural network may be merged into a new operator node and handed over to a third-party efficient calculation engine (such as *Nvidia TensorRT*, *Intel Ngraph*, etc.) to execute. The new operator node and the corresponding variable node can be expressed by a sub-computation graph. + +Each computational graph contains its own operator nodes, parameter and value. The operator nodes correspond to the calculation operators in the network, the parameter corresponds to the network weights and other temporary variables, and the value corresponds to a constant value in the network. + +**Table 1 – Graph definition** + +| Field | Type | Description | +|-----------------------|--------------------------|--------------------------------------------| +| Node (optional) | Operator Node (repeated) | Operator nodes of the computational graph | +| Value (optional) | Value (repeated) | Values of the computational graph | +| Parameters (optional) | Variable (repeated) | Parameters of the computational graph | +| id | string | The unique name of the computational graph | +| Return (optional) | Operator Node | Return value | + +IR uses Value to represent the constant expression. Each Value carries a constant value. The attribute value of a Value indicates the constant value the user defined. The value can be a tensor, primitive function (Primitive), a graph or some other type. (Such as string, scalar and so on.) + +**Table 2 – Value definition** + +| Field | Type | Description | +|-------|-------|-------------| +| Value | Value | Value | + +#### **8.2.1.3 Definition of operator node** + +The operator node definition is shown in Table 3. The *operator* defines the name of the operator node, such as *abs*, *relu*, etc. *Input* and *output* are the input variable node and output variable node of the operator node. *attribute* is the attributes of the operator node. + +**Table 3 – The definition of operator node** + +| Operator Node | Parameters | | +|----------------------|-------------------------|--------------------| +| Field | Types | Definition | +| name | string | Node's name | +| operator | string | Operator's name | +| input | map | Node's input | +| output | map | Node's output | +| attribute | map | Node's attributes | +| doc_string(optional) | string | Node's description | +| Definition(optional) | string | Node's definition | + +#### **8.2.1.4 Definition of variable node** + +Variable node as defined in Table 4. The name of the variable node represents the name of the variable; *variable\_type* represents the type of the variable, such as dense tensor or sparse tensor, etc. which are represented by integer values. + +**Table 4 – The definition of variable node** + +| Variable Node | Parameters | | +|---------------|------------|-------------| +| Field | Types | Definition | +| name | string | Node's name | +| variable_type | int64 | Node's type | + +### **8.2.2 Model structure representation** + +#### **8.2.2.1 The definition of contributor list** + +The list of contributors is defined in Table 5. + +**Table 5 – The definition of contributor list** + +| ContributorList | Parameters | | +|-----------------|-------------------|---------------------------| +| Field | Types | Definition | +| name | string (repeated) | Contributor's name | +| email | string (repeated) | Contributor's Email | +| institute | string (repeated) | Contributor's institution | + +#### 8.2.2.2 Definition of model structure + +The model structure definition is shown in Table 6. + +**Table 6 – The definition of model structure** + +| Model | Parameters | | +|-------------------|-------------------|-------------------------------------------| +| Field | Types | Definition | +| Version | int64 | The standard version | +| contributors | ContributorsList | The list of contributors | +| framework_name | string | The name of initial training framework | +| framework_version | string | The version of initial training framework | +| model_name | string | Model name | +| model_version | string | Model version | +| doc_url | string | The document link of model description | +| graph | graph | The calculation graph of the model | + +#### 8.2.2.3 Definition of attribute + +The attribute is shown in Table 7. + +**Table 7 – The definition of attribute** + +| Attribute | Specify Parameters | | +|------------------|---------------------------|--------------------| +| Field | Types | Definition | +| val | Type | Attribute value | +| list | list(Type) | Attribute sequence | + +#### 8.2.2.4 Definition of other type + +Other types are shown in Table 8. and refers to The IEEE Standard for Floating-Point Arithmetic [b-IEEE 754]. + +**Table 8 – The definition of other type** + +| Type | Parameters | | +|--------------|-------------------|----------------------------| +| Field | Types | Definition | +| i | int64 | Integer type | +| f | float | Single floating point type | +| d | double | Double floating point type | +| s | bytes | Bit type | +| b | bool | Boolean type | +| vt | vector | Array type | +| c | callable | Call type | +| v | ValueType | Tensor data type | +| t | Tensor | Tensor type | +| shape | TensorShape | Tensor size type | + +#### 8.2.2.5 Definition of tensor type + +The definition of tensor data type is shown in Table 9. + +**Table 9 – The definition of tensor type** + +| ValueType | | +|-----------|------------| +| Label | Types | +| 1 | UNDEFINED | +| 2 | BINARY | +| 3 | INT2 | +| 4 | INT4 | +| 5 | INT8 | +| 6 | INT16 | +| 7 | INT32 | +| 8 | INT64 | +| 9 | UINT2 | +| 10 | UINT4 | +| 11 | UINT8 | +| 12 | UINT16 | +| 13 | UINT32 | +| 14 | UINT64 | +| 15 | FLOAT16 | +| 16 | FLOAT32 | +| 17 | FLOAT64 | +| 18 | DOUBLE | +| 19 | BOOL | +| 20 | STRING | +| 21 | COMPLEX64 | +| 22 | COMPLEX128 | + +#### 8.2.2.6 Definition of tensor + +The definition of tensor is shown in Table 10. + +**Table 10 – The definition of tensor** + +| Tensor | Specify parameters | | +|----------------------|---------------------|------------------------------------| +| Field | Types | Definition | +| Name(optional) | string | Tensor's name | +| Version(optional) | string | Tensor's version information | +| Shape | TensorShape | Tensor's size | +| data_type | ValueType | Tensor's data type | +| content | ValueType(repeated) | Tensor's data area | +| int32_data(optional) | int32_data | Tensor 32 -bit signed integer data | + +**Table 10 – The definition of tensor** + +| Tensor | Specify parameters | | +|-----------------------|---------------------------|---------------------------------------------| +| Field | Types | Definition | +| uint32_data(optional) | uint32_data | Tensor 32 -bit unsigned integer data | +| int64_data(optional) | int64_data | Tensor 64 -bit signed integer data | +| uint64_data(optional) | uint64_data | Tensor 64 -bit unsigned integer data | +| float_data(optional) | float_data | Tensor single-precision floating point data | +| double_data(optional) | double_data | Tensor double-precision floating-point data | +| bool_data(optional) | bool_data | Tensor boolean data | +| string_data(optional) | string_data | Tensor string data | + +#### 8.2.2.7 Definition of tensor size + +Tensor size is shown in Table 11. + +**Table 11 – The definition of tensor size** + +| TensorShape | Parameters | | +|--------------------|-------------------|-----------------------------------| +| Field | Types | Definition | +| unknown | bool | Whether there is size information | +| dim | dimension | Dimension | + +## 8.3 Operator representation + +This clause defines the operators of the neural network, each operator includes input, output, type and attribute, and all the definitions are described in detail. + +The definition of basic operator: Basic operators are the fundamental mathematical or transformational operations in neural networks, used to perform basic transformations and computations on input data. They possess lower computational complexity and parameter quantity usually, serving as the fundamental computational units in constructing neural networks usually. + +The definition of advanced operator: Advanced operators are complex operations that are further combined and expanded upon the basic operators. Advanced operators are capable of handling more complex and structured data, and they possess higher computational complexity and parameter quantity, enabling the implementation of more powerful and sophisticated functionalities. + +Principle of use: + +The Operators contained in the neural network model are implemented by a specific computing system and the following principles should be followed: + +- a) The elements involved in the calculation should be considered, including: + - 1) Supportable parameters; + - 2) Supportable types. +- b) Adjustments should be made according to actual needs, including but not limited to: + - 1) Keyword (parameter) naming; + - 2) Operator naming; + - 3) Data type support range. + +### 8.3.1 Basic operator unit representation + +**Table 12 – abs operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-----------------------------------------------------------------------------|--------|---------|---------------|---------------------------| +| abs | Calculate the absolute value of the input data and apply it to each element | Input | X | Input tensor | float16, float32, float64 | +| | | Output | Y | Output tensor | float16, float32, float64 | + +**Table 13 – add operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|--------------------------------------------------------------------------------------------------------------------------------------|------------|-----------|-------------------------------|---------------------------| +| add | Perform the addition of two inputs and broadcast the parameters on the right to match the size of the input on the left if necessary | Input | X | Input operand 1 | float16, float32, float64 | +| | | | Y | Input operand 2 | float16, float32, float64 | +| | | Output | Z | Output tensor | float16, float32, float64 | +| | | Attributes | axis | Calculated dimension | int | +| | | | broadcast | Whether to allow broadcasting | bool | + +**Table 14 – and operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|--------------------------------------------------------------|------------|-----------|-------------------------------|-----------| +| and | Perform logical operators and operators on the input X and Y | Input | X | Input operand 1 | bool | +| | | | Y | Input operand 2 | bool | +| | | Output | Z | Output tensor | bool | +| | | Attributes | axis | Calculated dimension | int | +| | | | broadcast | Whether to allow broadcasting | bool | + +**Table 15 – argmax operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-----------------------------------------------------------------------------------------------------------|--------|---------|---------------|---------------------------| +| argmax | Calculate the index value of the largest element of the input tensor according to the provided dimensions | Input | X | Input tensor | float16, float32, float64 | +| | | Output | Y | Output tensor | int64 | + +**Table 15 – argmax operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-------------|------------|---------|----------------------|-----------| +| | | Attributes | axis | Calculated dimension | int | + +**Table 16 – argmin operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-----------------------------------------------------------------------------------------------------------|------------|---------|----------------------|---------------------------| +| argmin | Calculate the index value of the smallest element of the input tensor according to the provided dimension | Input | X | Input tensor | float16, float32, float64 | +| | | Output | Y | Output tensor | int64 | +| | | Attributes | axis | Calculated dimension | int | + +**Table 17 – argsort operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------|------------|---------------------------------------------------------------------------------------------|------------------| +| argsort | Arrange the input variables in ascending order along a given axis, and output the sorted data and the corresponding index, the dimensions of which are the same as the input | Input | X | Input tensor | float32, float64 | +| | | Output | Y | Output tensor | float32, float64 | +| | | | Z | Output index | int64 | +| | | Attributes | axis | Operate the input tensor on the axis | int | +| | | | descending | Optional, algorithm sorting direction | bool | +| | | | name | Optional, the prefix identifier of the network layer output and weight, the default is None | string | + +**Table 18 – size operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|--------------------------------------------------|------------|---------|---------------------------------------------------------------------------------------------|--------------------------------------| +| size | Count the number of elements of the input tensor | Input | X | Input tensor | float32, float64, int32, int64, bool | +| | | Output | Y | Output tensor | int64 | +| | | Attributes | name | Optional, the prefix identifier of the network layer output and weight, the default is None | string | + +**Table 19 – assign operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|------------------------------------|--------|---------|-----------------------------|--------------------------------------| +| assign | Copy input tensor to output tensor | Input | X | Input tensor or numpy array | float32, float64, int32, int64, bool | +| | | Output | Y | Output tensor | float32, float64, int32, int64, bool | + +**Table 20 – cast operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|--------------------------------------------------------------------|------------|---------|---------------------------|-----------| +| cast | Convert the input tensor element to the data type pointed to by to | Input | X | Input tensor | all | +| | | Output | Y | Output tensor | all | +| | | Attributes | to | Target type of conversion | string | + +**Table 21 – ceil operator** + +| Operator | Description | Field | Keyword | Definition | Type of data | +|----------|-------------|--------|---------|---------------|---------------------------| +| ceil | Rounded up | Input | X | Input tensor | float16, float32, float64 | +| | | Output | Y | Output tensor | float16, float32, float64 | + +**Table 22 – concat operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|---------------------------------------|------------|---------|-------------------|---------------------------| +| concat | Concatenate the list of input tensors | Input | X | Input tensor list | float16, float32, float64 | +| | | Output | Y | Output tensor | float16, float32, float64 | +| | | Attributes | axis | Merge dimensions | Int | + +**Table 23 – constant operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|--------------------------------|------------|---------|----------------------------------------|---------------------------| +| constant | Output a constant value tensor | Output | Output | Output tensor | float16, float32, float64 | +| | | Attributes | value | The value of the output tensor element | tensor | + +**Table 24 – cos operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-----------------|------------|---------|---------------------------------------------------------------------------------------------|------------------| +| cos | Cosine function | Input | X | Input tensor | float32, float64 | +| | | Output | Y | Output tensor | float32, float64 | +| | | Attributes | name | Optional, the prefix identifier of the network layer output and weight, the default is None | string | + +**Table 25 – cos\_sim operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|----------------------------|--------|---------|---------------|--------------| +| cos_sim | Cosine similarity operator | Input | X | Input tensor | float32 | +| | | | Y | Input tensor | float32 | +| | | Output | Z | Output tensor | int
float | + +**Table 26 – crop operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|----------------------------------------------------------|------------|---------|------------------------------------------------------------|------------------| +| crop | According to the offset and shape, crop the input tensor | Input | X | Input tensor | float32, float64 | +| | | Output | Y | Output tensor | float32, float64 | +| | | Attributes | shape | Dimensions of the output tensor | list of ints | +| | | | offsets | Optional, the offset of the crop in each dimension | list of ints | +| | | | name | Optional, used when developers print debugging information | string | + +**Table 27 – greater\_equal operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|---------------|-----------------------------------------------------------|------------|-----------|-------------------------------|--------------------------------| +| greater_equal | Apply the logical operator greater_equal to input X and Y | Input | X | Input tensor | float32, float64, int32, int64 | +| | | | Y | Input tensor | float32, float64, int32, int64 | +| | | Output | Z | Output tensor | bool | +| | | Attributes | axis | Calculated dimension | int64 | +| | | | broadcast | Whether to allow broadcasting | bool | + +**Table 28 – data operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|---------------------------------------------------------------------------------------------------------------------------------------------------------|------------|---------|---------------------------------|---------------------------------------------------------------------------------| +| data | The input tensor of the graph, which can be accessed by the operators in the calculation graph. This tensor can be used as a placeholder for data input | Output | Y | Output tensor | bool,
float16,
float32,
float64,
int8,
int16,
int32,
uint8 | +| | | Attributes | shape | Dimensions of the output tensor | list of ints | +| | | | dtype | Type of output tensor | string | + +**Table 29 – diag operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|---------------------------------------------------------------------------------------------------------------------|--------|---------|---------------|-----------------------------------------| +| diag | Create a square matrix and use the input tensor to specify the values of the diagonal elements of the square matrix | Input | X | Input tensor | float32,
float64,
int32,
int64 | +| | | Output | Y | Output tensor | float32,
float64,
int32,
int64 | + +**Table 30 – div operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|----------------------------------------------|------------|-----------|-------------------------------|---------------------------------| +| div | Arithmetic division of input tensors X and Y | Input | X | Input operand 1 | float16,
float32,
float64 | +| | | | Y | Input operand 2 | float16,
float32,
float64 | +| | | Output | Z | Output tensor | float16,
float32,
float64 | +| | | Attributes | axis | Calculated dimension | int | +| | | | broadcast | Whether to allow broadcasting | bool | + +**Table 31 – mod operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|------------------------------------------------------------------------------------------------------------------|------------|---------|---------------------------------------------------------------------------------------------|--------------------------------| +| mod | Take the modulo operator element by element, and save the output element at each position to the returned result | Input | X | Input tensor | float32, float64, int32, int64 | +| | | | Y | Input tensor | float32, float64, int32, int64 | +| | | Output | Z | Output tensor | float32, float64, int32, int64 | +| | | Attributes | axis | Optional, the index when the Y dimension corresponds to the X dimension | int32 | +| | | | name | Optional, the prefix identifier of the network layer output and weight, the default is None | string | + +**Table 32 – elementwise\_min operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|-----------------|-------------------------------------------------------------------------------------------------------------------------------------------|------------|---------|---------------------------------------------------------------------------------------------|--------------------------------| +| elementwise_min | Compare the two input multi-dimensional tensors element by element, and save the smaller elements in each position to the returned result | Input | X | Input tensor | float32, float64, int32, int64 | +| | | | Y | Input tensor | float32, float64, int32, int64 | +| | | Output | Z | Output tensor | float32, float64, int32, int64 | +| | | Attributes | axis | Optional, the index when the Y dimension corresponds to the X dimension | int32 | +| | | | name | Optional, the prefix identifier of the network layer output and weight, the default is None | string | + +**Table 33 – elementwise\_max operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|-----------------|----------------------------------------------------------------------------------------------------------------------------------------|------------|---------|-----------------------------------------------------------------------------------|-----------------------------------------| +| elementwise_max | Compare the two input multi-dimensional tensors element by element, and save the elements with larger positions in the returned result | Input | X | Input tensor | float32,
float64,
int32,
int64 | +| | | | Y | Input tensor | float32,
float64,
int32,
int64 | +| | | Output | Z | Output tensor | float32,
float64,
int32,
int64 | +| | | Attributes | axis | The index when the Y dimension corresponds to the X dimension | int32 | +| | | | name | The prefix identifier of the network layer output and weight, the default is None | string | + +**Table 34 – equal operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|---------------------------------------------------------------|------------|-----------|-------------------------------|---------------------------------| +| equal | Apply the logical operator equal to the input tensors X and Y | Input | X | Input operand 1 | float16,
float32,
float64 | +| | | | Y | Input operand 2 | float16,
float32,
float64 | +| | | Output | Z | Output tensor | bool | +| | | Attributes | axis | Calculated dimension | Int | +| | | | broadcast | Whether to allow broadcasting | bool | + +**Table 35 – exp operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|------------------------------------------------------------|--------|---------|---------------|---------------------------| +| exp | Return the exp index according to the input tensor element | Input | X | Input tensor | float16, float32, float64 | +| | | Output | Y | Output tensor | float16, float32, float64 | + +**Table 36 – expand operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|---------------------------------------------------------------------------------------------------------------------------------------|------------|--------------|---------------------------------------------------------------------------------------------|--------------------------------------| +| expand | Copy each dimension of the input according to the parameters, and set the number of times of copying for each dimension of the input. | Input | X | Input tensor | float32, float64, int32, int64, bool | +| | | | expand_times | The number of times to expand on each dimension, tensor or constant List | int32 | +| | | Output | Y | Output tensor | int32, int64, list of ints | +| | | Attributes | name | Optional, the prefix identifier of the network layer output and weight, the default is None | string | + +**Table 37 – flatten operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|--------------------------------------------------------|------------|---------|----------------------|---------------------------| +| flatten | Flatten the input tensor into a two-dimensional matrix | Input | X | Input tensor | float16, float32, float64 | +| | | Output | Y | Output tensor | float16, float32, float64 | +| | | Attributes | axis | Calculated dimension | int | + +**Table 38 – floor operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-------------|--------|---------|---------------|---------------------------| +| floor | Round down | Input | X | Input tensor | float16, float32, float64 | +| | | Output | Y | Output tensor | float16, float32, float64 | + +**Table 39 – gather operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|--------------------------------------------------------|------------|---------|-------------------------------|------------------------------| +| gather | Return input tensor X
Content indexed
as indices | Input | X | Input tensor | float16,
float32, float64 | +| | | | indices | Input integer tensor
index | int32, int64 | +| | | Output | Y | Output tensor | float16,
float32, float64 | +| | | Attributes | axis | Calculated dimension | int | + +**Table 40 – greater operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-------------------------------------------------------|------------|-----------|----------------------------------|-----------| +| greater | Apply logical
operator greater to
input X and Y | Input | X | Input operand 1 | Tensor | +| | | | Y | Input operand 2 | Tensor | +| | | Output | Z | Output tensor | bool | +| | | Attributes | axis | Calculated
dimension | int | +| | | | broadcast | Whether to allow
broadcasting | bool | + +**Table 41 – less operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|----------------------------------------------------------------|------------|-----------|----------------------------------|------------------------------| +| less | Apply logical
operator less to the input
tensors X and Y | Input | X | Input operand 1 | float16,
float32, float64 | +| | | | Y | Input operand 2 | float16,
float32, float64 | +| | | Output | Z | Output tensor | bool | +| | | Attributes | axis | Calculated
dimension | int | +| | | | broadcast | Whether to allow
broadcasting | bool | + +**Table 42 – less\_equal operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|------------|----------------------------------------------------------|------------|-----------|----------------------------------|-----------------------------------| +| less_equal | Apply logical operator
less_equal to
input X and Y | Input | X | Input tensor | float32, float64,
int32, int64 | +| | | | Y | Input tensor | float32, float64,
int32, int64 | +| | | Output | Z | Output tensor | bool | +| | | Attributes | axis | Calculated
dimension | int | +| | | | broadcast | Whether to allow
broadcasting | bool | + +**Table 43 – is\_empty operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-----------------------------------|------------|---------|------------------------------------------------------------------------------------------------------|-----------------------------------------------------| +| is_empty | Test if the input tensor is empty | Input | X | Input tensor | float16,
float32,
float64,
int32,
int64 | +| | | Output | Y | Output tensor | bool | +| | | Attributes | cond | Optional, if this parameter is passed in, the test result of the given x is stored in this parameter | int, float | + +**Table 44 – isfinite operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|--------------------------------------------------------|--------|---------|---------------|-----------------------------| +| isfinite | Test whether the input tensor contains infinite values | Input | X | Input tensor | int32,
float,
float64 | +| | | Output | Y | Output tensor | bool | + +**Table 45 – log operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|--------------------------------------------------------------------|--------|---------|---------------|---------------------------------| +| log | Calculate the natural logarithm of a given input tensor by element | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | + +**Table 46 – mat\_mul operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-----------------------|------------|-------------|------------------------|---------------------------------| +| mat_mul | Matrix multiplication | Input | X | Input tensor 1 | float16,
float32,
float64 | +| | | | Y | Input tensor 2 | float16,
float32,
float64 | +| | | Output | Z | Output tensor | float16,
float32,
float64 | +| | | Attributes | transpose_x | Whether to transpose X | bool | +| | | | transpose_y | Whether to transpose Y | bool | + +**Table 47 – mul operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-----------------------------------|------------|-----------|-------------------------------|---------------------------------| +| mul | Perform arithmetic multiplication | Input | X | Input tensor 1 | float16,
float32,
float64 | +| | | | Y | Input tensor 2 | float16,
float32,
float64 | +| | | Output | Z | Output tensor | float16,
float32,
float64 | +| | | Attributes | broadcast | Whether to allow broadcasting | bool | +| | | | axis | Calculated dimension | int | + +**Table 48 – negative operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-----------------------------------------|--------|---------|---------------|---------------------------------| +| negative | Invert each element of the input tensor | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | + +**Table 49 – not operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-------------------------------------------|--------|---------|---------------|-----------| +| not | Return the negation of each input element | Input | X | Input tensor | bool | +| | | Output | Y | Output tensor | bool | + +**Table 50 – one\_hot operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|--------------------------------------------------|------------|--------------------|----------------------------------------------------------------------------------------|-----------------| +| one_hot | Convert each id in the input to a one_hot tensor | Input | X | Input tensor | int32,
int64 | +| | | Output | Y | Output tensor | float32 | +| | | Attributes | depth | Define the length of the output tensor | int | +| | | | allow_out_of_range | Whether the id value contained in the input tensor can be greater than the depth value | bool | + +**Table 51 – or operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-----------------------------------------------------------|------------|-----------|-------------------------------|-----------| +| or | Perform logical operator or to each input element X and Y | Input | X | Input tensor 1 | bool | +| | | | Y | Input tensor 2 | bool | +| | | Output | Z | Output tensor | bool | +| | | Attributes | axis | Calculated dimension | int | +| | | | broadcast | Whether to allow broadcasting | bool | + +**Table 52 – pad operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|--------------------------------------------------------------------------|------------|---------|------------------------------------------------------------------------------------------------|---------------------------| +| pad | Expand the input tensor according to the given dimensions and attributes | Input | X | Input tensor | float16, float32, float64 | +| | | Output | Y | Output tensor | float16, float32, float64 | +| | | Attributes | mode | Three modes: constant(default), reflect, edge | string | +| | | | pads | A list of integers represents the count of fill elements at the beginning and end of each axis | list of ints | +| | | | value | Floating point number, which means the value to be filled in | float | + +**Table 53 – print operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|---------------------------------------------------------------------------|------------|-----------|-------------------------------------------------------------------------------------|-----------------------------------------------| +| print | Create a print operator to print the content of the tensor being accessed | Input | X | Input tensor | float16, float32, float64, int32, int64, bool | +| | | Output | Y | Output tensor | float16, float32, float64, int32, int64, bool | +| | | Attributes | summarize | Print the number of elements in the tensor, if the value is -1 , print all elements | int | + +**Table 53 – print operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-------------|-------|-------------------|-------------------------------------------------------------------------|-----------| +| | | | message | Custom string type message before printing tensor information | string | +| | | | first_n | Print the first n values of the tensor | int | +| | | | print_tensor_name | Optional, specify whether to print the tensor name, the default is True | bool | +| | | | print_tensor_type | Optional, specify whether to print the tensor type, the default is True | bool | + +**Table 54 – pow operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|--------------------------------|------------|-----------|-------------------------------|---------------------------| +| pow | Exponentiation of input tensor | Input | X | Input tensor 1 | float16, float32, float64 | +| | | | Y | Input tensor 2 | float16, float32, float64 | +| | | Output | Z | Output tensor | float16, float32, float64 | +| | | Attributes | axis | Calculated dimension | int | +| | | | broadcast | Whether to allow broadcasting | bool | + +**Table 55 – random\_normal operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|---------------|-----------------------------------------------|------------|---------|--------------------------------------------|------------------------------| +| random_normal | Generate a normally distributed random tensor | Output | Y | Output tensor | float16, float32, float64 | +| | | | dtype | The data type of the output tensor element | string | +| | | Attributes | mean | Mean of normal distribution | float | +| | | | scale | Standard deviation of normal distribution | float | +| | | | seed | Random generator seed | float | +| | | | shape | Output tensor size | list of ints, list of tensor | +| | | | | | | + +**Table 56 – random\_uniform operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------------|------------------------------------------------|------------|---------|-------------------------------------------------------|---------------------------------| +| random_uniform | Generate a uniformly distributed random tensor | Output | Y | Output tensor | float16,
float32,
float64 | +| | | Attributes | dtype | The data type of the output tensor element | string | +| | | | high | The upper limit of the output value, the default is 1 | float | +| | | | low | The lower limit of the output value, the default is 0 | float | +| | | | seed | Random generator seed | float | +| | | | shape | Output tensor size | list of ints
list of tensor | + +**Table 57 – range operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|--------------------------------------------------------------------------------------------------------------|------------|---------|-----------------------------------------------|-----------------------------------------| +| range | Separate the given numerical interval [X, Y] evenly according to the step , and return the separation result | Input | start | Start of interval (can be tensor or constant) | int32,
int64,
float32,
float64 | +| | | | end | End of interval (can be tensor or constant) | int32,
int64,
float32,
float64 | +| | | | step | Step size (can be tensor or constant) | int32,
int64,
float32,
float64 | +| | | Output | Y | Output tensor | int32,
int64,
float32,
float64 | +| | | Attributes | dtype | The data type of the output tensor | string | + +**Table 58 – reshape operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|--------------------------------------------------------------|-------|---------|--------------|-----------------------------------------| +| reshape | Adjust the size of the input tensor X according to the shape | Input | X | Input tensor | float16,
float32,
int32,
int64 | + +**Table 58 – reshape operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-------------|--------|---------|-------------------|--------------------------------| +| | | | shape | Transformed shape | int32,
int64 | +| | | Output | Y | Output tensor | float16,
float32,
double | + +**Table 59 – reverse operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|----------------------------------------------------------------------------------------|------------|---------|------------------------|-----------------------------------------| +| reverse | The input tensor performs the reverse order operator of the data on the specified axis | Input | X | Input tensor | int32,
int64,
float32,
float64 | +| | | Output | Y | Output tensor | int32,
int64,
float32,
float64 | +| | | Attributes | axis | Axis for reverse order | int
list of ints | + +**Table 60 – sin operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|---------------------------------|------------|---------|------------------------------------------------------------------------------------------------|---------------------------------| +| sin | Calculate the sine of the input | Attributes | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | +| | | Attributes | name | Optional, the prefix identifier of the output weight of the network layer, the default is None | string | + +**Table 61 – slice operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|--------------------------------------------------------------------|------------|---------|------------------------------------------|---------------------------------| +| slice | Divide the input tensor according to the given dimension and index | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | +| | | Attributes | axes | Applicable dimensions of ends and starts | list of ints | +| | | | ends | End index | list of ints | +| | | | starts | Start index | list of ints | + +**Table 62 – split operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|-----------------|------------------------------------------------------------|--------------|----------------|-------------------------|---------------------------------| +| split | Separate the part of the input tensor X according to split | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor list | float16,
float32,
float64 | +| | | Attributes | axis | Calculated dimension | ints | +| | | | split | The size of each output | list of ints | + +**Table 63 – sqrt operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|-----------------|---------------------------------|--------------|----------------|-------------------|---------------------------------| +| sqrt | Square root of the input tensor | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | + +**Table 64 – square operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|-----------------|---------------------------|--------------|----------------|-------------------|---------------------| +| square | Squaring the input tensor | Input | X | Input tensor | float16,
float32 | +| | | Output | Y | Output tensor | float16,
float32 | + +**Table 65 – stack operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|-----------------|---------------------------------------|--------------|----------------|----------------------------------------|-----------------------------------------| +| stack | Stack the input tensor along the axis | Input | X | Input tensor list | float16,
float32,
int32,
int64 | +| | | Output | Y | Output tensor | float16,
float32,
int32,
int64 | +| | | Attributes | axis | The axis for stacking the input tensor | int | + +**Table 66 – sub operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|------------------------------------------------------|------------|-----------|-------------------------------|---------------------------------| +| sub | Apply Operator and subtraction to the input tensor X | Input | X | Input operand 1 | float16,
float32,
float64 | +| | | | Y | Input operand 2 | float16,
float32,
float64 | +| | | Output | Z | Output tensor | float16,
float32,
float64 | +| | | | axis | Calculated dimension | int | +| | | Attributes | broadcast | Whether to allow broadcasting | bool | + +**Table 67 – round operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-------------|--------|---------|---------------|---------------------------------| +| round | Round up | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | int | + +**Table 68 – shape operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|--------------------------------------------|--------|---------|-------------------|-----------| +| shape | Returns the shape size of the input tensor | Input | X | Input tensor | tensor | +| | | Output | Y | Output shape size | int64 | + +**Table 69 – topk operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|------------------------------------------------------------|------------|---------|---------------------------------------------|---------------------------------| +| topk | Return the data with the largest K before the input tensor | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | values | The top K values of the output tensor | float16,
float32,
float64 | +| | | | indices | Index the top K values of the output tensor | float16,
float32,
float64 | +| | | Attributes | axis | Calculated dimension | int | +| | | | k | The size of the first k to find | int | + +**Table 70 – squeeze operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-----------------------------------------------------------------------|------------|---------|-------------------------------|---------------------------------| +| squeeze | Dimension compression of the input tensor according to the given axes | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | +| | | Attributes | axes | List of calculated dimensions | list of ints | + +**Table 71 – transpose operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|-----------|------------------|------------|---------|---------------------------------------------------------------------|---------------------------------| +| transpose | Matrix transpose | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | +| | | Attributes | perm | Transpose by default, otherwise adjust according to the given value | list of ints | + +**Table 72 – unsqueeze operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|-----------|---------------------------------------------------------------------------|------------|---------|-------------------------------|---------------------------------| +| unsqueeze | Decompress the dimensions of the input tensor according to the given axes | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | +| | | Attributes | axes | List of calculated dimensions | list of ints | + +**Table 73 – unstack operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|------------------------------------------------------------------------------------|------------|---------|----------------------------------------------------------------------|-----------------------------------------------------| +| unstack | Single dimension D tensor along axis of dismantling num dimensions of (D-1) tensor | Input | X | Input tensor | float16,
float32,
float64,
int32,
int64 | +| | | Output | Y | Output tensor | float16,
float32,
float64,
int32,
int64 | +| | | Attributes | axis | Optionally, the axis on which the input tensor is to be disassembled | int | +| | | | num | Optional, the length of the axis | int | + +**Table 74 – xor operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|---------------------------------------|------------|-----------|-------------------------------|-----------| +| xor | Perform logical exclusive OR operator | Input | X | Input tensor 1 | bool | +| | | | Y | Input tensor 2 | bool | +| | | Output | Z | Output tensor | bool | +| | | Attributes | axis | Calculated dimension | int | +| | | | broadcast | Whether to allow broadcasting | bool | + +**Table 75 – scale operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-----------------------------------------------------------|------------|---------|---------------|---------------------------------| +| scale | Perform $W \times X$ operator, W is the learned parameter | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | +| | | Attributes | W | coefficient | tensor | +| | | | shape | W size | list of ints | + +**Table 76 – shift operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|------------------------------------------------------|------------|---------|---------------|---------------------------------| +| shift | Perform $X + b$ operator, b is the learned parameter | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | +| | | Attributes | B | Bias | float | +| | | | shape | B size | list of ints | + +**Table 77 – zeros operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|--------------------------------------------------------------------------------|------------|---------|------------------------------------|-------------------------------------------| +| zeros | Create a tensor whose shape is shape, data type is dtype, and all values are 0 | Output | Shape | Tensor of the size to be output | list of ints | +| | | | Y | Output tensor | float16,
float32,
float64,
int32 | +| | | Attributes | dtype | The data type of the output tensor | string | + +**Table 78 – clip operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-------------------------------------------------------------------|------------|---------|---------------------------------------------------------------------------------------------|--------------------------------------| +| clip | Limit the size of input tensor elements according to MAX and MIN. | Input | X | Input tensor | float16,
float32,
float64, int | +| | | Output | Y | Output tensor | float16,
float32,
float64, int | +| | | Attributes | MAX | The maximum value. Anything greater than the maximum value is replaced by the maximum value | float,
int | +| | | | MIN | The minimum value. Anything less than the minimum value is replaced by the minimum value | float,
int | + +**Table 79 – tanh operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|----------------------------------------------------------------------|--------|---------|---------------|---------------------------------| +| tanh | Calculate the hyperbolic tangent of each element of the input tensor | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | + +### **8.3.2 Advanced operator unit representation** + +#### **8.3.2.1 The definition of advanced operator** + +**Table 80 – accuracy operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|---------------------------------------------|------------|---------|----------------------------------------------------------|---------------------| +| accuracy | Use inputs and labels to calculate accuracy | Input | X | Input tensor | float32,
float64 | +| | | | label | Data set label | int32,
int64 | +| | | Output | Y | Output tensor | float32 | +| | | Attributes | k | Take K predicted values in each category for calculation | int32,
int64 | +| | | | correct | Number of correct predictions | int32,
int64 | +| | | | total | Total predicted value | int32,
int64 | + +**Table 81 – affine\_channel operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------------|--------------------------------------------------------------|------------|-------------|---------------------------------------------------------------------------------------------|---------------------| +| affine_channel | Apply a separate affine transformation to each input channel | Input | X | Input tensor | float32,
float64 | +| | | Output | Y | Output tensor | float32,
float64 | +| | | Attributes | scale | Scale factor for affine transformation | float32,
float64 | +| | | | bias | Bias for affine transformation | float32,
float64 | +| | | | data_format | Optional, image channel format, default is "NCHW" | string | +| | | | name | Optional, the prefix identifier of the network layer output and weight, the default is None | string | +| | | | act | Optional, activation function, default is None | string | + +**Table 82 – eft operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|------------------------------------------------------------------------|------------|---------|---------------------------------------------------------------------------------------------|------------------| +| erf | Calculate the error function value of each element of the input tensor | Input | X | Input tensor | float32, float64 | +| | | Output | Y | Input tensor | float32, float64 | +| | | Attributes | name | Optional, the prefix identifier of the network layer output and weight, the default is None | string | + +**Table 83 – auc operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|------------------------------------|------------|----------------|----------------------------------------------------------------------------------------|------------------| +| auc | Calculation model evaluation index | Input | X | Input tensor | float32, float64 | +| | | | label | Training data label | int32, int64 | +| | | Output | Y | Output tensor | float32, float64 | +| | | Attributes | curve | Curve type, the default is ROC | string | +| | | | num_thresholds | The number of critical values used when discretizing the ROC curve, the default is 200 | int | +| | | | topk | Predicted output topk | int | +| | | | slide_steps | When calculating batch auc, use not only the current step but also the previous step | int | + +**Table 84 – average\_pool operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|--------------|--------------------------------------------------------------------------------------|------------|--------------|----------------------------------------------------------|---------------------------| +| average_pool | Apply average pooling on the input tensor according to kernel_shape, strides and pad | Input | X | Input tensor | float16, float32, float64 | +| | | Output | Y | Output tensor | float16, float32, float64 | +| | | Attributes | auto_pad | valid or Same, and cannot be pads used with | string | +| | | | kernel_shape | Core size | list of ints | +| | | | pads | For the beginning and end, the padding of each dimension | list of ints | + +**Table 84 – average\_pool operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-------------|-------|-------------|------------------------------|--------------| +| | | | strides | Step size for each dimension | list of ints | +| | | | data_format | Input and output data format | string | + +**Table 85 – case operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-----------------------------|------------|---------------|---------------------------------------------------------------------------------------------|------------------------------------| +| case | Implement if-else statement | Output | Y | Output tensor | float 32, float64, int, uint, bool | +| | | Attributes | pred_fn_pairs | pred a Boolean tensor, the Fn is callable a Callable, return tensor | list of (pred, fn) | +| | | | Default | Optional, callable object | callable | +| | | | Name | Optional, the prefix identifier of the network layer output and weight, the default is None | string | + +**Table 86 – cond operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|----------------------------------------------------------------|------------|----------|---------------------------------------------------------------------------------------------|-----------| +| cond | If pred is True, return true_fn(), otherwise return false_fn() | Input | pred | Input tensor | bool | +| | | Output | true_fn | Called when pred is true | callable | +| | | | false_fn | Called when pred is false | callable | +| | | Attributes | name | Optional, the prefix identifier of the network layer output and weight, the default is None | string | + +**Table 87 – conv2d operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------|---------|---------------|---------------------------| +| conv2d | Construct a two-dimensional convolutional layer, and calculate the output feature map according to the input, filter parameters, step size, filling and other parameters | Input | X | Input tensor | float16, float32, float64 | +| | | | W | Weight tensor | float16, float32, float64 | +| | | | B | Bias tensor | float16, float32, float64 | + +**Table 87 – conv2d operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-------------|------------|--------------|---------------------------------------------------------------------------|---------------------------------| +| | | Output | Y | Output tensor | float16,
float32,
float64 | +| | | Attributes | num_channels | The number of channels of the image | int | +| | | | num_fliters | Number of filters | int | +| | | | filter_size | Filter size | list of ints
int | +| | | | stride | Optional, step size | list of ints
int | +| | | | padding | Optional, fill size | list of ints
int | +| | | | dilation | Optional, expansion coefficient size | list of ints
int | +| | | | groups | Optional, the number of groups of the two-dimensional convolutional layer | int | +| | | | dtype | Optional, output tensor data type | string | + +**Table 88 – conv3d operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------|--------------|-------------------------------------------|---------------------------------| +| conv3d | Construct a three-dimensional convolutional layer, and calculate the output feature map according to the input, filter parameters, step size, filling and other parameters | Input | X | Input tensor | float16,
float32,
float64 | +| | | | W | Weight tensor | float16,
float32,
float64 | +| | | | B | Bias tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | +| | | Attributes | num_channels | The number of channels of the input image | int | +| | | | num_fliters | Number of filters | int | +| | | | filter_size | Filter size | int
list of ints | +| | | | stride | Optional, step size | int | + +**Table 88 – conv3d operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-------------|-------|----------|--------------------------------------------------------------|---------------------| +| | | | | | list of ints | +| | | | padding | Optional, fill size | int
list of ints | +| | | | dilation | Optional, expansion ratio | int
list of ints | +| | | | groups | Optional, the number of groups of the 3D convolutional layer | int | +| | | | dtype | Optional, input and output tensor format | string | + +**Table 89 – conv2d\_transpose operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------|--------------|----------------------------------------------|---------------------------------| +| conv2d_transpose | Construct a two-dimensional convolution transposed layer in the neural network, and output the feature map according to the input, filter parameters, step size and other parameters | Input | X | Input tensor | float16,
float32,
float64 | +| | | | W | Weight tensor | float16,
float32,
float64 | +| | | | B | Bias tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | +| | | | | | | +| | | Attributes | num_channels | The number of channels of the image | int | +| | | | num_filters | Number of filters | int | +| | | | filter_size | Filter size | int
list of ints | +| | | Attributes | output_size | Optional, the size of the output feature map | int
list of ints | +| | | | padding | Optional, fill size | int
list of ints | +| | | | stride | Optional, step size | int
list of ints | +| | | | dilation | Optional, expansion ratio | int
list of ints | + +**Table 89 – conv2d\_transpose operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-------------|-------|---------|---------------------------------------------------------------------------|-----------| +| | | | groups | Optional, the number of groups of the two-dimensional convolutional layer | int | +| | | | dtype | Optional, input and output tensor format | string | + +**Table 90 – conv3d\_transpose operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------|--------------|-------------------------------------------|---------------------------| +| conv3d_transpose | Construct a three-dimensional convolution transposed layer in the neural network, and output the feature map according to the input, filter parameters, step size and other parameters | Input | X | Input tensor | float16, float32, float64 | +| | | | W | Weight tensor | float16, float32, float64 | +| | | | B | Bias tensor | float16, float32, float64 | +| | | Output | Y | Output tensor | float16, float32, float64 | +| | | Attributes | num_channels | The number of channels of the input image | int | +| | | | num_filters | Number of filters | int | +| | | | filter_size | Filter size | int list of ints | +| | | | output_size | Optional, the size of the output image | int list of ints | +| | | | padding | Padding size | int list of ints | +| | | | stride | Step size | int list of ints | +| | | | dilation | Optional, expansion ratio | int list of ints | +| | | | groups | Optional, the number of groups of the 3D | int | + +**Table 90 – conv3d\_transpose operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-------------|-------|---------|---------------------------------------------------------------------------------------------|-----------| +| | | | | convolutional layer | | +| | | | name | Optional, the prefix identifier of the network layer output and weight, the default is None | string | +| | | | dtype | Optional, input and output tensor format | string | + +**Table 91 – cross\_entropy operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|---------------|--------------------------------------------------------------------------|------------|--------------|------------------------------------------------------------------------------------------|------------------| +| cross_entropy | Calculate the cross entropy between the input tensor and the label label | Input | X | Input tensor | float32, float64 | +| | | | label | Enter the label tensor value corresponding to X | int64 | +| | | Output | Y | Output tensor | float32, float64 | +| | | Attributes | soft_label | Specify whether the label is a soft label | bool | +| | | | ignore_index | Specify an ignored label value, this label value does not participate in the calculation | int | + +**Table 92 – reciprocal operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|------------|--------------------------------------------------------------|------------|---------|------------------------------------------------------------------------------------------------|---------------------------------| +| reciprocal | Calculate the reciprocal of each element in the input tensor | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | +| | | Attributes | name | Optional, the prefix identifier of the output weight of the network layer, the default is None | string | + +**Table 93 – depth\_to\_space operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------------|-------------------------------------------------------|------------|-----------|-----------------------------------|---------------------------------| +| depth_to_space | Rearrange the input tensor from depth to spatial data | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | +| | | Attributes | blocksize | The size of the block being moved | int | + +**Table 94 – dropout operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|--------------------------------------------------------------------------|------------|-------------|-------------------------------------|---------------------------------| +| dropout | Randomly lose part of the input tensor or simply output the input tensor | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | +| | | Attributes | is_training | Whether to train | bool | +| | | | ratio | Random probability (default is 0.5) | float | + +**Table 95 – embedding operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|-----------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------|----------------|----------------------------------------------------------------------------------------------------------|---------------------| +| embedding | Query the corresponding embedded information from the embedded tensor according to the input information. The function generates a two-dimensional embedded tensor according to the input tensor dimension and data type | Input | X | Input tensor | int64 | +| | | | W | Weight tensor | float32, float64 | +| | | Output | Y | Output tensor | float32, float64 | +| | | Attributes | size | Dimensions of the embedding matrix | list of ints | +| | | | is_sparse | Whether to use sparse update method | bool | +| | | | is_distributed | Whether to store the embedding tensor in a distributed manner | bool | +| | | | padding_idx | Set the embedding information corresponding to padding_idx in the input tensor to 0, the default is None | int
long
none | +| | | | dtype | Type of output tensor | string | + +**Table 96 – floordiv operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-------------------------------------------------------------------------------------------------------------|------------|---------|-------------------------------------------------------------------------|--------------------------------| +| floordiv | Divide the operator element by element, and save the output element at each position to the returned result | Input | X | Input tensor | float32, float64, int32, int64 | +| | | | Y | Input tensor | float32, float64, int32, int64 | +| | | Output | Z | Output tensor | float32, float64, int32, int64 | +| | | Attributes | axis | Optional, the index when the Y dimension corresponds to the X dimension | int32 | + +**Table 96 – floordiv operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-------------|-------|---------|---------------------------------------------------------------------------------------------|-----------| +| | | | name | Optional, the prefix identifier of the network layer output and weight, the default is None | string | + +**Table 97 – gather\_nd operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|-----------|-----------------------------------------------------------------------------------------|------------|---------|----------------------------------------|--------------------------------------| +| gather_nd | High-dimensional promotion of gather operator, support multi-axis simultaneous indexing | Input | X | Input tensor | int32, int64, float32, float64, bool | +| | | | index | Index tensor | int32, int64 | +| | | Output | Y | Output tensor | int32, int64, float32, float64, bool | +| | | Attributes | name | The name of the layer, default is None | string | + +**Table 98 – group\_norm operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|------------|---------------------|------------|-------------|-------------------------------------------------------------|------------------| +| group_norm | Group normalization | Input | X | Input tensor | float32, float64 | +| | | | W | Weight tensor | float32, float64 | +| | | | B | Bias tensor | float32, float64 | +| | | Output | Y | Output tensor | float32, float64 | +| | | Attributes | groups | Number of groups separated from the channel | int32 | +| | | | epsilon | Optional to help prevent the variance from dividing by zero | float32 | +| | | | act | Optional, activation function | string | +| | | | data_format | Optional, image channel format, default is "NCHW" | string | + +**Table 98 – group\_norm operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-------------|-------|---------|---------------------------------------------------------------------------------------------|-----------| +| | | | name | Optional, the prefix identifier of the network layer output and weight, the default is None | string | + +**Table 99 – hard\_max operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|--------------------------------------------------------------------------------------------------------------------------------|------------|---------|----------------------|---------------------------| +| hard_max | Returns the corresponding size map of the maximum value of a given dimension (1 means the maximum value, 0 means other values) | Input | X | Input tensor | float16, float32, float64 | +| | | Output | Y | Output tensor | float16, float32, float64 | +| | | Attributes | axis | Calculated dimension | int | + +**Table 100 – label\_smooth operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|--------------|--------------------------|------------|------------|---------------------------------------------------------------------------------------------|------------------| +| label_smooth | Label smoothing function | Input | X | Input tensor | float32, float64 | +| | | Output | Y | Output tensor | float32, float64 | +| | | Attributes | prior_dist | Optional, used to smooth the prior distribution of labels | int
float | +| | | | epsilon | Optional, mix the weights of the original true distribution and the fixed distribution | float | +| | | | dtyp | Optional, input tensor type | string | +| | | | name | Optional, the prefix identifier of the network layer output and weight, the default is None | string | + +**Table 101 – layer\_norm operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|------------|---------------------|--------|---------|---------------|------------------| +| layer_norm | Layer normalization | Input | X | Input tensor | float32, float64 | +| | | | W | Weight tensor | float32, float64 | +| | | | B | Bias tensor | float32, float64 | +| | | Output | Y | Output tensor | float32, float64 | + +**Table 101 – layer\_norm operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-------------|------------|-----------------|------------------------------------------------------------------------------------------------|-----------| +| | | Attributes | scale | Optional, whether to learn adaptive gain after normalization | bool | +| | | | shift | Optional, whether to learn adaptive deviation after normalization | bool | +| | | | begin_norm_axis | Optional, normalized execution dimension | int | +| | | | epsilon | Optional to help prevent the variance from dividing by zero | float32 | +| | | | act | Optional, activation function | string | +| | | | name | Optional, the prefix identifier of the output weight of the network layer, the default is None | string | + +**Table 102 – scatter\_nd operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|------------|------------------------------------------------------------------------|------------|---------|-----------------------------------------------------------------------------------------------|------------------| +| scatter_nd | The index value in the updated data is added to a new amount of sheets | Input | index | index | int 32, int64 | +| | | | updates | Update tensor | float32, float64 | +| | | Output | Y | Output tensor | float32, float64 | +| | | Attributes | shape | The shape of the output tensor | List of int | +| | | | name | Optional, the prefix identifier of the output weight of the network layer, the default is Non | string | + +**Table 103 – log\_loss operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|------------------------------------------------------------------------------------------------|------------|---------|-------------------------------------------------|-----------| +| log_loss | Calculate the input prediction result and target label, and return the negative log loss value | Input | X | Input tensor | float32 | +| | | | label | Enter the label tensor value corresponding to X | float32 | +| | | Output | Y | Output tensor | float32 | +| | | Attributes | epsilon | Prevent the variance from dividing by zero | float32 | + +**Table 103 – log\_loss operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-------------|-------|---------|------------------------------------------------------------------------------------------------|-----------| +| | | | name | Optional, the prefix identifier of the output weight of the network layer, the default is None | string | + +**Table 104 – lpnorm operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|------------------------------------------------------------------------------------------|------------|---------|--------------------------------------------------|---------------------------| +| lpnorm | Apply Lp normalization according to a given dimension (the p value is limited to 1 or 2) | Input | X | Input tensor | float16, float32, float64 | +| | | Output | Y | Output tensor | float16, float32, float64 | +| | | Attributes | axis | Calculated dimension | int | +| | | | p | Normalized p- value (only 1 and 2 are supported) | int | + +**Table 105 – lstm operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|--------------------------------------|------------|------------|-------------------------------------------|------------------| +| lstm | Long and short-term memory operators | Input | X | Input tensor | float32, float64 | +| | | | weight_ih | Input weight tensor | float32, float64 | +| | | | weight_hh | Hidden state weight tensor | float32, float64 | +| | | | bias_ih | Input bias tensor | float32, float64 | +| | | | bias_hh | Hidden state bias tensor | float32, float64 | +| | | Output | rnn_out | LSTM hidden output | float32, float64 | +| | | | last_h | The hidden state of the last step of LSTM | float32, float64 | +| | | | last_c | Cell state in the last step of LSTM | float32, float64 | +| | | Attributes | max_len | Maximum length of LSTM | int | +| | | | input_size | The dimensions of the input tensor | int | + +**Table 105 – lstm operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-------------|-------|------------------|---------------------------------------------------|-----------| +| | | | hidden_size | LSTM hidden state dimensions | int | +| | | | num_layers | The total number of LSTM layers | int | +| | | | dropout_prob | Optional, ignore the ratio | float | +| | | | is_bidirectional | Optional, whether it is a two-way LSTM | bool | +| | | | time_major | Optional, whether the time dimension is the first | bool | +| | | | is_test | Optional, whether in the testing phase | bool | + +**Table 106 – max\_pool operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|------------------------------------------------------------------------------------------------------|------------|----------------|-----------------------------------------------------------------------------|---------------------------| +| max_pool | Apply the maximum pooling operator to the input tensor X according to kernel_shape, pads and strides | Input | X | Input tensor | float16, float32, float64 | +| | | Output | Y | Output tensor | float16, float32, float64 | +| | | Attributes | auto_pad | valid or Same, and cannot be pads used with | string | +| | | | kernel_shape | Convolution kernel size | list of ints | +| | | | pads | For the beginning and end, the padding of each dimension | list of ints | +| | | | strides | Convolution step size per dimension | list of ints | +| | | | return_indices | If equal to True, the serial number of the maximum output value is returned | bool | +| | | | data_format | channel_first or channel_last | string | + +**Table 107 – max\_roi\_pool operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|--------------|-------------------------------------------------------------------------------------------------------------------|------------|---------------|----------------------|---------------------------| +| max_roi_pool | According to the input tensor X and the region of interest, maximum pooling is applied to each region of interest | Input | X | Input tensor | float16, float32, float64 | +| | | | rois | Area of interest | float16, float32, float64 | +| | | Output | Y | Output tensor | float16, float32, float64 | +| | | Attributes | pooled_shape | ROI pool output size | list of ints | +| | | | spatial_scale | Size of space | float | + +**Table 108 – pool2d operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|------------------------------------------------------------------------------------------------------------------------------|------------|----------------|-------------------------------------------------------------------------------------|---------------------| +| pool2d | Construct a two-dimensional pooling layer in the neural network, and get the output according to the parameter configuration | Input | X | Input tensor | float32, float64 | +| | | Output | Y | Output tensor | float32, float64 | +| | | Attributes | pool_size | Optional, pooled core size | int
list of ints | +| | | | pool_type | Optional, pooling type | string | +| | | | pool_stride | Optional, the step size of the pooling layer | int
list of ints | +| | | | pool_padding | Optional, fill size | int
list of ints | +| | | | global_pooling | Optional, whether to use global pooling | bool | +| | | | ceil_mode | Optional, whether to use the ceil function to calculate the output height and width | bool | +| | | | exclusive | Optional, whether to ignore the fill value in the average pooling model | bool | + +**Table 109 – pool3d operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|---------------------------------------------------------------------------------------------------------------------------------|------------|----------------|------------------------------------------------------------------------------------------------|-------------------------------| +| pool3d | Three-dimensional space pooling operator, pooling configuration of input parameters, calculating output according to parameters | Input | X | Input tensor | float32, float64 | +| | | Output | Y | Output tensor | float32, float64 | +| | | Attributes | pool_size | Pooling core size | int
list of ints | +| | | | pool_type | Pooling type | string | +| | | | pool_stride | The step size of the pooling layer | int
list of ints | +| | | | pool_padding | Pooled filling | int
list of ints
string | +| | | | global_padding | Whether to use global pooling | bool | +| | | | ceil_mode | Whether to use the ceil function to calculate the depth, height and width of the output | bool | +| | | | name | Optional, the prefix identifier of the output weight of the network layer, the default is None | string | +| | | | exclusive | Whether to ignore the fill value in flat pooling mode | bool | +| | | | data_format | Input and output data format | string | + +**Table 110 – print operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|---------------------------------------------------------------------------|------------|--------------------|------------------------------------------------------------------------------|-----------| +| print | Create a print operator to print the content of the tensor being accessed | Attributes | print_tensor_shape | Optional, specify whether to print the tensor dimension, the default is True | bool | +| | | | print_phase | Specify the printing stage, including, and the default is | string | + +**Table 111 – reduce\_all operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|------------|-------------------------------------------------------------------------------------------------------------------------------------------|------------|----------|------------------------------------------------------------|--------------| +| reduce_all | Perform AND logic ( & ) calculations on the tensor elements in the specified dimension , and output the corresponding calculation results | Input | X | Input tensor | bool | +| | | Output | Y | Output tensor | bool | +| | | Attributes | dim | And the dimension of logical operators | list of ints | +| | | | keep_dim | Whether to keep the reduced dimension in the output tensor | bool | + +**Table 112 – reduce\_any operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|------------|-----------------------------------------------------------------------------------------------------------------------------------------|------------|----------|------------------------------------------------------------|--------------| +| reduce_any | Perform an OR logic ( ) calculation on the tensor element in the specified dimension, and output the corresponding calculation result | Input | X | Input tensor | bool | +| | | Output | Y | Output tensor | bool | +| | | Attributes | dim | Or the dimension of logical operators | list of ints | +| | | | keep_dim | Whether to keep the reduced dimension in the output tensor | bool | + +**Table 113 – reduce\_mean operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|-------------|-----------------------------------------------------------------------------------|------------|----------|-------------------------------|---------------------------| +| reduce_mean | Calculate the average value of the input tensor according to the given dimensions | Input | X | Input tensor | float16, float32, float64 | +| | | Output | Y | Output tensor | float16, float32, float64 | +| | | Attributes | axis | Calculated dimension, - 1 | int | +| | | | keepdims | Whether to keep the dimension | bool | + +**Table 114 – reduce\_max** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|------------|----------------------------------------------------------------------------------|------------|----------|-------------------------------|---------------------------------| +| reduce_max | Calculate the maximum value of the input tensor according to the given dimension | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | +| | | Attributes | axis | Calculated dimension, -1 | int | +| | | | keepdims | Whether to keep the dimension | bool | + +**Table 115 – reduce\_min operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|------------|----------------------------------------------------------------------------------|------------|----------|-------------------------------|---------------------------------| +| reduce_min | Calculate the minimum value of the input tensor according to the given dimension | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | +| | | Attributes | axis | Calculated dimension, -1 | int | +| | | | keepdims | Whether to keep the dimension | bool | + +**Table 116 – reduce\_prod operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|-------------|-----------------------------------------------------------------------------------|------------|----------|-------------------------------|---------------------------------| +| reduce_prod | Calculate the inner product of the input tensor according to the given dimensions | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | +| | | Attributes | axis | Calculated dimension, -1 | int | +| | | | keepdims | Whether to keep the dimension | bool | + +**Table 117 – reduce\_sum operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|------------|----------------------------------------------------------------------|------------|----------|-------------------------------|---------------------------| +| reduce_sum | Calculate the sum of input tensors according to the given dimensions | Input | X | Input tensor | float16, float32, float64 | +| | | Output | Y | Output tensor | float16, float32, float64 | +| | | Attributes | axis | Calculated dimension, -1 | int | +| | | | keepdims | Whether to keep the dimension | bool | + +**Table 118 – reduce\_variance operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|-----------------|------------------------------------------------------------------------------|------------|----------|-------------------------------|---------------------------| +| reduce_variance | Calculate the variance of the input tensor according to the given dimensions | Input | X | Input tensor | float16, float32, float64 | +| | | Output | Y | Output tensor | float16, float32, float64 | +| | | Attributes | axis | Calculated dimension, -1 | int | +| | | | keepdims | Whether to keep the dimension | bool | + +**Table 119 – relu operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-------------------------------------------|--------|---------|---------------|---------------------------| +| relu | Apply $y = \max(x)$ to the input tensor X | Input | X | Input tensor | float16, float32, float64 | +| | | Output | Y | Output tensor | float16, float32, float64 | + +**Table 120 – relu6 operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|---------------------|------------|-----------|-------------------------------------|------------------| +| relu6 | Activation function | Input | X | Input tensor | float32, float64 | +| | | Output | Y | Output tensor | float32, float64 | +| | | Attributes | threshold | Threshold, the default value is 6.0 | float | + +**Table 121 – rsqrt operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|---------------------|--------|---------|---------------|------------------| +| rsqrt | Activation function | Input | X | Input tensor | float32, float64 | +| | | Output | Y | Output tensor | float32, float64 | + +**Table 122 – scatter operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|--------------------------------------------------------------|------------|-----------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--------------| +| scatter | The index value is updated to the input data updating tensor | Input | X | Input tensor | float32 | +| | | | index | index | int32, int64 | +| | | | updates | Update tensor | float32 | +| | | Output | Y | Output tensor | float32 | +| | | Attributes | name | Optional, the prefix identifier of the output weight of the network layer, the default is None | string | +| | | | overwrite | Alternatively, if the index values are repeatedly and overwrite is True, the old value is updated to be overwritten with new updated value; if False, the new value updates the old updated with the adding value. The default value is True | bool | + +**Table 123 – softmax\_with\_cross\_entropy operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------------------------|-----------------------------------------------|--------|---------|--------------------|------------------| +| softmax_with_cross_entropy | Implement softmax cross entropy loss function | Input | X | Input tensor | float32, float64 | +| | | | label | Input label tensor | float32, float64 | +| | | Output | Y | Output tensor | float32, float64 | + +**Table 124 – softmax\_with\_cross\_entropy operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------------------------|-----------------------------------------------|------------|------------|--------------------------------------------------------------------|-----------| +| softmax_with_cross_entropy | Implement softmax cross entropy loss function | Attributes | soft_label | Optional, specify whether to treat the input label as a soft label | bool | + +**Table 124 – softmax\_with\_cross\_entropy operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-------------|-------|---------------------|-----------------------------------------------------------------------------------|-----------| +| | | | ignore_index | Optional, specify the target value to be ignored | int | +| | | | numeric_stable_mode | Optional, indicate whether to use an algorithm with better mathematical stability | bool | + +**Table 125 – space\_to\_depth operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------------|-----------------------------------------------------------|------------|-----------|-----------------------------------|---------------------------| +| space_to_depth | Rearrange the input tensor from space to depth space data | Input | X | Input tensor | float16, float32, float64 | +| | | Output | Y | Output tensor | float16, float32, float64 | +| | | Attributes | blocksize | The size of the block being moved | int32, int64 | + +**Table 126 – switch\_case operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|-------------|--------------------------------------------|-------------|--------------|------------------------------------------------------------------------------------------------|-----------------------------------------| +| switch_case | Implement the switch/case statement of C++ | InputOutput | Y | Output tensor | float16, float32, float64, int32, int64 | +| | | Attributes | branch_index | Specify the branch to be executed | int32, int64, uint8 | +| | | | brach_fns | The key of the callable object | list of tuple (int, callable) | +| | | | default | Optional, callable object | callable | +| | | | name | Optional, the prefix identifier of the output weight of the network layer, the default is None | string | + +**Table 127 – while\_loop operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|------------|----------------------------------------------------|------------|-----------|------------------------------------------------------------------------------------------------|----------------| +| while_loop | Realize the loop control function similar to while | Out put | Y | Output tensor | list of tensor | +| | | Attributes | cond | Determine whether to execute the loop | callable | +| | | | body | Loop execution structure | callable | +| | | | loop_vars | List or tuple containing tensor | list of tensor | +| | | | is_test | Optional, whether to execute during the testing phase | bool | +| | | | name | Optional, the prefix identifier of the output weight of the network layer, the default is None | string | + +**Table 128 – full\_connected operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------------|-----------------------|------------|---------|--------------------------------------|---------------------------| +| full_connected | Fully connected layer | Input | X | Input tensor | float16, float32, float64 | +| | | | W | Tensor weights applied to the output | float16, float32, double | +| | | Ou t put | Y | Output tensor | float16, float32, float64 | +| | | Attributes | axis | batch_size dimension, default is 0 | int | +| | | | units | Number of hidden units | int | + +**Table 129 – expand\_dims operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|-------------|--------------------|------------|-------------|-----------------------------------------------------------------------|---------------------------| +| expand_dims | Extended dimension | Input | X | Input tensor | float16, float32, float64 | +| | | | ExpandTimes | Input tensor, if this input exists, ignore the expand_times attribute | int32, int64 | +| | | Output | Y | Output tensor | float16, float32, float64 | +| | | Attributes | axis | Calculated dimension | int | + +**Table 130 – batch\_normalization operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|---------------------|-----------------------|------------|-----------------------|----------------------------------------------------------------------------------|---------------------------| +| batch_normalization | Batch standardization | Input | X | Input tensor | float16, float32, float64 | +| | | | W | Tensor weights applied to the output | float16, float32, float64 | +| | | | B | Tensor bias applied to the output | float16, float32, float64 | +| | | | means | Average at runtime or average at forecast | float16, float32, float64 | +| | | | var | Variance at runtime or variance at prediction | float16, float32, float64 | +| | | Output | Y | Output tensor | float16, float32, float64 | +| | | | mean (optional) | Average running after operator | float16, float32, float64 | +| | | | var (optional) | Running variance after operator | float16, float32, float64 | +| | | | saved_mean (optional) | The average value saved during training, used to accelerate gradient calculation | float16, float32, float64 | +| | | | saved_var (optional) | Variance saved during training, used to accelerate gradient calculation | float16, float32, float64 | +| | | Attributes | epsilon | Minimal value used to help prevent division by 0 | float | +| | | | is_training | Determine whether the operator is used for training or prediction | bool | +| | | | momentum | Factors of mean and variance when calculating predictions | float | +| | | | spatial | Calculate the whole space or the mean and variance of each dimension | bool | +| | | | data_format | channel_first or channel_last | string | + +**Table 131 – elu operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|----------------------------------------------------------------------------------------------------|------------|---------|--------------------------------------------|---------------------------------| +| elu | Apply $f(x) = \alpha * (\exp(x) - 1.)$ for $x < 0$ , $f(x) = x$ for $x \geq 0$ to the input tensor | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | +| | | Attributes | alpha | The coefficient of elu, the default is 1.0 | float | + +**Table 132 – gemm operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|--------------------------------------------------------------------------------|------------|-----------|----------------------------------------|---------------------------------| +| gemm | General matrix multiplication, calculate $Y \times \alpha * A * B + \beta * C$ | Input | X | Input tensor 1 | float16,
float32,
float64 | +| | | | Y | Input tensor 2 | float16,
float32,
float64 | +| | | | Z | Input tensor 3 | float16,
float32,
float64 | +| | | Output | output | Output tensor | float16,
float32,
float64 | +| | | Attributes | alpha | $X * Y$ coefficient | float | +| | | | beta | Enter the coefficient of Z | float | +| | | | broadcast | Enter Z whether to allow broadcasting | int | +| | | | transX | Whether input X needs to be transposed | int | +| | | | transY | Whether input Y needs to be transposed | int | + +**Table 133 – hard\_sigmoid operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|--------------|----------------------------------------------------------------------|------------|---------|--------------------|---------------------------------| +| hard_sigmoid | Apply $y = \max(0, \min(1, \alpha * x + \beta))$ to the input tensor | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | +| | | Attributes | alpha | The default is 2.0 | float | +| | | | beta | The default is 0.5 | float | + +**Table 134 – instance\_normalization operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|------------------------|----------------------------------------------------|------------|---------|--------------------------------------------------|---------------------------------| +| instance_normalization | Perform batch standardization as described in [19] | Input | X | Input tensor | float16,
float32,
float64 | +| | | | W | Tensor weights applied to the output | float16,
float32,
float64 | +| | | | B | Tensor bias applied to the output | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | +| | | Attributes | epsilon | Minimal value used to help prevent division by 0 | float32 | + +**Table 135 – log\_softmax operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|-------------|-------------------------------------------------------------|------------|---------|----------------------|---------------------------------| +| log_softmax | Input X for log(X) operator, and then further preprocessing | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | +| | | Attributes | axis | Calculated dimension | Int | + +**Table 136 – lrn operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|---------------------------------------------------------------------------------------------------------------|------------|---------|-------------------------------------|---------------------------------| +| lrn | Apply $(\text{bias} + (\text{alpha} / \text{size}) * \sum (\text{xi}^2)^{\text{beta}})$ to each input element | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | +| | | Attributes | alpha | Scale parameter | float | +| | | | beta | index | float | +| | | | bias | The default is 0.1 | float | +| | | | k | LpNorm parameters | int | +| | | | size | Number of channels to be calculated | int | + +**Table 137 – leaky\_relu operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|------------|-----------------------------------------------------------------------------------------------|------------|---------|---------------------|---------------------------------| +| leaky_relu | Apply $f(x) = \text{alpha} * x$ for $x < 0$ , $f(x) = x$ for $x \geq 0$ to each input element | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | +| | | Attributes | alpha | The default is 0.01 | float | + +**Table 138 – selu operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|------------------------------------------------------------------------------------------------------------------------------------------|------------|---------|-----------------------|---------------------------------| +| selu | Apply $y = \text{gamma} * (\text{alpha} * e^x - \text{alpha})$ for $x \leq 0$ , $y = \text{gamma} * x$ for $x > 0$ to input each element | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | +| | | Attributes | alpha | The default is 1.6732 | float | +| | | | gamma | The default is 1.0507 | float | + +**Table 139 – prelu operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-----------------------------------------------------------------------------------------------|--------|---------|--------------------|---------------------------------| +| prelu | Apply $f(x) = \text{slope} * x$ for $x < 0$ , $f(x) = x$ for $x \geq 0$ to each input element | Input | X | Input tensor | float16,
float32,
float64 | +| | | | slope | Input slope tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | + +**Table 140 – sigmoid operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|------------------------------------------------------|--------|---------|---------------|---------------------------------| +| sigmoid | Apply $y = 1 / (1 + \exp(-x))$ to the input tensor X | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | + +**Table 141 – sigmoid\_cross\_entropy\_with\_logits operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|-----------------------------------|--------------------------------------------|------------|--------------|----------------------------------------------------------------------------------------------------------------------------------|---------------------| +| sigmoid_cross_entropy_with_logits | Calculate the probability error by element | Input | X | Input tensor | float32,
float64 | +| | | | label | Enter the label corresponding to the tensor | float32,
float64 | +| | | Output | Y | Output tensor | float32,
float64 | +| | | Attributes | ignore_index | Ignored target value | int | +| | | | name | Optional, the prefix identifier of the output weight of the network layer, the default is None | string | +| | | | normalize | If true, the output is divided by the number of targets excluding the target corresponding to ignore_index, the default is False | string | + +**Table 142 – softmax operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|-----------------------------------------------------------|------------|---------|----------------------|---------------------------------| +| softmax | Apply $y = \exp(x) / \sum(\exp(x))$ to the input tensor x | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | +| | | Attributes | axis | Calculated dimension | int | + +**Table 143 – softplus operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|----------------------------------------------------|--------|---------|---------------|---------------------------------| +| softplus | Apply $y = \ln(\exp(x) + 1)$ to the input tensor X | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | + +**Table 144 – softsign operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------|---------------------------------------------|--------|---------|---------------|---------------------------------| +| softsign | Apply $(x/(1 + x ))$ to the input tensor X | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | + +**Table 145 – reduce\_log\_sum operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------------|-------------------------------------------------------------------------------|--------|---------|---------------|---------------------------------| +| reduce_log_sum | According to the given dimensions, calculate the LogSum of each input element | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | + +**Table 146 – reduce\_log\_sum operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------------|-------------------------------------------------------------------------------|------------|----------|-------------------------------|-----------| +| reduce_log_sum | According to the given dimensions, calculate the LogSum of each input element | Attributes | axis | Calculated dimension | int | +| | | | keepdims | Whether to keep the dimension | bool | + +**Table 147 – reduce\_log\_sum\_exp operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|--------------------|-------------------------------------------------------------------------------|------------|----------|-------------------------------|---------------------------------| +| reduce_log_sum_exp | According to the given dimensions, calculate the LogSum of each input element | Input | X | Input tensor | float16,
float32,
float64 | +| | | Output | Y | Output tensor | float16,
float32,
float64 | +| | | Attributes | axis | Calculated dimension | int | +| | | | keepdims | Whether to keep the dimension | bool | + +**Table 148 – resize\_bilinear operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|-----------------|---------------------------------------------------------------------------------|--------|---------------|-----------------------------------------------------------------------------------------------------------------------------|-------------------------------| +| resize_bilinear | Use bilinear interpolation to resize the input image data to the specified size | Input | X | Input tensor | float32,
float64,
uint8 | +| | | | Shape | Enter the size tensor after resize | int32,
int64 | +| | | Output | Y | Output tensor | float32,
float64,
uint8 | +| | | | name | Used when developers print debugging information | string | +| | | | align_corners | If True, align the center of the 4 corner pixels of the input and output tensors, and retain the value of the corner pixels | bool | +| | | | align_mode | Bilinear interpolation option | int | +| | | | data_format | Specify the input data format | string | + +**Table 149 – resize\_nearest operator** + +| Operator | Description | Field | Keyword | Definition | Data Type | +|----------------|----------------------------------------------------------------------------------------------------|------------|---------------|-----------------------------------------------------------------------------------------------------------------------------|-------------------------| +| resize_nearest | Use the nearest neighbor interpolation method to resize the input image data to the specified size | Input | X | Input tensor | float32, float64, uint8 | +| | | | Shape | Enter the size tensor after resize | int32, int64 | +| | | Output | Y | Output tensor | float32, float64, uint8 | +| | | Attributes | name | Used when developers print debugging information | string | +| | | | align_corners | If True, align the center of the 4 corner pixels of the input and output tensors, and retain the value of the corner pixels | bool | +| | | | data_format | Specify the input data format | string | + +#### 8.3.2.2 The definition of training operator + +##### 8.3.2.2.1 Loss function + +This clause defines the operators required in the process of neural network training. + +**Table 150 – compute\_loss operator** + +| Operator | Description | Field | Keyword | Definition | +|--------------|-------------------------------------------------------------------------------------|--------|------------------|--------------------------------------------------------| +| compute_loss | By passing an existing loss function name. Return a scalar loss for each data point | Input | y_true | True label | +| | | | y_pred | Predictive value | +| | | | loss | Loss function | +| | | Output | l | Each sample has a tensor of scalar loss | +| | | | Type Constraints | tensor(float16, float32, float64) | +| | | | | Limit input and output types to floating-point tensors | + +The available loss functions include: mean\_squared\_error, mean\_absolute\_error, mean\_absolute\_percentage\_error, mean\_squared\_logarithmic\_error, squared\_hinge, hinge, categorical\_hinge, logcosh, corss\_entropy. + +**Table 151 – apply\_optimize operator** + +| Operator | Description | Field | Keyword | Definition | +|----------------|----------------------------------------------------------------------------------------------------------------------------|------------------|-----------------------------------|--------------------------------------------------------| +| apply_optimize | By passing the optimized tensor, parameter list and optimizer, the gradient is calculated and the parameters are optimized | Input | loss | Tensor that needs to be optimized | +| | | | val_list | parameter list | +| | | | optimizer | Optimizer | +| | | Output | grads | gradient | +| | | Type Constraints | tensor(float16, float32, float64) | Limit input and output types to floating-point tensors | + +##### **8.3.2.2.2 The definition of inverse operator** + +This clause defines the inverse operator for the training process of the training of neural networks. The inverse operator describes the gradient calculation behaviour of the neural network calculation graph structure and is an important part of the neural network structure. The neural network trains the parameters through the gradient update method, and the key operator involves solving the gradient of the loss function with respect to the parameters. The calculation of the gradient is completed by backpropagation. Given the input features and standard answers, the forward operator is used to calculate the loss function at first. Then the inverse operator (operator) is used to solve the gradient of the model parameters through the backpropagation algorithm. This clause uses some operators to illustrate the correspondence between forwarding operators and inverse operators. Most inverse operators and forward operators have a one-to-one correspondence. The input of the inverse operator is the input and output (optional) of the forward operator, and the derivative of the loss function value with respect to the output. The output of the inverse operator is the derivative of the loss function value with respect to the input variable of the forward operator. Other reverse operators can refer to these examples to transform the forward operators. If there are no special instructions, the above rules are used to define the inverse operators by default. + +**Table 152 – forward operator** + +| Operator | Description | Field | Keyword | Definition | +|----------|------------------------------------------------------------------------|------------------|-----------------------------------|--------------------------------------------------------| +| forward | Perform the feedforward computation of input operand and model weights | Input | X | Input operand | +| | | | Y | Tensor weights applied to the output | +| | | Output | Z | Output tensor | +| | | Type Constraints | tensor(float16, float32, float64) | Limit input and output types to floating-point tensors | + +**Table 153 – backward operator** + +| Operator | Description | Field | Keyword | Definition | +|----------|---------------------------------------------------------------------|------------------|-----------------------------------|-------------------------------------------------------------| +| backward | Perform the backward propagation of input operand and model weights | Input | X | Input tensor | +| | | | Y | Tensor weights applied to the output | +| | | | dZ | The derivative of the loss function value with respect to Z | +| | | Output | dX | The derivative of the loss function value with respect to X | +| | | | dY | The derivative of the loss function value with respect to Y | +| | | Type Constraints | tensor(float16, float32, float64) | Limit input and output types to floating-point tensors | + +**Table 154 – add operator** + +| Operator | Description | Field | Keyword | Definition | +|----------|--------------------------------------------------------------------------------------------------------------------------------------|------------------|-----------------------------------|--------------------------------------------------------| +| add | Perform the addition of two inputs and broadcast the parameters on the right to match the size of the input on the left if necessary | Input | X | Input operand 1 | +| | | | Y | Input operand 2 | +| | | Output | Z | Output tensor | +| | | Type Constraints | tensor(float16, float32, float64) | Limit input and output types to floating-point tensors | +| | | Attributes | axis (int) | Calculated dimension | +| | | | broadcast (bool) | Whether to allow broadcasting | + +**Table 155 – add\_grad operator** + +| Operator | Description | Field | Keyword | Definition | +|----------|--------------------------------------------------------------------------------------------------------------------------------------|--------|---------|-------------------------------------------------------------| +| add_grad | Perform the addition of two inputs and broadcast the parameters on the right to match the size of the input on the left if necessary | Input | X | Input operand 1 | +| | | | Y | Input operand 2 | +| | | | dZ | The derivative of the loss function value with respect to Z | +| | | Output | dX | The derivative of the loss function value with respect to X | +| | | | dY | The derivative of the loss function value with respect to Y | + +**Table 155 – add\_grad operator** + +| Operator | Description | Field | Keyword | Definition | +|----------|-------------|------------------|-----------------------------------|--------------------------------------------------------| +| | | Type Constraints | tensor(float16, float32, float64) | Limit input and output types to floating-point tensors | +| | | Attributes | axis (int) | Calculated dimension | +| | | | broadcast (bool) | Whether to allow broadcasting | + +**Table 156 – mat\_mul operator** + +| Operator | Description | Field | Keyword | Definition | +|----------|-----------------------|------------------|-----------------------------------|--------------------------------------------------------| +| mat_mul | Matrix multiplication | Input | X | Input tensor 1 | +| | | | Y | Input tensor 2 | +| | | Output | Z | Output tensor | +| | | Type Constraints | tensor(float16, float32, float64) | Limit input and output types to floating-point tensors | + +**Table 157 – mat\_mul\_grad** + +| Operator | Description | Field | Keyword | Definition | +|--------------|----------------------------------------|------------------|-----------------------------------|-------------------------------------------------------------| +| mat_mul_grad | Matrix multiplication reverse operator | Input | X | Input tensor 1 | +| | | | Y | Input tensor 2 | +| | | | dZ | The derivative of the loss function value with respect to Z | +| | | Output | dX | The derivative of the loss function value with respect to X | +| | | | dY | The derivative of the loss function value with respect to Y | +| | | Type Constraints | tensor(float16, float32, float64) | Limit input and output types to floating-point tensors | + +**Table 158 – fc operator** + +| Operator | Description | Field | Keyword | Definition | +|-------------------------|-----------------------|------------------|-----------------------------------|--------------------------------------------------------| +| fc
(fully_connected) | Fully connected layer | Input | X | Input tensor | +| | | | W | Tensor weights applied to the output | +| | | | b | Bias | +| | | Output | Y | Output tensor | +| | | Type Constraints | tensor(float16, float32, float64) | Limit input and output types to floating-point tensors | +| | | Attributes | axis (int) | batch_size dimension, default is 0 | +| | | | units (int) | Number of hidden units | + +**Table 159 – fc\_grad operator** + +| Operator | Description | Field | Keyword | Definition | +|----------|-----------------------|------------------|-----------------------------------|-------------------------------------------------------------| +| fc_grad | Fully connected layer | Input | X | Input tensor | +| | | | W | Tensor weights applied to the output | +| | | | b | Bias | +| | | | dY | The derivative of the loss function value with respect to Y | +| | | Output | dX | The derivative of the loss function value with respect to X | +| | | | dW | The derivative of the loss function value with respect to W | +| | | | db | The derivative of the loss function value with respect to b | +| | | Type Constraints | tensor(float16, float32, float64) | Limit input and output types to floating-point tensors | +| | | Attributes | axis (int) | batch_size dimension, default is 0 | +| | | | units (int) | Number of hidden units | + +# 9 AI model compression + +## 9.1 Compact representation + +The compact model representation format is required to define the serialization format of the compressed AI model, without regarding the technical details of specific neural network compression algorithms. + +This document builds upon [ISO/IEC 15938-17] which specifies a compressed representation of the parameters/weights of a trained neural network and a decoding process for the compressed representation, complementing the description of the network topology in existing (exchange) formats for neural networks. + +The syntax format, semantics, associated decoding process requirements, parameter sparsification, parameter transformation methods, parameter quantization, entropy coding method and integration/signalling within existing exchange formats are specified in [ISO/IEC 15938-17]. + +The externally observable output behaviour is required to conform to [ISO/IEC 15938-17]. + +### 9.1.1 Multiple models + +The compression technology for multiple models can reduce the amount of data stored by sharing the information in multiple models. One method of compression technology for multiple models is to reduce the total number of model parameters by sharing part of the weight of the model in multiple models, so as to achieve the purpose of overall compression for multiple models. In addition to reducing the storage space of the model, this method only needs to read the weight of the shared part of the model when the computing unit switches among multiple models in the calculation process. In this way, this method reduces the number of parameters that need to be processed and achieves the purpose of saving bandwidth. This method can be compatible with other methods such as structured matrix, pruning, quantization and so on, which can be used to compress the compressed model. + +The residual quantization for multiple models refers to the quantization based on the model on the increment of the pre-training model. At present, the neural network usually uses the pre-training model as the initial way to accelerate the training and improve the network performance. However, the traditional quantization method does not make full use of the pre-trained model, and the direct quantization of the original weight often leads to high performance loss in the practical application. The residual quantization method is aimed at quantizing the increment of the model on the pre-trained model. Under the premise of achieving the same quantization effect, the original performance can be guaranteed as far as possible. + +#### 9.1.1.1 Compression + +The input of the operation is the multiple topology consistent model, including the weight tensor list Weights\_list, the optional bias vector list Bias\_list and a compression method option M\_Method. The output of the operation includes a weight tensor W\_share, an optional bias vector B\_share, a weight tensor list W\_specific, an optional bias vector list B\_specific, and an error indicator Error. In W\_share and B\_share, the shared information between multiple models is stored. In W\_specific and B\_specific, the additional information across multiple models is stored. The compression operators for multiple models are defined in Table 160. + +**Table 160 – The compression for multiple models** + +| Operator | Description | Parameters | Keywords | Definition | Data type | Data format | +|--------------------|------------------------------------|------------------|--------------|-------------------------------------|-----------------|-------------| +| MultimodelCompress | Compress the input multiple models | Input Attributes | Weights_list | Input weight tensor | List of tensors | NS | +| | | | Bias_list | Input bias tensor | List of vectors | NS | +| | | | M_Method | Specific compression method options | value | int | +| | | Output | W_share | Output shared weight tensor | tensor | NS | +| | | | B_share | Output shared bias tensor | vector | NS | + +**Table 160 – The compression for multiple models** + +| Operator | Description | Parameters | Keywords | Definition | Data type | Data format | +|----------|-------------|------------|------------|----------------------------------------|-----------------|-------------| +| | | | W_specific | An output list of unique weight tensor | List of tensors | NS | +| | | | B_specific | An output list of unique bias tensor | List of vectors | NS | +| | | | Error | Error indicator | value | int | + +The definition of M\_method is shown in Table 161. + +**Table 161 – M\_method definition** + +| M_method | | +|----------|--------------------------------------------| +| label | type | +| 1 | UNDEFINED | +| 2 | multiple models weight sharing compression | +| 3 | Residual quantization compression | + +- If the label of M\_method is 2, see clause 9.2.1.1.1. +- If the M\_method label is 3, see clause 9.2.1.1.2. + +#### 9.1.1.2 Decompression + +The process of decompression for multiple models is to decode compressed data for multiple models into an uncompressed data format that conforms to the specification of neural network representation. + +The inputs of decompression for multiple models operator are a common weight tensor W\_share, an optional common bias vector B\_share, a list of specific weight tensors W\_specific, an optional list of specific bias vectors B\_specific, and a compression method option Q\_Method. The output is multiple topologically consistent models, including the weight tensor list Weights\_list and the optional bias vector list Bias\_list, along with the error indicator Error. In W\_share and B\_share, the shared information between multiple models is stored. W\_specific and B\_specific hold specific attachment information across multiple models. See Table 160 and Table 161 for detailed definitions. The definition of decompression for multiple models operator data is shown in Table 162. + +**Table 162 – The decompression for multiple models operator** + +| Operator | Description | Parameters | Keywords | Definition | Data type | Data format | +|----------------------|---------------------------------------------|------------|------------|-----------------------------|-----------------|-------------| +| DecompressMultimodel | Decompression for the input multiple models | Input | W_share | Shared weight tensor | tensor | NS | +| | | | B_share | Shared bias tensor | vector | NS | +| | | | W_specific | Specific weight tensor list | List of tensors | NS | + +**Table 162 – The decompression for multiple models operator** + +| Operator | Description | Parameters | Keywords | Definition | Data type | Data format | +|----------|-------------|------------|--------------|---------------------------------------|-----------------|-------------| +| | | | B_specific | Specific bias tensor list | List of vectors | NS | +| | | | Q_Method | Specific decompression method options | value | int | +| | | Output | Weights_list | List of weight tensor | List of tensors | NS | +| | | | Bias_list | List of bias tensor | List of vectors | NS | +| | | | Error | Error indication | value | int | + +The decompressed\_Method definition is shown in Table 163. + +**Table 163 – The decompressed method definition** + +| Q_Method | | +|----------|--------------------------------------------------| +| Label | type | +| 1 | UNDEFINED | +| 2 | Multiple models weight sharing | +| 3 | Residual quantization method for multiple models | + +- If the label of Q\_Method is equal to 2, refer to the clause of decompression of multiple models weight sharing (see clause 9.2.2.1.1). +- If the Q\_Method label is equal to 3, refer to the clause of decompression of residual quantization (see clause 9.2.2.1.2). + +#### **9.1.1.3 Data generation** + +To apply multiple models weight sharing compression technique, it is necessary to impose restrictions on the weights of multiple models during their training, to achieve the weight structure that complies with this approach. Multiple models weight update operator is defined in Table 164. + +**Table 164 – Multiple models weight update operator** + +| Operator | Description | Parameters | Key words | Definition | Data type | Data format | +|---------------------|-----------------------------------------------|------------|-----------|-----------------------------|-----------|-------------| +| Mul_m_weight_update | Update multiple models weight during training | Input | x | Training input sample | tensor | NS | +| | | | Y | Training input sample label | tensor | NS | +| | | | Graph | Network graph | function | | +| | | | Optimizer | Optimizer | function | | +| | | | W_share | Shared weight tensor | tensor | NS | + +**Table 164 – Multiple models weight update operator** + +| Operator | Description | Parameters | Key words | Definition | Data type | Data format | +|----------|-------------|------------|--------------------|--------------------------------------------|-----------------|-------------| +| | | | B_share | Shared bias tensor | vector | NS | +| | | | W_specific | Specific weight tensor list | List of tensors | NS | +| | | | B_specific | Specific bias tensor list | List of vectors | NS | +| | | | model_indicator | Model index corresponding to training data | value | Int | +| | | | loss_factor | Error normalization coefficient list | vector | float | +| | | Attributes | axe | Weight aggregation dimension | int | Int | +| | | Output | W_share_updated | Updated shared weight tensor | tensor | NS | +| | | | B_share_updated | Updated shared bias tensor | vector | NS | +| | | | W_specific_updated | Updated specific weight tensor list | List of tensors | NS | +| | | | B_specific_updated | Updated specific weight tensor list | List of vectors | NS | + +For neural networks that utilize this compression technique, the approaches in clause 9.2.2.1.1.3 can be applied for decompression. The result model has the same structure as the origin and therefore is able to support the forward propagation process (inference). Regarding the training of multiple models, there are two approaches discussed in the following clauses. + +### 9.1.2 Quantization + +#### 9.1.2.1 Compression + +##### 9.1.2.1.1 Definition + +Quantization compression is a technique used to change the bitwidth $b$ of parameter $\theta$ in the original high-precision complex model $M$ from a high bit representation (such as 32-bit floating point) to a lower bit width. In this process, quantization can be viewed as a mapping of model weight from continuous space to discrete space, $q: \theta \rightarrow \theta'$ . The goal is to find $q$ that reduces the bitwidth $b$ as low as possible with no significant loss in model accuracy after quantization. In the quantization process, it may be necessary to retrain the network parameters, and the internal parameters will change. + +Quantization methods are highly compatible. The combined use of other compression methods included in the standard is described below. + +###### a) Multiple models compression + +Multiple models weight sharing can achieve the overall compression of multiple models. Quantization can also be applied in this way. For quantization methods without retraining, the shared weights can be compressed directly. For the quantization methods that need to be retrained, the retraining should be executed according to the training method used in weight sharing. For shared + +weights and specific weights, different bitwidth can be used to improve the accuracy in the quantization process according to its importance. + +###### b) Compression of pruning + +For pruning methods, quantization compression can be applied to the pruned model for further optimization. After unstructured pruning, the corresponding sparse mask table can be obtained, and corresponding weight of 1 (that is, the weight not pruned) in the mask table can be quantized. Structural pruning can quantize the pruned model directly because it deletes part of the structure. + +###### c) Structured matrix + +For the structured matrix compression method, it is necessary to quantize first to reduce the contents of the matrix, then proceed with the structured matrix method. However, since the quantization method of codebook involves additional index information of the quantization table, the operator of the original weight will be transferred to the quantization table index. + +In application scenarios, model quantization can be used for different tasks, including machine vision tasks (image classification, target detection, semantic segmentation, etc.), video processing and analysis tasks (behaviour recognition, target tracking), speech processing tasks, etc. Most of the original models are trained with 32-bit floating-point precision, so there is space for quantized compression. Quantization can be widely used in convolution layers, full connection layers, cell structures in recurrent neural networks, and so on. In terms of training, low-bit models can be trained directly from scratch or compressed with trained models. In some previous tests, the quantization method can maintain the original accuracy level without any retraining and has about four times the compression rate. + +This clause first introduces two basic quantization operators and then proposes different quantization schemes for parameters and activation data, respectively. + +##### **9.1.2.1.2 Basic quantization operator** + +Given an input tensor of a continuous type, the quantization operator changes it into a discrete quantized tensor by using different methods. The definition and description of the parent class operator method are shown in Table 161 and Table 162. The definition of the basic quantization operator is shown in Table 165. + +**Table 165 – Definition of basic quantitative operator** + +| Operator | Description | Field | Keyword | Definition | Data type | Data format | +|-----------------------------|--------------------------------------|--------|-------------|--------------------------|---------------|-------------| +| Basic quantization operator | The input tensor is sparse quantized | Input | X | Input tensor | List [tensor] | NS | +| | | | granularity | Quantization granularity | Value | int | +| | | | uniformity | Quantization uniformity | Value | int | +| | | | static | Offline Quantization | Value | int | +| | | | Q_Method | Quantization method | Value | int | +| | | | kwarg | Parameters | Dict | Dict | +| | | Output | Y | Output tensor | list [tensor] | NS | + +The quantization granularity refers to the quantization range of different hierarchies in the input vector, including unified quantization, layer quantization, channel quantization and other granularity, as shown in Table 166. + +**Table 166 – Definition of granularity** + +| label | Definition | +|--------------|----------------------| +| 1 | Unified quantization | +| 2 | Layer quantization | +| 3 | Channel quantization | +| 4 | Other granularity | + +Among the labels above: + +- a) If the label of granularity is 1, all the input tensors (assuming there are $n$ ) are uniformly quantized. +- b) If the label of granularity is 2, all the input tensors (assuming $n$ ) are quantized according to the layer granularity. +- c) If the label of granularity is 3, all the input tensors (assuming $n$ ) are quantized according to the channel granularity. +- d) If the label of granularity is 4, the input tensors are quantized according to the granularity given by the user. + +Quantization can be divided into uniform and non-uniform in terms of uniformity. Uniform quantization includes symmetric quantization and asymmetric quantization, as shown in Table 167. + +**Table 167 – Definition of uniformity** + +| label | Definition | +|--------------|--------------------------| +| 1 | Symmetric quantization | +| 2 | Asymmetric quantization | +| 3 | Non-uniform quantization | + +Among the labels above: + +- a) If the label of uniformity is 1, the value range of the output tensors is symmetric about 0, called symmetric quantization. +- b) If the label of uniformity is 2, the value range of the output tensors is symmetric about a certain non-zero value, called asymmetric quantization. +- c) If the label of uniformity is 3, the step size of quantization is variable, called non-uniform quantization. + +Offline quantization means the parameters of the operator are calculated in advance. The opposite is online quantization, both shown in Table 168. In contrast, dynamic quantization calculates parameters online and is specific for each forward activation, as shown in Table 168. + +**Table 168 – Definition of offline and online quantization** + +| label | Definition | +|--------------|----------------------| +| 1 | Static quantization | +| 2 | Dynamic quantization | + +Among the labels above: + +- a) If the label is 1, parameters of operator in static quantization are pre-calculated and fixed for each forward activation; +- b) If the label is 2, parameters of operator in dynamic quantization are calculated online and specific for each forward activation. + +Quantization method Q\_Method represents the specific quantization method, and the corresponding label meaning is shown in Table 165. + +**Table 169 – Definition of Q\_ method quantization method** + +| Label | Object | Type | +|-------|--------------------|--------------------------------------| +| 1 | Weight, activation | Linear quantization | +| 2 | Weight | Codebook quantization | +| 3 | | Non-linear function mapping | +| 4 | | INT4 quantization | +| 5 | | Bounded rectifier units quantization | +| 6 | Activation | Trainable alpha quantization | +| 7 | | INT4 quantization | +| 8 | | Bounded rectifier units quantization | +| 9 | | Proportional alignment quantization | + +#### 9.1.2.2 Decompression + +##### 9.1.2.2.1 Definition + +Dequantization aims at recovering those compressed data from integers to their original precision. It is designed to benefit the procedure of computation and resolve the incompatibility among different platforms and operator phases. Most of dequantization operators come from their quantization counterparts and detailed information can be found in the last clause. In this clause, two dequantization operators are introduced at first. Then, two distinct methods are raised to apply the inverse quantization to parameters and activations respectively. + +##### 9.1.2.2.2 Basic dequantization operator + +Given a discrete input tensor, inverse quantization adopts different methods to generate a high-precision output tensor. The definition and description of parent operators in inverse quantization can be found in Table 170 and Table 172. The basic inverse quantization operators are given in Table 170. + +**Table 170 – Basic inverse quantization operators** + +| Operator | Description | Field | Keyword | Definition | Type | Format | +|---------------------------|-----------------------------------|--------|-------------|------------------------------------|------------------|--------| +| Dequantizati
on_tensor | Dequantize
the input
tensor | Input | X | Input tensor | List
[tensor] | NS | +| | | | granularity | Dequantiza-
tion
granularity | Value | int | +| | | | uniformity | Quantization
uniformity | Value | bool | +| | | | D_Method | Dequantiza-
tion method | Value | int | +| | | Output | Y | Output
tensor | List
[tensor] | NS | + +Dequantization can be divided into uniform and non-uniform in terms of uniformity. Uniform dequantization contains symmetric dequantization and asymmetric dequantization, shown in Table 165. + +**Table 171 – Definition of uniformity** + +| label | Definition | +|-------|----------------------------| +| 1 | Symmetric dequantization | +| 2 | Asymmetric dequantization | +| 3 | Non-uniform dequantization | + +Among above labels: + +- a) If the uniformity label is 1, the value range of the input tensors is symmetric about 0, which is called symmetric dequantization. +- b) If the uniformity label is 2, the value range of the input tensors is symmetric about a certain non-zero value, which is called asymmetric dequantization. +- c) If the uniformity label is 3, the dequantization step size is variable, which is called non-uniform dequantization. + +The granularity is the dequantization range of different hierarchical structures in the input vector of the dequantization method, including uniform dequantization, hierarchical dequantization, channel dequantization, and other granular dequantization, as shown in Table 172. + +**Table 172 – Dequantization granularity** + +| Label | Definition | +|-------|-------------------------------| +| 1 | Uniform dequantization | +| 2 | Hierarchical dequantization | +| 3 | Channel dequantization | +| 4 | Other granular dequantization | + +In these indexes: + +- If the label of granularity is 1, all ( $N$ ) input tensors are uniformly dequantized. +- If the label of granularity is 2, all ( $N$ ) input tensors are dequantized according to hierarchical granularity. +- If the label of granularity is 3, all ( $N$ ) input tensors are dequantized according to channel granularity. +- If the label of granularity is 4, all ( $N$ ) input tensors are dequantized according to a user-given granularity. + +D\_Method denotes the specific dequantization methods, and their corresponding label and definition can be found in Table 173. + +**Table 173 – Definition of D\_method** + +| Label | Objective | Type | +|-------|------------------------|--------------------------------------| +| 1 | Parameters/activations | Linear dequantization | +| 2 | Parameters | Codebook dequantization | +| 3 | | Non-Linear dequantization | +| 4 | | | +| 5 | | INT4 dequantization | +| 6 | activations | Trainable Alpha value dequantization | +| 7 | | | +| 8 | | INT4 dequantization | + +### 9.1.3 Pruning + +#### 9.1.3.1 Compression + +Pruning has strong compatibility, which is briefly analysed by combining the use of other compression methods below. + +##### a) Multiple models compression + +Joint multiple models compression can be achieved by sharing the weights of multiple models. Two pruning methods can also be applied to this technique. First, leverage the pruning methods to compress the original multiple models. Then, use multiple models compression algorithms to achieve further compression. + +##### b) Quantization + +Quantization methods can be used to further optimize pruned models. For unstructured pruning, quantize the weights corresponding to value 1 in the sparse mask table (unpruned weights). For structured pruning, the compressed model can be directly quantized as structured pruning removes entire structures in the original model. + +##### c) Structured matrix + +For the structured matrix method, the content of the matrix data shall be pruned first. After that, process it according to the structured matrix. + +### 9.1.4 Pruning operator + +Deep neural network pruning aims at arranging the weights of pre-trained models with respect to given criteria, e.g., for convolutions, channels can be removed based on their low saliency as described in subclause G.1.4.2 of [ISO/IEC 15938-17]. Additionally, data dependent sparsification + +(subclause G.1.4.3 of [ISO/IEC 15938-17]) is another technique employed in this process. this module considers 3 different factors in the loss to select the weights of a layer, including a task-based loss, a compressibility loss and a diversity loss term that encourages the diversity between filters. According to the pruning/accelerating rate set by the user, remove the corresponding proportion of unimportant weights. Specific definitions can be found in Table 174 and Table 175. + +**Table 174 – Definition of pruning** + +| Operator | Description | Field | Keyword | Definition | Data type | Data format | +|----------|---------------------------------------------|--------|-------------|----------------------|-----------------|-------------| +| Pruning | Prune input weights of deep neural networks | Input | X | Input weight tensor | List of vectors | NS | +| | | | granularity | Pruning granularity | Value | Int | +| | | | Rate | Pruning rate | Value | Float | +| | | | P_Method | Pruning method | Value | UINT | +| | | Output | W | Output weight tensor | List of vectors | NAS | + +**Table 175 – Definition of pruning granularity** + +| Pruning granularity | | +|---------------------|----------------------| +| Label | Type | +| 1 | UNDEFINED | +| 2 | Structured Pruning | +| 3 | Unstructured Pruning | + +#### 9.1.4.1 Decompression + +##### 9.1.4.1.1 Definition + +Inverse sparsity of deep neural networks involves recovering the pruned weights to the networks original size and filling the vacant position in the networks with 0. The operator can be referred to Table 176. + +**Table 176 – Definition of pruning** + +| Operator | Description | Field | Keyword | Definition | Data type | Data format | +|---------------|-------------------------------------------|--------|--------------|----------------------|--------------------|-------------| +| Inverse Prune | Structural recover the weights of network | Input | W | Sparse weight tensor | list[tensor] | NS | +| | | | B (optional) | Sparse bias tensor | list[tensor] | NS | +| | | | S | Mask | list[vector ] | NS | +| | | | granularity | granularity | Value | Int | +| | | Output | W_u | Pruned weight tensor | list[tensor] | NS | + +The definition of pruning granularity can be referred in Table 173. + +**Table 177 – Definition of pruning granularity** + +| Pruning granularity | | +|----------------------------|----------------------| +| Label | Type | +| 1 | UNDEFINED | +| 2 | Structured Pruning | +| 3 | Unstructured Pruning | + +##### 9.1.4.1.2 Inverse sparsity + +The process of inverse sparsity is as follow: + +Input parameters: Sparse tensor list W to be recovered; Sparse bias tensor list to be recovered; The dimension of N dim (considering that the memory format of the tensor may not be NCHW, the dimension of N should be set by users). + +#### 9.1.4.2 Data generation + +This clause mainly describes the mask generation in network pruning, which involves outputting the corresponding new neural network weights and mask tensor based on the given the deep neural network weight. The pruning data generation is shown in Table 178. + +**Table 178 – Pruning data generation** + +| Operator | Description | Field | Keyword | Definition | Data type | Data format | +|-----------------|------------------------------------------------|--------------|----------------|---------------------|------------------|--------------------| +| Mask generation | Output pruning mask with given network weights | Input | X | Input weight tensor | list[tensor] | NS | +| | | | P_Method | Pruning method | value | INT | +| | | | granularity | Pruning granularity | Value | Int | +| | | Output | Y | Output mask tensor | list[tensor] | NS | + +The definition of pruning granularity is shown in Table 179. + +**Table 179 – Definition of pruning granularity** + +| Pruning granularity | | +|----------------------------|------------------------------------| +| Label | Type | +| 1 | UNDEFINED | +| 2 | Structured pruning | +| 3 | Unstructured Pruning | +| 4 | Manual defined pruning granularity | + +The definition of P\_Method is shown in Table 180. + +**Table 180 – Definition of P\_method** + +| P_Method | | +|-----------------|---------------------------------| +| Label | Type | +| 1 | UNDEFINED | +| 2 | ALUM pruning | +| 3 | Incremental regularized pruning | + +### 9.1.5 Structured matrix + +#### 9.1.5.1 Compression + +The structured matrix compression is an operator that compresses the parameters of a neural network into a structured matrix with a certain format. Relevant definitions can be found in Table 181 to Table 183. The structured matrix compression operator is defined in Table 181. + +**Table 181 – Structured matrix compression operator definition** + +| Operator | Description | Parameters | Key words | Definition | Data type | Data format | +|--------------------------|----------------------------------------------------------------------------|-------------------|------------------|----------------------------------------------|------------------|--------------------| +| CompressStructuralMatrix | An operator compressing an input parameter matrix into a structured matrix | Input | X | Input tensor | Tensor | NS | +| | | | granularity | Compression granularity of structured matrix | Value | UINT | +| | | | S_Method | Method for the structured matrix compression | Value | UINT | +| | | Output | Y | Output tensor | Tensor | NS | + +**Table 182 – Structured matrix compression granularity definition** + +| granularity | | +|--------------------|-----------------------------------------------------------| +| Index | Type | +| 1 | UNDEFINED | +| 2 | Structured matrix compression applied to all input tensor | +| 3 | Layered structured matrix compression | +| 4 | Channelled structured matrix compression | +| 5 | Customized structured matrix compression | + +The method for compression (S\_Method) is defined in Table 183. + +**Table 183 – Compression method (S\_method) definition** + +| S_Method | | +|----------|----------------------------------------------------------| +| Index | Type | +| 1 | UNDEFINED | +| 2 | Block circulant matrix compression | +| 3 | Low-rank sparse decomposed structured matrix compression | +| 4 | Re-parameterization | + +#### 9.1.5.2 Decompression + +The decompression operator of the structured matrix is used to convert the structured matrix representation of a neural network to its original parameter matrix representation. The structured matrix technique can be used to represent a neural network by a certain matrix with a smaller magnitude of parameters. The decompression operator is defined in Table 184. + +**Table 184 – Structured matrix decompression operator definition** + +| Operator | Description | Parameter | Key words | Definitions | Data types | Data formats | +|--------------------------------|------------------------------------------------------------------------------------------------------------------|-----------|-------------|---------------------------------|------------|--------------| +| DecompressStructural
Matric | Convert a neural network from its structured matrix representation to its origin parameter matrix representation | Input | X | Input tensor | Tensor | NS | +| | | | granularity | Decompression granularity | Value | UINT | +| | | | D_Method | Decompression method to be used | Value | UINT | +| | | Output | Y | Output tensor | Tensor | NS | + +The granularity of decompression is shown in Table 185. + +**Table 185 – Structured matrix decompression granularity** + +| granularity | | +|-------------|-------------------------------------------------------------| +| Index | Type | +| 1 | UNDEFINED | +| 2 | Structured matrix decompression applied to all input tensor | +| 3 | Layered structured matrix decompression | +| 4 | Channelled structured matrix decompression | +| 5 | Customized structured matrix decompression | + +The method for decompression (D\_Method) is defined in Table 186. + +**Table 186 – Structured matrix D\_method definition** + +| D_Method | | +|----------|------------------------------------------------------------| +| Index | Type | +| 1 | UNDEFINED | +| 2 | Block circulant matrix decompression | +| 3 | Low-rank sparse decomposed structured matrix decompression | + +#### 9.1.5.3 Data generation + +The data generation of the structured matrix is used to update the weight tensor of a neural network layer, based on training data, network graph and optimizer. The aforementioned layer was compressed using the structured matrix. The structured matrix weight operator for data update is defined in Table 187. + +**Table 187 – Structured matrix weight update operator** + +| Operator | Description | Parameters | Key words | Definition | Data types | Data formats | +|------------------------------------|---------------------------------------------------------|------------|--------------|---------------------------|------------|--------------| +| update_st
ruc_matri
x_weight | Structured matrix weight update process during training | Input | x | Input sample for training | data | NS | +| | | | y | Truth label of sample | label | NS | +| | | | Graph | Network graph | Function | / | +| | | | Optimizer | Optimizer | function | / | +| | | | W_ori | Original weight tensor | tensor | NS | +| | | Output | B_ori | Original bias tensor | vector | NS | +| | | | W_structural | Updated weight tensor | tensor | NS | + +## 9.2 Compression workflow + +The model compression process adopts the following different methods to compress the AI model according to different needs: + +Quantization: First, quantize the weights in the pre-trained model to target bit width b. Then, train the quantized model till convergence to recover the model performance. Neural network compression (NNC) supports three types of parameter quantization: uniform quantization, codebook-based quantization and dependent scalar quantization, described in clauses 9.2.1, 9.2.2 and 9.2.3, of [ISO/IEC 15938-17] respectively. + +Pruning: e.g., for convolutions, channels can be removed based on their low saliency as described in clause G.1.4.2 of [ISO/IEC 15938-17]. + +Data dependent sparsification (clause G.1.4.3 of [ISO/IEC 15938-17]): this module considers 3 different factors in the loss to select the weights of a layer, including a task-based loss, a compressibility loss and a diversity loss term encourages the diversity between filters. + +Parameter unification (clause G.1.7 of [ISO/IEC 15938-17]): Refers to modifying neighbouring weights to create structured correlations and help entropy coding. + +Tensor decomposition (clause G.1.8 of [ISO/IEC 15938-17]): Based on low-rank approximation, the method derives smaller tensors from the original and transmits them separately. At the receiver, after decoding, the original tensor can be reconstructed or the sub-tensors can be multiplied at the inference stage if the system permits. + +Batch norm folding (clause G.1.9 of [ISO/IEC 15938-17]) + +Local scaling adaptation (clause G.1.10 of [ISO/IEC 15938-17]) + +Further to parameters reduction and quantization, which can impact the accuracy obtained with the decompressed model, NNC includes entropy coding based on the context-based adaptive binary arithmetic coding (CABAC) arithmetic engine used in several ITU/MPEG video compression standards. The binarization module has been adapted to neural network parameters and renamed DeepCABAC, which is described in clause 10 of [ISO/IEC 15938-17]. + +[ISO/IEC 15938-17] supports the compression of incremental representation of neural networks. + +In addition, the following methods can be defined or extended from NNC: + +- Multiple models: Utilizing the quantized first model as the guidance for the quantization of subsequent models. +- Adjustment to decomposing matrices: First, decompose the weight matrix of each layer into multiple matrices. Then, construct new layers which are parameterized by the decomposed weight matrices. Repeat these two steps until all layers are decomposed and replaced by new layers. + +### 9.2.1 Compression process + +#### 9.2.1.1 Multiple models + +##### 9.2.1.1.1 Shared compression operator for weights of multiple model layers + +###### 9.2.1.1.1.1 Definition + +The shared compression operator works on the layers with identical structures but across multiple models. The weights of a part of output neurons in these layers are shared among multiple models, while the weights of the other part of the output neurons are adjusted based on specific tasks (refer to Figure 8). This approach enables the filtering of data process information and in the meanwhile also performing optimization for specific tasks. + +![Diagram illustrating weight sharing between layers in multiple models. It shows three input neurons (blue circles) on the left and three output neurons (grey circles) on the right. The connections are labeled: 'Specific weights of models' (orange lines) and 'Shared weights among multiple models' (green lines). The diagram is labeled F.748.53(25).](906c1ba8ae95f501d4cd79eb08485307_img.jpg) + +The diagram shows three input neurons (blue circles) on the left and three output neurons (grey circles) on the right. The connections are labeled: 'Specific weights of models' (orange lines) and 'Shared weights among multiple models' (green lines). The diagram is labeled F.748.53(25). + +Diagram illustrating weight sharing between layers in multiple models. It shows three input neurons (blue circles) on the left and three output neurons (grey circles) on the right. The connections are labeled: 'Specific weights of models' (orange lines) and 'Shared weights among multiple models' (green lines). The diagram is labeled F.748.53(25). + +Figure 8 – Weight sharing between layers in multiple models + +###### 9.2.1.1.1.2 Weight aggregation + +The weight aggregation process aggregates $M$ $N$ -dimensional weight tensors with identical structures into an $N$ -dimensional shared weight tensor and a list containing $M$ $N$ -dimensional specific weight tensors. This is done by using index value-ruled dimensions and the information of aggregation points. Weight aggregation can be performed by the following steps: + +- a) Check the number of tensors in the input weight tensor list *Weight\_list* and the number of tensors the input bias tensor list *Bias\_list*. If they are not identical, indicate an error (value 1) and set the other output values to NULL, then close this operation. +- b) Check the shapes of the tensors in the input weight tensor list *Weight\_list* and bias tensor list *Bias\_list*. In case that they are not identical, indicate error (value 2) and set the other output values with NULL, close this operation. +- c) If the dimension axis of the aggregated weights is identical to the dimension of the output feature layer, set the status to 1; otherwise, set it to 2. +- d) If the status is 1, split each vector in *Bias\_list*, into two vectors *C* and *D*, at the aggregation indicator *slice\_indicator*. Reconstruct and output a new list *W\_specific* using the slices from the weight tensor list *Weights\_list*, and output *A* as *W\_share*. Reconstruct and output a new list *B\_specific* based on *D* from *Bias\_list*, and output *A* as *B\_share*. Output Unit as *Unit\_specific*. Error is 0. +- e) If the status is 2, reconstruct and output a new list *W\_specific* using the slices from the weight tensor list *Weights\_list*. Output *A* as *W\_share*. Output *Bias\_list* as *B\_specific*. *B\_share* is empty. Output Unit as *Unit\_specific*. Error is 0. + +Weight aggregation operator is defined in Table 188. + +**Table 188 – Weight aggregation among multiple models** + +| Operator | Description | Parameters | Key words | Definition | Data type | Data format | +|---------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------|------------------------|------------------------------------|-----------------|-------------| +| Multilayer_weight_regroup | aggregates M N-dimensional weight tensors with identical structure into a list containing an N-dimensional shared weight tensor and an M-dimensional specific weight tensor | Input | Weights_list | Input weight tensor list | List of tensors | NS | +| | | | Bias_list | Input bias vector list | List of vectors | NS | +| | | Attributes | axe | Aggregation dimension of weights | value | int | +| | | | slice_indicator | Slice indicator | value | int | +| | | Output | W_share | Output shared weight tensor list | tensor | NS | +| | | | B_share | Output shared weight tensor list | vector | NS | +| | | | W_specific | Output specific weight tensor list | List of tensors | NS | +| | | | B_specific | Output specific bias tensor list | List of vectors | NS | +| | | | Error | Error identifier | value | int | +| | | | | | | | + +The pseudocode of multiple model weight aggregation operator is shown in Table 189. + +**Table 189 – The pseudocode of multiple model weight aggregation** + +| multiple model weight aggregation | Descriptor | +|------------------------------------------------------------------------------------------------------|-------------------| +| Multilayer_weight_regroup(Weights_list, Bias_list, axe, slice_indicator, Error) { | | +| W_specific = [] | | +| B_specific = [] | | +| W_share = NULL | | +| B_share = NULL | | +| Error = 0 | | +| if(lens(Weights_list)!= lens(Bias_list)), | | +| Error=1 | | +| else{ | | +| for (i=0; i< lens(Weights_list)-1; i++){ | | +| if (Weights_list[i].shape != Weights_list[i+1].shape) ( Bias_list[i].shape != Bias_list[i+1].shape), | | +| Error=2 | | +| } | | +| } | | +| if (Error==1) { | | +| W_share = slice(Weights_list[0], axe, 0, slice_indicator) | | +| for (i=0; i< lens(Weights_list); i++){ | | +| W_specific.append(slice(Weights_list[0], axe, slice_indicator,-1)) | | +| } | | +| if(axe = dimension_of_out_feature_map) { | | +| B_share = slice(Bias_list[0], axe, 0, slice_indicator) | | +| for (i=0; i< lens(Bias_list); i++){ | | +| B_specific.append(slice(Bias_list[0], axe, slice_indicator,-1)) | | +| } | | +| } | | +| else { | | +| B_specific = Bias_list | | +| } | | +| return W_share, W_specific, B_share, B_specific, Error | | +| } | | + +##### 9.2.1.1.2 The residual quantization compression + +###### 9.2.1.1.2.1 The definition of residual quantization for multiple models + +First, each pre-trained model is retained to obtain the target model. It is assumed that the weight of the pre-training model is $W\_0$ , and the target model obtained by retraining is $W\_i = W\_0 + \Delta W$ . Then apply the weight sharing to each residual weight $\Delta W$ , which means the number of weight categories is $2^n$ in one convolution layer ( $n$ is the quantization bit, and weight sharing is applied for each convolution layer, so there is no weight sharing between different convolution layers). Through the above methods, the neural network convolutional layer can establish code storage to reduce the storage of the model through weight sharing. + +###### 9.2.1.1.2.2 Weight sharing + +Weight sharing is the quantization implementation of the residual quantization method for multiple models. The weight sharing method refers to the clustering method of deep compression. After the pre-trained model is retained and the increment is obtained, cluster the input model residuals (by layer). Meanwhile, fine-tuning of the model is carried out. For the same convolutional layer, the gradient of the weight is replaced with the mean value of the gradient of this class. + +Description of algorithm flow operator: input the weight tensor of the pre-trained model, DF\_model corresponds to the residual weight of the network model, n determines the quantization bit, and Weight\_list is the list of tensors. + +The weight of the pre-trained model is the untrained part (set attribute trainable = False). For the input residual weight of the input, K-means clustering (cluster function) is performed according to the given quantization bit. For each class, only the clustering centre needs to be stored. During the retraining phase (DFQuantization\_Weight\_Update, described in clause 9.1.2), the weight gradient belonging to the same class is updated to the mean of these gradients. After the retraining, the corresponding shared weights and the corresponding clustering results are generated, and the weights corresponding to each position can be restored according to the clustering results. (The residual quantization only quantizes the weight, and does not quantize the bias.) + +The residual quantization for multiple models operator is defined in Table 190. + +**Table 190 – Residual quantization for multiple models operator** + +| Operator | Description | Parameters | Key words | Definition | Data type | Data format | +|----------------|--------------------------------------|------------|--------------|---------------------------------|-----------------|-------------| +| DFQuantization | Residual quantization on input model | Input | Base_model | The weight of pre-trained model | tensor | NS | +| | | | DF_model | Weight residual | tensor | NS | +| | | | Weights_list | Input weight tensor list | List of tensors | NS | +| | | Attributes | n | Quantization bit | value | int | +| | | Output | W_share | Output shared weight tensor | tensor | NS | +| | | | W_specific | code storage | List | NS | +| | | | Error | Error indicator | value | int | + +The weight sharing operator is defined in Table 191. + +**Table 191 – Weight sharing operator** + +| Operator | Description | Parameters | Key words | Definition | Data type | Data format | +|----------------|------------------------------|------------|------------|-----------------------------|-----------|-------------| +| Weight Cluster | Weight sharing by clustering | Input | DF_model | Weight residual | tensor | NS | +| | | Attributes | n | Quantization bit | value | int | +| | | Output | W_share | Output shared weight tensor | tensor | NS | +| | | | W_specific | code storage | List | NS | + +**Table 191 – Weight sharing operator** + +| Operator | Description | Parameters | Key words | Definition | Data type | Data format | +|----------|-------------|------------|-----------|-----------------|-----------|-------------| +| | | | Error | Error indicator | value | int | + +The pseudocode of residual quantization is shown in Table 192. + +**Table 192 – The pseudocode of residual quantization** + +| Residual quantization for multiple models | Descriptor | +|-------------------------------------------------------------------------------|--------------------| +| DFquantization(Base_model, DF_model, n, Weights_list ) { | | +| W_specific = [] | | +| W_share = NULL | | +| Error = 0 | | +| if(DF_model == NULL), | | +| Error=1 | | +| else{ | | +| for (i=0; i< lens(Weights_list)-1; i++){ | | +| W_share[i] W_specific[i], Error_1=Cluster(DF_model[i],n) | K-means clustering | +| if(Error==1), | | +| exit() | | +| } | | +| } | | +| if (Error==1) | | +| exit() | | +| W_share=DFQuantization_weight_update(W_share,Base_model,gradients,W_specific) | Gradient updating | +| return W_share, W_specific, Error | | +| } | | + +#### **9.2.1.2 Quantization** + +##### **9.2.1.2.1 Basic quantization operator** + +###### **9.2.1.2.1.1 Linear quantization** + +Linear quantization sets the quantization bits at first, and then quantizes the full-precision weight of 32-bits to the low-bit value by calculating the corresponding scale transformation factor and offset factor for the weight of the quantization layer. Parameter tensors are uniformly quantized using a fixed step-size represented by the factors [ISO/IEC 15938-17]. + +Specific data for linear quantization is shown in Table 193. + +**Table 193 – Definition of linear quantization parameter** + +| Operator | Description | Field | Keyword | Definition | Data type | Data format | +|---------------------|---------------------------------------------------------------------------------------|--------|-------------|----------------------------------------------------|---------------|-------------| +| Linear quantization | The continuous high-precision network weight are mapped to the low-precision by layer | Input | X | Granularity index corresponds to the weight tensor | List [tensor] | NS | +| | | | X_scale | Weight quantization scale factor | List [tensor] | NS | +| | | | X_offset | Weight quantization bias factor | List [tensor] | NS | +| | | | N | Bitwidth | value | int | +| | | | granularity | Quantization granularity | Value | int | +| | | Output | Y | Weight tensor after linear mapping | List [tensor] | NS | + +The pseudocode of linear quantization parameter is shown in Table 194. + +**Table 194 – The pseudo code of linear quantization parameter** + +| Weight linear quantization | Descriptor | +|----------------------------------------------------------------|------------| +| def Linear_quantization (X, X_scale, X_offset, granularity,N){ | | +| if(granularity==1): { | | +| Y = round(add(mul(X_scale, X),X_offset), 0) | | +| } | | +| elseif (granularity==2): { | | +| for(x in X): { | | +| y = round(add(mul(x_scale, x),x_offset), 0) | | +| } | | +| } | | +| else: { | | +| for(x in X): { | | +| for(i in [0: x.channels]): { | | +| y[i] = round(add(mul(x_scale[i], x[i]),x_offset[i]), 0) | | +| } | | +| } | | +| } | | +| return Y | | +| } | | + +###### **9.2.1.2.1.2 Codebook quantization** + +Codebook quantization is used to change the original high-precision floating-point model into a tensor represented by an index value with a given quantization centre. The principle of quantization is the + +proximity quantization, which means that the original parameters are quantized to the nearest quantization centre value. The parameter tensors are represented as a codebook and quantized index tensor, with the latter having the same shape as the original tensors. The size of the codebook is chosen at the encoder and is transmitted as a metadata parameter. The codebook is composed of integer values that are strictly monotonically increasing [ISO/IEC 15938-17]. The quantized index tensor only has the range of the number of quantization centres, which is convenient for further compression. This method is mainly applicable to the compression of data generated by interval shrinkage quantization. Related data generation methods can be found in clause 9.2.3. The bias tensor is treated in the same way as the weight tensor. The specific data definition of the codebook method is shown in Table 195. + +**Table 195 – Definition of codebook quantization** + +| Operator | Description | Field | Keyword | Definition | Data type | Data format | +|-----------------------|----------------------------------------------------------------------------|--------|-------------|----------------------------------------------------|---------------|-------------| +| Codebook quantization | The discrete tensor is transformed into the tensor of index representation | Input | X | Granularity index corresponds to the weight tensor | List [tensor] | NS | +| | | | X_center | Quantization centre | List [tensor] | NS | +| | | | granularity | Quantization granularity | Value | int | +| | | Output | Y | Index tensor after quantization | List [tensor] | NS | + +X is the input tensor with high precision. X\_Center is the quantization centre, and each tensor list of quantization granularity corresponds to a group of centre points. Y is the index tensor after quantization, which is of an integer data type. The pseudocode is shown in Table 196. + +**Table 196 – The pseudo code of codebook quantization** + +| Codebook quantization | Descriptor | +|------------------------------------------|------------| +| def Codebook_quantization(X, X_center):{ | | +| X_shape = shape(X) | | +| if (granularity == 1):{ | | +| X= reshape(X, (-1,1)) | | +| for n in range(0, len(X_center)):{ | | +| dis[n] = abs(X –X_center [n]) | | +| } | | +| Y = argmin(dis, axis=0) | | +| Y = reshape(Y, X_shape) | | +| } | | +| if (granularity == 2):{ | | +| X = reshape(X, (X_shape[0],-1,1)) | | +| for n in range(0,len(X_center)):{ | | +| for i in range(0, len(X_center[n])):{ | | +| dis[n][i] = abs(X[n] –X_center[n][i]) | | +| } | | +| } | | +| } | | + +**Table 196 – The pseudo code of codebook quantization** + +| Codebook quantization | Descriptor | +|-------------------------------------------------|------------| +| Y[n] = argmin(dis[n], axis=0) | | +| } | | +| Y = reshape(Y, X_shape) | | +| } | | +| if (granularity == 3){ | | +| X = reshape(X, (X_shape[0], X_shape[1],-1,1)) | | +| for n in range(0,len(X_center): { | | +| for i in range(0, len(X_center[n]): { | | +| for j in range(0, len(X_center[n][i]): { | | +| dis[n][i][j] = abs(X[n][i] –X_center [n][i][j]) | | +| } | | +| } | | +| Y[n][i] = argmin(dis[n][i], axis=0) | | +| } | | +| Y = reshape(Y, X_shape) | | +| } | | +| return Y | | +| } | | + +**9.2.1.2.2 Parameter quantization operator** +**9.2.1.2.2.1 Non-linear function mapping** + +In the non-linear weight quantization, first, a non-linear mapping transformation is made to map the originally non-linear network weight to a new dimensional space, thus presenting an approximate linear distribution. Then, the weight after the mapping is linearly quantized. Linear quantization requires setting the bitwidth first, followed by calculating the corresponding scale transformation factor and offset factor for the weight of the quantization layer. The original 32-bit full precision weight is then quantized to the low bit value. The algorithm does not rely on any data and performs quantization operators by pre-trained model weights, allowing it to process all convolution layers and full connection layers that need to be quantized in parallel. The flow chart of the specific algorithm is shown in Figure 9. + +![Figure 9 – Algorithm flow chart](3f8bb1be8f73c7623f1380af78dcf8c6_img.jpg) + +``` + + graph LR + A[Original model] --> B[Weight non-linear function mapping] + B --> C[Setting bitwidth] + C --> D[Calculating scale factor and bias factor] + D --> E[Calculating quantization weight] + E --> F[Quantization model] + +``` + +F.748.53(25) + +Figure 9 – Algorithm flow chart + +**Figure 9 – Algorithm flow chart** + +According to the flow chart, the weights are mapped non-linearly at first. After setting the bitwidth, different scaling factors and bias factors are calculated according to different bitwidth. Scaling factors and bias factors determine the function that transforms full precision value to low bit value. Therefore, after calculating these two factors, the low bit value can be obtained and the original full precision weight can be replaced to get a smaller model. The specific steps are as follows: + + a) Weight mapping $\widetilde{W} = f(W)$ : In this clause, there are following three mapping methods: + + a) Linear mapping: $\widetilde{W} = W$ + +90 + +**Rec. ITU-T F.748.53 (03/2025)** + +- b) Log mapping: $\tilde{W} = \log_2 |W|$ +- c) Tanh mapping: $\tilde{W} = \tanh(\alpha W)$ +- b) Set bitwidth: Different bitwidth $n$ is set according to different network structures, and its value range is generally in the positive integer of [1,8]. +- c) Calculate scale factor $S$ and bias factor $Z$ corresponding to weight $\tilde{W}$ of quantization layer after non-linear transformation, where $n$ is the bitwidth. + +The formula of scale factor $S$ is equation (11): + +$$S = (\max(\tilde{W}) - \min(\tilde{W})) / (2^n - 1) \quad (11)$$ + +In this formula: + +$\tilde{W}$ weight after non-linear transformation + +$S$ scale factor + +$n$ bitwidth + +The formula of bias factor $Z$ is equation (12): + +$$Z = -\min(\tilde{W})/S \quad (12)$$ + +In this formula: + +$Z$ bias factor + +$S$ scale factor + +The specific operator definitions are shown in Table 197 and Table 198, and the definition of non-linear weight quantization is shown in Table 197. + +**Table 197 – Definition of non-linear weight quantization** + +| Operator | Description | Field | Keyword | Definition | Data type | Data format | +|--------------------------------|--------------------------------------------------------------------------------------|--------|-------------|----------------------------------------------------|---------------|-------------| +| Weight non-Linear quantization | The discrete high-precision network weight are mapped to the low precision by layers | Input | W | Granularity index corresponds to the weight tensor | List [tensor] | NS | +| | | | granularity | Quantization granularity | Value | int | +| | | | function | Non-Linear function | Value | int | +| | | | N | Bitwidth | Value | int | +| | | Output | W_q | Weight tensor after linear mapping | List [tensor] | NS | + +Label definition of non-linear function can be found in Table 198. + +**Table 198 – Optional non-linear quantization functions** + +| Label | Type | +|-------|-----------------| +| 1 | Linear function | +| 2 | Log function | +| 3 | Tanh function | +| 4 | Other function | + +Relevant pseudocode can be found in Table 199. + +**Table 199 – The pseudo code of weight non-linear quantization** + +| Weight non-linear quantization | Descriptor | +|-----------------------------------------------------------------|-------------------| +| def Weight_non_Linear_quantization (W, granularity,function,N){ | | +| if(function==1): { | | +| W = W | | +| } | | +| elif (function ==2): { | | +| W = abs(log(W)) | | +| } | | +| elif(function ==3): { | | +| W = tanh(W) | | +| W_scale = (max(W)-min(W))/(2^n-1) | | +| W_offset = -1*min(W)/W_scale | | +| W_q = Linear_quantization (W, W_scale, W_offset, granularity,N) | | +| } | | +| return W_q | | +| } | | + +###### 9.2.1.2.2.2 INT4 parameter quantization + +INT4 quantization refers to the quantization process that converts a tensor from its high-precision representation to an INT4 tensor or multiple INT4 tensors. Generally, Dual INT4 quantization is applied if the result contains multiple INT4 tensors. INT4 activation quantization is corresponds to INT4 parameter quantization. The latter includes neural network weight quantization and bias quantization. During the INT4 parameter quantization process, the bias of the network is not processed. For a neural network, quantization can be performed layer by layer, following these steps: + +- a) Load the weight tensor W to be quantized from the current layer in a network and the quantization scale factor list, then determine the range of quantization values based on the weight quantization bias identifier offset\_w. +- b) Determine the type of network layer and apply the quantization mode with the proper granularity: + - 1) If the layer is convolutional, the quantization with convolution granularity is applied. It quantizes the 3D convolution kernel tensor corresponding to each output channel. In detail, for each convolution kernel tensor W[k], determine the quantization mode, which is either single INT4 quantization or multiple INT4 quantization (commonly Dual INT4 quantization), based on the quantization scale factor list sf\_w\_list[k]. On top of this, compute the quantized convolution kernel tensor W\_quantized[k]. + - 2) If the layer is fully connected, the quantization with layer-granularity is applied. It quantizes the weight tensor W of input layer. In detail, based on the quantization scale factor list sf\_w\_list, determine the quantization mode, which is either single INT4 quantization or multiple INT4 quantization (commonly Dual INT4 quantization). On top of this, compute the quantized convolution kernel tensor W\_quantized[k]. + - 3) When performing Dual INT4 quantization, the error can be determined by using the minimal mean square error (MMSE). For a tensor T to be quantized and a given quantization scale factor, let $\tilde{T}_j^1, \tilde{T}_j^2$ the j-th element in quantized tensors $\tilde{T}^1, \tilde{T}^2$ ; $\mathbb{Z}_q$ the quantization scope. These can be computed by Formula 18. + +$$\begin{aligned} +\text{c)} \quad & \min_{\tilde{T}_j^1, \tilde{T}_j^2} (T_j - \alpha_1 \tilde{T}_j^1 - \alpha_2 \tilde{T}_j^2)^2 \Leftrightarrow \min_{\tilde{t} \in \mathbb{Z}_q} (\min_{\tilde{T}_j^2} (T_j - \alpha_1 \tilde{t} - \alpha_2 \tilde{T}_j^2)^2) \\ +\text{d)} \quad & \Leftrightarrow \min_{\tilde{t} \in \mathbb{Z}_q} \left( T_j - \alpha_1 \tilde{t} - \alpha_2 \left[ \frac{T_j - \alpha_1 \tilde{t}}{\alpha_2} \right]_{\mathbb{Z}_q} \right)^2 +\end{aligned} \tag{18}$$ + +INT4 parameter quantization operator is defined in Table 200. + +**Table 200 – INT4 parameter quantization operator definition** + +| Operator | Description | Parameters | Key words | Definition | Data type | Data format | +|-------------------------|-------------------------------------------------------------------------------------------------------|------------|------------------------|-------------------------------------|--------------|-------------| +| int4_quantize_parameter | Quantize the neural network weight from its high-precision representation (e.g., FP32) to INT4 tensor | Input | W | Input weight tensor to be quantized | tensor | float32 | +| | | | B (optional) | Bias tensor | tensor | NS | +| | | | sf_w_list | Quantization scale factor list | list | float32 | +| | | | offset_w | Weight quantization bias identifier | value | Boolean | +| | | Attribute | range_min | Lower bound of quantization value | value | INT4 | +| | | | range_max | Upper bound of quantization value | value | INT4 | +| | | Output | W_quantized | Quantized weight tensor | list[tensor] | INT4 | +| | | | B_quantized (optional) | Quantized bias tensor | tensor | NS | + +The pseudocode of INT4 parameter quantization is in Table 201. + +**Table 201 – The pseudocode of INT4 parameter quantization** + +| INT4 parameter quantization | Descriptor | +|------------------------------------------------------------------------------|------------| +| def int4_quantize_parameter(W,sf_w_list,offset_w,bit_w=4, B): { | | +| if offset_w==True: { | | +| range_max=2**bit_w-1 | | +| range_min=0 | | +| W-=min(W) | | +| } | | +| else: { | | +| range_max=2**(bit_w-1)-1 | | +| range_min= -2**(bit_w-1) | | +| } | | +| if len(W.shape)>2: { | | +| W_quantized=[] | | +| for k in range(len(W)): { | | +| if len(sf_w_list[k])==1: | | +| W_quantized[k]=clip(round(W[k]/sf_w_list[k][0]),range_min,range_max) | | +| if len(sf_w_list[k])==2: | | +| W_quantized[k]=get_params_double_grid(W[k],sf_w_list[k],range_min,range_max) | | +| } | | +| } | | +| else: { | | +| if(len(sf_w_list)==1: | | +| W_quantized=clip(round(W/sf_w_list[0]),range_min,range_max) | | +| if len(sf_w_list)==2: | | +| W_quantized=get_params_double_grid(W,sf_w_list,range_min,range_max) | | +| } | | +| B_quantized=B | | +| return W_quantized, B_quantized | | +| } | | + +The pseudocode of the operator get\_params\_double\_grid() is in Table 202. + +**Table 202 – The pseudocode of get\_params\_double\_grid() operator** + +| get_params_double_grid() operator | Descriptor | +|-----------------------------------------------------------------|------------| +| def get_params_double_grid(T,sf,range_min,range_max,bit_w=4): { | | +| w1=linspace(range_min,range_max,2**bit_w).reshape(1,-1) | | +| w1=tile(w1,[len(T),1]) | | +| w2=round((T-(w1*sf[0]))/sf[1]) | | +| w2=clip(w2,range_min,range_max) | | +| t_temp=(T-w1*sf[0]-w2*sf[1])**2 | | +| index=argmin(t_temp,1) | | + +**Table 202 – The pseudocode of get\_params\_double\_grid() operator** + +| get_params_double_grid() operator | Descriptor | +|------------------------------------------|-------------------| +| range_vec=range(w1.shape[0]) | | +| T1=w1[range_vec, index].reshape(T.shape) | | +| T2=w2[range_vec, index].reshape(T.shape) | | +| Tq=[T1,T2] | | +| return Tq | | +| } | | + +###### **9.2.1.2.2.3 Parameter quantization for bounded ReLU** + +Linear quantization is recommended for parameter quantization for bounded ReLU. When parameter quantization for bounded ReLU is called, it invokes linear quantization in the basic quantization clause. The definition is described in Table 203. + +**Table 203 – Definition of parameter quantization for bounded ReLU** + +| Operator | Description | Key group | Keyword | Definition | Data type | Data format | +|----------------------------------------------------------------------|------------------------------------------------------------------------------|------------------|----------------|----------------------------------|------------------|--------------------| +| brelu_quantizer_parameter
Parameter Quantization for Bounded ReLU | Map continuous high-precision parameters into low-precision range layer-wise | Input | X | Weight tensor | List [tensor] | NS | +| | | | X_scale | Weight quantization scale | List [tensor] | NS | +| | | | X_offset | Weight quantization offset | List [tensor] | NS | +| | | | N | Number of quantization bits | Value | int | +| | | Output | granularity | Quantization granularity | Value | int | +| | | | Y | Weight tensor after quantization | List [tensor] | NS | + +The pseudocode of parameter quantization for bounded ReLU is given in Table 204. + +**Table 204 – The pseudocode of parameter quantization for bounded ReLU** + +| Parameter Quantization for Bounded ReLU | Descriptor | +|----------------------------------------------------------------------|-------------------| +| def brelu_quantizer_parameter (X, X_scale, X_offset, granularity,N): | | +| Return Linear_quantization (X, X_scale, X_offset, granularity,N) | | + +##### **9.2.1.2.3 Activate quantization operator** + +###### **9.2.1.2.3.1 Trainable alpha quantization** + +Quantization shall handle outliers beyond the low-bit representation space. The gradient-based trainable truncation values can make the quantization truncation values more generalized, stable and adaptive. + +The specific definition of training truncation value quantization can be found in Table 205. + +**Table 205 – Definition of trainable alpha quantization** + +| Operator | Description | Field | Keyword | Definition | Data type | Data format | +|------------------------------|---------------------------------|--------|-------------|--------------------------------------|---------------|-------------| +| Trainable alpha quantization | The alpha are trained by layers | Input | X | Activation tensor to be quantified | List [tensor] | NS | +| | | | Alpha | Trainable truncation upper bound | List [tensor] | NS | +| | | | Granularity | Quantization granularity | Value | int | +| | | | N | Bitwidth | Value | int | +| | | Output | X_q | Activation tensor after quantization | List [tensor] | NS | + +X is the input tensor with high precision, Alpha is the trainable truncated quantization parameter and X\_q is the quantized tensor. The pseudocode of the algorithm is shown in Table 206. + +**Table 206 – The pseudocode of trainable alpha quantization** + +| Trainable alpha quantization | Descriptor | +|-----------------------------------------------------------------|------------| +| def Trainable_alpha_quantization(X, Alpha, Granularity ,N):{ | | +| X_Scale = 2**(n-1)/Alpha | | +| X_offset = 0 | | +| X = max(min(X, Alpha) , -Alpha) | | +| X_q = Linear_quantization (X, X_scale, X_offset, Granularity,N) | | +| return X_q | | +| } | | + +###### 9.2.1.2.3.2 INT4 activation quantization + +In the INT4 quantization operation, a high-precision tensor is quantized into an INT4 tensor or a combination of several INT4 tensors. In general, dual INT4 quantization is used when multiple INT4 tensors are quantized. The parameter quantization mode corresponding to INT4 activation quantization is INT4 parameter quantization. The activation of the neural network is quantized by INT4 weight layer by layer: + +- a) Read the activation tensor X, which is to be quantized, and the list of quantization scale factors layer\_act\_List. Determine the quantitative value range according to the activation quantization offset list offset\_act\_List. +- b) Carry out linear quantization successively according to the elements in the quantization scale factor list until the end of the cycle: + - 1) If the length of the scale factor list is 1, quantize the activation tensor directly. + - 2) If the length of the scale factor list is greater than 1, quantize the residual tensor of the activation tensor for the second time. + +The data definition of INT4 activation quantization is shown in Table 207. + +**Table 207 – Int4 activation quantization data definition** + +| Operator | Description | Field | Keyword | Definition | Data type | Data format | +|--------------------------|--------------------------------------------------------------------------------------------------------|-----------|-----------------|-----------------------------------------|--------------|-------------| +| int4_quantize_activation | The neural network activation of high precision representation is quantized into tensor of int4 format | Input | X | Input activation tensor to be quantized | tensor | float32 | +| | | | sf_act_list | List of quantitative scaling factors | list | float32 | +| | | | offset_act_list | Activate quantization bias list | list | float32 | +| | | Attribute | range_min | Lower bound of quantization range | value | INT4 | +| | | | range_max | Upper bound of quantization range | value | INT4 | +| | | Output | X_quantized | Activation tensor after quantization | list[tensor] | INT4 | + +The pseudocode for int4 activation quantization are shown in Table 208. + +**Table 208 – The pseudocode for INT4 activation quantization** + +| INT4 Activate Quantization | Descriptor | +|--------------------------------------------------------------------------------------------|------------| +| def int4_quantize_activation(X,sf_act_list,offset_act_list=None,bit_w=4,X_quantized=[]): { | | +| if offset_act_list is not None: { | | +| range_max=2**bit_w-1 | | +| range_min=0 | | +| x_quantized=clip(round((X-offset_act_list[0])/sf_act_list[0]),range_min,range_max) | | +| out=sf_act_list[0]*x_quantized+offset_act_list[0] | | +| } | | +| else: {} | | +| range_max=2**(bit_w-1)-1 | | +| range_min= -2**(bit_w-1) | | +| x_quantized=clip(round(X/sf_act_list[0]), range_min, range_max) | | +| out=sf_act_list[0]*x_quantized | | +| } | | +| X_quantized.append(x_quantized) | | +| if len(sf_act_list)>1: { | | +| X=X-out | | +| if offset_act_list[1:] is not None: | | + +**Table 208 – The pseudocode for INT4 activation quantization** + +| INT4 Activate Quantization | Descriptor | +|--------------------------------------------------------------------------------------------------|-------------------| +| X_quantized=int4_quantize_activation(X,sf_act_list[1:],offset_act_list[1:],bit_w=4, X_quantized) | | +| else: | | +| X_quantized=int4_quantize_activation(X,sf_act_list[1:],None,bit_w=4, X_quantized) | | +| } | | +| return X_quantized | | +| } | | + +###### **9.2.1.2.3.3 Activation quantization for bounded ReLU** + +The definition of activation quantization for bounded ReLU is given in Table 209. + +**Table 209 – Definition of activation quantization for bounded ReLU** + +| Operator | Description | Key group | Keyword | Definition | Data type | Data format | +|-------------------------------|-----------------------------------------|------------------|----------------|-------------------------------|------------------|--------------------| +| brelu_quantization_activation | Quantize activations using bounded ReLU | Input | X | Input tensor | Tensor | NS | +| | | | N_sigma | Set clip percentage | value | float | +| | | | Scale_X | Scale of X | value | float | +| | | | Stage | Quantization stage | value | int | +| | | Attribute | Bound_f | Bound before quantization | value | float | +| | | | n = 0 | Number of samples | value | int | +| | | | Bound_i | Bound after quantization | value | int | +| | | | Mul | Multiplier | value | int | +| | | | Shift | Bit shifts | value | int | +| | | Output | Y | Activation after quantization | List | NS | +| | | | Y_scale | Scale of Y | value | float | + +Definition of Stage is given in Table 210. + +**Table 210 – Definition of stage** + +| Value | Definition | +|--------------|--------------------------------| +| 1 | Before quantization, no clip | +| 2 | Before quantization, with clip | +| 3 | Quantized | + +Algorithm description: + +During Stage 1, compute bound\_f and n. + +This method requires scale factor of input activation tensor X (scale\_X) and offset (offset\_X). + +Given input tensor X, sort all values within X in ascending order and record the number at a certain percentage. The recorded number is noted as $b_f$ . The percentage of decided by N of N-sigma range is given in Table 211. + +**Table 211 – Definition of N\_sigma** + +| Percentage | N_sigma | +|------------|---------| +| 99.865% | 3 | +| 99.921% | 3.2 | +| 99.977% | 3.5 | + +Multiple training iterations results in multiple X tensors as well as multiple $b_f$ 's. Multiple $b_f$ 's are averaged. $b_f$ is a high precision number. + +Quantization of $b_f$ : according to scaling factor and offset of X (scale\_X and offset\_X), $b_f$ is quantized to get $b_i$ . $b_i$ is of same precision with $X_i$ . Then, $b_i$ is de-quantized to modify corresponding $b_f$ , as given in equations (19) and (20). + +$$b_i = \text{round}(b_f * X_{\text{scale}} + X_{\text{offset}}) \quad (19)$$ + +$$b_f = (b_i - X_{\text{offset}})/X_{\text{scale}} \quad (20)$$ + +When Stage is equal to 1, $b_f$ does not clip activations but count itself. + +When Stage is equal to 2, high precision tensor is clipped by $b_f$ as in equation (21). + +$$Y_f = \text{clip}(X, \max = b_f) \quad (21)$$ + +When Stage is equal to 3, low precision tensor is clipped by $b_i$ as in equation (22) and then quantized. + +$$Y_i = \text{clip}(X, \max = b_i) \quad (22)$$ + +The initial value of N is 3. If validation set is available, multiple Ns can be tried and selected a better N. + +When $b_i$ is available, $Y_i$ can be quantized. Linear quantization in basic quantization methods is invoked to quantize $Y_i$ . The pseudocode of Activation Quantization for Bounded ReLU is given in Table 212. + +**Table 212 – The pseudocode of activation quantization for bounded ReLU** + +| BReLU | Descriptor | +|-------------------------------------------------|------------| +| def BReLU(X,N_sigma=3,Stage,Scale){ | | +| if stage==1:{ | | +| Bound=n_sigma_of(X,N_sigma) | | +| self.Bound_f+=(self.Bound_f*self.n+Bound)/(n+1) | | +| Self.n+=1 | | +| } | | +| if self.n==50:{ | | +| self.Bound_i=round(self.Bound_f*X_scale/self.n) | | +| self.mul,self.shift=mul_shift(self.bound_i) | | +| Y_scale=X_scale*mul/(2**shift) | | +| self.Bound_f=127/Y_scale | | +| } | | +| } | | + +**Table 212 – The pseudocode of activation quantization for bounded ReLU** + +| BReLU | Descriptor | +|------------------------------------------|-------------------| +| } | | +| return X, Y_scale | | +| elif stage==2: | | +| return clip(X,self.Bound_f), 1 | | +| else: { | | +| X = clip(X,self.Bound_i) | | +| return ((X+2***(shift-1))*mul)>>shift, 1 | | +| } | | +| | | +| def n_sigma_of(x,n_sigma):{ | | +| if n_sigma==3: | | +| rate=0.99865 | | +| elif n_sigma==3.5: | | +| rate=0.99977 | | +| else: | | +| rate=0.99865 | | +| sorted_x_ =torch.sort(x.view(-1)) | | +| index=round(sorted_x_.size()[0]*rate) | | +| bound=sorted_x_[index].item() | | +| return bound | | +| } | | +| | | +| def mul_shift(bi) { | | +| for i in range(1,28): | | +| max_int=(2^(n-1)+0.5)*2**i | | +| if max_int>bi: { | | +| mul=round(max_int/bi) | | +| temp=max_u*mul/2**i | | +| if(abs(temp-2^(n-1))<0.5): { | | +| shift=i | | +| return mul,shift | | +| } | | +| } | | +| return mul,shift | | +| } | | + +###### 9.2.1.2.3.4 Ratio synchronization quantization + +Definition of ratio synchronization quantization is given in Table 213. + +**Table 213 – Definition of ratio synchronization quantization** + +| Operator | Description | Key group | Keyword | Definition | Data type | Data format | +|---------------------------------------------------------------|--------------------------------------------------------------------------------------------------------------------------|-----------|-------------|--------------------------|---------------|-------------| +| Scale_syncronization
Ratio synchronization
quantization | Input multiple tensors with different scales, output multiple tensors with the same scale, compute the rescaling factor. | Input | X | Input tensors | List [tensor] | int | +| | | | X_scales | Scale of X | List [tensor] | float | +| | | | Stage | Quantization stage | Value | int | +| | | | Method | Quantization algorithm | Value | int | +| | | | Granularity | Quantization granularity | Value | int | +| | | Attribute | params | Quantization parameter | List [tensor] | NS | +| | | Output | Y | Output tensors | List [tensor] | NS | + +The parameter "Method" only defines 1, which is the multiply and right-shift quantization. The parameter "Granularity" only defines layer-wise quantization. + +This method is used for multiple input tensors when they are going to be added, subtracted or concatenated. This method only works on multiple input tensors. If only one tensor is inputted, this method is skipped. + +During Stage 1, the quantization parameter is calculated. At this stage, scaling factors of multiple input tensors shall be provided, based on the calculated attribute. + +Denote n input tensors as $Y_1, Y_2, \dots, Y_n$ in the input list of tensors X, and their scaling factors as $Y_{scale1}, Y_{scale2}, \dots, Y_{scaletn}$ , and the scaling factors of the weights of their last convolution. + +$Y_0$ is selected from one of the input tensors (first tensor for default, otherwise designated). Its scaling factor is $Y_{scale0}$ . + +For each other tensor $Y_j$ ( $0 < j < n$ ), + +Calculate its scale difference with $Y_0$ : + +$$scale_{diff} = Y_{scale0} / Y_{scalej} \quad (23)$$ + +Calculate multiplier mul and shift for right shift: + +$$mul, shift = mul\_shift\_f(scale_{diff}) \quad (24)$$ + +mul\_shift\_f function is defined in Table 205. + +For each input tensors $Y_j$ ( $0 \leq j < n$ ) record their mul and shift: ((1,0) for $Y_0$ ) + +When the value of Stage is equal to 1, calculate new $Y_{scalej}$ , noted as $Y_{scalej}'$ : + +$$Y_{scalej}' = Y_{scale0} * 2^{shift} / mul \quad (25)$$ + +And update $W_{scalej}$ : + +$$W_{scalej}' = Y_{scalej}' * W_{scalej} / Y_{scalej} \quad (26)$$ + +And $Y_{scalej}$ : + +$$Y_{scalej} = Y_{scalej'} \quad (27)$$ + +During high-precision training stage these are not needed. + +During low-precision inference stage, quantize all tensors except $Y_0$ . + +$$Y_{qj} = (Y_j * mul + 2^{shift-1}) \gg shift \quad (28)$$ + +The pseudocode is given in Table 214. + +**Table 214 – The pseudocode for ratio synchronization** + +| Scale_synchronization | Descriptor | +|----------------------------------------------------------------|------------| +| Def Scale_synchronization (X,scale,stage,method,granularity) { | | +| if stage==1: { | | +| Scale0,pos=scale[0] | | +| for i in range(len(scale)-1): { | | +| Mul,shift=mul_shift_f(scale0/scale[i+1]) | | +| Self.params.append([mul,shift]) | | +| } | | +| return X | | +| } | | +| elif stage==2: { | | +| Return X | | +| } | | +| else: { | | +| For i in range(len(X)-1): { | | +| Mul,shift=self.params[i+1] | | +| X[i+1]=(X[i+1]*mul+2^(shift-1))>>shift | | +| } | | +| Return X | | +| } | | +| } | | +| def mul_shift_f(scale,precision=0.02): { | | +| for i in range(0,28): { | | +| max_int=2**i | | +| if max_int>scale: { | | +| temp=max_int/ratio | | +| mul=round(temp) | | +| if(abs(max_int/mul-scale)Pruning | Descriptor | +|-------------------------------------------------------------------|-------------------| +| b.reshape(w_shape[dim],-1); | | +| If(granularity==2){ | | +| for(j=0;j 1e-5): { | | +| for i in range(groups): { | | +| sum_ = zeros(shape(weight)) | | +| for j in range(groups): { | | +| sum_ = sum_ + W2[j] ×_1 W1[j] ×_2 W3[j] | × n : n-mode of tensor times scikit-tensor==0.1 | +| } | | +| res = W - sum_ + W2[i] ×_1 W1[i] ×_2 W3[i] | × n : n-mode of tensor times scikit-tensor==0.1 | +| W1[i], W2[i], W3[i] = HOSVD(res, r1, r2) | | +| } | | +| } | | + +**Table 224 – The pseudocode for the convolutional layer low-rank sparse decomposition compression operator** + +| Convolutional layer low-rank sparse decomposition compression | Descriptor | +|---------------------------------------------------------------|------------| +| } | | +| d_error = norm(res)/norm(W) | | +| } | | +| w1 = concat(W1, axis = 0) | | +| w2 = concat(W2, axis = 1) | | +| w3 = concat(W3, axis = 1) | | +| return w1, w2, w3 | | +| } | | + +The pseudocode of high-dimensional SVD decomposition is in Table 225. + +**Table 225 – The pseudocode of high-dimensional SVD decomposition** + +| High-dimensional SVD decomposition | Descriptor | +|---------------------------------------------------------------------|-----------------------------------------------------------| +| def HOSVD(weight, r1, r2): { | | +| U = [] | | +| for order_idx in range(3): { | | +| U[order_idx], _, _ = svd(unfold (weight, order_idx)) | svd: numpy==1.18.1; unfold:
scikit-tenor==0.1 | +| } | | +| Core = weight $\times_1$ transpose(U[0]) $\times_2$ transpose(U[1]) | $\times_n$ : n-mode of tensor times
scikit-tensor==0.1 | +| $\times_3$ transpose(U[2]) | | +| return U[0], Core $\times_3$ U[2], U[1] | | +| } | | + +###### 9.2.1.4.2.3 Compression operator of fully connected or $1 \times 1$ convolutional layer in low-rank sparse decomposed structured matrix + +The process of the decomposition compression for a fully connected layer or a $1 \times 1$ convolutional layer is as follows: + +- Determine the size (core\_size) of the output of the middle layers by calculating the compression rate (ref. clause 9.4.2.1), and load the weight of the current layer. +- Decompose the weight matrix via SVD and return approximated tensors w1 and w2. + +The compression operator for the fully connected and $1 \times 1$ convolutional layer in low-rank sparse decomposed structured matrix is defined in Table 226. + +**Table 226 – Definition of the operator for the fully connected layer compression in low-rank sparse decomposed structured matrix** + +| Operator | Description | Parameters | Key words | Definitions | Data type | Data format | +|-------------|-------------------------------------------------------------------------------------------|------------|-----------|---------------------------------------------------|-----------|-------------| +| SVD_process | Compress the weight matrix of the current layer by applying low-rank sparse decomposition | Input | weight | Input tensor of weights to be compressed | matrix | NS | +| | | | core_size | Size of the output corresponding to middle layers | value | UINT | +| | | Output | w1 | The first weight tensor out of compression | matrix | NS | +| | | | w2 | The second weight tensor out of compression | matrix | NS | + +The pseudocode for the fully connected layer compression in low-rank sparse decomposed structured matrix is shown in Table 227. + +**Table 227 – The pseudocode for the fully connected layer compression in low-rank sparse decomposed structured matrix** + +| fully connected layer compression in low-rank sparse decomposed structured matrix | descriptor | +|-----------------------------------------------------------------------------------|-----------------------| +| def SVD_process(weight, core_size): { | | +| u, sigma, v = svd(weight) | svd:
numpy==1.18.1 | +| w2 = mat_mul(u[:, :core_size], sigma[:core_size]) | | +| w1 = v[:core_size, :] | | +| return w1, w2 | | +| } | | + +##### 9.2.1.4.3 Re-parameterization + +###### 9.2.1.4.3.1 Definition + +Re-parameterization is defined as the construction of a unique neural network module. During training, this module has a great capacity for learning, but during testing, it can be equated to extremely basic neural network operations. The process of converting complex neural network structures into simple operations is called reparameterization. + +**Table 228 – Definition of re-parameterization** + +| Operator | Description | Field | Keyword | Definition | Data Format | +|---------------------|--------------------------------|--------|---------|----------------------------|-------------| +| Re-parameterization | Re-parameterize a given module | Input | M | The original module | NS | +| | | | Method | Re-parameterization method | String | +| | | Output | Y | Re-parameterized module | NS | + +The re-parameterization process is defined as follows: + +- 1) Train the network normally. +- 2) Re-parameterize the modules in the network. +- 3) Test the Re-parameterized network. + +**Table 229 – The pseudocode for the re-parameterization process** + +| Step-by-step reparameterization according to the methods | descriptor | +|----------------------------------------------------------|------------| +| def reparameterization(M, Method): { | | +| Y=M | | +| for m in Method: { | | +| Y = m(Y) | | +| } | | +| return Y | | +| } | | + +###### 9.2.1.4.3.2 Conv-BN reparameterization + +A $3 \times 3$ convolution layer followed by a batchnorm layer can be combined into a single convolution layer. The initial two operators combine to form one operator after the combination, resulting in network compression. The full combination process is defined as follows. + +Suppose the input feature is $x$ , the weight and bias of the convolution layer are $w$ and $b$ respectively. The parameters in the batchnorm layer are expected mean ( $\mu$ ), expected variance ( $\sigma$ ), scale factor ( $\gamma$ ) and bias factor ( $\beta$ ). The convolution layer can be formalized as: + +$$x2 = conv(x) = w \otimes x + b,$$ + +where $\otimes$ represent the convolution operation. The batchnorm layer can be formalized as: + +$$x3 = bn(x2) = \frac{x2 - \mu}{\sqrt{\sigma}} \gamma + \beta.$$ + +A convolution layer followed by a batchnorm layer can be expressed formally as: + +$$\begin{aligned} x3 &= bn(conv(x)) = \frac{w \otimes x + b - \mu}{\sqrt{\sigma}} \gamma + \beta \\ &= \frac{w\gamma}{\sqrt{\sigma}} \otimes x + \frac{\gamma}{\sqrt{\sigma}} (b - \mu) + \beta. \end{aligned}$$ + +$\frac{w\gamma}{\sqrt{\sigma}}$ is defined as the weight in a new convolution layer, and $\frac{\gamma}{\sqrt{\sigma}} (b - \mu) + \beta$ is defined as the bias, which means: + +$$w2 = \frac{w\gamma}{\sqrt{\sigma}},$$ + +$$b2 = \frac{\gamma}{\sqrt{\sigma}} (b - \mu) + \beta.$$ + +Thus, the convolution layer followed by a batchnorm layer can be represented as: + +$$x3 = w2 \otimes x + b2,$$ + +which is just one convolution layer. In this way, the amount of computation is significantly decreased because the convolution layer and batchnorm layer are merged into one convolution layer. + +###### 9.2.1.4.3.3 Conv-Conv reparameterization + +###### 1 $3 \times 3$ and $1 \times 1$ Conv + +A $3 \times 3$ convolution layer and a $1 \times 1$ convolution layer can be combined into a single convolution layer. The weight and bias of the $3 \times 3$ convolution layer are defined as $w1$ and $b1$ respectively, while the weight and bias of the $1 \times 1$ convolution layer are $w2$ and $b2$ respectively. There are two forms of combination here: + +$$x2 = conv^1(conv^3(x)),$$ + +and + +$$x2 = conv^3(x) + conv^1(x).$$ + +The first form can be formalized as + +$$\begin{aligned} x2 &= conv^1(conv^3(x)) \\ &= w2 \otimes (w1 \otimes x + b1) + b2 \\ &= (w2 @ w1) \otimes x + w2 \otimes b1 + b2, \end{aligned}$$ + +where $@$ means the combination of two weights of the convolution layer. Here, because of the uniqueness of $1 \times 1$ convolution, $w2$ has the shape $c1 \times c2 \times 1 \times 1$ and $w1$ has the shape $c1 \times c2 \times 3 \times 3$ . The shape of the combined result is still $c1 \times c2 \times 3 \times 3$ . + +$w2 @ w1$ is defined as the weight in a new convolution layer, and $w2 \otimes b1 + b2$ is defined as the bias, which means: + +$$\begin{aligned} w3 &= w2 @ w1, \\ b3 &= w2 \otimes b1 + b2. \end{aligned}$$ + +Two convolutional layers can be combined into one: + +$$x2 = w3 \otimes x + b3.$$ + +The second form can be formalized as: + +$$\begin{aligned} x2 &= conv^3(x) + conv^1(x) \\ &= w1 \otimes x + b1 + w2 \otimes x + b2 \\ &= (w1 \oplus w2) \otimes x + (b1 + b2), \end{aligned}$$ + +where $\oplus$ means combine $w1$ and $w2$ with an operation similar to addition. Since $w1$ and $w2$ have different shapes, $w2$ is extended to match $w1$ 's shape before the element-wise addition. $w1 \oplus w2$ is defined as the weight, and $(b1 + b2)$ is defined as the bias, which means: + +$$\begin{aligned} w3 &= w1 \oplus w2, \\ b3 &= b1 + b2. \end{aligned}$$ + +Two convolutional layers can be combined into one: + +$$x2 = w3 \otimes x + b3.$$ + +###### 2 $3 \times 3$ and $3 \times 3$ Conv + +Two $3 \times 3$ convolution layers can also be combined into one $3 \times 3$ convolution layer. Only two parallel $3 \times 3$ convolutions are applicable here. More specifically, this procedure is formalized as: + +$$\begin{aligned} x2 &= conv^3(x) + conv^3(x) \\ &= w1 \otimes x + b1 + w2 \otimes x + b2 \\ &= (w1 + w2) \otimes x + (b1 + b2). \end{aligned}$$ + +Because the shape of $w1$ and $w2$ is the same, the addition operation can be performed directly. + +$w1 + w2$ is defined as the weight, and $(b1 + b2)$ is defined as the bias, which means + +$$w3 = w1 + w2,$$ + +$$b3 = b1 + b2.$$ + +Two convolutional layers can be combined into one: + +$$x2 = w3 \otimes x + b3.$$ + +###### 9.2.1.4.3.4 Conv-pooling reparameterization + +The parallel $3 \times 3$ convolution layer and $3 \times 3$ average pooling layer can be combined into a single convolution layer. The $3 \times 3$ average pooling layer can be viewed as a special $3 \times 3$ convolution layer, so the process is similar to merging two $3 \times 3$ convolutions. In addition, a $3 \times 3$ convolution layer with residual connection can also be merged into one $3 \times 3$ convolution layer. + +###### 9.2.1.4.3.5 Multiple-layer reparameterization + +The combination of two layers is defined above. Multiple-layer reparameterization is defined as merging every two layers into a $3 \times 3$ convolution layer step by step. In this way, special modules can be designed, which can use residual connection, inception structure, etc. during training, and can be re-parameterized into one convolution layer during testing. The calculation of the model is greatly reduced. More importantly, this compression process will not affect the accuracy of the network, because the re-parameterization process is an equivalent operation. The modules before and after the re-parameterization are completely equivalent during the testing. In addition, the network structure upon re-parameterization is typically quite simple, without skip connections, etc. which greatly increases running efficiency. + +### 9.2.2 Decompression process + +#### 9.2.2.1 Multiple models + +##### 9.2.2.1.1 Decompression operator for weights of multiple model layers + +###### 9.2.2.1.1.1 Decompression for weights of multiple model layers + +The decompression of the compressed shared weights in multiple models includes 3 operators: + +- 1) Decompress multiple models: For each of the $N$ tasks, this is to obtain shared weights and their corresponding specific weight values. +- 2) Decompress a specific model: This is to obtain shared weight values and weight values in a specific model. It does not need to obtain the specific weight values of other specific models. +- 3) Switch from one kind of specific model to another kind of specific model: This is to obtain the specific weight values corresponding to targeted tasks, rather than re-obtain shared weight values. + +###### 9.2.2.1.1.2 Decompression output multiple models + +This operator refers to generating a weight tensor list containing $M$ $N$ -dimensional specific weight tensors with identical structure from an $N$ -dimensional shared weight tensor and a list containing $M$ $N$ -dimensional specific weight tensors. This is performed by the following steps: + +- 1) If the weight aggregation dimension axis points to the number of output feature layers, check whether the number of the tensors in $W\_specific$ is equal to the number of tensors in $B\_specific$ . If not, set the error indicator to 1, other outputs with NULL, and complete this operation. + +- 2) Check whether the shapes of the tensors in $W\_specific$ and $B\_specific$ are equal, respectively. If they are not equal, set the error indicator to 2, other outputs with NULL, and complete this operation. +- 3) Obtain the shared weights and specific weights from models that are not compressed. Obtain the shared bias and specific bias from models that are not compressed. +- 4) In the aggregation dimension axis, aggregate the shared weight tensor $W\_share$ with each specific weight tensor in list $W\_specific$ . The results are added into the output weight tensor list $Weight\_list$ . In the aggregation dimension axis, aggregate the shared bias vector $B\_share$ with each specific bias vector in list $B\_specific$ . The results are added into the output bias vector list $Bias\_list$ . Output, as the aggregation point index $slice\_indicator$ , the number of weight tensors $W\_share$ that in the dimension axis. + +The operator of decompression and output multiple models is defined in Table 230. + +**Table 230 – Decompression and output multiple models** + +| Operator | Description | Parameters | Key words | Definition | Data type | Data format | +|------------------------------|-------------------------------------------------------------------------------------------------------------------------------|------------|--------------------|-----------------------------------|-----------------|-------------| +| Decompress_Multilayer_weight | Decompress models into M N-dimensional weight tensors with identical structure, based on given dimension and number of models | Input | $W\_share$ | Input shared weight tensor | tensor | NS | +| | | | $B\_share$ | Input shared bias vector | vector | NS | +| | | | $W\_specific$ | Input specific weight tensor list | List of tensors | NS | +| | | | $B\_specific$ | Input specific bias vector list | List of vectors | NS | +| | | Attributes | $axe$ | Weight aggregation dimension | value | int | +| | | Output | $Weights\_list$ | Output weight tensor list | List of tensors | NS | +| | | | $Bias\_list$ | Output bias vector list | List of vectors | NS | +| | | | $slice\_indicator$ | Aggregation point index | value | int | + +The pseudocode of decompression and output multiple models is in Table 231. + +**Table 231 – The pseudocode of decompression and output multiple models** + +| Decompression and output multiple models | Descriptor | +|-------------------------------------------------------------------------------------------|------------| +| Decompress_Multilayer_weight( $W\_share$ , $B\_share$ , $W\_specific$ , $B\_specific$ ) { | | +| $Weights\_list = []$ | | +| $Bias\_list = []$ | | +| $slice\_indicator = 0$ | | +| $Error = 0$ | | +| if ( $axe = dimension\_of\_out\_feature\_map$ ) { | | + +**Table 231 – The pseudocode of decompression and output multiple models** + +| Decompression and output multiple models | Descriptor | +|------------------------------------------------------------------|-------------------| +| if(lens(W_specific)! = lens(B_specific)), Error = 1 | | +| } | | +| if(Error ==0) { | | +| for (i = 0; i < lens(W_specific)-1; i++){ | | +| if (W_specific [i].shape != W_specific [i + 1].shape), Error = 2 | | +| } | | +| } | | +| if(Error ==0) { | | +| for (i = 0; i < lens(W_specific); i++) { | | +| Weights_list.append(Concat(W_share, W_specific[i], axe)) | | +| Bias_list.append(Concat(B_share, B_specific[i], axe)) | | +| } | | +| slice_indicator = W_share.shape(axe) | | +| } | | +| return Weights_list, Bias_list, slice_indicator, Error | | +| } | | + +###### 9.2.2.1.1.3 Decompression output specific model + +This operator generates a specific decompressed neural network model corresponding to a given model index, from an N-dimensional shared weight tensor and a list of M N-dimensional specific weight tensors. It can be performed by the following steps: + +- 1) Determine the model indicator Unit\_indicator based on the task index. +- 2) Obtain shared weight and the corresponding specific weights: This is to the obtain specific weight tensor W from the list W\_specific, and obtain the specific bias vector B from the specific bias vector list B\_specific. +- 3) In the dimension axis, aggregate W\_share and W, then outputs it to Weights. +- 4) In the dimension axis, aggregate B\_share and B, then outputs it to Bias. + +The operator for decompression and output specific model is defined in Table 232. + +**Table 232 – Decompression and output specific model operator** + +| Operator | Description | Parameters | Key words | Definition | Data type | Data format | +|------------------------|---------------------------------------------------------------------------|-------------------|------------------|-----------------------------------|------------------|--------------------| +| indicated_layer_weight | Decompress a specific model, based on given dimension and model indicator | Input | W_share | Input shared weight tensor | tensor | NS | +| | | | B_share | Input shared bias vector | vector | NS | +| | | | W_specific | Input specific weight tensor list | List of tensors | NS | + +**Table 232 – Decompression and output specific model operator** + +| Operator | Description | Parameters | Key words | Definition | Data type | Data format | +|----------|-------------|------------|----------------|---------------------------------|-----------------|-------------| +| | | | B_specific | Input specific bias vector list | List of vectors | NS | +| | | Attributes | axe | Weight aggregation dimension | value | int | +| | | | Unit_indicator | Model indicator | value | int | +| | | Output | Weights | Output weight tensor | tensors | NS | +| | | | Bias | Output bias vector | vectors | NS | + +The pseudocode of decompression and output specific model is in Table 233. + +**Table 233 – The pseudocode of decompression and output specific model** + +| Decompression and output specific model | Descriptor | +|----------------------------------------------------------------------------------------------|------------| +| Get indicated_layer_weight (W_share, B_share, W_specific, B_specific, Unit_indicator, axe) { | | +| Weights = Concat(W_share, W_specific[Unit_indicator], axe) | | +| Bias = Concat(B_share, B_specific[Unit_indicator], axe) | | +| return Weights, Bias | | +| } | | + +###### 9.2.2.1.1.4 Decompression output switched specific models + +Switching refers to outputting a decompressed specific model 2 from a decompressed specific model 1. Such models are the decompression result out of compressed models via weight sharing compression approach. This kind of compressed model contains M layers (M is a positive integer). The i-th (i is an integer, $1 \leq i \leq M$ ) layer in M contains shared weights and N groups of specific weights. The shared weight is used to execute each task out of N. Each of the N groups is used when executing its corresponding task in N. + +Since the weight is shared among N models that are not compressed, it does not need to update the i-th layer when performing a switch. The only task is to obtain the specific weight corresponding to a specific model, and replace the original weight with this specific weight. + +The operator can be performed by the following steps: + +- 1) Obtain the specific weight values from the model corresponding to the indicator: this is to fetch the corresponding weight tensor W\_specific\_idx out of the weight tensor list W\_specific, and fetch the corresponding specific bias vector B\_specific\_idx out of the bias vector list B\_specific; +- 2) Replace the specific weight in the indicated model: This is to replace the specific weights in the input weight tensor Weights with W\_specific\_idx, and replace the specific bias part in the input bias vector Bias with B\_specific\_idx. + +The operator obtaining the specific weight from an indicated model is defined in Table 234. + +**Table 234 – Obtaining specific weight from an indicated model operator** + +| Operator | Description | Parameters | Key words | Definition | Data type | Data format | +|-----------------------------------------|-------------------------------------------------|------------|----------------|-----------------------------------|-----------------|-------------| +| Get_indicated_layer_weight_specific_idx | obtains specific weight from an indicated model | Input | W_specific | Input specific weight tensor list | List of tensors | NS | +| | | | B_specific | Input specific bias vector list | List of vectors | NS | +| | | Attributes | Unit_indicator | Model indicator | value | int | +| | | Output | W_specific_idx | Output specific weight tensor | tensors | NS | +| | | | B_specific_idx | Output specific bias vector | vectors | NS | + +The pseudocode obtaining the specific weight from an indicated model is defined in Table 235. + +**Table 235 – The pseudocode obtaining the specific weight from an indicated model** + +| Obtaining the specific weight from an indicated model | Descriptor | +|------------------------------------------------------------------------------------|------------| +| Get_indicated_layer_weight_specific_idx (W_specific, B_specific, Unit_indicator) { | | +| W_specific_idx = W_specific[Unit_indicator] | | +| B_specific_idx = B_specific[Unit_indicator] | | +| return W_specific_idx, B_specific_idx | | +| } | | + +The operator replacing the specific weight in an indicated model is defined in Table 236. + +**Table 236 – Operator replacing the specific weight in an indicated model** + +| Operator | Description | Parameters | Key words | Definition | Data type | Data format | +|---------------------------------------------|---------------------------------------------------|------------|----------------|------------------------------|-----------|-------------| +| Replace_indicated_layer_weight_specific_idx | replace the specific weight in an indicated model | Input | Weights | Input weight tensor | tensors | NS | +| | | | Bias | Input bias tensor | vectors | NS | +| | | | W_specific_idx | Input specific weight tensor | tensors | NS | +| | | | B_specific_idx | Input specific bias vector | vectors | NS | +| | | Output | Weights | Output weight tensor | tensors | NS | +| | | | Bias | Output bias vector | vectors | NS | + +The pseudocode replacing the specific weight in an indicated model is defined in Table 237. + +**Table 237 – The pseudocode replacing the specific weight in an indicated model** + +| Replacing the specific weight in an indicated model | Descriptor | +|-----------------------------------------------------------------------------------------------|------------| +| Replace indicated_layer_weight_specific_idx (Weights, Bias, W_specific_idx, B_specific_idx) { | | +| Weights[len(W_specific_idx): -1] = W_specific_idx | | +| Bias[len(B_specific_idx): -1] = B_specific_idx | | +| return Weights, Bias | | +| } | | + +##### **9.2.2.1.2 The decompression of residual quantization for multiple models** + +###### **9.2.2.1.2.1 Definition of decompression** + +The decompression operator of residual quantization consists of the following parts: First, obtain the weight of the pre-trained model; then, obtain the residual weight of the incremental model at a specific position; finally, add the two weights for the target network model. + +###### **9.2.2.1.2.2 Decompression of the output target model** + +Decompressing the output target model means to obtain the weight of the target model incrementally based on the weight of the existing pre-training model. The process is as follows: First, decompress the weight of the existing pre-trained model; then, decompress the corresponding residual weight of the target model; finally, obtain the weight of the target model by combining the two models. + +The data definitions for the unzipped output target model are shown in Table 238. + +**Table 238 – Decompression of the output target model operator** + +| Operator | Description | Parameters | Key words | Definition | Data type | Data format | +|------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------|------------|----------------|-------------------------------|-----------------|-------------| +| Decompression of the output target model | 26/5000
The weight of the target model was recovered by combining the weight of the target model with the weight of the residual model | Input | W_share | Shared weight tensor | tensor | NS | +| | | | W_specific | Clustering results | List | NS | +| | | | Base_model | Weights of pre-trained model | Tensor | NS | +| | | | Weights_list | List of weight tensor | List of tensors | NS | +| | | Output | Weights_target | Weight tensor of target model | Tensor | NS | +| | | | Bias | Output bias tensor | vectors | NS | + +The pseudocode of decompression for residual quantization is defined in Table 239. + +**Table 239 – The pseudocode of decompression for residual quantization (based on pre-trained model)** + +| Decompression for residual quantization | Descriptor | +|--------------------------------------------------------------------------|------------| +| Decompress_Multilayer_weight(W_share, B_share, W_specific, B_specific) { | | +| Weights_list = [] | | +| If (len(W_specific)==len(Weights_list)) | | +| Error = 0 | | +| else | | + +**Table 239 – The pseudocode of decompression for residual quantization +(based on pre-trained model)** + +| Decompression for residual quantization | Descriptor | +|-----------------------------------------------------------|-------------------| +| Error = 1 | | +| if(Error == 0) { | | +| for (i = 0; i < lens(W_specific) - 1; i++){ | | +| Weight_target[i] = Base_model[i] + W_share[W_specific[i]] | | +| } | | +| } | | +| return Weights_target, Error | | +| } | | + +#### 9.2.2.2 Dequantization + +##### 9.2.2.2.1 Basic dequantization operator + +###### 9.2.2.2.1.1 Linear dequantization + +Linear dequantization restores the original 8-bit data to a full-precision tensor by using the scaling factor and the offset factor on the quantized data according to the input bits. The reconstructed values in the decoded tensor are integer multiples of the step size [ISO/IEC 15938-17]. The data definition of linear dequantization is shown in Table 240. + +**Table 240 – Definition of linear dequantization** + +| Operator | Description | Field | Keyword | Definition | Type | Format | +|-----------------------|-----------------------------------|--------------|----------------|---------------------------------------|---------------|---------------| +| Linear_dequantization | Restore high-precision parameters | Input | X | Weight tensor | List [tensor] | NS | +| | | | X_scale | Scaling factor | List [tensor] | NS | +| | | | X_offset | Bias factor | List [tensor] | NS | +| | | | Granularity | Quantized granularity | Value | int | +| | | | N | Quantization bit number | Value | int | +| | | Output | Y | Weight tensor after linear projection | List [tensor] | NS | + +The pseudocode of linear dequantization can be found in Table 241. + +**Table 241 – Pseudo-code of linear dequantization** + +| Linear Dequantization | Symbol | +|-------------------------------------------------------------------|---------------| +| def Linear_dequantization (X, X_scale, X_offset, granularity,N):{ | | +| if(granularity==1): { | | +| Y = (X - X_offset)/X_scale | | +| } | | +| elif (granularity==2): { | | + +**Table 241 – Pseudo-code of linear dequantization** + +| Linear Dequantization | Symbol | +|--------------------------------------------|--------| +| for (x in X): { | | +| $y = (x - x\_offset)/x\_scale$ | | +| } | | +| } | | +| else: { | | +| for(x in X): { | | +| for(i in [0: x.channels]): { | | +| $y[i] = (x[i] - x\_offset[i])/x\_scale[i]$ | | +| } | | +| } | | +| } | | +| return Y | | +| } | | + +###### 9.2.2.2.1.2 Codebook dequantization + +The codebook dequantization refers to the process of recovering the original discrete high-precision parameters based on the given codebook index and corresponding value. The reconstructed integer tensors are the values of codebook elements referred to by their index value, and the reconstructed tensors are derived by multiplying the reconstructed integer tensors with a step size [ISO/IEC 15938-17]. This method is suitable for the dequantization of interval shrinkage. For related data generation methods, please see clause 9. The codebook dequantization finds the corresponding high-precision parameter value according to the index value of the input index tensor matrix and assigns it to the same position in the output tensor. The data definition of codebook dequantization is shown in Table 242. + +**Table 242 – Definition of codebook dequantization** + +| Operator | Description | Field | Keyword | Definition | Type | Format | +|-------------------------|-----------------------------------|--------|-------------|---------------------------------------------|---------------|--------| +| Codebook_dequantization | Restore high-precision parameters | Input | X_i | Input Tensor | List [tensor] | NS | +| | | | X_center | Index value of the weights | List [tensor] | NS | +| | | | granularity | Granularity | Value | int | +| | | Output | Y | Corresponding output tensor in the Codebook | List [tensor] | NS | + +The pseudocode of codebook dequantization can be found in Table 243. + +**Table 243 – Pseudo-code of codebook dequantization** + +| Codebook Dequantization | Symbol | +|-----------------------------------------------------------|--------| +| def Codebook_dequantization(X_i, X_center, granularity):{ | | +| Y = [] | | +| if (granularity == 1) or (granularity == 2):{ | | +| for i in range(0, len(X_center)):{ | | +| Y[i] = X_center[X_i[i]] | | +| } | | +| } | | +| if (granularity == 3):{ | | +| for i in range(0, len(X_center)):{ | | +| for j in range(0, len(W[i])):{ | | +| Y[i][j] = X_center[i][j][X_i[i][j]] | | +| } | | +| } | | +| } | | +| return Y | | +| } | | + +##### **9.2.2.2.2 Parameter dequantization operator** + +###### **9.2.2.2.2.1 Non-linear function mapping dequantization** + +When the model is read for calculation, formula (29) is used to restore the full-precision weight, where the $S$ is the scaling factor, $Z$ is the offset factor, $Q$ is the weight value after quantization, and $f$ is the corresponding non-linear function: + +$$W' = f^{-1}[S(Q - Z)] \tag{29}$$ + +Here, + +- $Z$ denotes the Scaling factor, +- $S$ denotes the offset factor, +- $Q$ denotes the weight value after quantization, +- $f$ denotes the corresponding non-linear function, +- $W'$ denotes the recovered weight value after dequantization. + +Specifically, $f$ is the inverse function with respect to the label. The specific operators can be found in Table 244. + +**Table 244 – Definition of non-linear dequantization** + +| Operator | Description | Field | Keyword | Definition | Type | Format | +|-----------------------------------|-----------------------------------|-------|----------|----------------------------------|--------------|--------| +| Weight non Linear_dequant ization | Restore high-precision parameters | Input | W_q | Weight tensor after quantization | list[tensor] | NS | +| | | | W_scale | Scaling factor | list[tensor] | NS | +| | | | W_offset | Offset factor | list[tensor] | NS | + +**Table 244 – Definition of non-linear dequantization** + +| Operator | Description | Field | Keyword | Definition | Type | Format | +|----------|-------------|--------|-------------|--------------------------|--------------|--------| +| | | | Granularity | Quantization Granularity | value | int | +| | | | Function | Non-linear function | value | Int | +| | | | N | Quantization bit number | value | int | +| | | Output | W | Recovered weight tensor | list[tensor] | NS | + +Some optional non-linear functions are listed in Table 245. + +**Table 245 – Optional non-linear quantization function** + +| Label | Type | +|-------|----------------------| +| 1 | Linear function | +| 2 | Exponential function | +| 3 | Tanh function | +| 4 | Other function | + +The pseudocode of non-linear dequantization can be found in Table 246. + +**Table 246 – Pseudo-code of non-linear dequantization** + +| Non-linear Dequantization | Symbol | +|----------------------------------------------------------------------------------------|--------| +| def Weight_non_Linear_dequantization (W_q, W_scale, W_offset, granularity,function,N){ | | +| if(function==1): { | | +| W_q=W_q | | +| } | | +| elif (function ==2): { | | +| W_q = e^(W_q) | | +| } | | +| elif(function ==3): { | | +| W_q = atanh(W_q) | | +| W = Linear_dequantization (W, W_scale, W_offset, granularity,N) | | +| } | | +| return W | | +| } | | + +###### **9.2.2.2.2.2 INT4 parameter dequantization** + +The INT4 dequantization operator converts a single or a set of INT4 tensor(s) to a tensor with high-precision representation (e.g., FP32). Commonly, the dequantization of multiple INT4 tensors refers to Dual INT4 dequantization. In comparison with INT4 activation quantization, INT4 activation dequantization is the opposite operator. Parameter dequantization includes the network weight and + +bias dequantization. When performing INT4 parameter dequantization, the network bias is not processed. For network weights, quantization can be performed layer after layer by the following steps: + +- 1) Load the quantized weight tensor $W\_quantized$ from the current layer in a network, quantization scale factor list and weight quantization offset list $offset\_w\_list$ . +- 2) Determine the type of the current layer based on the quantized weight tensor $W\_quantized$ , and apply a proper dequantization mode: + - a) If the layer is a convolutional, perform dequantization to 3D convolution kernel tensor $W\_quantized[k]$ corresponding to each output channel, and compute convolution kernel reconstruction tensor $W\_recon[k]$ . + - b) If the layer is fully connected, perform a layer-by-layer dequantization to quantized weight tensor $W\_quantized$ , and compute weight reconstruction tensor $W\_recon$ . + +INT4 parameter dequantization operator is defined in Table 247. + +**Table 247 – INT4 parameter dequantization operator definition** + +| Operator | Description | Parameters | Key words | Definition | Data type | Data format | +|--------------------------------|---------------------------------------------------------------|------------|---------------------------|-----------------------------------------------------------------|--------------|-------------| +| int4_recon_quantized_parameter | Dequantize weight(s) in INT4 to high-precision representation | Input | $W\_quantized$ | Quantized weight tensor in INT4 | list[tensor] | INT4 | +| | | | $B\_quantized$ (optional) | Quantized bias tensor in INT4 | list[tensor] | NS | +| | | | $sf\_w\_list$ | Quantization scale factor list | list | float32 | +| | | | $offset\_w\_list$ | Weight quantization bias identifier | list | float32 | +| | | Output | $W\_recon$ | Reconstruction weight tensor out of this dequantization process | list[tensor] | float32 | +| | | | $B\_recon$ (optional) | Bias tensor out of this dequantization | list[tensor] | NS | + +The pseudocode of INT4 parameter dequantization is defined in Table 248. + +**Table 248 – The pseudocode of INT4 parameter dequantization** + +| INT4 parameter dequantization | Descriptor | +|--------------------------------------------------------------------------------------------------------|------------| +| def int4_recon_quantized_parameter( $W\_quantized, sf\_w\_list, offset\_w\_list, B\_quantized$ ):
{ | | +| if len( $W\_quantized.shape$ )>2: { | | +| for k in range(len( $W\_quantized$ )): { | | +| if len( $sf\_w\_list[k]$ )==1: | | +| $W\_recon[k]=W\_quantized[k]*sf\_w\_list[k][0]+offset\_w\_list[k]$ | | +| elif len( $sf\_w\_list[k]$ )==2: | | +| $W\_recon[k]=W\_quantized[k][0]*sf\_w\_list[k][0]+/$ | | + +**Table 248 – The pseudocode of INT4 parameter dequantization** + +| INT4 parameter dequantization | Descriptor | +|------------------------------------------------------------------------------------------|------------| +| $W\_quantized[k][1]*sf\_w\_list[k][1]+offset\_w\_list[k]$ | | +| } | | +| } | | +| else: { | | +| if len(sf_w_list)==1: | | +| $W\_recon=W\_quantized*sf\_w\_list[0]+offset\_w\_list$ | | +| elif len(sf_w_list)==2: | | +| $W\_recon=W\_quantized[0]*sf\_w\_list[0]+W\_quantized[1]*sf\_w\_list[1]+offset\_w\_list$ | | +| } | | +| $B\_recon=B\_quantized$ | | +| return $W\_recon, B\_recon$ | | +| } | | + +##### **9.2.2.2.3 Activate dequantization operator** + +###### **9.2.2.2.3.1 Trainable alpha value dequantization** + +In the dequantization operators, the quantized value X is linearly mapped from a lower-bit space to a new floating-point space. Corresponding to the quantization part in the previous clause, when the model is read for calculation, the weights are reverted to full-precision with the dequantization function $X = X\_Scale(X\_q - X\_offset)$ . + +The data definition of training truncated value dequantization can be found in Table 249. + +**Table 249 – Definition of trainable alpha dequantization** + +| Operator | Description | Field | Keyword | Definition | Type | Format | +|--------------------------------|-----------------------------------------------|--------|-------------|----------------------------------------|---------------|--------| +| Trainable Alpha Dequantization | Train the quantized truncated value in layers | Input | $X\_q$ | Activation tensor after quantization | List [tensor] | NS | +| | | | Alpha | Trainable truncated upper-bound value | List [tensor] | NS | +| | | | Granularity | Quantization Granularity | Value | int | +| | | | N | Quantization bit number | Value | int | +| | | Output | X | Activation tensor after dequantization | List [tensor] | NS | + +The pseudocode of trainable alpha dequantization can be found in Table 250. + +**Table 250 – Pseudo-code of training truncated value dequantization** + +| Training truncated value dequantization | Symbol | +|------------------------------------------------------------------------|--------| +| def Trainable_Alpha_DeQuantization( $X\_q, Alpha, Granularity, N$ ):{ | | +| $X\_Scale = (2^{n-1}-1)/Alpha$ | | +| $X\_offset = 0$ | | +| $X= Linear\_dequantization (X\_q, X\_scale, X\_offset, Granularity,N)$ | | + +**Table 250 – Pseudo-code of training truncated value dequantization** + +| Training truncated value dequantization | Symbol | +|-----------------------------------------|--------| +| return X | | +| } | | + +###### **9.2.2.2.3.2 INT4 activation dequantization** + +The INT4 dequantization operator converts a single or a set of INT4 tensor(s) to a tensor with high-precision representation (e.g., FP32). Dual INT4 quantization results in multiple INT4 tensors. Oppositely, dual INT4 dequantization is used for the dequantization of multiple INT4 tensors. INT4 parameter dequantization can be applied by the following steps: + +- 1) Load the quantized activation tensor X\_quantized of the current layer, quantization scale factor list sf\_act\_list and quantization offset list offset\_w\_list. +- 2) Perform linear dequantization to the activation tensor. + +INT4 activation dequantization operator is defined in Table 251. + +**Table 251 – INT4 activation dequantization operator definition** + +| Operator | Description | Parameters | Key words | Definition | Data type | Data format | +|---------------------------------|--------------------------------------------------------------------------|------------|-----------------|--------------------------------------------------------|--------------|-------------| +| int4_recon_quantized_activation | Dequantize activation tensor(s) in INT4 to high-precision representation | Input | X_quantized | quantized activation tensor | list[tensor] | INT4 | +| | | | sf_act_list | quantization scale factor list | list | float32 | +| | | | offset_act_list | Activation quantization offset list | list | float32 | +| | | Output | X_recon | Reconstructed weight vector out of this dequantization | tensor | float32 | + +The pseudocode of INT4 activation dequantization is in Table 252. + +**Table 252 – The pseudocode of INT4 activation dequantization** + +| INT4 activation dequantization | Descriptor | +|---------------------------------------------------------------------------------|------------| +| def int4_recon_quantized_activation(X_quantized,sf_act_list,offset_act_list): { | | +| X_recon=zeros(X_quantized[0].shape) | | +| for(i=0;i] | NS | +| | | | N_dim | Dimension of N | value | int | +| | | | granularity | Pruning granularity | value | int | +| | | Output | W_u | Inversed weight tensor | list[tensor] | NS | +| | | | B_o | Inversed bias tensor | list[tensor] | NS | + +The pseudocode can be referred in Table 254. + +**Table 254 – Structure sparse data generation** + +| Structure sparse data generation | Descriptor | +|--------------------------------------------------------------|------------| +| Un_structured_sparse(W,B,S,N_dim, Skip_Column, granularity){ | | +| for(i=0;iStructure sparse data generation | Descriptor | +|-------------------------------------------------------------|-------------------| +| New_tensor_shape_w[dim] = length(S[i]); | | +| New_tensor_shape_b = b_shape; | | +| New_tensor_shape_b[dim] = length(S[i]); | | +| W_o[i] = Tensor(New_tensor_shape_w); | | +| B_o[i] = Tensor(New_tensor_shape_b) | | +| k = 0; | | +| for(j=0;jlen(vector)-1: { | | +| shift_vector[mod(i+shift, len(vector))] = vector[i] | | +| } | | +| else: { | | +| shift_vector[i+shift] = vector[i] | | +| } | | +| return shift_vector | | +| } | | + +###### 9.2.2.4.1.2 Disturbance vector generation operator + +For the layers based on signed vector and block circulant matrix techniques, the input vector needs to be converted to the correction input vector before calculating the output of the network layer in training and reasoning. Each element in a calibrated input vector is the multiplication of the corresponding elements in its original vector and a disturbance vector. The disturbance vector is the result of the random vector extension process, which process is discussed as follows: + +- 1) Based on the reference random vector (base\_sign\_vector\_network) and the compressed layer vector dimension list (len\_list), generate a random vector for each layer. That involves cutting a vector with the length of len\_list[i] from the reference random vector to serve as the random vector for the i-th layer to be compressed. +- 2) Based on the random vector, apply cyclic shift or replication approaches to generate the disturbance vector. Its number of dimensions is equal to the number of dimensions of compression layer input vector. + +The disturbance vector generation operator is defined in Table 259. + +**Table 259 – Disturbance vector generation operator** + +| Operator | Description | Parameters | Key words | Definition | Data types | Data formats | +|---------------------------|-------------------------------------------------------------------|------------|--------------------------|-------------------------------------------|------------|--------------| +| generate_sign_vector_list | Generate a disturbance vector corresponding to a compressed layer | Input | len_list | Random vector length list | list[int] | UINT | +| | | | base_sign_vector_network | Reference random vector of neural network | vector | UINT | + +**Table 259 – Disturbance vector generation operator** + +| Operator | Description | Parameters | Key words | Definition | Data types | Data formats | +|----------|-------------|------------|------------------|------------------------------------------------------------------|--------------|--------------| +| | | | extend_list | Mode of disturbance vector generation | list[int] | UINT | +| | | | len_target | Number of dimensions of disturbance vector of a compressed layer | list[int] | UINT | +| | | Output | sign_vector_list | Signed vector of a compressed layer | list[vector] | UINT | + +The pseudocode of disturbance signed vector generation is in Table 260. + +**Table 260 – The pseudocode of disturbance signed vector generation** + +| Disturbance signed vector generation | Descriptor | +|--------------------------------------------------------------------------------------------------|------------| +| def generate_sign_vector_list(len_list, base_sign_vector_network, extend_list, len_target): | | +| { | | +| sign_vector_list = [] | | +| for i in range(len(len_list)): { | | +| base_sign_vector = base_sign_vector_network[:len_list[i]] | | +| sign_vector_list.append(extend_base_to_sign(base_sign_vector, len_target[i],
extend_list[i])) | | +| } | | +| return sign_vector_list | | +| } | | + +The operator extend\_base\_to\_sign (in Table Table 263) is defined in Table 261. + +**Table 261 – Extended signed vector generation operator** + +| Operator | Description | Parameters | Key words | Definition | Data types | Data formats | +|---------------------|---------------------------------------------|------------|------------------|-------------------------------------|------------|--------------| +| extend_base_to_sign | Generate disturbance signed vector based on | Input | base_sign_vector | Random signed vector base | vector | Int | +| | | | len_extend | Length of disturbance signed vector | value | Int | + +**Table 261 – Extended signed vector generation operator** + +| Operator | Description | Parameters | Key words | Definition | Data types | Data formats | +|----------|----------------------|------------|--------------------|---------------------------------------------------------|------------|--------------| +| | random signed vector | | method | Method of disturbance signed vector generation | value | Int | +| | | Output | extend_sign_vector | Disturbance signed vector of the layer to be compressed | vector | Int | + +The methods of disturbance signed vector generation are defined in Table 262. + +**Table 262 – Methods of disturbance signed vector generation** + +| method | | +|--------|--------------| +| Index | Type | +| 1 | Cyclic shift | +| 2 | Replication | + +The pseudocode of the operator extend\_base\_to\_sign is in Table 263. + +**Table 263 – The pseudocode of extended signed vector generation** + +| Extended signed vector generation | Descriptor | +|--------------------------------------------------------------------|------------| +| def extend_base_to_sign(bae_sign_vector, len_extend, method): { | | +| extend_sign_vector = zeros(len_extend) | | +| if method == 1: { | | +| for i in range(0, len_extend, len(base_sign_vector)): { | | +| extend_sign_vector[i:i+len(base_sign_vector)-1] = circ_shift(\ | | +| base_sign_vector, i/len(base_sign_vector)) | | +| } | | +| } | | +| elif method == 2: { | | +| for i in range(0, len_extend, len(base_sign_vector)): { | | +| extend_sign_vector[i:i+len(base_sign_vector)-1] = base_sign_vector | | +| } | | +| } | | +| else: { | | +| return error | | +| } | | +| return extend_sign_vector | | +| } | | + +###### 9.2.2.4.1.3 Operator on the layers using signed vector and block circulant matrix techniques + +For the layer using symbol vector and block circulant matrix technology, the input vector is the correction input vector when calculating the output of the network layer in training and reasoning. Each element in a calibrated input vector is the multiplication of the corresponding elements in the original vector and disturbance vector. Accordingly, the computation of a layer can be performed by the following steps: + +- 1) Compute the calibrated input vector. Each element in the calibrated input vector is the multiplication of the corresponding elements in the original vector and disturbance vector. +- 2) Feed the network layer with the calibrated input vector, weight matrix, bias and compute the output. + +The layer computation operator is defined in Table 264. + +**Table 264 – Layer computation operator** + +| Operator | Description | Parameters | Key words | Definition | Data types | Data formats | +|---------------|-----------------------------------------------------------------------------------|------------|-------------|----------------------------|------------|--------------| +| cal_sign_circ | Compute the output of a layer which uses signed vector and block circulant matrix | Input | layer_in | Input vector | vector | NS | +| | | | sign_vector | Disturbance vector | vector | UINT | +| | | | weight | Decompressed weight tensor | matrix | UINT | +| | | | B(optional) | Bias vector | vector | NS | +| | | Output | layer_out | Output vector of the layer | vector | NS | + +The pseudocode of the layer computation is in Table 265. + +**Table 265 – The pseudocode of layer computation** + +| Layer computation computation | Descriptor | +|--------------------------------------------------------|------------| +| def cal_sign_circ(layer_in, weight, sign_vector, B): { | | +| sign_layer_in = mul(layer_in, sign_vector) | | +| layer_out = mul(sign_layer_in, weight) | | +| if B: { | | +| layer_out = layer_out + B | | +| } | | +| return layer_out | | +| } | | + +##### 9.2.2.4.2 Methods for the decompression of low rank sparse decomposed structured matrix + +###### 9.2.2.4.2.1 Overview + +The tensors corresponding to $3 \times 3$ convolutional layers are generated by high-dimensional SVD decomposition. Therefore, the decompression of them requires an n-mode multiplication on tensors, which yields the decompression result. + +The tensors corresponding to $1 \times 1$ convolutional and fully connected layers are also generated by high-dimensional SVD decomposition. The decompression of them requires a matrix multiplication of these two tensors. + +###### 9.2.2.4.2.2 Decompression operator for low rank sparse decomposed structured matrix + +The decompression process of a convolutional layer is as follows: + +- 1) Restore each sub-tensor to a sub-tensor group. +- 2) Apply n-mode multiplication to each sub tensor group, and compute an approximated value to the original tensor. The decompression result is the sum of the approximated values. + +The decompression operator for low rank sparse decomposed structured matrix is defined in Table 266. + +**Table 266 – Decompression operator for the convolutional layers in low rank sparse decomposed structured matrix** + +| Operator | Description | Parameters | Key words | Definition | Data types | Data formats | +|-------------------------|-------------------------------------------------------------------------------------|------------|--------------|----------------------------------------------|------------|--------------| +| decompr_of_conv_in_lrsm | decompress the convolutional layers in low rank sparse decomposed structured matrix | Input | w1 | The first compressed weight tensor | matrix | NS | +| | | | w2 | The second compressed weight tensor | matrix | NS | +| | | | w3 | The third compressed weight tensor | matrix | NS | +| | | | Weight_shape | Dimension information of the original tensor | vector | UINT | +| | | | R1 | Rank of input channel | value | UINT | +| | | | R2 | Rank of output channel | value | UINT | +| | | Output | approx | Decompressed weight tensor | matrix | NS | + +The pseudocode of the operator is in Table 267. + +**Table 267 – The pseudocode of the decompression of the convolutional layers in low rank sparse decomposed structured matrix** + +| Decompression of the convolutional layers in low rank sparse decomposed structured matrix | Descriptor | +|-------------------------------------------------------------------------------------------|------------| +| def BTD_decompression(w1, w2, w3, Weight_shape, R1, R2, groups): { | | +| r1 = R1/groups | | +| r2 = R2/groups | | +| w1 = [w1[:,idx*r1:idx*r1+r1] for idx in range(groups)] | | +| w2 = [w2[idx*r1:idx*r1+r1,:] for idx in range(groups)] | | +| w3 = [w3[:,idx*r2:idx*r2+r2] for idx in range(groups)] | | + +**Table 267 – The pseudocode of the decompression of the convolutional layers in low rank sparse decomposed structured matrix** + +| Decompression of the convolutional layers in low rank sparse decomposed structured matrix | Descriptor | +|-------------------------------------------------------------------------------------------|----------------------------------------------------------------------| +| approx = zeros(Weight_shape) | | +| for i in range(groups): { | | +| approx += w2[i]×_1w1[i]×_2w3[i] | × n : the multiplication of n-mode and scikit-tensor==0.1 | +| } | | +| return approx | | +| } | | + +###### **9.2.2.4.2.3 Decompression operator for the fully connected and 1 × 1 layers in low rank sparse decomposed structured matrix** + +The decompression of a fully connected and 1 × 1 convolutional layer involves computing matrix multiplication over these two sub-tensors, which results in an approximation of the original tensor. The operator for this decompression is defined in Table 268. + +**Table 268 – Decompression operator for the fully connected and 1 × 1 layers in low rank sparse decomposed structured matrix** + +| Operator | Description | Parameters | Key words | Definition | Data types | Data formats | +|---------------------------------|----------------------------------------------------------------------------------------------|------------|-----------|----------------------------------------------------|------------|--------------| +| decompr_of_fc_1x1_con_v_in_lrsm | Decompress a fully connected and 1 × 1 layer in low rank sparse decomposed structured matrix | Input | w1 | Input weight tensor to be decompressed | matrix | NS | +| | | | w2 | Shape of the output corresponding the middle layer | matrix | NS | +| | | Output | approx | Decompressed weight tensor | matrix | NS | + +The pseudocode of this operator is in Table 269. + +**Table 269 – The pseudocode of the decompression for the fully connected and 1 × 1 layers in low rank sparse decomposed structured matrix** + +| Decompression for the fully connected and 1 × 1 layers in low rank sparse decomposed structured matrix | Descriptor | +|--------------------------------------------------------------------------------------------------------|------------| +| def SVD_decompression(w1, w2) { | | +| approx = mat_mul(w2, w1) | | +| return approx | | +| } | | + +### 9.2.3 Data generation + +#### 9.2.3.1 Definition + +For some neural network compression technologies specified in this document, the weights of the network model should meet the requirements of certain structures, data, and hyperparameters. To meet these requirements, a series of limitations and special training methods need to be added in the training process to obtain the information of neural network models that can be compressed. This clause describes the data generation methods for neural network model compression described in the preceding two clauses. + +This clause describes how to generate training data and weight data. Clause 9.2.3.2 describes the training data generation method, while clauses 9.2.3.3 to 9.2.3.7 describe the model weight data generation methods involved in different neural network compression methods. + +The following principles should be complied with when processing data used for neural network training and inference: + +- 1) Data shall be processed according to the computation implementation and user requirements before being used. +- 2) Elements involved in data processing methods shall be adjusted based on actual needs, including but not limited to: + - a) Method naming + - b) Selection and naming of parameters (keywords) + - c) Parameter type and format + +#### 9.2.3.2 Training data generation method + +##### 9.2.3.2.1 Method of generating training data based on real data + +###### 9.2.3.2.1.1 Overview + +This clause describes how to augment real data or use alternative methods to derive more real data when real training data is available. + +We could augment original images by rotating, translating and scaling them. Besides, a generative adversarial network (GAN) that is trained using real data can be used to convert noise into images. + +###### 9.2.3.2.1.2 Data augmentation method + +This clause describes basic operators such as rotation, translation, and scaling to augment the obtained real data. + +Specifically, given a real input image $X$ , it can be rotated with an angle of $\alpha$ , moved by $n$ positions on the x-axis (y-axis) as a whole, flipped along the x-axis (y-axis) as a whole, or scaled as a whole to obtain a new image $X'$ . + +Table 270 describes the definition of various data augmentation methods. + +**Table 270 – Definition of various data augmentation methods** + +| Computational Operator | Description | Field | Keyword | Definition | Type | Format | +|------------------------|---------------------------------|-------|----------|------------------------|--------|--------| +| Data augmentation | Augment the existing real data. | Input | X | Input real data | tensor | NS | +| | | | $\alpha$ | Rotation angle | int | UINT | +| | | | n | Translational distance | int | UINT | + +**Table 270 – Definition of various data augmentation methods** + +| Computational Operator | Description | Field | Keyword | Definition | Type | Format | +|------------------------|-------------|--------|---------|--------------------------------------------|--------|--------| +| | | | axis | Translational direction (x axis or y axis) | bool | BOOL | +| | | | n_row | Width after scaling | int | UINT | +| | | | n_col | Height after scaling | int | UINT | +| | | Output | X' | Output augmented data | tensor | NS | + +Table 271 describes the pseudocode for the data augmentation methods. + +**Table 271 – The pseudocode for the data augmentation methods** + +| Data augmentation method | Descriptor | +|-------------------------------------------------------------|------------| +| def data_augmentation(X, , n, axis , n_row, n_col, type): { | | +| if type == 'rotate': | | +| X' = rotate(X, $\alpha$ ) | | +| elif type == 'translation': | | +| X' = translate(X, n, axis=axis) | | +| elif type == 'scale': | | +| X' = scale(X, n_row, n_col) | | +| elif type == 'flip': | | +| X' = flip(X, axis=axis) | | +| return X' | | +| } | | +| | | +| def rotate(X, ): { | | +| X' = np.zero((X.row, X.col)) | | +| for i in range(X.row): { | | +| for j in range(X.col): { | | +| i' = round(i*cos+j*sin) $\alpha\alpha$ | | +| j' = round(-i*sin+j*cos) $\alpha\alpha$ | | +| X '[ i', j' ] = X[i,j] | | +| } | | +| } | | +| return X' | | +| } | | +| | | +| def translate(X, n, axis): { | | +| X' = np.zero((X.row, X.col)) | | +| if axis== 'x': | | + +**Table 271 – The pseudocode for the data augmentation methods** + +| Data augmentation method | Descriptor | +|---------------------------------------------------------|-------------------| +| for j in range(X.col): | | +| if j+n>=0 and j+n=0 and i+nmultiple models data generation approach 1 | Descriptor | +|-------------------------------------------------------------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------------------------------| +| Multilayer_weight_update_train(x, y, Graph, Optimizer, W_share, B_share, W_specific, B_specific, model_indicator, axe, batch_size, epoch )
{ | X and y are the sample set and label set, rather than individual sample; model_indicator is the corresponding list | +| for epoch in range(epochs): { | | +| for ite in range(int(len(trainX)/batch_size)): { | | +| batch_x=x[ite:ite+batch_size] | | +| batch_y=y[ite:ite+batch_size] | | +| batch_model_indicator=model_indicator[ite:ite+batch_size] | | +| for i in range(len(batch_x)): { | | +| idx=batch_model_indicator[i] | | +| W = Concat(W_share, W_specific[idx], axe) | | +| B = Concat(B_share, B_specific[idx], axe) | | +| Result = Graph(batch_x[i], W, B) | | +| Loss = compute_loss(Result, batch_y[i]) * loss_factor[idx] | | +| W_gradient, B_gradient = compute_gradient(W, B, Loss, Optimizer) | | +| W_gradient_share = slice(W_gradient[0], axe, 0, W_share.shape(axe)) | | +| W_gradient_specific = slice(W_gradient[0], axe, W_share.shape(axe), -1) | | +| B_gradient_share = slice(B_gradient[0], axe, 0, B_share.shape(axe)) | | +| B_gradient_specific = slice(B_gradient[0], axe, B_share.shape(axe), -1) | | +| W_share = W_share + W_gradient_share | | +| B_share=B_share + B_gradient_share | | +| W_specific[idx] += W_gradient_specific | | +| B_specific[idx] += B_gradient_specific | | +| } | | +| } | | +| } | | +| W_share_updated=W_share | | +| W_specific_updated=W_specific | | +| B_share_updated=B_share | | +| B_specific_updated=B_specific | | +| return W_share_updated, W_specific_updated, B_share_updated, B_specific_updated | | +| } | | + +###### 9.2.3.4.1.2 Multiple models weight shared data generation approach 2 + +To generate the data to be compressed, another option is to extract the shared output data and the model-specific output to the next layer, without extracting specific weight inside the current layer. This process is depicted in Figure 10. The training process is as follows: + +- a) Determine the layers that would contain shared weights in multiple models, with the constraint that those layers hold an identical structure. +- b) Obtain the training object, which extracts training sample batches from training input and labels, including multiple tasks. +- c) During an iteration of training, perform forward propagation for each sample. For an arbitrary layer: + - 1) Obtain the specific weights and shared weights. That is to determine the position indices which distinguish the shared weights ( $W\_share$ ) and independent weights in those layers. + - 2) Construct a layer $L_n$ , containing the sum of the weights (shared and specific) of multiple models. + - 3) Filter and distinguish the shared output data and specific output data. That is to set a filter, which only allows for the training data to pass the transitional layer if the data corresponds to a specific model and is taken as input of the next layer. +- d) Compute (perform statistics) loss. That is done by comparing the result of forward propagation based on the network graph with the corresponding labels and adjusting the loss value with $loss\_factor$ , based on the model index associated with training samples. +- e) Adjust the specific weight and shared weights. That is to perform backward propagation based on loss, then adjust the shared weights and specific weights associated with specific models. +- f) Repeat steps b) – e) to accomplish one epoch. +- g) Repeat steps b) – f). The training network updates parameters until the network converges or reaches the upper limit of training times. + +The pseudocode of multiple models data generation approach 2 is in Table 281. + +![Diagram of Figure 10 showing a neural network with input nodes connecting to a transitional layer with specific and shared weights, followed by a filter and output nodes.](ffe2d3e622dd30bea4caf15951c60e42_img.jpg) + +``` +graph LR + subgraph Input + I1(( )) + I2(( )) + I3(( )) + end + subgraph Transitional_layer [Transitional layer] + direction TB + W1(( )):::red_node + W2(( )):::yellow_node + W3(( )):::green_node + W4(( )):::green_node + end + subgraph Filter_Block [Output tensor filtering/Selection] + Filter[Filter]:::blue_rect + end + subgraph Output + O1(( )):::grey_node + O2(( )):::grey_node + O3(( )):::grey_node + end + + I1 -- red --- W1 + I2 -- red --- W1 + I3 -- red --- W1 + I1 -- yellow --- W2 + I2 -- yellow --- W2 + I3 -- yellow --- W2 + I1 -- green --- W3 + I2 -- green --- W3 + I3 -- green --- W3 + I1 -- green --- W4 + I2 -- green --- W4 + I3 -- green --- W4 + + W1 --> Filter + W2 --> O1 + W3 --> O2 + W4 --> O3 + Filter --> O1 + + classDef red_node fill:#ff0000,stroke:#333,stroke-width:1px; + classDef yellow_node fill:#ffcc00,stroke:#333,stroke-width:1px; + classDef green_node fill:#4caf50,stroke:#333,stroke-width:1px; + classDef grey_node fill:#9e9e9e,stroke:#333,stroke-width:1px; + classDef blue_rect fill:#2196f3,color:#fff,stroke:#333,stroke-width:1px; +``` + +Figure 10 illustrates the 'Convolution process' and 'Output tensor filtering/Selection'. It shows an 'Input' layer of three blue nodes connecting to a 'Transitional layer'. The transitional layer has a red node labeled 'Specific weights of model 1', a yellow node labeled 'Specific weights of model 2', and two green nodes labeled 'Shared weights'. Connections are color-coded to match the target nodes. A blue 'Filter' box processes the output from the red node before it reaches the 'Output' layer (three grey nodes). The yellow and green nodes connect directly to the output nodes. Reference: F.748.53(25). + +Diagram of Figure 10 showing a neural network with input nodes connecting to a transitional layer with specific and shared weights, followed by a filter and output nodes. + +**Figure 10 – Training approach of multiple models with shared weights** + +**Table 281 – The pseudocode of multiple models data generation approach 2** + +| Multiple models data generation approach 1 | Descriptor | +|-----------------------------------------------------------------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------------------------------| +| Multilayer_weight_update_train2(x, y, Graph, Optimizer, W_share, B_share, W_specific, B_specific, model_indicator, axe, batch_size, epoch, perc ) { | X and y are the sample set and label set, rather than individual sample; model_indicator is the corresponding list | +| for epoch in range(epochs): { | perc is the shared weight proportion | +| for ite in range(int(len(trainX)/batch_size)): { | | +| batch_x=x[ite:ite+batch_size] | | +| batch_y=y[ite:ite+batch_Size] | | +| batch_model_indicator=model_indicator[ite:ite+batch_size] | | +| for i in range(len(batch_x)): { | | +| idx=batch_model_indicator[i] | | +| layer_in=batch_x[i] | | +| for layer in Graph.layers: { | | +| layer_W= Concat(layer.W_share, layer.W_specific, axe) | | +| layer_B= Concat(layer.B_share, layer.B_specific, axe) | | +| layer_out=layer.operate(layer_in,layer_W,layer_B) | | +| #filtering | | +| share_end=perc*layer_out.shape(axe) | | +| len_specific=(1-perc)*layer_out.shape(axe)/len(set(model_indicator)) | | +| layer_out_share=slice(layer_out,axe,0,share_end) | | +| layer_out_specific=slice(layer_out,axe,share_end+idx*len_specific,share_end+(idx+1)*len_specific) | | +| layer_out=Concat(layer_out_share,layer_out_specific) | | +| layer_in=layer_out | | +| } | | +| Result = layer_out | | +| Loss = compute_loss(Result, batch_y[i]) * loss_factor[idx] | | +| W_gradient, B_gradient = compute_gradient(W, B, Loss, Optimizer) | | +| W_gradient_share = slice(W_gradient[0], axe, 0, W_share.shape(axe)) | | +| W_gradient_specific = slice(W_gradient[0], axe, W_share.shape(axe), -1) | | +| B_gradient_share = slice(B_gradient[0], axe, 0, B_share.shape(axe)) | | +| B_gradient_specific = slice(B_gradient[0], axe, B_share.shape(axe), -1) | | +| W_share = W_share + W_gradient_share | | +| B_share=B_share + B_gradient_share | | +| W_specific[idx] += W_gradient_specific | | +| B_specific[idx] += B_gradient_specific | | + +**Table 281 – The pseudocode of multiple models data generation approach 2** + +| Multiple models data generation approach 1 | Descriptor | +|---------------------------------------------------------------------------------|------------| +| } | | +| } | | +| } | | +| W_share_updated=W_share | | +| W_specific_updated=W_specific | | +| B_share_updated=B_share | | +| B_specific_updated=B_specific | | +| return W_share_updated, W_specific_uodated, B_share_updated, B_specific_updated | | +| } | | + +##### **9.2.3.4.2 Residual quantization training method for multiple models** + +When adopting residual quantization, corresponding training methods shall be adopted to improve the performance. The training method involves fixing the weight of the pre-training model and sharing the weight with the training to further improve the network performance. The gradient updating method replaces the gradient belonging to the same class with the sum of the mean values of the same class in the clustering results. The specific definition is defined in Table 282. + +**Table 282 – Residual quantization operator** + +| Operator | Description | Parameters | Key words | Definition | Data type | +|------------------------------|-------------------------------------------------------------------|------------------|------------------|------------------------------------------------|-----------------| +| DFQuantization_weight_update | Residual quantization improves network performance by fine-tuning | Input Attributes | W_share_original | The initial shared weight | tensor | +| | | | Base model | Weights of Pre-trained model | tensor | +| | | | W_gradients | The gradient corresponding to the weight | Tensor | +| | | | W_specific | The clustering results of weight (code storge) | List of tensors | +| | | Output | W_share_output | Shared weight tensor | tensor | + +The resulting process is shown in the pseudocode in Table 283. + +**Table 283 – The pseudocode of residual quantization** + +| Residual quantization training method for multiple models | descriptor | +|---------------------------------------------------------------------------|------------| +| DFQuantization_weight_update(W_share, Base_model,gradients,W_specific): { | | +| Epoch=100 | | +| W_share_output = W_share_original | | + +**Table 283 – The pseudocode of residual quantization** + +| Residual quantization training method for multiple models | descriptor | +|----------------------------------------------------------------------------------|-------------------------------------------------| +| for(j=0;jgradients =Gradients_compute( W_share_output , Base_model ) | Backpropagation weight gradient calculation | +| gradients =average( W_specific ) | Gradient calculation based on clustering result | +| Lr = 1e-4 | | +| for(i=0;iW_specific );i++) | | +| W_share_output [i]= W_share [i]-Lr* gradients [i] | Gradient descent | +| } | | +| Return W_share_output | | +| } | | + +#### 9.2.3.5 Quantization + +##### 9.2.3.5.1 Parameter quantization + +###### 9.2.3.5.1.1 Data generation for linear parameter quantization + +When the bit-width is fixed to b, other quantization parameters (scale factor s and the zero-point z) are critical for uniform quantization. Determining and updating these parameters can be done by some heuristic rules or gradient-based learning methods. + +At first, the rule-based updating methods are introduced. The most straightforward rules is the Min-max algorithm. The maximum value in the real number domain corresponds to the maximum value in the quantized number domain. And the minimum values in two domains also have a corresponding relationship. Scale factor s represents the mapping ratio from the real numerical range to the quantized numerical range. Scale factor s is determined as: + +$$s = \frac{(\text{maximum}(x_r) - \text{minimum}(x_r))}{(\text{maximum}(x_q) - \text{minimum}(x_q))}$$ + +where for unsigned setting, $\text{maximum}(x_q) = 2^b - 1$ , $\text{minimum}(x_q) = 0$ . Zero-point z determined as: + +$$z = \text{maximum}(x_q) - \frac{\text{maximum}(x_r)}{s}$$ + +For weight quantization, the maximum value and minimum value can be easily collected since the weights remain constant during the inference process. However, determining the scale factor for activation quantization requires a few calibration data. These calibration data are sampled from the training set. In order to guarantee the generalization for unseen data, the maximum and minimum values of activation are collected on calibration data, and then they are aggregated via exponential moving averages (EMA) with the smoothing parameter being close to 1. Although this min-max algorithm causes no clipping error, it may lead to excessive rounding errors especially when there are outliers. In order to deal with the outliers, some other methods provide a better trade-off between clipping errors and rounding errors. The Percentile algorithm uses the percentile $\alpha$ to eliminate outliers: + +$$s = \frac{(\text{percentile}(x_r, \alpha) - \text{percentile}(x_r, 1 - \alpha))}{(\text{maximum}(x_q) - \text{minimum}(x_q))}$$ + +where $\alpha$ is usually set to 99.99% or 99.999%. It chooses the largest/smallest values at the percentage $\alpha$ instead of the maximum/minimum values. Another algorithm to alleviate the problem of outliers is to update quantization parameters to reduce the MSE loss between distributions of $x_r$ and $\widehat{x}_r$ : + +$$\operatorname{argmin}_{s, z} ||x_r - \widehat{x}_r||^2$$ + +where $\widehat{x}_r = s \cdot (x_q - z)$ . Grid search, the golden section approach, or analytical approximations with a closed-form solution are frequently used to address optimization problems. For some layers, values in the quantized tensor $x_r$ may not be equally important. Due to the fact that MSE weighs all variables equally regardless of their order, it is probably not the best statistic to utilize. In this case, it is beneficial to minimize the following cross-entropy loss function: + +$$\operatorname{argmin}_{s, z} \text{CrossEntropy}(x_r - \widehat{x}_r)$$ + +Both the MSE algorithm and Cross-Entropy algorithm can effectively eliminate outliers but they require a large optimization time cost. The KL-Divergence algorithm tries to find the suitable range that the KL divergence of two distributions between $x_r$ and $\widehat{x}_r$ is the smallest. It is usual that $x_r$ and $\widehat{x}_r$ are divided into 2048 bins with the histogram (noted as hist) to obtain the distribution, then: + +$$\operatorname{argmin}_{s, z} D_{kl}(\text{hist}(x_r), \text{hist}(\widehat{x}_r))$$ + +The above are the commonly used rule-based methods, and there are some other methods that will not be discussed here. Methods of updating quantization parameters by gradient are called learning-based methods. The formula is as follows: + +$$\widehat{x}_r = s \cdot x_q = s \cdot \text{clamp}(\lfloor \frac{x_r}{s} \rfloor; 0, 2^b - 1) = \begin{cases} 0 & \lfloor \frac{x_r}{s} \rfloor < 0 \\ \lfloor \frac{x_r}{s} \rfloor \cdot s & 0 < \lfloor \frac{x_r}{s} \rfloor < 2^b - 1 \\ (2^b - 1) \cdot s & \lfloor \frac{x_r}{s} \rfloor > 2^b - 1 \end{cases}$$ + +Then, take the derivative of the parameters: + +$$\frac{d\widehat{x}_r}{ds} = \begin{cases} 0 & \lfloor \frac{x_r}{s} \rfloor < 0 \\ \lfloor \frac{x_r}{s} \rfloor + \frac{d\lfloor \frac{x_r}{s} \rfloor}{ds} \cdot s & 0 < \lfloor \frac{x_r}{s} \rfloor < 2^b - 1 \\ 2^b - 1 & \lfloor \frac{x_r}{s} \rfloor > 2^b - 1 \end{cases}$$ + +where $d\lfloor \frac{x_r}{s} \rfloor$ is always approximate with $\frac{dx_r}{s} = -\frac{x_r}{s^2}$ . The derivative of $\widehat{x}_r$ with respect to $s$ is finally calculated as: + +$$\frac{d\widehat{x}_r}{ds} = \begin{cases} 0 & \lfloor \frac{x_r}{s} \rfloor < 0 \\ \lfloor \frac{x_r}{s} \rfloor + -\frac{x_r}{s} & 0 < \lfloor \frac{x_r}{s} \rfloor < 2^b - 1 \\ 2^b - 1 & \lfloor \frac{x_r}{s} \rfloor > 2^b - 1 \end{cases}$$ + +Based on the above formula, scale factor $s$ can be a parameter of the deep neural networks which updated by gradient. Considering about the zero point $z$ , the formula is: + +$$\widehat{x}_r = s \cdot (x_q - z)$$ + +Take the derivative of the parameter $z$ : + +$$\frac{d\widehat{x}_r}{dz} = s \cdot (\frac{dx_q}{dz} - 1) \approx \begin{cases} 0 & 0 < \lfloor \frac{x_r}{s} \rfloor + z < 2^b - 1 \\ -s & \text{otherwise} \end{cases}$$ + +Another learning-based method focuses on the activation quantization. Traditional ReLU operations are unbounded, which means their results can have a large range. Through the truncated RELU (e.g., ReLU6), adding an upper bound to the activation can reduce clip errors. But results of the truncated ReLU are still significantly inferior to the full-precision model. Learning the upper bound via the gradient for each layer can make a better trade-off between clip errors and rounding errors to obtain the further performance. + +There are no clear experiments to prove which of the two kinds of method is better than the other. But compared to learning-based methods, rule-based methods do not require an additional training stage. + +###### 9.2.3.5.1.2 Data generation for INT4 parameter quantization + +The compression of neural networks includes weight quantization, quantization scale factor adjustment and activation quantization. Once quantized, the output of a neural network is the result of forward propagation based on its quantized weight and activation tensors. + +During INT4 parameter quantization, the bias of the network is not processed. For the neural network, quantization can be performed layer by layer. INT4 parameter quantization data generation can be done by the following steps: + +- a) When performing INT4 quantization to the network layers, weight quantization scale factor can be computed by the following steps: + - 1) Perform INT4 quantization to each layer of the neural network. This yields to the quantized weight tensor and corresponding weight quantization scale factor. Here, convolution kernel granularity quantization is applied to convolutional layers, while layered granularity quantization is applied to fully connected layers. + - 2) Based on the reconstruction error and pre-defined error valve, determine the layers that need multiple INT4 quantization. + - 3) For those determined layers by step 2), it needs to compute quantization scale factor. This can be accomplished by checking the distance between the minimal original tensor and the reconstructed tensor (via multiple INT4 quantization). Let $W$ be the tensor to be quantized, $\alpha_i, \widetilde{W}_i, i = 1 \dots t$ the scale factor and tensor, $\alpha_i, \widetilde{W}_i$ can be computed by Formula 35. + +$$\min_{\alpha_1, \widetilde{W}_1 \in I_{k_1}} \left\{ \min_{\alpha_2, \widetilde{W}_2 \in I_{k_2}} \left\{ \dots \min_{\alpha_n, \widetilde{W}_n \in I_{k_n}} d(W, \sum_{i=1}^n \alpha_i \widetilde{W}_i) \right\} \right\} \quad (35)$$ + +Where, + +$d$ is a distance measure; + +$I_{k_i}$ determines the scope where the quantization is allowed. + +Normally, for Dual INT4 quantization, the scale factor can be realized via dual linear search over reconstruction error with quadratic terms. + +- b) The pre-requisite of adjusting the weight quantization scale factor is to determine the quantization factor adjustment coefficient. This can be done by determining a calibration data set, which is used for the comparison between the output between a model and its quantized version. Let $\gamma_l$ (computed by Formula 36) be the quantization adjustment factor for the $l$ -th layer. Adjusted quantized tensor can be $\widetilde{W}_l = \gamma_l(\alpha_l \widetilde{W}_l)$ . + +$$\gamma_l = \min_{\gamma} (\sum_l \beta_l d(f(X, W_l), f(X, \widetilde{W}_l(\gamma)))) \quad (36)$$ + +in which: + +$0 \leq \beta_l \leq 1$ refers to the weight of the $l$ -th layer. + +INT4 parameter quantization data generation operator is defined in Table 284. + +**Table 284 – INT4 parameter quantization data generation operator** + +| Operator | Description | Parameters | Key words | Definition | Data type | Data format | +|-----------------------------|----------------------------------------------------------------------|--------------|------------------------|------------------------------------------------|--------------|-------------| +| int4_quantization_parameter | perform quantization to layers of a neural network, without training | Input | W | Weight tensor to be quantized | list[tensor] | float32 | +| | | | B (optional) | Bias tensor | list[tensor] | NS | +| | | | Graph | Network group | function | / | +| | | | mse_thresh | Quantization reconstruction error threshold | value | float32 | +| | | | cali_sf | Scale factor optimization calibration data set | tensor | float32 | +| | | Attributions | offset_w | Weight quantization offset identifier | value | Boolean | +| | | | bit_w | Number of bits for quantization | value | int | +| | | | n_points | Number of samples of linear search | value | int | +| | | Output | W_quantized | Quantized weight tensor | list[tensor] | signed int4 | +| | | | B_quantized (optional) | Quantized bias tensor | list[tensor] | NS | +| | | | sf_w_list | Weight quantization scale factor list | list | float32 | +| | | | dual_list | Dual INT4 quantization identifier list | list | Boolean | + +The pseudocode of INT4 parameter quantization data generation is defined in Table 285. + +**Table 285 – The pseudocode of INT4 parameter quantization data generation** + +| INT4 parameter quantization data generation | Descriptor | +|--------------------------------------------------------------------------------------|------------| +| def int4_quantization_parameter(W, Graph, mse_thresh, offset_w=False, cali_sf, B): { | | +| W_quantized, sf_w_list, dual_list = quantize_weights(W, mse_thresh, offset_w=False) | | +| sf_w_list = optimize_sf(W, W_quantized, sf_w_list, Graph, cali_sf) | | + +**Table 285 – The pseudocode of INT4 parameter quantization data generation** + +| INT4 parameter quantization data generation | Descriptor | +|-------------------------------------------------------|------------| +| B_quantized=B | | +| return W_quantized, sf_w_list, B_quantized, dual_list | | +| } | | + +The former part of INT4 parameter quantization data generation is weight quantization, the pseudocode of the quantize\_weights() operator is defined in Table 286. + +**Table 286 – The pseudocode of quantize\_weights()** + +| quantize_weights() | Descriptor | +|-------------------------------------------------------------------------------------------------------------|------------| +| def quantize_weights(W,mse_thresh,offset_w=False,bit_w=4): { | | +| dual_list=[] | | +| sf_w_list=[] | | +| W_quantized=[] | | +| if offset: | | +| W-=min(W) | | +| for i,w in enumerate(W): { | | +| if len(w.shape)>2: | Conv layer | +| for k in range(len(w): | | +| w_quantized[k],scaling_factor[k]=quantize(w[k],bit_w=4,offset_w,\n                n_points=500) | | +| else: | FC layer | +| w_quantized,scaling_factor=quantize(w,bit_w=4,offset_w,n_points=500) | | +| w_recon=int4_recon_quantized_parameter(w_quantized,scaling_factor,\n            offset_w_list=None) | | +| mse=mean((w-w_recon)**2) | | +| if mse>mse_thresh: { | Dual INT4 | +| dual_list.append(True) | | +| if len(w.shape)>2: | Conv layer | +| for k in range(len(w): | | +| w_quantized[k],scaling_factor[k]=\n                        dual_quantize(w[k],bit_w=4,offset_w,n_points=30) | | +| else: | | +| w_quantized,scaling_factor=dual_quantize(w,bit_w=4,\n                    offset_w,n_points=30)} | FC layer | +| } | | +| else: { | | +| dual_list.append(False) | | +| } | | +| W_quantized.append(w_quantized) | | +| sf_w_list.append(scaling_factor) | | + +**Table 286 – The pseudocode of quantize\_weights()** + +| quantize_weights() | Descriptor | +|----------------------------------------|-------------------| +| } | | +| return W_quantized,sf_w_list,dual_list | | +| } | | + +In Table 286, the definition of int4\_recon\_quantized\_parameter() refers to dequantization clause in this document. The pseudocode of quantize() operator is defined in Table 287. + +**Table 287 – The pseudocode of quantize()** + +| quantize() | Descriptor | +|---------------------------------------------------------------|-------------------| +| def quantize(w,bit_w=4,offset_w,n_points=500): { | | +| if offset: { | | +| max_range_w = max(w) - min(w) | | +| range_max = 2. ** bit_w - 1. | | +| range_min = 0. | | +| start_alpha, end_alpha = 1e-16, 1.1 * max_range_w / range_max | | +| } | | +| else: { | | +| max_range_w = max(abs(w)) | | +| range_max = 2. ** (bit_w - 1) - 1 | | +| range_min = -2. ** (bit_w - 1) | | +| start_alpha, end_alpha = -1.1 * max_range_w / range_max, \ | | +| 1.1 * max_range_w / range_max | | +| } | | +| alpha_vec = linspace(start_alpha, end_alpha, n_points) | | +| res_vec=[] for i in range(len(alpha_vec)] | | +| for i,alpha in enumerate(alpha_vec): { | | +| wq=clip(round(w/alpha),range_min,range_max) | | +| res_vec[i]=max(abs(w-wq*alpha)) | | +| } | | +| scaling_factor=alpha_vec[argmin(res_vec)] | | +| w_quantized=clip(round(w/scaling_factor),range_min,range_max) | | +| return w_quantized,scaling_factor | | +| } | | + +The pseudocode of dual\_quantize() operator is defined in Table 288. + +**Table 288 – The pseudocode of dual\_quantize()** + +| dual_quantize() | Descriptor | +|-----------------------------------------------------------------------|-------------------| +| def dual_quantize(w,bit_w, offset_w, n_points): { | | +| if offset: { | | +| max_range_w=max(w)-min(w) | | +| range_max=2.**bit_w-1 | | +| range_min=0 | | +| w-=min(w) | | +| start_alpha, end_alpha = 1e-16, 1.1 * max_range_w / range_max | | +| } | | +| else: { | | +| max_range_w = max(abs(w)) | | +| range_max = 2. ** (bit_w - 1) - 1 | | +| range_min = -2. ** (bit_w - 1) | | +| start_alpha, end_alpha = -1.1 * max_range_w / range_max, \ | | +| 1.1 * max_range_w / range_max | | +| } | | +| alpha_vec = linspace(start_alpha, end_alpha, n_points) | | +| factors=[] | | +| for r0 in alpha_vec: | | +| for r1 in alpha_vec: | | +| factors.append([r0,r1]) | | +| res_vec=[[[]for i in range(len(factors))]] | | +| for i,fct in enumerate(factors): { | | +| wq=get_params_double_grid(w,fct,range_min,range_max,bit_w=4) | | +| res_vec[i]=mean((wq[0]*fct[0]+wq[1]*fct[1]-w)**2) | | +| } | | +| scaling_factor=factors[argmin(res_vec)] | | +| w_quantized=get_params_double_grid(w,fct,range_min,range_max,bit_w=4) | | +| return w_quantized,scaling_factor | | +| } | | + +In Table 202, get\_params\_double\_grid() operator is defined in clause of INT4 parameter quantization (clause 9.2.1.2.2.2). + +The remained part of INT4 parameter quantization data generation is about scale factor optimization. The pseudocode is defined in Table 289. + +**Table 289 – The pseudocode of optimize\_sf()** + +| optimize_sf() | Descriptor | +|-------------------------------------------------------------------------|-------------------| +| def optimize_sf(W,W_quantized,sf_w_list,Graph,cali_sf,epochs=25): { | | +| gamma=[] for i in range(len(sf_w_list)) | | +| W_recon=[] for i in range_len(sf_w_list) | | +| for l in range(len(sf_w_list)): | | +| W_recon[l]=int4_recon_quantized_parameter(W_quantized[l],sf_w_list[l],\ | | +| offset_w_list=None) | | +| W.trainable=False | | +| W_recon.trainable=False | | +| gamma.trainable=True | | +| optimizer=RMSprop(gamma*W_recon) | | +| for epoch in epochs: { | | +| for batch in cali_sf: { | | +| out_quantized_gamma=Graph(batch,gamma*W_recon) | | +| out_orig=Graph(batch,W) | | +| loss=mean((out_quantized_gamma-out_orig)**2) | | +| loss.backward() | | +| optimizer.step() | | +| } | | +| } | | +| for l in range(len(sf_w_list)): | | +| sf_w_list[l]=gamma[l]*sf_w_list[l] | | +| return sf_w_list | | +| } | | + +###### **9.2.3.5.1.3 Interval shrinkage quantization data generation** + +Interval shrinkage quantization refers to the process where the weights in a single interval gradually approaching to the quantization centre points during the training phase. This method is mainly used for low-bit quantization. The definition of interval shrinkage quantization is shown in Table 290. + +**Table 290 – Definition of interval shrinkage quantization data generation** + +| Operator | Description | Field | Keyword | Definition | Data type | Format | +|------------------------|---------------------------------------------------------------------------------------------|--------------|----------------|-------------------------------|------------------|---------------| +| Shrink_qu
ntization | Transfer the discrete high precision network parameters to quantized parameters by interval | Input | W | Weight tensor to be quantized | List [tensor] | float32 | +| | | | B (optional) | Layer bias tensor | List [tensor] | float32 | +| | | | Graph | Network graph | function | / | + +**Table 290 – Definition of interval shrinkage quantization data generation** + +| Operator | Description | Field | Keyword | Definition | Data type | Format | +|----------|-------------------------|--------|------------------------|------------------------------------|---------------|---------| +| | shrinkage quantization. | | num_centers | Number of quantization centres | Value | int | +| | | | method | Method to get quantization centres | Value | int | +| | | | optimizer | Network optimizer | class | / | +| | | | X | Training data | List [tensor] | float32 | +| | | | Y | Training label | List [tensor] | int | +| | | | get_center | User method | function | / | +| | | Output | W_quantized | Quantized weight tensor | List [tensor] | float32 | +| | | | B_quantized (optional) | Quantized layer bias tensor | List [tensor] | float32 | + +For the weight of each layer that needs to be quantized, according to its granularity, the quantization process based on interval shrinkage is as follows. + +a) Clustering: + +- 1) If the *method* is equal to 0, use the K-Means method for clustering, referring to the K-Means data generation process to obtain the lookup table value $W\_v$ of size $num\_centers$ , and return the lookup table index value $W\_i$ . If the *method* equals to 1, use a user-defined method to generate the lookup table and lookup table index value $W\_i$ . +- 2) Sort according to the absolute value of $W\_v$ as the quantization order of different intervals. +- 3) The weight is divided into contraction interval S, retraining interval R and freezing interval F. + +b) Shrinkage: + +Shrinkage is the main process of interval quantization. For the weights in the shrinkage interval, the L1 loss is used to make them close to the quantization centre during training. And as the training progresses, the weights in the interval are gradually assigned to the interval centre. The flow chart of L1 loss calculation in this process is shown in the Figure 11. + +![Figure 11: L1 loss calculation diagram. The diagram illustrates the iterative process of quantizing neural network weights. It starts with a 'Quantization Function for W_ij' graph showing a step function with centers C = {-0.45, 0, 0.45}. A 3x3 weight matrix is divided into intervals: green (shrinkage), blue (frozen), and yellow (retraining). The process involves 'Divide intervals', 'Get the interval centres', 'Computing L1 loss' (shown in a graph), and 'Update weights'. The weight matrix evolves from an initial state to a more quantized state where many weights are set to the interval centers.](12280970bb3eea7f6fbdc08200d444c2_img.jpg) + +Divide intervals + +Update weights + +Quantization Function for $W_{ij}$ + +$C = \{-0.45, 0, 0.45\}$ + +Get the interval centres + +Computing $L_1$ loss + +F.748.53(25) + +Figure 11: L1 loss calculation diagram. The diagram illustrates the iterative process of quantizing neural network weights. It starts with a 'Quantization Function for W\_ij' graph showing a step function with centers C = {-0.45, 0, 0.45}. A 3x3 weight matrix is divided into intervals: green (shrinkage), blue (frozen), and yellow (retraining). The process involves 'Divide intervals', 'Get the interval centres', 'Computing L1 loss' (shown in a graph), and 'Update weights'. The weight matrix evolves from an initial state to a more quantized state where many weights are set to the interval centers. + +**Figure 11 – L1 loss calculation** + +In Figure 11, the green part is the weights in the shrinkage interval, the blue part is the weights in the frozen interval, and the yellow part is the weights in the retraining interval. For the weights in the shrinkage interval of each layer, L1 loss is used, as shown in formula (37): + +$$L_1 = \sum |w - c| \quad \text{if } w \in S \quad (37)$$ + +Here: + +$w$ neural network model weights; + +$c$ quantization centre of the shrinkage part at the current stage; + +$S$ the range of the shrinkage interval. + +During the training process, the weight is assigned to the quantization centre value from near to far, based on the absolute distance between the weights in the interval and the quantization centre, as shown in formulas (38)-(41). + +$$e_{R0} = s|c - e_0| \quad (38)$$ + +$$e_{R1} = s|c - e_1| \quad (39)$$ + +$$Q = [c - e_{R0}, c + e_{R1}] \quad (40)$$ + +$$w_s(p, q) = \begin{cases} c, & \text{if } w \in Q \\ w(p, q), & \text{if } w \notin Q \end{cases} \quad (41)$$ + +Here: + +$e_0, e_1$ The end point of the shrinkage interval $S$ ( $e_0 < e_1$ ), that is, the midpoint of the quantization centers of two adjacent intervals; + +$e_{R0}, e_{R1}$ The distance between two shrinkage interval endpoints during the quantization process; + +$s$ Scale factor ( $0 \leq s \leq 1$ ); + +$c$ Quantization centre of the shrinkage interval at the current stage; + +$Q$ The range of the shrinkage quantization part; + +$w(p, q)$ Original network model, weight value located at position $(p, q)$ ; + +$w_s(p, q)$ After shrinkage operator, weight value located at position $(p, q)$ . + +For the quantized weights, the gradients of the backpropagation are set to zero, which means that the weight value will not change as the training progresses, so as to achieve the freezing effect. + +###### a) Retraining + +Retraining is an important method to compensate for the loss caused by quantization. For the weights in the retraining interval, normal training methods are used, such as cross-entropy loss, as defined in formula (42): + +$$L_R = L_{CE}(f(x^{(i)}; \theta), y^{(i)}) \quad (42)$$ + +Here: + +$x^{(i)}$ The data of the i-th batch; + +$y^{(i)}$ The label of the i-th batch; + +$\theta$ Neural network model weights; + +$f$ Neural network forward process; + +$L_{CE}$ Cross-entropy loss function; + +$L_R$ Retraining loss; + +In order to prevent the value of the retrained weights from being in the frozen interval after backpropagation, the value is restricted so that it cannot exceed the range of the retraining interval. + +###### b) Total loss and training + +The total loss is the cross-entropy loss of the category and the L1 weight loss. + +Select $t$ samples $\{x^{(1)}, x^{(2)}, \dots, x^{(t)}\}$ from the neural network training dataset for network learning optimization, where $x^{(i)}$ 's corresponding target is $y^{(i)}$ ; + +Calculate the classification error loss $L_R$ of the neural network. Then calculate the totalized loss $L_S$ . Lastly, add them together with a certain proportion to get the total loss $L$ , as shown in formula (43): + +$$L = L_R + L_S = L_{CE}(f(x^{(i)}; \theta), y^{(i)}) + \alpha \cdot \sum L_{1l} \quad (43)$$ + +Here: + +$x^{(i)}$ The data of the i-th batch; + +$y^{(i)}$ The label of the i-th batch; + +$\theta$ Neural network model weights; + +$f$ Neural network forward process; + +$L_{CE}$ Cross-entropy loss function; + +$L_S$ Quantization loss; + +$L_R$ Retraining loss; + +$\alpha$ Quantization loss factor; + +$L_{1l}$ The L1 loss of the l-th layer; + +$L$ The total loss. + +Calculate the gradient using formula (44) chain rule: + +$$g = \frac{1}{t} \nabla_{\theta} \sum_i^t L_i \quad (44)$$ + +Here: + +- $t$ The number of data in one batch; +- $L_i$ The total loss of the i-th batch of data; +- $\nabla_{\theta}$ Partial derivative of weight; +- $g$ Gradient of weight. + +Update the weight parameters: + +$$\theta = \theta - \epsilon g \quad (45)$$ + +Here: + +- $\theta$ Neural network model weights; +- $\epsilon$ Learning rate. + +The pseudocode of interval shrinkage quantization is shown in Table 291. + +**Table 291 – The pseudocode of Interval shrinkage quantization** + +| Interval shrinkage quantization | Descriptor | +|------------------------------------------------------------------------------------------------------|------------| +| def shrink_weight(weight, num_center, method, alpha, d_s, X, Y, optimizer, graph, get_center=None):{ | | +| S, R, F = zeros(shape(weight)), ones(shape(weight)), zeros(shape(weight)) | | +| for i in range(0, num_center):{ | | +| if (method == 0):{ | | +| C_i = k_means(weight, num_center) | | +| c_arg = argsort(abs(C_i)) | | +| } | | +| elif (method == 1):{ | | +| C_i = get_center(weight, num_center) | | +| c_arg = argsort(abs(C_i)) | | +| } | | +| e_0, e_1 = get_endpoints(C_i, i) | | +| s = 0 | | +| c = C_i[c_arg == i] | | +| S = weight[weight>e_0 and weight c - e_R0 and S < c - e_R1] | | +| weight[M_f] = c | | +| y_p = graph(x) | | +| l_c = compute_loss(y_p, y, loss) | | +| l_all = l_c + alpha * L1 | | +| w_g = compute_gradient(l_all, [weight[R], S-weight[M_f]], optimizer) | | +| weight = weight + w_g | | + +**Table 291 – The pseudocode of Interval shrinkage quantization** + +| Interval shrinkage quantization | Descriptor | +|---------------------------------|------------| +| $s = s + d\_s$ | | +| $F = F + R$ | | +| } | | +| } | | +| } | | +| } | | + +##### 9.2.3.5.2 Activate quantization + +###### 9.2.3.5.2.1 Data generation for INT4 activate quantization + +When using INT4 quantization to configure a neural network, the compression process of the neural network model includes weight quantization, quantization scale factor correction and activation quantization. The output of the neural network is calculated according to the weight tensor and activation tensor after quantization. The corresponding parameter quantization data generation mode for INT4 activation quantization data generation is INT4 parameter quantization data generation. + +For the activation of the neural network, INT4 quantization is performed layer by layer: + +- a) The calibration set is obtained. By minimizing the error between the original activation tensor and the reconstructed activation tensor on the calibration set, the INT4 quantization is performed, and the corresponding activation quantization scale factor is determined. Make the calibration set size be $M$ , $A_i^l$ be the activation tensor of layer $l$ corresponding to the $i$ -th sample. The calculation method for the first scaling factor is shown in equation (46): + +$$\beta_{l,1} = \underset{\beta}{\operatorname{argmin}} \left\{ \sum_{i=1}^M \|A_i^l - \beta[A_i^l/\beta]\|^2 \right\} \quad (46)$$ + +Here, + +$[\cdot] \rightarrow [\cdot]_{\mathbb{Z}_q}$ , + +$\mathbb{Z}_q$ —Scope of quantization. + +- b) When the weight quantization is obtained, the network layer for multiple INT4 quantization is determined, usually dual INT4 quantization. The second scaling factor is obtained from the quantized residuals. The calculation method is shown in equation (47): + +$$\beta_{l,2} = \underset{\beta}{\operatorname{argmin}} \left\{ \sum_{i=1}^M \left\| A_i^l - \beta_{l,1}[A_i^l/\beta_{l,1}] - \beta \left[ \frac{A_i^l - \beta_{l,1}[A_i^l/\beta_{l,1}]}{\beta} \right] \right\|^2 \right\} \quad (47)$$ + +Here, + +$[\cdot] \rightarrow [\cdot]_{\mathbb{Z}_q}$ , + +$\mathbb{Z}_q$ —Scope of quantization. + +The data definition of INT4 activation quantization data generation is shown in Table 292. + +**Table 292 – Int4 activate quantization data to generate data definition** + +| Operator | Description | Field | Keyword | Definition | Data type | Format | +|------------------------------|-----------------------------------------------------------------|-----------|-------------|-------------------------------------------------------------|--------------|----------------------| +| int4_quantization_activation | The neural network is quantized layer by layer without training | Input | X | Activation tensor to be quantified | tensor | float32 | +| | | | Graph | network diagram | function | / | +| | | | dual_list | Dual INT4 quantization identification list of network layer | list | Boolean | +| | | | cali_act | Activate quantization calibration set | tensor | float32 | +| | | | offset_act | Activate quantization bias flag | value | Boolean | +| | | Attribute | bit_w | Quantization Bits | value | int | +| | | | n_points | Number of linear search samples | value | int | +| | | Output | X_quantized | Activation tensor after quantization | list[tensor] | signed/unsigned int4 | +| | | | sf_act_list | Activate the list of quantitative scaling factors | list | float32 | + +The pseudocode for INT4 to activate the quantization data generation operation is shown in Table 293. + +**Table 293 – Int4 activates the pseudocode of quantization data generation operation** + +| INT4 activates the quantization data generation operation | Descriptor | +|----------------------------------------------------------------------------------------|------------| +| def int4_quantization_activation(X,offset_act,dual_list,cali_act, bit_w=4,n_points): { | | +| sf_act_list=[] | | +| for l in range(cali_act.layers): { | | +| x=cali_act[l] | | +| x_quantized,sf=cal_sf(x,offset_act,dual_flag=False, bit_w=4,n_points) | | +| x_recon=int4_recon_quantized_activation(x_quantized,sf,offset_act_list=None) | | +| if dual_list[l]==True: { | | +| residual_x=x-x_recon | | +| residual_quantized,residual_sf=cal_sf(residual_x,offset_act,\ | | +| dual_flag=True, bit_w=4,n_points) | | +| x_quantized.append(residual_quantized) | | +| sf.append(residual_sf) | | + +**Table 293 – Int4 activates the pseudocode of quantization data generation operation** + +| INT4 activates the quantization data generation operation | Descriptor | +|---------------------------------------------------------------------------------------------------------|------------| +| } | | +| sf_act_list.append(sf) | | +| } | | +| X_quantized=[] | | +| for i,x in enumerate(X): { | | +| x_quantized=int4_quantize_activation(x,sf_act_list[i],offset_act_list\
=None,bit_w=4,x_quantized=[]) | | +| X_quantized.append(x_quantized) | | +| } | | +| return X_quantized,sf_act_list | | +| } | | + +Among them, int4\_recon\_quantized\_Activation() operates INT4 to activate the inverse quantization section. For the operation of int4\_quantize\_activation (), please refer to the clause of activation quantization in INT4. The pseudocode of Cal\_SF () operation is shown in Table 294. + +**Table 294 – cal\_sf() operation pseudocode** + +| cal_sf() operator | Descriptor | +|-------------------------------------------------------------------------------------|------------| +| def cal_sf(x, offset_act, dual_flag, bit_w=4, n_points): { | | +| if offset_act: { | | +| max_range_x=max(x)-min(x) | | +| range_min=0 | | +| range_max=2**bit_w-1 | | +| beta_vec = linspace(0.1 *max_range_x/range_max, max_range_x/range_max,
n_points) | | +| } | | +| else: { | | +| max_range_x=max(abs(x)) | | +| range_max = 2. ** (bit_w - 1) - 1 | | +| range_min = -2. ** (bit_w - 1) | | +| if dual_flag: | | +| beta_vec = linspace(-max_range_x/range_max, max_range_x/range_max, 2 *
n_points) | | +| else: | | +| beta_vec = linspace(-max_range_x/range_max, max_range_x/range_max, 2 *
n_points) | | +| } | | +| res_vec=[][for i in range(beta_vec)] | | +| for i, beta in enumerate(beta_vec): | | +| res_vec[i]=(clip(round(x/beta),range_min,range_max)-x)**2 | | + +**Table 294 – cal\_sf() operation pseudocode** + +| cal_sf() operator | Descriptor | +|----------------------------------------------------------------|------------| +| scaling_factor=beta_vec[argmin(res_vec)] | | +| x_quantized=clip(round(x/scaling_factor),range_min,range_max)\ | | +| return x_quantized, scaling_factor | | +| } | | + +###### 9.2.3.5.2.2 Trainable alpha quantization training data generation + +To parameterize the truncated value, the parameters shall be initialized first. In order to make the parameters converge faster, statistically estimation of the maximum value of the quantization layer is needed based on the pre-trained model and the given training data, according to the granularity requirements. After a period of training, the training truncated value of each granularity can be obtained, and this value can be directly used in the inference process, as described above. The algorithm relies on pre-trained models and training data, and can process all convolutional layers and fully connected layers that need to be quantized in parallel. The algorithm flow chart is shown in Figure 12. + +![Algorithm flow chart showing the steps: Original model -> EMA statistics -> Alpha and model parameter initialization -> Quantized model training -> Alpha is quantized truncated value -> Quantized model. A label F.748.53(25) is present below the last step.](b0a5d736df5f6fe2c0c6dd800bb0ff9a_img.jpg) + +``` + +graph LR + A[Original model] --> B[EMA statistics] + B --> C[Alpha and model parameter initialization] + C --> D[Quantized model training] + D --> E[Alpha is quantized truncated value] + E --> F[Quantized model] + F --- F7485325[F.748.53(25)] + +``` + +Algorithm flow chart showing the steps: Original model -> EMA statistics -> Alpha and model parameter initialization -> Quantized model training -> Alpha is quantized truncated value -> Quantized model. A label F.748.53(25) is present below the last step. + +**Figure 12 – Algorithm flow chart** + +The training procedure is listed as follows. + +- a) Given a pre-trained model, the value range of the layer that needs to be quantized in the network is counted to obtain the initial value of the training truncation parameter Alpha, according to the granularity requirement with the statistical method of EMA. The specific statistical method is shown in the formula (48): + +$$\alpha_t^l = \beta \cdot \alpha_{t-1}^l + (1 - \beta) \cdot \text{mean}(\max(X_1^l), \max(X_2^l), \dots, \max(X_m^l)) \quad (48)$$ + +Here, + +$t$ denotes the iteration times of a mini-batch, + +$l$ denotes the quantized tensor instructed by granularity requirement, + +$X$ denotes the output of a convolution layer, + +$\alpha$ denotes the trainable truncated value, + +$\beta$ denotes the smoothing coefficient. + +- b) Use the initial value of Alpha in step a) to assign a value to Alpha. +- c) Start training with the given training epoch number, quantization bit number, Alpha and pre-training model. +- d) Repeat step c) until the Loss of the network no longer changes (convergence), then go to step e) +- e) Obtain Alpha as the truncated value and generate quantization scale with formula $X\_Scale = \frac{2^{n-1}-1}{Alpha}$ . + +The data definition of trainable alpha quantization data generation can be found in Table 295. + +**Table 295 – Definition of trainable alpha quantization data generation** + +| Operator | Description | Field | Keyword | Definition | Type | Format | +|--------------------|---------------------------------------------------------------------------------------------------------|--------|-------------|------------------------------|---------------|--------| +| Alpha_quantization | Convert the static statistical quantization truncated value to a trainable quantization truncated value | Input | X | Input number to be quantized | List [tensor] | NS | +| | | | M | Pre-trained model | List [tensor] | NS | +| | | | Gamma | EMA constant | Value | Float | +| | | | Epoch | Training epoch number | Value | Int | +| | | | Granularity | Quantization granularity | Value | int | +| | | | N | Quantization bit number | Value | int | +| | | Output | X_Scale | Quantization scale factor | List [tensor] | NS | + +The psuedo-code of trainable alpha quantization data generation can be found in Table 296. + +**Table 296 – Psuedo-code of training alpha quantization data generation** + +| Trainable alpha quantization data generation | Symbol | +|-----------------------------------------------------------------------|--------------------------------------------------------------------------------------------------------| +| def Alpha_quantization (M, X, Epoch, Gamma, Alpha, Granularity, N) :{ | | +| for epoch in range(epoch): | | +| if (epoch==0): | | +| Alpha = EMA(X,Gamma) | EMA=function 1 | +| else: | | +| M.forward(Alpha,X, Granularity ,N) | In the forward process of M, Alpha based quantization and inverse quantization operations are inserted | +| X_Scale =(2**(n-1) -1)/Alpha | | +| return X_Scale | | +| } | | + +#### 9.2.3.6 Pruning + +##### 9.2.3.6.1 Sparse data generation method + +The process of structure sparsity is as follows. + +Input parameters: A four-dimensional tensor X, which represents the training set of the input network; a two-dimensional tensor Y, where X.tensorshape[0] == Y.tensorshape[0]; the definition graph of the network model; a weight tensor list of W; the layer-wise sparsity hyper-parameter lambda; and the corresponding sparsity requirement epsilon; + +Given the input data X, Y and W with the corresponding network structure and the weight W, the aim is to achieve a high row sparsity of W as high as possible, while losing the least performance of the corresponding dataset. In particular, two auxiliary variables F and K are introduced, where the size of the two is the same. The optimal value of W, F and K is hopefully solved when the input data are X and Y with respect to the following loss function: + +$$\mathcal{L}_{W,F,K}(X, Y) = \mathcal{L}_{X,Y}(W) + \sum_{l=1}^L \lambda \|F^l\|_{2,1} + \sum_{l=1}^L \text{trace}(K^l(W^l - F^l)) + \sum_{l=1}^L \frac{1}{2} \|W^l - F^l\|_F^2 \quad (49)$$ + +Here: + +$\mathcal{L}_{X,Y}(W)$ The per-layer softmax loss of corresponding network with input data X,Y + +$W^l$ The network weight of $l$ -th layer + +$F^l, K^l$ The auxiliary variable of $l$ -th layer + +$X, Y$ Input data + +We optimize corresponding $W^l, F^l, K^l$ , respectively, + +- a) Initialize $K=0, F=W$ +- b) Sample a batch $x, y$ in input data $X$ and $Y$ +- c) For the network weights $W$ , after deleting items unrelated to $W$ , $K, F$ can be fixed. Introduce intermediate matrix $T^l = F^l - \frac{1}{2}K^l$ , directly leverage $\mathcal{L}_{x,y}(W^l) + \frac{1}{2}\|W^l - T^l\|_F^2$ , input compute\_loss to calculate the loss function, and use compute\_grade to obtain the gradient of $W^l$ for gradient updating: +- d) For the $F$ matrix, fix $W$ and $K$ , introduce intermediate matrix $U^l = W^l + \frac{1}{2}K^l$ , use equation (50) is used directly to update each row of $F$ : +- e) +$$F_i^{l*} = U_i^l (\max\{\|U_i^l\|_2 - \lambda, 0\}) / \|U_i^l\|_2 \quad (50)$$ +- f) where, +- g) $i$ – Row index of $F$ matrix, that is, $F_i$ corresponds with the $i$ -th row of $F$ +- h) $F^l$ – Intermediate variable of the $l$ -th layer +- i) $U_i^l$ – Calculating intermediate matrix of the $l$ -th layer +- j) After getting new $F$ and $W$ , the update manner of $K$ is: $K^{l\{k+1\}} = K^{l\{k\}} + W^l - F^l$ +- k) When $\|W^{l\{k+1\}} - F^{l\{k+1\}}\|_F \leq \varepsilon$ , or $\|F^{l\{k+1\}} - F^{l\{k\}}\|_F \leq \varepsilon$ , $l=l+1$ , return a), else return b) +- l) When traversed all layers, jump out of the loop and end the algorithm. + +The definition of operator can be referred in Table 297. + +**Table 297 – Structure pruning data generation** + +| Compute operator | Description | Field | Keyword | Definition | Type | Data format | +|-----------------------------------|-----------------------------------------------------------------------------|--------|-----------|----------------------------|--------------|-------------| +| Structure pruning data generation | Given the training set and network structure, generate sparse data and mask | Input | X | Input data | tensor | NS | +| | | | Y | Input label | vector | NS | +| | | | graph | Network structure | graph | | +| | | | optimizer | Optimizer | optimizer | | +| | | | W | Weight tensor | list | NS | +| | | | lambda | Layer-wise hyper-parameter | vector | NS | +| | | | epsilon | Sparsity level | value | FLOAT | +| | | Output | W | Weight tensor | list | NS | +| | | | F | Mask | list | NS | + +The pseudocode can be referred in Table 298. + +**Table 298 – Row sparse data generation** + +| Row sparse data generation | Descriptor | +|-------------------------------------------------------------------------------|------------| +| Row_sparse(X,Y,graph,W,optimizer,lambda,epsilon){ | | +| List[tensor] K=0; | | +| List[tensor] F = W; | | +| for (i=0;i] | NS | +| | | | Sparsity | Target sparsity of each layer | list[float] | NS | +| | | | Target_reg | Regularization threshold | value | FLOAT | +| | | | Reg_increase | Regularization increment | list[vector] | NS | +| | | | Iter | Regularization increment adjustment interval | value | INT | +| | | | X | Input data | tensor | NS | +| | | | Y | Input label | vector | NS | +| | | | dtype | Data type | ValueType | | +| | | | granularity | Pruning granularity | value | INT | +| | | Output | W_2d_mask | Mask matrix | list[tensor] | NS | +| | | | B_mask | Mask tensor | list[vector] | NS | + +The definition of regularization increment function can be referred in Table 300. + +**Table 300 – Definition of regularization increment function** + +| Compute operator | Description | Field | Keyword | Definition | Type | Data format | +|------------------|-----------------------------------|--------|--------------|--------------------------|---------------|-------------| +| Reg_func | Regularization Increment Function | Input | W_2d | Weight matrix | tensor | NS | +| | | Output | Reg_increase | Regularization Increment | vector | NS | + +The parameter zeroing operator can be referred in Table 301. + +**Table 301 – Parameter zeroing operator** + +| Compute operator | Description | Field | Keyword | Definition | Type | Data format | +|------------------|-----------------------------------------------------------------------------|--------|----------|-------------------------|---------------|-------------| +| Offer_mask | Parameter zeroing operator. Given mask, set the value of parameter to zero. | Input | A | Parameter matrix | tensor | NS | +| | | | A_mask | Mask matrix | tensor | NS | +| | | | B | Parameter tensor | vector | NS | +| | | | B_mask | Mask tensor | vector | NS | +| | | Output | A_sparse | Sparse parameter matrix | tensor | NS | +| | | | B_sparse | Sparse parameter tensor | vector | NS | + +Incremental regularization pruning only supports the pattern of granularity 3. All input weight tensors (assume n) are column-pruned in a layer-wise manner. The process of incremental regularization operator ID\_Reg is defined as follows: + +- a) Expand the weight parameter of each layers by reshape(N,C\*H\*W) to obtain a two-dimensional weight matrix, where the row is N and the column is C\*H\*W. Divide the C\*H\*W column group according to the number of columns. +- b) Use the incremental regularization operator to set the related hyper-parameters Sparsity, Target\_Reg and Reg\_func. Obtain the parameter W\_2d and B from each layer of the model and leverage ID\_Reg to prune the model at the row dimension. Obtain the mask W\_2d\_mask, B\_mask at the output and the new regularization value Reg\_val. +- c) Use Offer\_mask to prune the weight of each layer, and obtain the sparse weight W\_sparse and B\_sparse. Then, use the transform operator reshape(N, C, H, W) to recover the desired tensor format of the training operator. +- d) Update the model parameter using the dataset. + +The algorithm of incremental regularization pruning is defined as follows: + +- Step 1: Prepare the pre-trained model, network structure, training configuration file, dataset, and set the target sparsity level and pruning hyper-parameters Sparsity, Target\_Reg and Reg\_func. +- Step 2: Initialize the training process and initialize the model parameters with pre-trained weights. +- Step 3: Check the sparse level of the current layer. If the sparsity level has reached the target rate, jump to step 7. Otherwise, continue to the next step. + +- Step 4.1: Use the gradient descent method to train $m$ iterations iteratively. +- Step 4.2: Sort the weight column of each layer with respect to L1-norm, average the rank of the first $m$ iterations. Sort the average rank again to get the importance order of each column parameter group in each layer. +- Step 4.3: Use the importance rank of column parameter groups as the input of the incremental regularization function, use the output of the function as the regularization increments and update it to each column parameter group to obtain $\text{Reg\_val}$ . +- Step 4.4: Judge if the updated regularization value is larger than the threshold. If it is larger than the regularization threshold, set the corresponding column in the mask matrix to zero. Otherwise, do not process. +- Step 4.5: Use $\text{Offer\_mask}$ to delete weight by multiplying the mask matrix with the weight matrix. +- Step 5: Add the new regularization value to the loss function. +- Step 6: In the back propagation phase, use $\text{compute\_gradient}$ to calculate the gradient per-layer, multiply the sparse mask and gradient, update the weight and return to step 4. +- Step 7: End the pruning, save the model, and retrain the obtained model. +- Step 8: Output the model with the highest accuracy in the re-training phase. +- The pseudocode of incremental regularization pruning is shown in Table 302. + +**Table 302 – Incremental regularization pruning** + +| Incremental regularization pruning | Descriptor | +|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------------| +| [1] $\text{ID\_Reg}(\text{W\_2d}, \text{B}, \text{Sparsity}, \text{Reg\_func}, \text{Iter}, \text{Target\_reg}, \text{X}, \text{Y}, \text{dtype}, \text{granularity})$ : { | | +| Initialize $\text{W\_2d\_mask}$ , $\text{B\_mask}$ with ones for each layer | | +| while (1) { | | +| for layer = 1 to $\text{NumLayer}$ { | | +| Update $\text{L\_sparsity}$ | | +| if ( $\text{L\_sparsity} < \text{Sparsity}[\text{layer}]$ ) | | +| continue | | +| else { | | +| if (all $\text{L\_sparse} \geq \text{Sparsity}$ ) { | | +| save model | | +| break while | | +| } | | +| else | | +| continue | | +| } | | +| Get $\text{W\_2d}[\text{layer}]$ , $\text{W\_2d\_mask}[\text{layer}]$ , $\text{B}[\text{layer}]$ , $\text{B\_mask}[\text{layer}]$ | | +| $\text{W\_sparse}$ , $\text{B\_sparse} = \text{Offer\_mask}(\text{W\_2d}[\text{layer}], \text{W\_2d\_mask}[\text{layer}], \text{B}[\text{layer}], \text{B\_mask}[\text{layer}])$ | | +| $\text{forward}(\text{W\_sparse}, \text{B\_sparse})$ | | +| } | | +| if ( $\text{iter} \% \text{Iter} == 0$ ) { | | +| for layer = 1 to $\text{NumLayer}$ { | | +| Get $\text{W\_2d}[\text{layer}]$ , $\text{W\_2d\_mask}[\text{layer}]$ from layer | | + +**Table 302 – Incremental regularization pruning** + +| Incremental regularization pruning | Descriptor | +|----------------------------------------------------------------------------------------------------------------|-------------------| +| for col = 1 to W_2d.cols { | | +| Reg_sort = sort(abs(W_2d[:, col], iter) | | +| Reg_increase = Reg_func(Reg_sort) | | +| Use Reg_increase to update regularizer and get Reg_val | | +| if (Reg_val >= Target_reg) | | +| Update W_2d_mask[layer] | | +| loss += Reg_val | | +| } | | +| } | | +| grad = compute_gradient(loss, val_list, optimizer, graph) | | +| for layer = NumLayer to 1 { | | +| Calculate W_2d_grad[layer], B_grad[layer] with grad | | +| W_grad_sparse, B_grad_sparse = Offer_mask(W_2d_grad[layer],
W_2d_mask[layer], B_grad[layer], B_mask[layer]) | | +| W_2d[layer] = W_2d[layer] – learning_rate* W_grad_sparse | | +| B[layer] = B[layer] – learning_rate * B_grad_sparse | | +| } | | +| } | | +| iter = iter + 1 | | +| } | | +| W_2d_mask_list = list[], B_mask_list = list[] | | +| for layer = 1 to NumLayer { | | +| Get W_2d_mask[layer], B_mask[layer] | | +| W_2d_mask_list.append(W_2d_mask[layer]) | | +| B_mask_list.append(B_mask[layer]) | | +| } | | +| return W_2d_mask_list, B_mask_list | | +| } | | +| | | +| [2] Retrain(X, Y, graph) { | | +| for(x,y in X,Y) { | | +| W_sparse, B_sparse = Offer_mask(W_2d, W_2d_mask, B, B_mask) | | +| y_pred = graph(W_sparse, B_sparse , x) | | +| loss = compute_loss(y,y_pred,loss=cross_entropy) | | +| grad = compute_gradient(loss, W_sparse, B_sparse , optimizer) | | +| Calculate W_2d_grad, B_grad with grad | | +| W_grad_sparse, B_grad_sparse=Offer_mask(W_2d_grad, W_2d_mask, B_grad, B_mask) | | +| W_2d[layer] = W_2d[layer] – learning_rate* W_grad_sparse | | +| B[layer] = B[layer] – learning_rate * B_grad_sparse | | + +**Table 302 – Incremental regularization pruning** + +| Incremental regularization pruning | Descriptor | +|-------------------------------------------|-------------------| +| } | | +| } | | +| | | +| [3] Output retrained model | | + +##### **9.2.3.6.3 Evaluation metric** + +The following are typical evaluation metrics for parameter importance: + +- 1) The value of weight. The simplest solution is to evaluate a parameter's importance based on its weight. Although this approach is the most straightforward and common, it frequently yields positive outcomes. Calculating the convolution kernel's p-norm and using that value as the metric is a typical approach. +- 2) The value of gradient. Another highly popular method uses the gradient to determine the importance of network parameters. After network training is complete, the gradient approach typically estimates the gradient accumulation of parameters on a small data set. Afterward, examine the gradient using Taylor decomposition to determine the parameter's importance. +- 3) High-level semantic information. There are some methods that quantify the importance of the neurons/convolution kernel using higher-order semantic information in addition to directly using the weight or gradient. These methods, which are typically not intuitive, assess the significance of each parameter using the specific attributes of convolutional neural networks in order to complete the pruning process. + +##### **9.2.3.6.4 Uniform/non-uniform pruning** + +There are now two distinct pruning processes, namely uniform pruning and non-uniform pruning. + +- 1) Uniform pruning means that each layer of the convolutional neural network is pruned in the same proportion. The advantage of this scheme is that each layer's pruning proportion is easy to obtain. Different layers' evaluation measures may have different orders. The obvious drawback of uniform pruning is that it is not flexible enough, and that different layers may not have equal importance. +- 2) Non-uniform pruning means that the proportion of each layer can be arbitrary. The advantage of this scheme is that it is flexible, and the theoretical upper limit is very high. The drawback is that it is exceedingly challenging to measure how important different layers' parameters are. + +##### **9.2.3.6.5 Training strategy** + +There are various ways to train a network once it has been pruned, and each method has advantages and disadvantages. + +- 1) The most popular technique is to fine-tune the pruned network by loading the corresponding parameters from the large network after pruning. Typically, a smaller learning rate and a shorter iteration are used for fine-tuning. +- 2) Another option is to only keep the network structure after pruning, throw away all of the pre-trained weights of the original network, and retrain the pruned network from scratch. It typically requires more training time to fully train the pruned network because it gets smaller. +- 3) Another highly popular approach is to distil the small network using the original large network when training the pruned network. Distillation is a very suitable method for + +enhancing the tiny network since the structure of the network before and after pruning is fairly comparable. + +#### 9.2.3.7 Structured matrix + +##### 9.2.3.7.1 Approach for generating data to be compressed in block circulant matrix with signed vectors + +To compress a model by applying signed vector and block circulant matrix techniques, the structure of weights in layers of the model under training needs to be restricted. Based on this process, weights can be achieved. The training can be performed by the following steps: + +- a) Identify the layer on which the signed vector and block circulant matrix techniques will be applied. +- b) For a given input to the network, apply the forward computation and finally achieve the loss value, in particular: + - 1) If the signed vector and block circulant matrix techniques are not applied to a layer, the forward is processed as normal. + - 2) If the signed vector and block circulant matrix techniques are applied, invoke node\_forward\_sign\_circ() for forward propagation: + - i. Compute disturbance input vector by multiplying the input vector and disturbance signed vector. + - ii. Decompress the block circulant matrix weight. + - iii. Return the multiplication of disturbance input vector and decompressed matrix weight. +- c) Based on the loss achieved via forward propagation, prepare the gradient updates, in particular: + - 1) If the signed vector and block circulant matrix techniques are not applied to a layer, the backpropagation and gradient update are processed normally. + - 2) If the signed vector and block circulant matrix techniques are applied, invoke node\_backward\_sign\_circ\_for\_w\_circ() for the inverse gradient update. + - i) Compute the gradient matrix of the expanded weight matrix based on the loss. + - ii) Augment the gradient values corresponding to the same element in the block circulant matrix and compute the average. + - iii) Compute the weights of the block circulant matrix and update, based on the average. +- d) Repeat steps b and c, train the neural network, update parameters until the network convergence or achieve other termination criteria. + +The operator is defined in Table 303. + +**Table 303 – Operator for updating the weights of the network using signed vector and block circulant matrix techniques** + +| Operator | Description | Parameters | Key words | Definition | Data types | Data formats | +|---------------------------------------|-------------------------------------------------------------------|------------|-----------|---------------------------|------------|--------------| +| Block_circu
lant_weight
_update | updating the weights of the network using signed vector and block | Input | x | Input sample for training | data | NS | +| | | | y | Truth label of sample | label | NS | +| | | | Graph | Network graph | Function | / | + +**Table 303 – Operator for updating the weights of the network using signed vector and block circulant matrix techniques** + +| Operator | Description | Parameters | Key words | Definition | Data types | Data formats | +|----------|-----------------------------|------------|-----------------|------------------------------------------------------------------------------------|---------------|--------------| +| | circulant matrix techniques | | Optimizer | Optimizer | function | / | +| | | | W_ori | Original weight tensor | tensor | NS | +| | | | B_ori | Original bias tensor | vector | NS | +| | | Attributes | layer_indicator | The index of the layer in a network using signed vector and block circulant matrix | list | Int | +| | | | sign_vector | Signed vector list | list [vector] | NS | +| | | | circ_size | Block circulant matrix shape list | list | Int | +| | | Output | W_structural | Updated weight tensor | tensor | NS | + +The pseudocode of the operator is shown in Table 304. + +**Table 304 – The pseudocode of updating the weights in a network using signed vector and block circulant matrix techniques** + +| Updating the weights in a network using signed vector and block circulant matrix techniques | Descriptor | +|------------------------------------------------------------------------------------------------------------------|------------| +| def Block_circulant_weight_update(x, y, Graph, Optimizer, W_ori, layer_indicator): { | | +| { | | +| layer_in = x | | +| for layer in range(1,NumLayer): { | | +| if layer_indicator[layer] == False: { | | +| layer_out = forward(layer_in, W_ori[layer], B_ori[layer]) | | +| } | | +| else: { | | +| layer_out = node_forward_sign_circ(layer_in, W_ori[layer], \ sign_vector[layer], circ_size[layer], B_ori[layer]) | | +| } | | +| layer_in = layer_out | | +| } | | +| Result = layer_out | | +| Loss = compute_loss(Result, y) | | + +**Table 304 – The pseudocode of updating the weights in a network using signed vector and block circulant matrix techniques** + +| Updating the weights in a network using signed vector and block circulant matrix techniques | Descriptor | +|----------------------------------------------------------------------------------------------------|-------------------| +| for layer in range(NumLayer ,1,-1): { | | +| if layer_indicator[layer] == False: { | | +| W_ori[layer], B_ori[layer] = backward(Loss[layer], W_ori[layer], B_ori[layer]) | | +| } | | +| else: { | | +| W_structural[layer] = node_backward_sign_circ_for_w_circ(Loss[layer], \ | | +| W_ori[layer], circ_size[layer], Optimizer) | | +| } | | +| } | | +| return W_structural, W_ori, B_ori | | +| } | | + +The definition and the pseudocode of node\_forward\_sign\_circ() are in Table 305 and Table 306 respectively. + +**Table 305 – Forward propagation operator for a given node** + +| Operator | Description | Parameters | Key words | Definition | Data types | Data formats | +|------------------------|-------------------------------------------------------------------------------------------------|-------------------|------------------|--------------------------------------------------------------------|-------------------|---------------------| +| node_forward_sign_circ | forward propagation through the layer using signed vector and block circulant matrix techniques | Input | x | Input vector of the given node | vector | NS | +| | | | w_circ | Weight of the node after compression | matrix | NS | +| | | | sign_vector | Disturbance signed vector corresponding to the given node | vector | NS | +| | | | circ_size | Size of the block circulant matrix corresponding to the given node | value | Int | +| | | | B(optional) | Bias vector | vector | NS | +| | | Output | output | Output of the given node | vector | NS | + +**Table 306 – The pseudocode of forward propagation through the layer using signed vector and block circulant matrix techniques** + +| Forward propagation through the layer using signed vector and block circulant matrix techniques | Descriptor | +|--------------------------------------------------------------------------------------------------------|-------------------| +| def node_forward_sign_circ(x, w_circ, sign_vector, circ_size, B): { | | +| sign_x = mul(x, sign_vector) | | + +**Table 306 – The pseudocode of forward propagation through the layer using signed vector and block circulant matrix techniques** + +| Forward propagation through the layer using signed vector and block circulant matrix techniques | Descriptor | +|-------------------------------------------------------------------------------------------------|------------| +| output = mul(sign_x, decompress_W_circ_to_weight(w_circ, circ_size)) | | +| if B: { | | +| output = output + B | | +| } | | +| return output | | +| } | | + +The definition and the pseudocode of node\_backward\_sign\_circ\_for\_w\_circ() are in Table 307 and Table 308, respectively. + +**Table 307 – Backward propagation operator for a given node** + +| Operator | Description | Parameters | Key words | Definition | Data types | Data formats | +|------------------------------------|--------------------------------------------------------------------------------------------------|------------|-----------|--------------------------------------------------------------------|--------------|--------------| +| node_backward_sign_circ_for_w_circ | backward propagation through the layer using signed vector and block circulant matrix techniques | Input | loss | Loss value corresponding to the given node | list[tensor] | NS | +| | | | w_circ | Weight of the node after compression | matrix | NS | +| | | | circ_size | Size of the block circulant matrix corresponding to the given node | value | Int | +| | | | Optimizer | Optimizer | function | | +| | | Output | w_circ | Updated compressed weight matrix | matrix | NS | + +**Table 308 – The pseudocode of the backward propagation for a given node** + +| Backward propagation through the layer using signed vector and block circulant matrix techniques | Descriptor | +|--------------------------------------------------------------------------------------------------|------------| +| def node_backward_sign_circ_for_w_circ(loss, w_circ, circ_size, Optimizer): { | | +| gradient_weight = compute_gradient(loss, Optimizer) | | +| height_ori, width_ori = gradient_weight.shape() | | +| for i in range(0, height_ori, circ_size): | | +| for j in range(0, width_ori, circ_size): { | | + +**Table 308 – The pseudocode of the backward propagation for a given node** + +| Backward propagation through the layer using signed vector and block circulant matrix techniques | Descriptor | +|--------------------------------------------------------------------------------------------------|------------| +| for k in range(circ_size): { | | +| gradient_sum = 0 | | +| for l in range(circ_size): { | | +| if(k+l > circ_size): | | +| shift = k+l-circ_size | | +| else: | | +| shift = k+l | | +| gradient_sum += gradient_weight[i+l][j+shift] | | +| } | | +| w_circ[i/circ_size][j] = w_circ[i/circ_size][j] - gradient_sum/circ_size | | +| } | | +| } | | +| return w_circ | | +| } | | + +##### 9.2.3.7.2 Approach for generating the weight in low rank sparse decomposed structured matrix + +###### 9.2.3.7.2.1 Overview + +The computation of hyper-parameters is needed, when performing model compression by using low rank sparse decomposed structured matrix. This can be achieved by computing initial weights and fine-tuning. + +###### 9.2.3.7.2.2 Approaches for determining hyper-parameter R1, R2, groups and core\_size + +Let $C_{out} \times C_{in} \times K \times K$ shape a given convolutional layer. The dimension volume of output is $C_{out}$ . The dimension volume of input is $C_{in}$ . The convolution kernel size is noted as $K$ . After performing a low rank sparse decomposition, the convolutional layer is decomposed into three convolutional layers with smaller sizes: $R1 \times C_{in} \times 1 \times 1$ , $R2 \times r1 \times K \times K$ and $C_{out} \times R2 \times 1 \times 1$ , where $r1 = R1/groups$ . + +The volume of *groups* can be from 1 to 4, and the sizes of $R1$ and $R2$ are proportional to $C_{in}$ and $C_{out}$ , respectively. The computation satisfies Formula 51: + +$$C_{out} \times C_{in} \times K \times K \times ratio = R1 \times C_{in} \times 1 \times 1 + R2 \times r1 \times K \times K + C_{out} \times R2 \times 1 \times 1 \quad (51)$$ + +in which: + +$C_{in}$ – number of input channels. + +$C_{out}$ – number of output channels. + +$K$ – the width and height of convolution kernel. + +*ratio* – compression ratio. + +$R1$ – rank of input channel. + +$R2$ – rank of output channel. + +In case of the fully connected layers and 1x1 convolutional layers in a network, the compression process results in two approximated tensors with the shapes: $W \times core\_size$ and $core\_size \times H$ . Formula 52 can be used for calculating hyper parameter $core\_size$ . + +$$W \times H \times ratio = W \times core\_size + core\_size \times H \quad (52)$$ + +Where: + +$W$ – number of input channels (for 1x1 convolutional layer) or neurons (for fully connected layer). + +$H$ – number of output channels (for 1x1 convolutional layer) or neurons (for fully connected layer). + +$ratio$ – compression ratio. + +$core\_size$ – matrix decomposition rank. + +###### 9.2.3.7.2.3 Process for the generation of weights of low rank sparse decomposed structured matrix + +The training process is defined as follows: + +- a) Determine hyper parameters manually. +- b) Prepare a pre-trained model, parameters for training and data set. +- c) Extract the parameters from convolutional layers and fully connected layers, then apply the decomposed compression to them. +- d) Set parameters to the compressed network with the compressed parameters out of step c) and the ones within pre-trained model. +- e) Adjust learning rate per epoch. Perform forward propagation, compute loss, compute gradient, perform backward propagation and update parameters. +- f) Repeat e) until the preset number of epochs is met. + +The operator for the parameter generation of low rank sparse decomposed structured matrix is defined in Table 309. + +**Table 309 – Definition of the operator for parameter generation of low rank sparse decomposed structured matrix** + +| Operator | Description | Parameters | Key words | Definition | Data types | Data formats | +|-------------------|-----------------------------------------------------------------------------------------------------------|------------|-------------------|-------------------------------------------------------|------------|--------------| +| BTD_update_weight | Update the weights during training, when applying low rank sparse decomposed structured matrix techniques | Input | x | Input sample for training | Data | NS | +| | | | y | Truth label of sample | Label | NS | +| | | | Graph | Network graph after compression | Function | / | +| | | | Optimizer | Optimizer | Function | / | +| | | | epoch | Epoch of training | value | Int | +| | | | Pretrained_weight | Weights of pre-trained model | Tensor | NS | +| | | | R1 | Rank of the input channel of each convolutional layer | List | Int | + +**Table 309 – Definition of the operator for parameter generation of low rank sparse decomposed structured matrix** + +| Operator | Description | Parameters | Key words | Definition | Data types | Data formats | +|----------|-------------|------------|------------|--------------------------------------------------------------|------------|--------------| +| | | | R2 | Rank of the output channel of each convolutional layer | List | Int | +| | | | groups | Number of groups for each convolutional layer | List | Int | +| | | | core_size | Number of the intermediate output of a fully connected layer | List | Int | +| | | Output | BTD_weight | Updated weight tensor | Tensor | NS | + +The pseudocode of this operator is in Table 303. + +**Table 310 – The pseudocode of weight generation of low rank sparse decomposed structured matrix** + +| Weight generation of low rank sparse decomposed structured matrix | Descriptor | +|--------------------------------------------------------------------------------------------------|------------| +| def BTD_update_weight(Pretrained_weight, Graph, R1, R2, groups, core_size, batch_size, epoch): { | | +| svd_idx = 0 | | +| hosvd_idx = 0 | | +| Pretrained_layers = Pretrained_weight.keys | | +| BTD_weight = {} | | +| for layer_idx in range(len(Pretrained_layer)): { | | +| key = Pretrained_layers[layer_idx] | | +| weight = Pretrained_weight[key].weight | | +| if (Pretrained_weight[key].type == "FC"): { | | +| w1, w2 = SVD_process(weight, core_size[svd_idx]) | | +| svd_idx = svd_idx+1 | | +| BTD_weight[key+"w1"].weight = w1 | | +| BTD_weight[key+"w2"].weight = w2 | | +| } | | +| elif (Pretrained_weight[key].type == "Conv") and shape(weight)[3] == 1: { | | +| w1, w2 = SVD_process(weight, core_size[svd_idx]) | | +| svd_idx = svd_idx + 1 | | +| BTD_weight[key+"w1"].weight = w1 | | +| BTD_weight[key+"w2"].weight = w2 | | + +**Table 310 – The pseudocode of weight generation of low rank sparse decomposed structured matrix** + +| Weight generation of low rank sparse decomposed structured matrix | Descriptor | +|-----------------------------------------------------------------------------------|--------------------------| +| } | | +| elif (Pretrained_weight[key].type == "Conv"): { | | +| w1, w2, w3 = BTD_process(weight, R1[hosvd_idx], R2[hosvd_idx], groups[hosvd_idx]) | | +| hosvd_idx = hosvd_idx + 1 | | +| BTD_weight[key+"w1"].weight = w1 | | +| BTD_weight[key+"w2"].weight = w2 | | +| BTD_weight[key+"w3"].weight = w3 | | +| } | | +| else: { | | +| BTD_weight[key].weight = weight | | +| } | | +| } | | +| for epoch_idx in range(epoch): { | | +| local_lr = adjust_learning_rate(optimizer, epoch_idx) | Learning rate adjustment | +| for ite in range(len(x)/batch_size): { | | +| batch_x = x[ite*batch_size:ite*batch_size+batch_size] | | +| batch_y = y[ite*batch_size:ite*batch_size+batch_size] | | +| out = Graph(batch_x, BTD_weight) | | +| loss = compute_loss(out, batch_y) | | +| gradient = compute_gradient(BTD_weight, loss, optimizer) | | +| update_weight(BTD_weight, gradient) | | +| } | | +| } | | +| return BTD_weight | | +| } | | + +# **10 AI model encapsulation** + +## **10.1 Encapsulation representation** + +### **10.1.1 Encapsulation representation structure of model distribution** + +This clause introduces the composition of model encapsulation and the definition of each component, where the structure is illustrated in the following figure. Specifically, model encapsulation can be divided into the following components, where the name and the explanations are listed as follows: + +- a) Model encapsulation representation: The model encapsulation representations with specific data structures. + +- b) Program files: It is divided into two parts. The first part is the library required by the model execution. The second part is the script, which describes how to run the model to obtain the final output. +- c) Model description files: Model information, attributes and the other corresponding descriptions. + +See also Figure 13. + +![Figure 13 – Model encapsulation structure diagram. The diagram shows a large rectangle labeled 'Model encapsulation'. Inside this rectangle, at the top, are two dashed boxes. The left dashed box is labeled 'Model' and contains a stack of three rectangles, the front-most labeled 'Model representation'. The right dashed box is labeled 'Program files' and contains two stacks of rectangles: one labeled 'Library' and the other 'Script'. Below these dashed boxes, inside the 'Model encapsulation' rectangle, is a single wide rectangle labeled 'Model description files'. A small label 'F.748.53(25)' is at the bottom right of the diagram.](7d1b7ee3543f5a1a9b7ad98572d52f85_img.jpg) + +Figure 13 – Model encapsulation structure diagram. The diagram shows a large rectangle labeled 'Model encapsulation'. Inside this rectangle, at the top, are two dashed boxes. The left dashed box is labeled 'Model' and contains a stack of three rectangles, the front-most labeled 'Model representation'. The right dashed box is labeled 'Program files' and contains two stacks of rectangles: one labeled 'Library' and the other 'Script'. Below these dashed boxes, inside the 'Model encapsulation' rectangle, is a single wide rectangle labeled 'Model description files'. A small label 'F.748.53(25)' is at the bottom right of the diagram. + +**Figure 13 – Model encapsulation structure** + +### 10.1.2 Format definition of encapsulation model + +Encapsulation model files need to follow the directory structure to distinguish the functionality of the different modules. + +The directory includes: + +- Model: Store the encapsulation model data. +- Program: Store the algorithms/library/script files. +- Meta-info: Store the property and technical information of the model. + +**Table 311 – Directory structure** + +| Directory Name | Necessity | Files/Subdirectory | | Description | +|----------------|-----------|----------------------------------------|-----------------|--------------------------------------------------------------| +| Model | Required | (Model files) | | AI models | +| Program | Optional | Library | | The library files for running the models | +| | | Script | | Scripts for controlling program start, stop and check status | +| Meta-info | Required | Subdirectory named by model identifier | Management info | Model property information description file | +| | | | Technical info | Model technical information description File | + +The model file can be either an AI model representation or a compact representation. For AI model representation, the user can refer to clause 8. For compact representation, the standard includes multiple models (in clause 9.1.1), quantization (in clause 9.1.2), pruning (in clause 9.1.3), pruning operator (in clause 9.1.4) and structured matrix (in clause 9.1.5). + +### 10.1.3 Model data structure + +The encapsulation model data structure includes the file header information, the model header information, and the data for the representation/compact representation of the model as described in clauses 8 and 9. + +An encapsulation model file has a unique file header and can contain multiple models. Where the model header corresponds to the model data one by one, in order after the file header. See Table 312. + +**Table 312 – The order of header and data** + +| File Header | Model Header | Model Data | Model Header | Model Data | +|-------------|--------------|------------|--------------|------------| +|-------------|--------------|------------|--------------|------------| + +The file header is defined in Table 313. + +**Table 313 – File header structure** + +| Field | Data type | Value | Definition | +|--------------|--------------|--------------|-------------------------------| +| Start_code | unsigned int | 0x5352434D | File header start code (SRCM) | +| Magic_number | unsigned int | 0x47D02F93 | File header magic number | +| Version | unsigned int | Customizable | Version of file | +| Model_number | unsigned int | Customizable | Number of models in the file | + +The model header is defined in Table 314. + +**Table 314 – Model header structure** + +| Field | Data type | Value | Definition | +|------------------------------|--------------|--------------|---------------------------------------------------------------------------------------------------------------------------------------------------| +| Start_code | unsigned int | 0x486F4D52 | Model header start code (HoMR) | +| Identifier | unsigned int | Customizable | Model header unique identifier. A model can be divided into multiple parts of data with the same identifier if the size is too large. | +| Check_sum | unsigned int | Customizable | 32-bit MD5 file checksum. | +| Residual updating identifier | unsigned int | Customizable | If the value is not 0x00000000, declare that the model data is for residual updating. The target model identifier should be same with this value. | +| Data size | unsigned int | Customizable | The size of the model data. | + +The model data includes the specific data needed to be encapsulated and transmitted. In this standard, the default content is to adopt the format of model representation and model compact representation described in clauses 8 and 9. + +In addition to the model encapsulated representation formats described in the above clauses, other standard data formats can also be used such as NNR [ISO/IEC 15938-17]. The user needs to manually modify the model\_data\_format field in Management info as detailed in Table 316. + +### 10.1.4 Model program structure + +The model program file mainly includes two parts: Library and scripts. The library mainly includes the environment dependency library which is named environment and the self-built algorithm library which is named core. The converter is another class of program files used to convert uniform model representations into formats specific to different platforms (e.g., Pytorch, PaddlePaddle, etc.) or devices (e.g., GPUs, CPUs, NPUs, etc.). Scripts include related scripts that support the use of mode, as shown in Table 315. Table 318 describes the model task. + +**Table 315 – Model program structure** + +| | Directory Name | Necessity | Files/Subdirectory | Description | +|---------|----------------|-----------|--------------------|-------------------------------------------------------------------------------------------------------------------------| +| Program | Library | Optional | environment | The environment mainly includes the environment dependency library, such as PyTorch, MXNet, numpy, pandas, Scikit-learn | +| | | | core | The core includes the self-built algorithm library such as data pre-processing, validation, visualization | +| | | | converter | The converter path contains tools for converting standard representations to different devices for deployment. | +| | Scripts | Optional | Run.py | Run algorithm | +| | | | Stop.py | Stop algorithm | +| | | | Status.py | Check algorithm running status | +| | | | Convert.py | Convert the representations | +| | | | [Else] | Other scripts supporting model | + +### **10.1.5 Model description files** + +The model description file is used to describe the attribute information and technical information of the model. It consists of two files: managementinfo.json and technicalinfo.json. Managementinfo.json stores the attribute information of the model, as shown in Table 316. Technicalinfo.json stores the technical information of the model, as shown in Table 317. + +**Table 316 – The attribute information of model** + +| Name | Necessity | Type | Description | +|---------------------|-----------|----------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| model name | Required | string | The name of the model, such as ResNet101 | +| model_size | Required | dict | Describe the size of the model. It includes two parts: a. model params Unit is MB or GB; b. model FLOPs, unit is MFLOP or GFLOPS. Example: {"params": "10MB", "FLOPs": "1GFLOPs"} | +| model_data_format | Optional | string | If this field is empty, the data format described in 10.1.3 is used by default. If you want to use other data formats such as NNR, you need to fill in this field and provide the corresponding program files. | +| model_task | Optional | string | Describe the task of the model, such as image_classification, object_detection, text_to_image, speech_to_text, machine_translation and style_transfer. Examples are shown in Table 311 | +| model_metrics | Optional | dict | The metrics of the model, such as precision, recall, F1-score, accuracy, etc. | +| model_license | Optional | string | Open source license for the model, such as GPL-2.0, MIT, etc. | +| algorithm_doc | Optional | string | Description of the model design and algorithm development, such as backbone, Activation function, etc. | +| scripts_description | Optional | string | Describe the functions, usage and parameter settings of scripts | + +**Table 317 – The technical information of model** + +| Name | Necessity | Type | Description | +|-------------------|-----------|----------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| model version | Required | unsigned int | The version of the model | +| model_requirement | Required | string | Minimum hardware requirements to support model operation, such as PU/GPU minimum requirements, memory minimum requirements, etc. example: "GPU:typeA,typeB-GPUMem:8G-CPU:typeC-Mem:8G" | +| model_env | Required | string | The basic running environment requirements of the model, including the operating system requirements (windows/Linux) and environment dependency requirements, such as python 3.8, Python 1.7, numpy1.5, etc. example "Ubuntu18.04-PYTHON3.7-PyTorch1.7". | +| model_inputs | Required | list[string] | The description of the input data requirements of the model are shown in Table 312 | +| model_outputs | Required | list | Explain the output mode of the model | +| model_config | Optional | dict | The configs of the model that can be modified during model operation, such as input and output path, threshold, output format setting, etc. | +| model_framework | Optional | string | Framework for model development, such as pytorch and tensorflow | + +**Table 318 – Model task** + +| Name | Necessity | Type | Description | +|----------------------|-----------|--------|-------------------------------------------------------------------------------------------------------------------| +| image_classification | Optional | string | Distinguish different features in the image according to the given category | +| image_segmentation | | | Divide the image into several specific areas with unique properties | +| image_style_transfer | | | Convert the image from the original style to another style, while ensuring that the image content does not change | +| ocr_recognition | | | Recognize the text in the image and return the text content | +| object_detection | | | Locate the target object in the image and give the bounding box | +| translation | | | Translate the input text into the specified language | +| text_to_image | | | According to the input text, generate pictures that match the description | +| other | | | Other tasks | + +**Table 319 – The description of the input data** + +| Name | Necessity | Type | Description | +|-------------------|-----------|--------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| input_type | Required | string | The type of the input data, such as image, text, video, speech, other, etc. | +| input_name | Optional | string | The name of input variate | +| input_requirement | Optional | string | Specific requirements for input data, including the file format of input data such as PNG, CSV and PNG, input data size limit, input image size requirements, etc. | + +## 10.2 Encryption of model encapsulation + +To prevent unauthorized third parties from accessing the model, the model can be protected using public encryption algorithms. Common encryption algorithms can be divided into two types, symmetric and asymmetric encryption algorithms. + +Symmetric encryption means that the same key is used for both encryption and decryption. The symmetric encryption algorithm is faster and requires less resources. This makes it suitable for model data. Common symmetric encryption algorithms include AES, DES, SM4 and so on. See Tables 320 and 321. + +**Table 320 – Encryption side field definition** + +| Field | Source | Type | Description | +|---------------|-----------------------|--------|------------------------------------------------------------| +| SymmetricKey | Generated | string | The Symmetric Key for model encryption. | +| PublicKey | Generated | string | The Public Key to encrypt the Symmetric Key. | +| EncryptionKey | Asymmetric Encryption | string | The Encryption Key generated by asymmetric encryption. | +| SM | N/A | string | Symmetric encryption method, such as 'AES', 'DES', 'SM4'. | +| AM | N/A | string | Asymmetric encryption method, such as 'RSA', 'DSA', 'SM2'. | + +**Table 321 – Decryption side field definition** + +| Name | Source | Type | Description | +|---------------|-----------------------|--------|----------------------------------------------------------------------------------| +| PrivateKey | Generated | string | the private key to decrypt the encryption key. | +| EncryptionKey | Encapsulated model | string | The encryption key generated by asymmetric encryption. | +| SymmetricKey | Asymmetric Decryption | string | The symmetric key for model decryption. | +| SM | N/A | string | Symmetric decryption method, such as 'AES', 'DES', 'SM4', same with encryption. | +| AM | N/A | string | Asymmetric decryption method, such as 'RSA', 'DSA', 'SM2', same with encryption. | + +Asymmetric encryption uses a public key and a private key, where the public key is used for encryption and the private key is used for decryption. The asymmetric encryption algorithm is more reliable, but the encryption and decryption efficiency is low and suitable for small files. Common asymmetric encryption algorithms include RSA, DSA, SM2 and so on. + +![Figure 14: Two-step encryption system flowchart. The process starts with 'Model data' (rounded rectangle) and 'Symmetric key' (hexagon) entering a 'Symmetric encryption' block. The output of this block and a 'Public key' (hexagon) enter an 'Asymmetric encryption' block. The final outputs are 'Encryption model data' (rounded rectangle) and 'Encryption key' (hexagon), which are grouped together in a box. A label F.748.53(25) is present at the bottom right.](d9862ac8473902e7dfe441c5e88d47d8_img.jpg) + +``` + +graph TD + MD[Model data] --> SE[Symmetric encryption] + SK{Symmetric key} --> SE + SE --> EM[Encryption model data] + PK{Public key} --> AE[Asymmetric encryption] + SE --> AE + AE --> EK{Encryption key} + subgraph Box [ ] + EM + EK + end + style Box fill:none,stroke:none + +``` + +F.748.53(25) + +Figure 14: Two-step encryption system flowchart. The process starts with 'Model data' (rounded rectangle) and 'Symmetric key' (hexagon) entering a 'Symmetric encryption' block. The output of this block and a 'Public key' (hexagon) enter an 'Asymmetric encryption' block. The final outputs are 'Encryption model data' (rounded rectangle) and 'Encryption key' (hexagon), which are grouped together in a box. A label F.748.53(25) is present at the bottom right. + +**Figure 14 – Two-step encryption system** + +To balance efficiency and security, a two-step encryption system can be applied. + +First, the model data is encrypted using a symmetric encryption algorithm to obtain the Encrypted Model Data. Then generate the encryption key by using the public key to encrypt the symmetric key by an asymmetric encryption algorithm. The encryption side field definition is shown in Table 320. + +The decryption process decrypts the Symmetric Key by the Asymmetric Decryption method with the Private Key and the Encryption Key. Then use the Symmetric Key and the Encryption Model Data to decrypt the model and get the original Model Data. The encryption side field definition is shown in Table 321. + +![Figure 15: Two-step decryption system flowchart. The process starts with 'Encryption model data' (rounded rectangle) and 'Encryption key' (hexagon) entering an 'Asymmetric decryption' block. A 'Private key' (hexagon) also enters this block. The output is a 'Symmetric key' (hexagon), which then enters a 'Symmetric decryption' block along with the 'Encryption model data'. The final output is 'Model data' (rounded rectangle). A label F.748.53(25) is present at the bottom right.](f5dc5aa74ea5482955a1e8e161b5c549_img.jpg) + +``` + +graph TD + EM[Encryption model data] --> AD[Asymmetric decryption] + EK{Encryption key} --> AD + PK{Private key} --> AD + AD --> SK{Symmetric key} + EM --> SD[Symmetric decryption] + SK --> SD + SD --> MD[Model data] + +``` + +F.748.53(25) + +Figure 15: Two-step decryption system flowchart. The process starts with 'Encryption model data' (rounded rectangle) and 'Encryption key' (hexagon) entering an 'Asymmetric decryption' block. A 'Private key' (hexagon) also enters this block. The output is a 'Symmetric key' (hexagon), which then enters a 'Symmetric decryption' block along with the 'Encryption model data'. The final output is 'Model data' (rounded rectangle). A label F.748.53(25) is present at the bottom right. + +**Figure 15 – Two-step decryption system** + +## 10.3 Residual updating of model encapsulation + +In this clause, the process of models of residual updating and transmission are introduced. + +Model residual updating is a procedure where models need to be updated between the client and cloud. Generally speaking, the new model can be generated at the cloud side and then distributed to the edge side, such that knowledge is propagated to facilitate better intelligent applications. The distributed generation of the deep learning models at the edge end also requires the models to be further transmitted and fused. For the same task or intrinsically similar tasks, these models typically share the same architectures. In these similar models, there exists high correlation among different models. Therefore, model residual updating technology can be used to remove the information redundancy between the information transmissions. + +The process of updating and transmission includes the following five steps: + +- a) Train the base model A and transmit the model from cloud end to edge end. +- b) Retrain and fine-tune the base model A according to the practical need to obtain the target model B. Follow the training process in clause 9.2.1.1.2. +- c) Compute the difference between the base model A and target model B, to obtain the model residual $\Delta = B - A$ . +- d) Apply residual quantization to the model residual to obtain quantized model residual $Q(\Delta)$ . +- e) Transmit the quantized model residual $Q(\Delta)$ from cloud end to edge end. The edge end decapsulates and matches the same node name to get the sharing part. +- f) The edge end adds the residual part to update model to B. + +![Figure 16 – Residual updating workflow diagram. The diagram is split into two sections by a dashed line: Server (top) and Client (bottom). On the Server side, 'AI model A' has three outgoing arrows: 'Sharing parts' to a circle labeled 'Subtraction', 'Retrain' to 'AI model B', and a dotted line to a 'Compact representation' box. 'AI model B' also points to the 'Subtraction' circle. The 'Subtraction' circle outputs to another 'Compact representation' box, which then points to a 'Residual model' box. On the Client side, the 'Residual model' box points to a circle labeled 'Addition'. 'AI model A' (dotted line from server) and the 'Addition' circle both point to 'AI model B' on the client side. A small label 'F.748.53(25)' is at the bottom right.](9a8f8276c7199f1dd9f3754cf60555c6_img.jpg) + +Figure 16 – Residual updating workflow diagram. The diagram is split into two sections by a dashed line: Server (top) and Client (bottom). On the Server side, 'AI model A' has three outgoing arrows: 'Sharing parts' to a circle labeled 'Subtraction', 'Retrain' to 'AI model B', and a dotted line to a 'Compact representation' box. 'AI model B' also points to the 'Subtraction' circle. The 'Subtraction' circle outputs to another 'Compact representation' box, which then points to a 'Residual model' box. On the Client side, the 'Residual model' box points to a circle labeled 'Addition'. 'AI model A' (dotted line from server) and the 'Addition' circle both point to 'AI model B' on the client side. A small label 'F.748.53(25)' is at the bottom right. + +**Figure 16 – Residual updating workflow** + +## 10.4 Distribution workflow + +This clause specifies the distribution workflow of AI model from server to client. + +Inputs to this process are: + +- Compact representation model which generated from the previous clause, +- Program files which to run the models (optional), +- Model description files which contain the model information. + +Output of this process is the encapsulation representation model. + +The Encapsulation Representation Model is derived as follows: + +- Prepare the required input files. +- If the model requires residual updating: + - Use the residual updating process. +- Else skip the residual updating process. +- If the model data requires encryption: + - Use the encryption methods in clause 10.2. +- Else skip the encryption process. +- Encapsulate the model/encrypted model with file header and model header information. + +- Organize the files. +- Generate the encapsulation representation model. +- Distribute the encapsulation representation model to different devices. + +![Flowchart of the encapsulation workflow. Inputs include Program files, Compact representation model, and Model description files. The Compact representation model is processed through a 'Residual?' decision. If 'Yes', it goes to a 'Residual model'. If 'No', it goes to an 'Encryption?' decision. If 'Yes', it goes to 'Data encryption'. Both 'Data encryption' and 'Residual model' lead to 'Data encapsulation'. 'Program files' and 'Model description files' lead to 'Organization files'. 'Data encapsulation' and 'Organization files' both lead to the final 'Encapsulation representation model', which is then distributed to Device A, Device B, and Device C. A label F.748.53(25) is present at the bottom right of the diagram.](f449330259f8d44842bfe0829754c00f_img.jpg) + +``` + +graph TD + subgraph Inputs + direction TB + PF[Program files] + CRM[Compact representation model] + MDF[Model description files] + end + CRM --> Residual{Residual?} + Residual -- Yes --> RM[Residual model] + Residual -- No --> Encryption{Encryption?} + Encryption -- Yes --> DE[Data encryption] + Encryption -- No --> DE + DE --> DataEncaps[Data encapsulation] + RM --> DataEncaps + PF --> OrgFiles[Organization files] + MDF --> OrgFiles + DataEncaps --> OrgFiles + OrgFiles --> ERM[Encapsulation representation model] + ERM --> DA((Device A)) + ERM --> DB((Device B)) + ERM --> DC((Device C)) + F7485325[F.748.53(25)] + +``` + +Flowchart of the encapsulation workflow. Inputs include Program files, Compact representation model, and Model description files. The Compact representation model is processed through a 'Residual?' decision. If 'Yes', it goes to a 'Residual model'. If 'No', it goes to an 'Encryption?' decision. If 'Yes', it goes to 'Data encryption'. Both 'Data encryption' and 'Residual model' lead to 'Data encapsulation'. 'Program files' and 'Model description files' lead to 'Organization files'. 'Data encapsulation' and 'Organization files' both lead to the final 'Encapsulation representation model', which is then distributed to Device A, Device B, and Device C. A label F.748.53(25) is present at the bottom right of the diagram. + +**Figure 17 – Encapsulation workflow** + +Distribute the encapsulation representation model to different devices through the model transmission process. The model transmission process is based on a specific transmission protocol, which transfers the data stream from device A to device B or from Server to Client. The data stream during the model transmission process is bitstream described in clause 10.1.3. And all the control interfaces and data interfaces shall adopt the HTTPS protocol with the Representational State Transfer (REST) architecture style. The HTTPS protocol format is shown in Figure 12 of [IEEE 3161.9]. The data interface can refer to subclause 5 of [IEEE 3161.9]. + +![Diagram showing the transmission between Device A and Device B. Device A sends a 'Control flow' to Device B and receives a 'Bitstream' from Device B. A label F.748.53(25) is present at the bottom right of the diagram.](5cb79a4c597e9880d65949d9416dda5a_img.jpg) + +``` + +graph LR + DA[Device A] -- Control flow --> DB[Device B] + DB -- Bitstream --> DA + F7485325[F.748.53(25)] + +``` + +Diagram showing the transmission between Device A and Device B. Device A sends a 'Control flow' to Device B and receives a 'Bitstream' from Device B. A label F.748.53(25) is present at the bottom right of the diagram. + +**Figure 18 – Encapsulation representation model transmission** + +Once the client has received the model distributed from the server, decapsulation can begin. + +Inputs to this process are: + +- Encapsulation representation model which received from the server. + +Outputs of this process are programmes and model files that can be run on the device. + +The on-device model is derived as follows: + +- Prepare the required input files. +- If the model data requires decryption: + - Use the decryption methods in clause 10.2. +- Else skip the encryption process. +- If the model is a compact representation model: + - Use the decompression process. +- Else skip the decompression process. +- If a Model Converter is provided (optional): + - Convert the model to On-device Model. +- Else skip the conversion process. +- Loading models in the device runtime library. +- Get model inputs and get outputs after forward inference + +![Flowchart of the Decapsulation workflow. It starts with 'Encapsulation representation model' which branches into 'Program files', 'Model files', and 'Model description files'. 'Model files' lead to an 'Encryption?' decision. If 'Yes', it goes to 'Data decryption' and then to 'Compact representation?'. If 'No', it goes directly to 'Compact representation?'. From 'Compact representation?', if 'Yes', it goes to 'Decompression' and then to 'Model representation'. If 'No', it goes directly to 'Model representation'. Both 'Program files' and 'Model representation' lead to 'Model converter'. 'Model converter' leads to 'On-device model'. 'Core file' (from 'Program files') and 'On-device model' are combined at a summation node with 'Data input' to produce 'Model output'.](573293af67ee571aef04bf725fcefbe5_img.jpg) + +``` +graph TD; A[Encapsulation representation model] --> B[Program files]; A --> C[Model files]; A --> D[Model description files]; C --> E{Encryption?}; E -- Yes --> F[Data decryption]; E -- No --> G{Compact representation?}; F --> G; G -- Yes --> H[Decompression]; H --> I[Model representation]; G -- No --> I; B --> J[Core file]; I --> K[Model converter]; K --> L[On-device model]; J --> M((+)); L --> M; N[Data input] --> M; M --> O[Model output]; +``` + +Flowchart of the Decapsulation workflow. It starts with 'Encapsulation representation model' which branches into 'Program files', 'Model files', and 'Model description files'. 'Model files' lead to an 'Encryption?' decision. If 'Yes', it goes to 'Data decryption' and then to 'Compact representation?'. If 'No', it goes directly to 'Compact representation?'. From 'Compact representation?', if 'Yes', it goes to 'Decompression' and then to 'Model representation'. If 'No', it goes directly to 'Model representation'. Both 'Program files' and 'Model representation' lead to 'Model converter'. 'Model converter' leads to 'On-device model'. 'Core file' (from 'Program files') and 'On-device model' are combined at a summation node with 'Data input' to produce 'Model output'. + +F.748.53(25) + +**Figure 19 – Decapsulation workflow** + +# Bibliography + +- [b-ITU-T F.748.12] Recommendation ITU-T F.748.12 (2021), *Deep learning software framework evaluation methodology*. +- [b-ITU-T X.1235] Recommendation ITU-T X.1235 (2022), *Technologies in countering website spoofing for telecommunication organizations*. +- [b-ITU-T Y.3172] Recommendation ITU-T Y.3172 (2019), *Architectural framework for machine learning in future networks including IMT-2020*. +- [b-IEEE 2941] IEEE Std 2941-2021, *IEEE Standard for Artificial Intelligence (AI) Model Representation, Compression, Distribution, and Management*. +- [b-IEEE 754] IEEE Std 754-2019, *IEEE Standard for Floating-Point Arithmetic*. + + + + + +# SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.55-202503-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.748.55-202503-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c1629394c02f697f60ae954ad36bf274fa8082cb --- /dev/null +++ b/marked/F/T-REC-F.748.55-202503-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3763acd7a0ed56aa30f3f90468c1fb0cb54573eb9045bca3eddf8976b4f0f41d +size 7219 diff --git a/marked/F/T-REC-F.748.55-202503-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.748.55-202503-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7af875c5e702dccfa600ca90339bba0d143840e7 --- /dev/null +++ b/marked/F/T-REC-F.748.55-202503-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4af8236e418fd0241d678243240592b0e1e6cad977cb325f65cdf9c2fbb1f0eb +size 66014 diff --git a/marked/F/T-REC-F.748.55-202503-I_PDF-E/raw.md b/marked/F/T-REC-F.748.55-202503-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..051c83e9eedda5549b9210dd06245d061b66b6b9 --- /dev/null +++ b/marked/F/T-REC-F.748.55-202503-I_PDF-E/raw.md @@ -0,0 +1,773 @@ + + +# Recommendation **ITU-T F.748.55 (03/2025)** + +SERIES F: Non-telephone telecommunication services +Multimedia services + +--- + +## **Technical requirements and evaluation methods of robotic process automation systems** + +![ITU logo](390120de4fe440c42fea8154fcaad334_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.55 + +# Technical requirements and evaluation methods of robotic process automation systems + +## Summary + +Recommendation ITU-T F.748.55 provides the technical requirements and evaluation methods of robotic process automation (RPA) systems. Some key indicators to evaluate the capabilities of robotic process automation system are set. + +## History\* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.55 | 2025-03-16 | 21 | 11.1002/1000/16244 | + +### Keywords + +AI technologies, development and tests, management and monitoring, robotic process automation system, work execution. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2025 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|------|--------------------------------------------------------------------|----| +| 1 | Scope ..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 1 | +| 4 | Abbreviations and acronyms ..... | 1 | +| 5 | Conventions ..... | 2 | +| 6 | The technical structure for robotic process automation system..... | 2 | +| 7 | Development and test of RPA system ..... | 3 | +| 7.1 | Development mode..... | 3 | +| 7.2 | Development templates ..... | 4 | +| 7.3 | Functional components..... | 5 | +| 7.4 | Version management ..... | 6 | +| 7.5 | Process debugging ..... | 6 | +| 8 | Management and monitoring of RPA system..... | 7 | +| 8.1 | Deployment management..... | 7 | +| 8.2 | Robot management..... | 8 | +| 8.3 | Permission management..... | 9 | +| 8.4 | Asset management..... | 9 | +| 8.5 | Operation monitoring ..... | 11 | +| 8.6 | Task management..... | 12 | +| 8.7 | Log management ..... | 12 | +| 8.8 | System risk management..... | 13 | +| 8.9 | Environment risk management..... | 14 | +| 9 | Workflow execution of RPA system ..... | 15 | +| 9.1 | Web automation..... | 15 | +| 9.2 | Application automation ..... | 16 | +| 9.3 | Basic automation ..... | 17 | +| 9.4 | Human-RPA collaboration ..... | 18 | +| 9.5 | Queue and concurrency ..... | 19 | +| 9.6 | Dynamic adaptation..... | 20 | +| 9.7 | Exception handling..... | 20 | +| 9.8 | Analysis and insight ..... | 21 | +| 10 | Artificial intelligence empowerments..... | 22 | +| 10.1 | Intelligent development ..... | 22 | +| 10.2 | Intelligent management ..... | 23 | +| 10.3 | Intelligent execution ..... | 23 | + + + +## Technical requirements and evaluation methods of robotic process automation systems + +## 1 Scope + +This Recommendation specifies the technical structure, evaluation methods and indicators for robotic process automation (RPA) systems and other related aspects. The service providers and customers can use them for professional technical tests. + +This Recommendation lists the technical requirements and evaluation methods for robotic process automation system in five dimensions: + +- the development and tests; +- the management and monitoring; +- the work execution; +- the involved AI technologies. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +None. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 manned mode:** When, in the process of operation, the robot system needs human intervention by requiring manual command or input to carry out the complete task flow. + +**3.2.2 robotic process automation:** Automation that uses software robots based on certain rules of interactive action to automate repetitive software tasks on systems, software applications and web that once required human actions. + +**3.2.3 unattended mode:** When in the process of operation, there is no need for humans to be involved in the tasks, and the task flow is completely carried out by the robot autonomously. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +AI            Artificial Intelligence + +| | | +|-----|-----------------------------------| +| API | Application Programming Interface | +| BOM | Browser Object Model | +| CPU | Central Processing Unit | +| CRM | Customer Relationship Management | +| DOM | Document Object Model | +| ERP | Enterprise Resource Planning | +| GPU | Graphics Processing Unit | +| ID | Identity Document | +| MAC | Media Access Control | +| NLP | Natural Language Processing | +| OA | Office Automation | +| OCR | Optical Character Recognition | +| RPA | Robotic Process Automation | +| UI | User Interface | + +## 5 Conventions + +In this Recommendation: + +The keywords "**is required to**" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. + +The keywords "**is recommended**" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement needs not be present to claim conformance. + +The keywords "**can optionally**" and "**may**" indicate an optional requirement which is permissible, without implying any sense of being recommended. These terms are not intended to imply that the vendor's implementation must provide the option, and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +## 6 The technical structure for robotic process automation system + +A typical robot process automation (RPA) system includes three main parts: the controller, the designer and the software robots. The controller can provide the capabilities such as offering, deploying, triggering, monitoring, measuring, or tracking status of the workflows. The designer enables the development and optimization of software robots. The software robots are customized processes which can be used to perform specific tasks or collect essential data. In recent years, with the development of artificial intelligence (AI) technologies, some AI capabilities have been integrated into RPA system as functional components to expand the overall system capabilities. See Figure 1. + +![Figure 1: The technical structure of RPA system. The diagram shows three main components: Development and test, Work and execution, and Management and monitoring, all supported by Artificial intelligence empowerment. Development and test involves creating and modifying automated processes. Work and execution involves allocating and executing automated processes based on RPA systems, including Software robot 1, Software robot 2, ..., Software robot N. Management and monitoring involves managing and monitoring automated processes. Arrows indicate flow and feedback between components, with a common base of Artificial intelligence empowerment.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +F.748.55(25) + +Figure 1: The technical structure of RPA system. The diagram shows three main components: Development and test, Work and execution, and Management and monitoring, all supported by Artificial intelligence empowerment. Development and test involves creating and modifying automated processes. Work and execution involves allocating and executing automated processes based on RPA systems, including Software robot 1, Software robot 2, ..., Software robot N. Management and monitoring involves managing and monitoring automated processes. Arrows indicate flow and feedback between components, with a common base of Artificial intelligence empowerment. + +**Figure 1 – The technical structure of RPA system** + +## 7 Development and test of RPA system + +### 7.1 Development mode + +#### 7.1.1 Technical requirements + +Through the development of RPA processes, companies can closely align automation capabilities with diverse business scenarios. A good development mode can guarantee an efficient and high-quality process building. The capability classification methods and related descriptions are listed in Table 1: + +**Table 1 – The capability classification methods of development mode** + +| Competency levels | Descriptions | +|------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Basic functions |
  • a) It is recommended to support different development modes, such as visual development, coding development and recording development.
  • b) It is recommended to support the automatic optimization of existing processes.
  • c) It is recommended to support the visual development using low-code technologies.
  • d) It is recommended to support hotkey operations, such as using hotkeys to save, run, debug, or search for items.
  • e) It is recommended to support the basic management of third-party libraries, such as loading and unloading them.
  • f) It is recommended to support adjustments to development interfaces, such as size, theme, or location of the object.
  • g) It is recommended to support providing application programming interfaces (APIs) and code auto-complement function during the development phase.
  • h) It is recommended to support the online cooperation among different developers.
  • i) It is recommended to support the real-time transformation between the visual development and coding development.
| + +**Table 1 – The capability classification methods of development mode** + +| Competency levels | Descriptions | +|---------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Advanced functions | a) It is recommended to support detecting and analysing the user's actions on the user interface (UI).
b) It is recommended to support the transformation between recording development and visual development.
c) It is recommended to support exporting the entire or partial process for secondary development.
d) It is recommended to support developing the process via script editors. | + +#### **7.1.2 Evaluation methods** + +- a) The product information is recommended to be checked to determine whether the robotic process automation system supports the functions listed in Table 1. The product information is provided by the service providers. +- b) The listed items are recommended to be checked by setting product functional tests. The checked items involve the development model types, the system stability in each development mode, and the usability for developers. + +### **7.2 Development templates** + +#### **7.2.1 Technical requirements** + +The development templates are important tools for ensuring development efficiency. Development templates can guide the user to develop or modify similar processes rapidly. The capability classification methods and related descriptions are listed in Table 2: + +**Table 2 – The capability classification methods of development templates** + +| Competency levels | Descriptions | +|------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Basic functions | a) It is recommended to support a variety of development templates.
b) It is recommended to support quick creation of new processes using existing templates.
c) It is recommended to support displaying function descriptions and instructions for development templates.
d) It is recommended to support exporting the edited processes as new development templates, and to allow local storage or cloud storage.
e) It is recommended to support explaining the abilities of the templates through different ways, such as written explanations and video displays.
f) It is recommended to support providing the pre-set development templates on general applications, such as office software and office systems.
g) It is recommended to support providing development templates based on attributes and characteristics of different industries.
h) It is recommended to support providing development templates based on different business areas or positions.
i) It is recommended to support the centralized management for development templates.
j) It is recommended to support providing warning tips when upgrading the template's version. | + +**Table 2 – The capability classification methods of development templates** + +| Competency levels | Descriptions | +|---------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Advanced functions |
  • a) It is recommended to support customizing the new development templates.
  • b) It is recommended to support building the template stores for sharing, shelving, updating, off shelving the templates.
  • c) It is recommended to support detecting the exceptions when executing the same process under different visions.
  • d) It is recommended to support the intelligent recommendations for templates among different users.
| + +#### 7.2.2 Evaluation methods + +- a) The product information is recommended to be checked to determine whether the robotic process automation system supports the functions listed in Table 2. The product information is provided by the service providers. +- b) The listed items are recommended to be checked by setting product functional tests. The checked items involve the size of template library, the application scenarios covered by templates, and the reusability of templates. + +### 7.3 Functional components + +#### 7.3.1 Technical requirements + +Functional components are essential tools for process development. Compared to the development templates, the components focus on basic operations or simple functions. A rich and complete set of functional components can make process development more convenient and efficient. The capability classification methods and related descriptions are listed in Table 3: + +**Table 3 – The capability classification methods of functional components** + +| Competency levels | Descriptions | +|---------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Basic functions |
  • a) It is recommended to support basic management for functional components, such as creation, updating, browsing, classification, query and deletion.
  • b) It is recommended to support importing and loading the functional components from local or online resources.
  • c) It is recommended to support customizing the functional components, such as developing and encapsulating specialized functional components based on business needs.
  • d) It is recommended to support the hotkey operations for functional components, may alert and provide other choices when hotkeys are occupied.
  • e) It is recommended to support using third-party components or extensions to meet specific functional needs.
| +| Advanced functions |
  • a) It is recommended to support using the extension packages.
  • b) It is recommended to support privatization deployment.
  • c) It is recommended to support providing intelligent recommendations for functional components among different users.
| + +#### 7.3.2 Evaluation methods + +- a) The product information is recommended to be checked to determine whether the robotic process automation system supports the functions listed in Table 3. The product information is provided by the service providers. +- b) The listed items are recommended to be checked by setting product functional tests. The checked items involve the size of component library, the application scenarios covered by components, and the reusability of components. + +### 7.4 Version management + +#### 7.4.1 Technical requirements + +Version management is a necessary function for process of development and management. It helps the developer trace and compare the changes among different system visions. The capability classification methods and related descriptions are listed in Table 4: + +**Table 4 – The capability classification methods of version management** + +| Competency levels | Descriptions | +|--------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Basic functions |
  • a) It is recommended to support recording or checking the details of the version information, such as the time of creation and update, the publishers, and the version descriptions.
  • b) It is recommended to support ranking the versions automatically by creation date.
  • c) It is recommended to support some operations between multiple versions, such as switching and replacement.
  • d) It is recommended to support checking the historical versions and the system rollback to the specified historical version.
  • e) It is recommended to support adding comments and rich texts for versions' description.
  • f) It is recommended to support selecting and editing the version for exported or released processes.
| +| Advanced functions |
  • a) It is recommended to support checking the differences between the current version and the historical versions.
  • b) It is recommended to support the hierarchy management of versions.
  • c) It is recommended to support multiple publish ways of processes, such as publishing directly to the controller or publishing as local files.
| + +#### 7.4.2 Evaluation methods + +- a) The product information is recommended to be checked to determine whether the robotic process automation system supports the functions listed in Table 4. The product information is provided by the service providers. +- b) The listed items are recommended to be checked by setting product functional tests. The checked items involve version rollback capability, the version update capability, and the version comparison capability. + +### 7.5 Process debugging + +#### 7.5.1 Technical requirements + +Process debugging helps developers quickly find and target problems. It also helps users verify the accuracy of the RPA process and improve development efficiency. The capability classification methods and related descriptions are listed in Table 5: + +**Table 5 – The capability classification methods of process commissioning** + +| Competency levels | Descriptions | +|---------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Basic functions | a) It is recommended to support customizing the range of debugging.
b) It is recommended to support setting breakpoints for debugging.
c) It is recommended to support centralized management of breakpoints, such as batch activation or abandonment.
d) It is recommended to support targeting the errors in specific components and modes.
e) It is recommended to support testing the process in units.
f) It is recommended to support checking the code during the execution.
g) It is recommended to support pausing, stopping, and restarting of the process/tasks during debugging. | +| Advanced functions | a) It is recommended to support other abilities of debugging such as step over, step in, and step out.
b) It is recommended to support the switchover among breakpoints.
c) It is recommended to support displaying the variable information and contents of process during the execution. | + +#### **7.5.2 Evaluation methods** + +- a) The product information is recommended to be checked to determine whether the robotic process automation system supports the functions listed in Table 5. The product information is provided by the service providers. +- b) The listed items are recommended to be checked by setting product functional tests. The checked items involve the abilities of using breakpoints, adapting to multiple debugging modes, and tracking the entire debugging process. + +## **8 Management and monitoring of RPA system** + +### **8.1 Deployment management** + +#### **8.1.1 Technical requirements** + +Deployment management is a necessary step in transferring RPA technology to the digital workforce. As a kind of software system, the deployment should follow pre-set rules from deployment conditions, product characters to customer needs. The capability classification methods and related descriptions are listed in Table 6: + +**Table 6 – The capability classification methods of deployment management** + +| Competency levels | Descriptions | +|------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Basic functions | a) It is recommended to support multiple deployment modes, including but not limited to local deployment, cloud deployment and mixed deployment.
b) It is recommended to support being installed on multiple operating systems.
c) It is recommended to support the deployment of different kinds of robots, such as manned robots, unattended robots, and mixed robots.
d) It is recommended to support a variety of deployment methods, such as graphical deployment and code deployment.
e) It is recommended to support cluster deployment and disaster-tolerant deployment for large-scale servers.
f) It is recommended to support upgrading the robots. | + +**Table 6 – The capability classification methods of deployment management** + +| Competency levels | Descriptions | +|---------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| | g) It is recommended to support the deployment which does not rely on external libraries or environment. | +| Advanced functions | a) It is recommended to support installation on uncommon operating systems.
b) It is recommended to support installation on mobile terminal devices.
c) It is recommended to support installation on cloud platforms.
d) It is recommended to support development-oriented deployments.
e) It is recommended to support quick deployment, so that the products can be installed or uninstalled after copy or delete. | + +#### 8.1.2 Evaluation methods + +- a) The product information is recommended to be checked to determine whether the robotic process automation system supports the functions listed in Table 6. The product information is provided by the service providers. +- b) The listed items are recommended to be checked by setting product functional tests. It is recommended to check the abilities of adapting to multiple deployment environment. It is recommended to supporting multiple deployment modes and managing the system versions. + +### 8.2 Robot management + +#### 8.2.1 Technical requirements + +Robot management is a basic ability of the RPA management platform. It monitors the whole life cycle of software robots, such as registration, permission assignment, use and abandonment. The capability classification methods and related descriptions are listed in Table 7: + +**Table 7 – The capability classification methods of robot management** + +| Competency levels | Descriptions | +|---------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Basic functions | a) It is recommended to support identifying the different robots via the addresses and the user's accounts.
b) It is recommended to support the bulk management for robots, and the operations such as activating or deactivating or recycling the robots.
c) It is recommended to support the basic monitoring operations on the robots, such as configuration, modification, and inquiry.
d) It is recommended to support enabling or disabling the robots from local or cloud.
e) It is recommended to support remote control of RPA robots.
f) It is recommended to support distributing the robots into different groups.
g) It is recommended to support sending alerts for exceptions and errors. | +| Advanced functions | a) It is recommended to support unlocking the desktop via remote operations.
b) It is recommended to support docking the environment for specific RPA processes.
c) It is recommended to support identifying unauthorized access from the IP address or media access control (MAC) address and setting the login permissions.
d) It is recommended to support the management and monitoring of the robots from other brands. | + +#### 8.2.2 Evaluation methods + +- a) The product information is recommended to be checked to determine whether the robotic process automation system supports the functions listed in Table 7. The product information is provided by the service providers. +- b) The listed items are recommended to be checked by setting product functional tests. The checked items involve the abilities such as creating the robot account, supervising the operations of robots, and tracking the robot's status. + +### 8.3 Permission management + +#### 8.3.1 Technical requirements + +Permission management is a strategy which protects the system's security. Under permission management, system access can be distributed. The capability classification methods and related descriptions are listed in Table 8: + +**Table 8 – The capability classification methods of permission management** + +| Competency levels | Descriptions | +|---------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Basic functions |
  • a) It is recommended to support the basic management of users' accounts, such as creation, modification, and deletion.
  • b) It is recommended to support distinguishing the permissions for different accounts based on the operation objects.
  • c) It is recommended to support the management of different organizations, users, and tenants.
  • d) It is recommended to support the basic operations for password management, such as resetting the password and changing the password policy settings.
| +| Advanced functions |
  • a) It is recommended to support defining the permissions between different components, processes, tasks and data.
  • b) It is recommended to support defining the permissions between different resources such as robots, designers, and management sides.
  • c) It is recommended to support automatically requesting the necessary permissions when it is recommended to supported permissions are insufficient.
  • d) It is recommended to support the lightweight directory access protocols.
| + +#### 8.3.2 Evaluation methods + +- a) The product information is recommended to be checked to determine whether the robotic process automation system supports the functions listed in Table 8. The product information is provided by the service providers. +- b) The listed items are recommended to be checked by setting product functional tests. The checked items involve the abilities such as organization management, account management, and role management. + +### 8.4 Asset management + +#### 8.4.1 Technical requirements + +Assets include valuable hardware systems, software systems or data. With asset management, the manager can ensure the safety and efficiency of processes execution. The capability classification methods and related descriptions are listed in Table 9: + +**Table 9 – The capability classification methods of asset management** + +| Competency levels | Descriptions | +|---------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Basic functions |
  • a) It is recommended to support managing the different types of assets, such as MAC addresses, accounts, passwords and so on.
  • b) It is recommended to support managing the asset's status, such as creation, freezing and deletion.
  • c) It is recommended to support invoking the interface of third-party business systems when creating new assets.
  • d) It is recommended to support deleting the asset individually or deleting the assets in bulk.
  • e) It is recommended to support synchronizing updated changes of different assets.
  • f) It is recommended to support searching the target based on different criteria such as name, type, status, creator, and creation time.
  • g) It is recommended to support setting different permission assignments, including the limit of usage period and usage times.
  • h) It is recommended to support protecting the sensitive or confidential assets, such as encrypting information or blocking related information.
  • i) It is recommended to support setting blacklists and whitelists.
  • j) It is recommended to support the basic data management and usage specifications.
  • k) It is recommended to support data security management specifications and data security control mechanisms.
  • l) It is recommended to support life-cycle management mechanisms for sensitive data, such as dynamic and static data encryption, removing tracks.
  • m) It is recommended to support the mechanisms for encrypted storage, backup, and recovery of sensitive data.
  • n) It is recommended to support the specification assessment of data handled by RPA robots, and it is recommended to support monitoring whether RPA robots are eligible when handling sensitive data.
| +| Advanced functions |
  • a) It is recommended to support the configuration or management of biometric characters.
  • b) It is recommended to support managing the hardware assets.
  • c) It is recommended to support distinguishing data of the same asset among different users.
  • d) It is recommended to support distinguishing data of the same asset among different robots.
  • e) It is recommended to support automatic protection capabilities for sensitive data, such as automatic desensitization of sensitive data.
  • f) It is recommended to support automatic auditing capabilities for visiting sensitive data, such as automatic auditing of suspicious operations.
| + +#### 8.4.2 Evaluation methods + +- a) The product information is recommended to be checked to determine whether the robotic process automation system supports the functions listed in Table 9. The product information is provided by the service providers. +- b) The listed items are recommended to be checked by setting product functional tests. The checked items involve the types of the asset, the safety and security protection strategies, and the efficiency of asset application. + +### 8.5 Operation monitoring + +#### 8.5.1 Technical requirements + +Operational monitoring can ensure the ability of system self-robustness, task handling timeliness and abnormal problems handling. The capability classification methods and related descriptions are listed in Table 10: + +**Table 10 – The capability classification methods of operation monitoring** + +| Competency levels | Descriptions | +|---------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Basic functions |
  • a) It is recommended to support monitoring the basic characteristics of processes such as progress, node status and resource occupancy.
  • b) It is recommended to support remote control and other ways to review the current or historical operation logs of the robots from different locations.
  • c) It is recommended to support monitoring and recording the operations by videos or screen capture recorders.
  • d) It is recommended to support checking the robots by their conditions, such as executing, idle status and offline status.
  • e) It is recommended to support the monitoring of different dynamic events for robots.
  • f) It is recommended to support centralized storage management for different types of data.
  • g) It is recommended to supports retracing the information for executed tasks, such as task initiator, execution process and key nodes.
  • h) It is recommended to support alerting via different ways when exceptions exist, such as mails, messages, and calls.
| +| Advanced functions |
  • a) It is recommended to support associations and transformations between multiple operational monitoring methods.
  • b) It is recommended to support several ways to hide or monitor the sensitive information, such as blurring, masking, and transferring into background processing.
  • c) It is recommended to support the real-time analysis and alerts, which can prejudge the occurrence of anomalies or to report potentially dangerous nodes in advance.
  • d) It is recommended to support the joint monitoring of multiple management platforms to find the chain problems of robot clusters.
  • e) It is recommended to support customizing alarm conditions, such as setting specific thresholds and the frequency of use.
| + +#### 8.5.2 Evaluation methods + +- a) The product information is recommended to be checked to determine whether the robotic process automation system supports the functions listed in Table 10. The product information is provided by the service providers. +- b) The listed items are recommended to be checked by setting product functional tests. The checked items involve the abilities of resource monitoring, task monitoring and exception monitoring. + +### 8.6 Task management + +#### 8.6.1 Technical requirements + +Task management realizes the abilities of task launch, cycle scheduling, exception handling and so on. The task management can act in task creation, task release, and task monitoring. The capability classification methods and related descriptions are listed in Table 11: + +**Table 11 – The capability classification methods of task management** + +| Competency levels | Descriptions | +|---------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Basic functions |
  • a) It is recommended to support displaying the task lists and task details, such as describing task node relationships through a visual interface.
  • b) It is recommended to support automatic task creation, deletion, modification, update and query.
  • c) It is recommended to support different task scheduling modes, such as real-time task scheduling, assignment scheduling and custom calendar scheduling.
  • d) It is recommended to support multiple task assignment modes, such as task assignment to all robots, partial robots, or dynamic assignments.
  • e) It is recommended to support the priority control of tasks, in the same period, the tasks will be carried out according to priority.
  • f) It is recommended to support controlling the execution steps of tasks.
  • g) It is recommended to support locking the mouse and keyboard input during task execution to prevent accidental touch interference.
  • h) It is recommended to support the human-machine interaction for specific tasks.
| +| Advanced functions |
  • a) It is recommended to support setting specific event triggers to start tasks.
  • b) It is recommended to support tolerating faults and recording the exceptions during the task execution.
  • c) It is recommended to support the intelligent task assignment, such as automatically arranging the schedules of tasks according to user needs.
  • d) It is recommended to support the hotkey operations for controlling the task's statuses.
| + +#### 8.6.2 Evaluation methods + +- a) The product information is recommended to be checked to determine whether the robotic process automation system supports the functions listed in Table 11. The product information is provided by the service providers. +- b) The listed items are recommended to be checked by setting product functional tests. The checked items involve the abilities of task plan, task online and task scheduling. + +### 8.7 Log management + +#### 8.7.1 Technical requirements + +Log management is an important way for RPA tools to meet the basic audit and backtracking requirements of the processes. There is a variety of log information including server node performance, robot behaviours, task execution and product operations. The capability classification methods and related descriptions are listed in Table 12: + +**Table 12 – The capability classification methods of log management** + +| Competency levels | Descriptions | +|------------------------|--------------------------------------------------------------------------------------------------------------------------| +| Basic functions |
  • a) It is recommended to support the acquisition of robots' behaviours.
| + +**Table 12 – The capability classification methods of log management** + +| Competency levels | Descriptions | +|---------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| |

b) It is recommended to support the protection of logs.

c) It is recommended to support saving the robot's operational logs as local files or sent them to the cloud.

d) It is recommended to support a variety of log file formats, such as files, screenshots, and videos.

e) It is recommended to support real-time transforms of logs between the robots and the server.

f) It is recommended to support customizing the types of the logs on the management side.

g) It is recommended to support setting a variety of conditions to filter the logs.

h) It is recommended to support multiple types of information records, such as execution steps, operational status, run results and warning messages.

| +| Advanced functions |

a) It is recommended to support saving the logs in structured and retrievable formats, it is recommended to support reviewing the logs.

b) It is recommended to support the visual configuration of the dashboard for the logs.

c) It is recommended to support recording the performance indexes for the RPA robots.

d) It is recommended to support managing the alert records.

| + +#### 8.7.2 Evaluation methods + +- a) The product information is recommended to be checked to determine whether the robotic process automation system supports the functions listed in Table 12. The product information is provided by the service providers. +- b) The listed items are recommended to be checked by setting product functional tests. The checked items involve the log storage structure, the log granularity, and the management mechanisms. + +### 8.8 System risk management + +#### 8.8.1 Technical requirements + +The system risk management focuses on regulating the security capabilities of the RPA tool itself, and the main indicators include the security configuration and management, the security of functions and technologies. The capability classification methods and related descriptions are listed in Table 13: + +**Table 13 – The capability classification methods of system risk management** + +| Competency levels | Descriptions | +|------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Basic functions |

a) It is recommended to support setting the access control and configuring the user permissions.

b) It is recommended to support the hierarchical management policy of user permission.

c) It is recommended to support password management policies or mechanisms, such as setting password update cycles or controlling the password strength.

d) It is recommended to support the log management and configuration, such as querying and filtering.

e) It is recommended to support the statistics and analysis of operational status.

| + +**Table 13 – The capability classification methods of system risk management** + +| Competency levels | Descriptions | +|---------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| |

f) It is recommended to support operational security analysis based on configuration policy.

g) It is recommended to support alarming for operational security issues and alarm notification through work order system or other systems.

h) It is recommended to support patch management functions.

i) It is recommended to support basic anti-attack functions like cracking.

j) It is recommended to support anti-hook mechanism.

| +| Advanced functions |

a) It is recommended to support automatic security management, such as automatic analysis, automatic warning, and automatic auditing of security configurations.

b) It is recommended to support multi-dimensional security auditing, such as for processes, employees, and applications.

c) It is recommended to support setting customizable storage time and storage addresses of logs.

d) It is recommended to support implementing the security alerts mechanisms.

e) It is recommended to support automatically detecting the operating system environment.

f) It is recommended to support the unified patch management platform.

g) It is recommended to support the anti-reversal mechanisms.

| + +#### **8.8.2 Evaluation methods** + +- a) The product information is recommended to be checked to determine whether the robotic process automation system supports the functions listed in Table 13. The product information is provided by the service providers. +- b) The listed items are recommended to be checked by setting product functional tests. The tests should cover the system functions' safety and the system technologies' safety. + +### **8.9 Environment risk management** + +#### **8.9.1 Technical requirements** + +Environmental risk management focuses on protecting the security of operating system environment. The capability classification methods and related descriptions are listed in Table 14: + +**Table 14 – The capability classification methods of environment risk management** + +| Competency levels | Descriptions | +|------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Basic functions |

a) It is recommended to support the basic specifications of environment safety management.

b) It is recommended to support the basic requirements of system operation environment.

c) It is recommended to support the standardized security policies of operation environment and steps.

d) It is recommended to support the multi-dimensional security baseline for RPA application environment, such as network, operating system, middleware, and database.

e) It is recommended to support the system environment monitoring, such as alerts and analysis for threats, alarm-processing mechanisms, and threat treatment options.

| + +**Table 14 – The capability classification methods of environment risk management** + +| Competency levels | Descriptions | +|---------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| | f) It is recommended to support the ability to conduct comprehensive tests of the operation environment. | +| Advanced functions | a) It is recommended to support the capabilities of automatic security monitoring and auditing.
b) It is recommended to support the automatic threat analysis.
c) It is recommended to support the automatic operation capabilities, it is recommended to support the interactions of data flows among various systems, such as work order system, knowledge base management system, and office automation (OA) system. | + +#### **8.9.2 Evaluation methods** + +- a) The product information is recommended to be checked to determine whether the robotic process automation system supports the functions listed in Table 14. The product information is provided by the service providers. +- b) The listed items are recommended to be checked by setting product functional tests. The tests should cover the security mechanisms and risk analysis capabilities. + +## **9 Workflow execution of RPA system** + +### **9.1 Web automation** + +#### **9.1.1 Technical requirements** + +Web automation refers specifically to the operational take-over capability of the RPA system when it operates automatic workflows on websites. RPA should provide secure and stable take-over technologies with good compatibility for different browsers. The capability classification methods and related descriptions are shown in Table 15: + +**Table 15 – The capability classification methods of web automation** + +| Competency levels | Descriptions | +|------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Basic functions | a) It is recommended to support taking over the Web application systems securely, be compatible with different desktop resolutions, work properly when browser window is covered, and run the process after the operating system locks the screen.
b) It is recommended to support the automatic operations on mainstream browsers.
c) It is recommended to support automatic text inputting such as account identity documents (IDs) or passwords in browser applications.
d) It is recommended to support basic operations on multiple browser windows or multiple browser pages, such as adding, accessing, and refreshing browsers.
e) It is recommended to support detecting screen elements in a variety of ways.
f) It is recommended to support capturing the screen elements and triggering the process on the web pages.
g) It is recommended to support accurately capturing the information on pop-up forms or dialog boxes.
h) It is recommended to support the automatic operations on mainstream Web page frameworks. | + +**Table 15 – The capability classification methods of web automation** + +| Competency levels | Descriptions | +|---------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| |
  • i) It is recommended to support the screenshots for screen elements, the page where the element is placed will be on top of the form and the screenshots will automatically focus on the target.
  • j) It is recommended to support capturing the locations of the error-reported screen elements in browser, including the elements themselves, locations, and reasons.
  • k) It is recommended to support monitoring the dynamic changes of screen elements, such as emergence and disappearance.
| +| Advanced functions |
  • a) It is recommended to support taking over the security of embedded browsers.
  • b) It is recommended to support getting cookies from browsers.
  • c) It is recommended to support the asynchronous operations, such as performing the action without the results of the previous action.
  • d) It is recommended to support the migration of automation capabilities among multiple browsers.
  • e) It is recommended to support calling browser API interfaces, such as browser object model (BOM) and document object model (DOM).
  • f) It is recommended to support dragging or dropping the element on web pages, such as setting relative end-point coordinates and recording the tracks.
  • g) It is recommended to support automatically recording the screenshots when the pages are scrolling.
| + +#### 9.1.2 Evaluation methods + +- a) The product information is recommended to be checked to determine whether the robotic process automation system supports the functions listed in Table 15. The product information is provided by the service providers. +- b) The listed items are recommended to be checked by setting product functional tests. The tests should cover the compatibility of different browsers and technology framework, the ability of capturing the screen elements on the website. + +### 9.2 Application automation + +#### 9.2.1 Technical requirements + +Application automation refers specifically to the take-over capability of the RPA system for desktop applications. RPA may be able to capture screen elements on desktop applications with different technology architectures, accurately obtain their attribute information and trigger related actions to complete the required business operations. The capability classification methods and related descriptions are shown in Table 16: + +**Table 16 – The capability classification methods of application automation** + +| Competency levels | Descriptions | +|------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Basic functions |
  • a) It is recommended to support the secure takeover of applications.
  • b) It is recommended to support adapting to different desktop resolutions.
  • c) It is recommended to support accommodating the application windows which are covered.
  • d) It is recommended to support operating the system after the screen is locked.
| + +**Table 16 – The capability classification methods of application automation** + +| Competency levels | Descriptions | +|---------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| |
  • e) It is recommended to support capturing the screen element and triggering the process on desktop applications.
  • f) It is recommended to support the use of common interface frameworks.
  • g) It is recommended to support locating the screen elements' position on application interface, matching and obtaining the user interface elements by pre-labelled features.
  • h) It is recommended to support the basic operations on screen elements, such as touching, hovering, scrolling, and typing.
  • i) It is recommended to support taking screenshots for user interface elements, and automatically focusing on the screen elements for screenshot saving.
  • j) It is recommended to support using hotkeys to send elements.
| +| Advanced functions |
  • a) It is recommended to support the takeover of non-standard programs.
  • b) It is recommended to support the silent operations for tasks, such as information capture and doing the tasks without opening the application interfaces.
  • c) It is recommended to support the migration and rapid updates for different versions of applications.
| + +#### 9.2.2 Evaluation methods + +- a) The product information is recommended to be checked to determine whether the robotic process automation system supports the functions listed in Table 16. The product information is provided by the service providers. +- b) The listed items are recommended to be checked by setting product functional tests. The tests should cover the automatic operations on different applications and technology framework, the ability of capturing the elements on the applications. + +### 9.3 Basic automation + +#### 9.3.1 Technical requirements + +The basic automation is for operating systems. All the automation operations which are based on operating system instructions, scripts, and network protocols are defined as basic automation. Basic automation is a core component of automation abilities. The capability classification methods and related descriptions are shown in Table 17: + +**Table 17 – The capability classification methods of basic automation** + +| Competency levels | Descriptions | +|------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Basic functions |
  • a) It is recommended to support the system-level operating instructions under mainstream operating systems.
  • b) It is recommended to support executing the common operations on different types of files.
  • c) It is recommended to support executing the operations on windows such as focusing, maximizing, minimizing, displaying, and hiding.
| + +**Table 17 – The capability classification methods of basic automation** + +| Competency levels | Descriptions | +|---------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Advanced functions |

a) It is recommended to support commanding operations for background system such as parameter adjustment, registry management, installation and uninstallation for applications and processes.

b) It is recommended to support calling the multiple programming languages, such as VB/F#, Java and Go.

| + +#### 9.3.2 Evaluation methods + +- a) The product information is recommended to be checked to determine whether the robotic process automation system supports the functions listed in Table 17. The product information is provided by the service providers. +- b) The listed items are recommended to be checked by setting product functional tests. The tests should cover the compatibility of different operation objects and the abundance of automatic execution actions. + +### 9.4 Human-RPA collaboration + +#### 9.4.1 Technical requirements + +Specific processes and scenarios need human interventions. General RPA has reliable performance on repetitive jobs, but the tasks still rely on human beings to make decisions. The capability classification methods and related descriptions are shown in Table 18: + +**Table 18 – The capability classification methods of human-RPA collaboration** + +| Competency levels | Descriptions | +|---------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Basic functions |

a) It is recommended to support variety forms of human-machine collaboration, such as providing manned robots, setting mails, and providing remote scheduling.

b) It is recommended to support inserting human-computer interaction nodes, such as adding nodes in the specific positions of the workflow.

c) It is recommended to support providing multiple human-computer interactive interfaces.

d) It is recommended to support executing the automatic process by detecting the user's actions on interface.

e) It is recommended to support the transformation between different execution modes such as the unattended mode and the manned mode.

| +| Advanced functions |

a) It is recommended to support the real-time detection and discovery of user's actions during interactions.

b) It is recommended to support batch-based approvals for specific interaction steps.

c) It is recommended to support building complex and flexible human-machine interaction interfaces.

d) It is recommended to support triggering the task execution by user's specific actions.

e) It is recommended to support cross-terminal, cross-system, cross-platform, cross-network collaborations, such as collaboration in different environments.

f) It is recommended to support controlling the task execution from the user side.

| + +#### 9.4.2 Evaluation methods + +- a) The product information is recommended to be checked to determine whether the robotic process automation system supports the functions listed in Table 18. The product information is provided by the service providers. +- b) The listed items are recommended to be checked by setting product functional tests. The checked items involve the interaction parameter settings, the interaction forms, and the effects of interaction. + +### 9.5 Queue and concurrency + +#### 9.5.1 Technical requirements + +The queue is a list of tasks waiting for RPA robots to execute. The tasks in queue can be executed by multiple robots and will be processed according to the urgency. The user can dynamically add new tasks to the queue. The capability classification methods and related descriptions are shown Table 19: + +**Table 19 – The capability classification methods of queue and concurrency** + +| Competency levels | Descriptions | +|---------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Basic functions |
  • a) It is recommended to support setting different robot groups.
  • b) It is recommended to support calling multiple robot groups to execute the same queue.
  • c) It is recommended to support the concurrent executions for different queues.
  • d) It is recommended to support the task executions for all robots in one robot group at the same time.
  • e) It is recommended to support setting different priorities for queues.
  • f) It is recommended to support adding new tasks into a running queue.
  • g) It is recommended to support distributing the multi-steps task to robots in different groups.
  • h) It is recommended to support the real-time scheduling and temporary transferring more robots to high-priority tasks.
  • i) It is recommended to support making schedules of task execution, such as setting repetition cycles and period for specific tasks.
| +| Advanced functions |
  • a) It is recommended to support the intelligent task execution.
  • b) It is recommended to support avoiding adverse effects between different tasks.
  • c) It is recommended to support concurrent execution for tasks, it is recommended to support providing differentiated execution abilities for different task sizes and complexities.
  • d) It is recommended to support setting the threshold to monitoring the number of executing tasks.
  • e) It is recommended to support the visual presentations for queues and the task list can be graphically demonstrated in the reports.
| + +#### 9.5.2 Evaluation methods + +- a) The product information is recommended to be checked to determine whether the robotic process automation system supports the functions listed in Table 19. The product information is provided by the service providers. +- b) The listed items are recommended to be checked by setting product functional tests. The checked items involve the abilities of adapting all the dynamic changes. + +### 9.6 Dynamic adaptation + +#### 9.6.1 Technical requirements + +The operating environment in which software robots are installed is often diverse and can constantly change. Stable operation is a fundamental requirement for RPA products. The system should adapt to the dynamic changes. The capability classification methods and related descriptions are shown in Table 20: + +**Table 20 – The capability classification methods of dynamic adaptation** + +| Competency levels | Descriptions | +|---------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Basic functions |
  • a) It is recommended to support the dynamic adaptation to resolution changes.
  • b) It is recommended to support the dynamic adaptation to GUI changes.
  • c) It is recommended to support the dynamic adaptation to elements' characters.
  • d) It is recommended to support the dynamic adaptation to colour changes.
  • e) It is recommended to support the dynamic adaptation to refresh rate changes.
  • f) It is recommended to support the dynamic adaptation to human interventions.
  • g) It is recommended to support the dynamic adaptation to the changes of the number of robots.
| +| Advanced functions |
  • a) It is recommended to support the dynamic adaptation to changes of computing resources such as Central Processing Unit (CPU) and Graphics Processing Unit (GPU) resources.
  • b) It is recommended to support the dynamic adaptation to changes in network resources such as network speed and stability.
  • c) It is recommended to support the dynamic adaptation to changes in memory resources such as memory being occupied by other programs.
  • d) It is recommended to supports for the dynamic adaptation to changes in cloud service resources.
| + +#### 9.6.2 Evaluation methods + +- a) The product information is recommended to be checked to determine whether the robotic process automation system supports the functions listed in Table 20. The product information is provided by the service providers. +- b) The listed items are recommended to be checked by setting product functional tests. The checked items involve the abilities of adapting all the dynamic changes. + +### 9.7 Exception handling + +#### 9.7.1 Technical requirements + +When RPA handles complex business processes, it is inevitable that there will be various systems and business anomalies. The handling of anomalies is particularly important in the production environment. The timeliness and effectiveness of abnormal prevention and handling are related to the execution's accuracy. The capability classification methods and related descriptions are shown in Table 21: + +**Table 21 – The capability classification methods of exception handling** + +| Competency levels | Descriptions | +|------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Basic functions |
  • a) It is recommended to support the various ways to report the robots' anomalies, such as pop-up windows and mail.
  • b) It is recommended to support several options to hand the abnormal process.
| + +**Table 21 – The capability classification methods of exception handling** + +| Competency levels | Descriptions | +|---------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| |

c) It is recommended to support continuing the process after human interactions.

d) It is recommended to support calling other robots to complete the follow-up process when the previous robot is busy and ensure the integrity of the process.

e) It is recommended to support recording or alerting the anomalies in different forms.

f) It is recommended to support labelling the abnormal information, the robot can accurately feedback the cause of abnormal information after identifying it.

g) It is recommended to support a variety of fault-tolerant modes such as showing pop-up tips for exceptions and capturing the exceptions in code lines or modules.

| +| Advanced functions |

a) It is recommended to support dealing with the redundancy of robots during the process.

b) It is recommended to support the rollback of different steps of process.

c) It is recommended to support choosing the node where the process will continue to run after human intervention.

d) It is recommended to support monitoring and recording the indexes of processes and execution environment such as execution time and logs.

| + +#### **9.7.2 Evaluation methods** + +- a) The product information is recommended to be checked to determine whether the robotic process automation system supports the functions listed in Table 21. The product information is provided by the service providers. +- b) The listed items are recommended to be checked by setting product functional tests. The checked items involve the abilities of monitoring and recording the exceptions. + +### **9.8 Analysis and insight** + +#### **9.8.1 Technical requirements** + +Analysis and insight provide a data-based, visual presentation for the indicators running on the RPA platform. It also can provide multi-dimensional query and analysis of task execution, task accuracy and process stability. The capability classification methods and related descriptions are shown in Table 22: + +**Table 22 – The capability classification methods of analysis and insight** + +| Competency levels | Descriptions | +|------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Basic functions |

a) It is recommended to support the statistics of various types of system's data, such as the number of users, process, and tasks.

b) It is recommended to support providing the reports to show the robot's task execution statuses.

c) It is recommended to support the comprehensive statistics, such as visual display of the number of robots and processing tasks, task proportion, running time, operational success rate and average processing speed.

d) It is recommended to support the basic analysis of data, and outputting the reports and the key indicators.

e) It is recommended to support the automatic generation of electronic files, such as the reports for a single robot or the robot's cluster.

| + +**Table 22 – The capability classification methods of analysis and insight** + +| Competency levels | Descriptions | +|---------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Advanced functions |
  • a) It is recommended to support comprehensive statistical analysis of robots and data, and it is recommended to support the operational analysis.
  • b) It is recommended to support the API services and the statistical analysis for external data.
  • c) It is recommended to support the generation and output for custom reports.
  • d) It is recommended to support providing the revenue analysis for users, and it is recommended to support recording the indicators such as time savings and cost savings.
| + +#### 9.8.2 Evaluation methods + +- a) The product information is recommended to be checked to determine whether the robotic process automation system supports the functions listed in Table 22. The product information is provided by the service providers. +- b) The listed items are recommended to be checked by setting product functional tests. The checked items involve the abilities of task plan, task online and task scheduling. + +# 10 Artificial intelligence empowerments + +### 10.1 Intelligent development + +#### 10.1.1 Technical requirements + +Intelligent development focuses on the artificial intelligence used in the development stage. By applying intelligent technologies, the development will be more convenient and efficient. The capability classification methods and related descriptions are shown in Table 23: + +**Table 23 – The capability classification methods of Intelligent development** + +| Competency levels | Descriptions | +|---------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Basic functions |
  • a) It is recommended to support tracking the user's actions and recording the critical nodes.
  • b) It is recommended to support categorizing the user's behaviours.
  • c) It is recommended to support simulating and reproducing the recorded operations.
  • d) It is recommended to support developing the process by using the chatbots.
  • e) It is recommended to support identifying or locating screen elements by using perception technologies such as optical character recognition and natural language processing.
  • f) It is recommended to support the script optimization, and automatically optimizing the operation's steps according to the existing processes and improve the efficiency.
  • g) It is recommended to support developing the process by using the chatbots.
  • h) It is recommended to support the reservation of AI components.
| +| Advanced functions |
  • a) It is recommended to support process development based on video analysis.
  • b) It is recommended to support the multi-modal human-computer interactions used in development which are based on various large model technologies.
| + +#### 10.1.2 Evaluation methods + +- a) The product information is recommended to be checked to determine whether the robotic process automation system supports the functions listed in Table 23. The product information is provided by the service providers. +- b) The listed items are recommended to be checked by setting product functional tests, the tests should consider several aspects such as: intelligent recording development, intelligent pickup, AI components, intelligent assistance. + +### 10.2 Intelligent management + +#### 10.2.1 Technical requirements + +Intelligent management focuses on AI used in system and stuff management. By applying intelligent technologies, the management will be more convenient and efficient. The capability classification methods and related descriptions are shown in Table 24: + +**Table 24 – The capability classification methods of Intelligent management** + +| Competency levels | Descriptions | +|---------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Basic functions |
  • a) It is recommended to support intelligent task scheduling and distribution.
  • b) It is recommended to support monitoring and recording the robot's operating status with AI technologies.
  • c) It is recommended to support the intelligent script analysis as well as automatic error correction and script optimization.
  • d) It is recommended to support the intelligent analysis and risk warning.
| +| Advanced functions |
  • a) It is recommended to support automatically generating the reports according to users' actions, the robots' operations, and the generated data.
  • b) It is recommended to support intelligent task mining and process mining.
  • c) It is recommended to support creating flowcharts and calculating the probability of every execution path.
  • d) It is recommended to support key points extraction from logs, it is recommended to support the visual display of these key points.
| + +#### 10.2.2 Evaluation methods + +- a) The product information is recommended to be checked to determine whether the robotic process automation system supports the functions listed in Table 24. The product information is provided by the service providers. +- b) The listed items are recommended to be checked by setting product functional tests, the tests should consider the task scheduling, the intelligent monitoring, the intelligent analysing. + +### 10.3 Intelligent execution + +#### 10.3.1 Technical requirements + +Intelligent execution focuses on AI used in process execution. By applying intelligent technologies, the convenience and efficiency of process execution will be improved. The capability classification methods and related descriptions are shown in Table 25: + +**Table 25 – The capability classification methods of Intelligent execution** + +| Competency levels | Descriptions | +|---------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Basic functions |
  • a) It is recommended to support the simple intelligent human-computer interactions during process execution.
  • b) It is recommended to support the recognition of several kinds of unformatted data, such as text, characters, and pictures.
  • c) It is recommended to support capturing the event data from a variety of business systems, such as enterprise resource planning (ERP) and customer relationship management (CRM).
  • d) It is recommended to support intelligent data classification, verification, error correction, and error comparison.
| +| Advanced functions |
  • a) It is recommended to support multiple ways to trigger robot processes, such as voice control and intelligent message conversation.
  • b) It is recommended to support the task transfer between humans and computers, it is recommended to support adding manual feedback during the process execution.
  • c) It is recommended to support using multiple perception technologies during process execution, such as optical character recognition (OCR) and natural language processing (NLP).
| + +#### **10.3.2 Evaluation methods** + +- a) The product information is recommended to be checked to determine whether the robotic process automation system supports the functions listed in Table 25. The product information is provided by the service providers. +- b) The listed items are recommended to be checked by setting product functional tests. The checked items involve the abilities of setting intelligent triggers, intelligent recognition, intelligent interaction. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.56-202510-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/F/T-REC-F.748.56-202510-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..25f8fbfaef317942ed3102c74b40a7d42f80608b --- /dev/null +++ b/marked/F/T-REC-F.748.56-202510-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:83cab7dcabfe939b223d97242d8523e461968d21f7c5432693c30a64f2d95521 +size 7050 diff --git a/marked/F/T-REC-F.748.56-202510-I_PDF-E/41a438d7e4adc17c3a4005e7c9500091_img.jpg b/marked/F/T-REC-F.748.56-202510-I_PDF-E/41a438d7e4adc17c3a4005e7c9500091_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..0e64bd91aa1d48d9b022df2f66474f80b59a082b --- /dev/null +++ b/marked/F/T-REC-F.748.56-202510-I_PDF-E/41a438d7e4adc17c3a4005e7c9500091_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:47fb904005861adc137a94c3add5bfedc542c1eda7f5e7c19b0175e988055710 +size 33559 diff --git a/marked/F/T-REC-F.748.56-202510-I_PDF-E/ddc7460821484f1ae2835c67955c554c_img.jpg b/marked/F/T-REC-F.748.56-202510-I_PDF-E/ddc7460821484f1ae2835c67955c554c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..571c135b5f80183f693aed6454165db70f4190c4 --- /dev/null +++ b/marked/F/T-REC-F.748.56-202510-I_PDF-E/ddc7460821484f1ae2835c67955c554c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b701bb9bc9da4a9c5c5f92b1e31e19a77db060a66dfbaee770b9cd951aaa8bfb +size 65011 diff --git a/marked/F/T-REC-F.748.56-202510-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.748.56-202510-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..13490665c37f40fdfc551f9171f25420df554ba0 --- /dev/null +++ b/marked/F/T-REC-F.748.56-202510-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cc6dcc1254736a0da311eaf0c1bf63c09d28bd97362a36f52065d91ef67cf6c5 +size 101654 diff --git a/marked/F/T-REC-F.748.56-202510-I_PDF-E/f57a881fcbcaf494a7dedc4f61224991_img.jpg b/marked/F/T-REC-F.748.56-202510-I_PDF-E/f57a881fcbcaf494a7dedc4f61224991_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..e6938c864c0c6aaf5bae526d378b64a729ea444f --- /dev/null +++ b/marked/F/T-REC-F.748.56-202510-I_PDF-E/f57a881fcbcaf494a7dedc4f61224991_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fc4d0907ebada5c7dad0c5be15971210b08fb8bc4587e004f6b05cce73525e21 +size 50213 diff --git a/marked/F/T-REC-F.748.56-202510-I_PDF-E/raw.md b/marked/F/T-REC-F.748.56-202510-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..13e660d9d3a105d3541f0b168926a3d57cdad6ba --- /dev/null +++ b/marked/F/T-REC-F.748.56-202510-I_PDF-E/raw.md @@ -0,0 +1,640 @@ + + +# Recommendation **ITU-T F.748.56 (10/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +## --- **Artificial intelligence generated content – General framework and requirements** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with white lines representing latitude and longitude, and the letters 'ITU' in a bold, blue, sans-serif font superimposed on the globe. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.56 + +## Artificial intelligence generated content – General framework and requirements + +## Summary + +Recommendation ITU-T F.748.56 defines a general three-layer framework for the artificial intelligence generated content\* (AIGC) system. The application layer is divided into unimodal/multimodal tasks from the perspective of content generation modalities. The model layer covers generative artificial intelligence (AI) technologies and multimodal foundational models related to AIGC. The infrastructure layer includes big data, cloud platforms, and computational modules. It forms an overall framework for the AIGC system, along with the functionality requirements, model capability benchmark testing scenarios, evaluation methods, evaluation metrics, and workflows. This Recommendation is intended to guide enterprises in building relevant AIGC capabilities for text, audio, image, video, 3D, and other modalities, and to provide a reference for testing, selecting, or evaluating AIGC technological capabilities. + +### History \*\* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.56 | 2025-10-17 | 21 | 11.1002/1000/16253 | + +### Keywords + +AIGC, AIGC model benchmark, content generation, requirements of AIGC system. + +--- + +\* Generation, creation, and regulation of content is outside the scope of this Recommendation. + +\*\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2026 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|------|---------------------------------------------------|----| +| 1 | Scope..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 1 | +| 4 | Abbreviations and acronyms ..... | 1 | +| 5 | Conventions ..... | 2 | +| 6 | General framework of the AIGC system ..... | 2 | +| 6.1 | Infrastructure layer ..... | 3 | +| 6.2 | Model support layer..... | 3 | +| 6.3 | Application layer ..... | 4 | +| 7 | Requirements of AIGC system functionalities ..... | 4 | +| 7.1 | Requirement for AI generated text ..... | 5 | +| 7.2 | Requirements for AI generated audio..... | 5 | +| 7.3 | Requirements for AI generated image..... | 6 | +| 7.4 | Requirements for AI generated video..... | 6 | +| 7.5 | Requirements for AI generated 3D content..... | 6 | +| 8 | Framework of AIGC models benchmark ..... | 7 | +| 8.1 | Workload ..... | 7 | +| 8.2 | Benchmarks ..... | 8 | +| 8.3 | Metrics..... | 8 | +| 8.4 | Workflow of AIGC model benchmark..... | 9 | +| | Appendix I – Use case of AIGC task workflow ..... | 10 | +| I.1 | Single modality task ..... | 10 | +| I.2 | Multi-modality task ..... | 10 | +| | Appendix II – Metrics calculation ..... | 11 | +| II.1 | Text generation metrics ..... | 11 | +| II.2 | Audio generation metrics ..... | 11 | +| II.3 | Image generation metrics ..... | 12 | +| II.4 | Video generation metrics..... | 12 | +| II.5 | 3D generation metrics..... | 13 | +| | Bibliography..... | 14 | + + + +# Recommendation ITU-T F.748.56 + +## Artificial intelligence generated content – General framework and requirements + +# 1 Scope + +This Recommendation provides the general framework and requirements for AIGC system capability, and the framework for the AIGC model benchmark. It addresses the following subjects: + +- The framework of the AIGC system; +- The requirements of AIGC system functionalities; +- The framework of the AIGC models benchmark. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[\[ITU-T F.748.57\]](#) Recommendation ITU-T F.748.57 (2025), *Technical requirements and assessment methods of generative artificial intelligence enabled multimedia applications*. + +# 3 Definitions + +### 3.1 Terms defined elsewhere + +None. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following term: + +**3.2.1 artificial intelligence generated content system:** A system that uses generative artificial intelligence to add, change, or create various types of content, including text, code, images, audio, videos, 3D content and other multimedia formats. + +NOTE – By analysing input data such as keywords, prompts, or other multimedia content, artificial intelligence generated content systems generate content that aligns with the user's specifications. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|-------------------------------------------| +| 3D | Three Dimensional | +| AI | Artificial Intelligence | +| AIGC | Artificial Intelligence Generated Content | +| ARM | Autoregressive Model | +| BLEU | Bilingual Evaluation Understudy | +| DGM | Deep Generative Model | + +| | | +|-------|-----------------------------------------| +| DM | Diffusion Model | +| FBM | Flow-Based Model | +| FID | Fréchet Inception Distance | +| GAN | Generative Adversarial Network | +| GPU | Graphics Processing Unit | +| KID | Kernel Inception Distance | +| MDR | Matching Distance Ratio | +| MOS | Mean Opinion Score | +| MPJPE | Mean Per Joint Positioning Error | +| PESQ | Perceptual Evaluation of Speech Quality | +| PSNR | Peak Signal-to-Noise Ratio | +| SSIM | Structural Similarity Index Measure | +| TPU | Tensor Processing Unit | +| VAE | Variational Autoencoder | + +## 5 Conventions + +The following conventions are used in this Recommendation: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted, if conformance to this Recommendation is to be claimed. +- The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. + +# 6 General framework of the AIGC system + +As illustrated in Figure 1, the general framework of the artificial intelligence generated content (AIGC) system involves the following three layers: + +- the infrastructure layer (including cloud platforms, computational modules, and big data), +- the model support layer (the model of AIGC is based on the multimodal foundation and generative techniques), and +- the application layer (including single modality task and multi-modality task, which are the two main modes of AIGC for content generation such as text generation, audio generation, image generation, video generation, and 3D content generation). (For the requirements see [ITU-T F.748.57]). + +![Figure 1 – General framework of the AIGC system. The diagram shows three layers: Application, Model support, and Infrastructure. The Application layer includes Content generation (Text, Audio, Image, Video, 3D), Single modality tasks (Chatbot, Audio inpainting, Image restoration, Partial 3D data completion and transformation), and Multimodality tasks (Text-to-image, Text-to-video, Text-to-audio, Text-to-3D, Image-to-text). The Model support layer includes Multimodal foundation model and Generative techniques. The Infrastructure layer includes Cloud platforms, Computational modules, and Big data.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +The diagram illustrates the general framework of the AIGC system, organized into three horizontal layers: + +- Application Layer:** This top layer is divided into three sub-sections: + - Content generation:** Includes Text generation, Audio generation, Image generation, Video generation, and 3D generation. + - Single modality task:** Includes Chatbot, Audio inpainting, Image restoration, and Partial 3D data completion and transformation. + - Multimodality task:** Includes Text-to-image, Text-to-video, Text-to-audio, Text-to-3D, and Image-to-text. +- Model support Layer:** This middle layer includes Multimodal foundation model and Generative techniques. +- Infrastructure Layer:** This bottom layer includes Cloud platforms, Computational modules, and Big data. + +F.748.56(25) + +Figure 1 – General framework of the AIGC system. The diagram shows three layers: Application, Model support, and Infrastructure. The Application layer includes Content generation (Text, Audio, Image, Video, 3D), Single modality tasks (Chatbot, Audio inpainting, Image restoration, Partial 3D data completion and transformation), and Multimodality tasks (Text-to-image, Text-to-video, Text-to-audio, Text-to-3D, Image-to-text). The Model support layer includes Multimodal foundation model and Generative techniques. The Infrastructure layer includes Cloud platforms, Computational modules, and Big data. + +**Figure 1 – General framework of the AIGC system** + +### 6.1 Infrastructure layer + +This layer comprises the foundational elements required for the AIGC system, including cloud platforms, computational modules, and big data. + +#### 6.1.1 Cloud platforms + +The cloud platforms equipped with computational capabilities and massive storage, enable efficient training of the foundation AIGC model and offer scalable resources and infrastructure for model development and deployment. + +#### 6.1.2 Computational modules + +The computational modules provide the necessary processing power for training and running AI models, such as graphics processing units (GPUs), tensor processing units (TPUs), or AI chips. + +#### 6.1.3 Big data + +The involved AIGC big data can be annotated or collected from the Internet, which can be unstructured, semi-structured, and reconstructed data, and based on the purpose it can be further classified as training data, fine-tuning data, and testing data. + +### 6.2 Model support layer + +The model support layer encompasses various components related to AI models used in the AIGC system. It includes both non-open-sourced foundational models and open-source foundational models. Non-open-sourced foundational models refer to proprietary or commercially developed models that serve as the basis for specific AIGC tasks. Open-source foundational models are publicly available widely used models that can be customized or extended for AIGC purposes. + +#### 6.2.1 Multimodal foundation model + +The multimodal foundation model serves as the engine of the AIGC system, which empowers upper-layer AIGC applications with increasingly strong real-time learning capabilities. + +#### 6.2.2 Generative techniques + +Generative techniques are introduced in training AIGC models. Classic techniques such as Transformer [b-Transformer] and others can be used to train AIGC models. The basic model + +structures could be autoregressive models (ARMs), variational autoencoders (VAEs), flow-based models (FBMs), generative adversarial networks (GANs), and diffusion models (DMs), etc. + +### **6.3 Application layer** + +The application layer focuses on specific content generation, including text generation, audio generation, image generation, video generation, and 3D content generation. AIGC models from the model layer are utilized in these tasks to generate the desired content, and these models can perform various unimodal or multimodal tasks. + +#### **6.3.1 Single modality task** + +Single modality tasks include implementing the generation of large amounts of data with similar characteristics based on existing data: + +- Text-to-text tasks such as chatbots are to provide better communication between humans, according to whether the task is specified in the applications, dialogue systems can be divided into two categories: task-oriented dialogue systems and open-domain dialogue systems. +- Audio-to-audio tasks such as audio inpainting are audio restoration tasks that reconstruct missing or corrupted portions of a digital audio signal. +- Image-to-image tasks such as image editing refer to changing the image semantics to meet a certain need in various forms, such as content, style, and object attributes. +- The 3D-to-3D tasks include completion from partial 3D data and transformation, with 3D object retrieval as a representative transformation task. + +#### **6.3.2 Multi-modality task** + +The multi-modality AIGC task generates multimodal data with corresponding characteristics based on the input demand, especially from textual descriptions to generate different content: + +- Text-to-image task aims to generate images from textual descriptions, the methods can be categorized into three branches, including GAN-based methods, autoregressive methods, and diffusion-based methods. +- Text-to-video aims to use natural language prompts as input to generate a video. This needs AIGC models to understand the context and semantics of the input text and then generate a corresponding video sequence. +- Text-to-audio aims to generate general audio given user description. +- Text-to-3D task aims to generate three-dimensional (3D) models from text descriptions while using wireless augmented reality (AR) applications. Typically, generating 3D models requires higher computational resources than generating two-dimensional images. +- Image-to-text, also known as image captioning, refers to describing a given image's content in natural language. + +#### **6.3.3 Content generation** + +Generative artificial intelligence technology typically provides AIGC services through product forms such as Internet systems, web pages, and platforms. Examples include text-generation chatbot products, image-generation platforms, video-generation platforms, and 3D content-generation platforms. + +## **7 Requirements of AIGC system functionalities** + +Requirements on the tasks of content generation among the AIGC applications layer as described in clause 6.3, are discussed in this clause. + +### 7.1 Requirement for AI generated text + +Text generation is required to reflect appropriate semantics and style based on the input provided. The requirements for AI-generated text include high-quality output. Table 1 describes the requirements of text generation. + +**Table 1 – Requirements of AI generated text** + +| Requirement | Description | +|----------------------------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| (Req. 7.1.1) Natural language coherence | The generated text is required to exhibit fluency and logical consistency, following natural language conventions. | +| (Req. 7.1.2) Diversity of output | The text generation system is recommended to allow the adjustment of diversity in text generation, ranging from highly creative outputs to more deterministic responses. | +| (Req. 7.1.3) Context awareness | The generated text is required to be relevant to the context, whether it is within a dialogue, storytelling, or technical documentation. | +| (Req. 7.1.4) Logical consistency | The generated text is required to maintain internal consistency, especially when generating longer content, ensuring coherent argumentation or storytelling. | +| (Req. 7.1.5) Real-time processing | The text generation system is required to ensure low-latency text generation that is suitable for real-time applications like interactive dialogue systems. | +| (Req. 7.1.6) Customizability | The text generation system is required to allow for custom inputs, enabling users to direct the tone, style, or length of the generated content. | +| (Req. 7.1.7) Multilingual generation support | The text generation system is recommended to support generating text in multiple languages. | +| (Req. 7.1.8) Domain-specific text generation | The text generation system is recommended to support fine-tuning for generating domain-specific text. | + +### 7.2 Requirements for AI generated audio + +The requirements for AI-generated audio include that it be natural-sounding, and free from artifacts or distortions. Table 2 describes the requirements of audio generation. + +**Table 2 – Requirements of AI generated audio** + +| Requirement | Description | +|--------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------| +| (Req. 7.2.1) Natural sound quality | Generated audio is required to sound lifelike, with clear, accurate pronunciation in speech or high-quality tones and rhythm in music generation. | +| (Req. 7.2.2) Seamless transitions | Generated audio is required to flow naturally without abrupt breaks or irregularities. | +| (Req. 7.2.3) Tone and style adaptability | The audio generation system is required to adapt the style of the audio to suit various contexts, such as professional speeches or casual conversations. | +| (Req. 7.2.4) Emotional expressiveness | Speech generation is recommended to incorporate tonal variations to convey appropriate emotions in context. | +| (Req. 7.2.5) Handling complex audio mixing | The audio generation system is recommended to handle advanced audio mixing tasks, including multi-speaker generation and cross-lingual voice synthesis. | + +### 7.3 Requirements for AI generated image + +The requirements for AI-generated image content involve aspects such as visual quality, semantic accuracy, diversity, context relevance, and resolution control. Table 3 describes the requirements of image generation. + +**Table 3 – Requirements of AI generated image** + +| Requirement | Description | +|--------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| (Req. 7.3.1) High visual quality | The generated images are required to be clear, detailed, and visually appealing, avoiding blurriness or distortion. | +| (Req. 7.3.2) Semantic accuracy | The generated images are required to accurately represent the input description or context, ensuring that generated content aligns with the intended meaning. | +| (Req. 7.3.3) Diversity of output | The image generation system is recommended to generate multiple variations of an image based on the same input, offering creative diversity. | +| (Req. 7.3.4) Contextual relevance | The generated images are required to maintain consistency with the input data or prompts, ensuring that the visual output accurately matches the user's request. | +| (Req. 7.3.5) Multi-modal integration | The image generation system is recommended to support multi-modal generation that combines text, images, and other media, as well as offering enhanced control over image properties like lighting, texture, and resolution. | + +### 7.4 Requirements for AI generated video + +The requirements for AI-generated video content involve aspects such as visual quality, and storytelling coherence. Table 4 describes the requirements of video generation. + +**Table 4 – Requirements of AI generated video** + +| Requirement | Description | +|--------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| (Req. 7.4.1) Visual quality | The generated videos are required to be visually coherent, with smooth transitions between scenes and consistent colour schemes. | +| (Req. 7.4.2) Storytelling coherence | The generated videos are required to follow a logical narrative, especially when generated from a script or input text, ensuring a consistent storyline. | +| (Req. 7.4.3) Frame consistency | The generated videos are required to maintain frame consistency with no noticeable artifacts or distortions, ensuring a smooth and immersive viewing experience. | +| (Req. 7.4.4) Real-time generation | The video generation system is required to support real-time video generation, ensuring quick generation times while maintaining quality. | +| (Req. 7.4.5) Dynamic resolution adjustment | The video generation system is recommended to dynamically adjust video resolution based on application needs. | + +### 7.5 Requirements for AI generated 3D content + +The requirements for AI-generated 3D content encompass aspects such as visual quality, semantic accuracy, realism, customization options, integration with other modalities, complexity control, and evaluation. Table 5 describes the requirements of 3D generation. + +**Table 5 – Requirements of AI generated 3D content** + +| Requirement | Description | +|------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| (Req. 7.5.1) Realism and detail | The generated 3D content is required to reflect accurate physical characteristics with attention to detail in texture, lighting, and depth. | +| (Req. 7.5.2) Semantic relevance | The generated 3D content is required to correspond precisely to the input prompt, whether it is text, image, or a combination of modalities. | +| (Req. 7.5.3) Customizability | The 3D content generation system is required to allow users to fine-tune 3D model attributes, such as size, shape, and material properties, to suit specific application needs. | +| (Req. 7.5.4) Integration with other modalities | The 3D content generation system is recommended to integrate seamlessly with other content types such as text, video, and augmented reality environments. | +| (Req. 7.5.5) Real-time 3D rendering | The 3D content generation system is recommended to support real-time 3D model rendering, with physics-based simulation. | + +## 8 Framework of AIGC models benchmark + +AIGC models must meet diverse user experience requirements to ensure user satisfaction. Requirements on the multimodal foundation model and generative techniques among the AIGC model support layer as presented in clause 6.2, are discussed in this clause. Benchmarking is also used to reflect the requirements. + +As shown in Figure 2, the benchmarking framework consists of three parts: + +- workload as the input part, +- the AIGC models benchmark with a specific evaluation environment, and +- metrics as the output part with the benchmark score. + +![Figure 2: Framework of AIGC models benchmark. The diagram shows a flow from Workload to Benchmark to Outputs. Workload includes Scenarios, Datasets (prompt), Evaluation object (AIGC models), Single modality task, and Multimodality task. Benchmark includes Evaluation program and AI framework. Outputs include Metrics, Text generation score, Audio generation score, Image generation score, Video generation score, and 3D generation score. A small code F.748.56(25) is visible at the bottom right of the diagram.](ddc7460821484f1ae2835c67955c554c_img.jpg) + +``` + +graph LR + subgraph Workload + W1[Scenarios] + W2[Datasets (prompt)] + W3[Evaluation object (AIGC models)] + W4[Single modality task] + W5[Multimodality task] + end + subgraph Benchmark + B1[Evaluation program] + B2[AI framework] + end + subgraph Outputs + O1[Metrics] + O2[Text generation score] + O3[Audio generation score] + O4[Image generation score] + O5[Video generation score] + O6[3D generation score] + end + Workload --> Benchmark + Benchmark --> Outputs + style F.748.56(25) fill:none,stroke:none + F.748.56(25) + +``` + +Figure 2: Framework of AIGC models benchmark. The diagram shows a flow from Workload to Benchmark to Outputs. Workload includes Scenarios, Datasets (prompt), Evaluation object (AIGC models), Single modality task, and Multimodality task. Benchmark includes Evaluation program and AI framework. Outputs include Metrics, Text generation score, Audio generation score, Image generation score, Video generation score, and 3D generation score. A small code F.748.56(25) is visible at the bottom right of the diagram. + +**Figure 2 – Framework of AIGC models benchmark** + +### 8.1 Workload + +The workload should be a representative AIGC application, ensuring a comprehensive and representative evaluation that aligns with the requirements of the industry-oriented application. Key principles include: + +- Wide variety of tasks: The workload requirements cover a variety of AIGC tasks such as text, image, audio, video, and 3D content generation depending on the single modality task and in multi-modality tasks, covering as many generative tasks as possible to reflect the model's capabilities. + +- Wide variety of data: The requirement for data sets to include a variety of inputs ensures that the model can handle multiple contexts, languages, domains, and levels of complexity. +- Task complexity: Workload requirements include both simple and complex tasks, allowing the benchmark to evaluate models of varying difficulty. +- Industry-oriented applications: Suggested workloads reflect common use cases such as content creation, entertainment, gaming, and education. + +### 8.2 Benchmarks + +The AIGC models benchmark is a comprehensive evaluation framework designed to assess the performance and effectiveness of AI models specifically focused on content generation. The benchmark evaluates various aspects of AIGC models, including the accuracy, fluency, coherence, relevance, and diversity of the generated content. It incorporates both automated evaluation metrics and human judgment-based assessments to ensure a comprehensive evaluation process. The evaluation is required to cover the following dimensions: + +- Accuracy and fluency: Models are required to generate content that closely aligns with the input prompts, ensuring grammatical correctness, coherence, and logical consistency. +- Creativity and diversity of output: Models are required to demonstrate the ability to generate novel and diverse outputs, avoiding repetitive or overly deterministic content. +- Speed and efficiency: The benchmark is required to evaluate how quickly and efficiently a model can generate outputs, particularly in real-time or time-sensitive applications. +- Adaptability and scalability: Models are recommended to be assessed based on their ability to adapt to new data or tasks and scale across different computational infrastructures. +- Human evaluation: Human judgment-based assessments are recommended for subjective quality attributes, such as creativity and context relevance. + +### 8.3 Metrics + +A combination of qualitative and quantitative metrics is utilized to evaluate the performance of models across various tasks. Quantitative metrics provide objective and numerical measures to assess different aspects of model performance. Qualitative metrics play a crucial role by capturing subjective assessments and human judgments. These metrics involve human evaluation, expert reviews, and user feedback to evaluate factors such as visual appeal, coherence, naturalness, readability, and user satisfaction. By considering both quantitative and qualitative metrics, a comprehensive evaluation approach is achieved, ensuring that the models perform effectively and meet both objective standards and subjective expectations across a wide range of tasks. Key principles include: + +- Text generation metrics, such as language translation that rely on metrics like bilingual evaluation understudy (BLEU), accuracy, and F1 score, and that measure the accuracy, fluency, and alignment of translations, are recommended. +- Audio generation metrics, like text-to-speech synthesis, utilize metrics such as mean opinion score (MOS), perceptual evaluation of speech quality (PESQ) to evaluate the quality, intelligibility and real-time performance of synthesized speech are recommended. +- Image generation metrics, like text-based image synthesis or generation, or metrics such as Fréchet inception distance (FID), peak signal-to-noise ratio (PSNR), and structural similarity index measure (SSIM) quantify the quality, fidelity and similarity of generated images are recommended. +- Video generation metrics, such as SSIM, GenVideo-RefVideo CLIP (corresponding frames), and frame count are recommended. +- 3D generation metrics, such as matching distance ratio (MDR), mean per joint positioning error (MPJPE) and kernel inception distance (KID) are recommended. + + + +## Appendix I + +### Use case of AIGC task workflow + +(This appendix does not form an integral part of this Recommendation.) + +This appendix describes use cases for each type of AIGC application, as described in clause 6.3. + +### I.1 Single modality task + +Single modality (unimodal) tasks include implementing the generation of large amounts of data with similar characteristics based on existing data transforming the characteristics or modality of existing data based on a demand. Taking image generation as an example, the training data for a generative model is a single modality. The inputs are the images to be generated or transformed, and the task could be generating a large number of images of similar style based on a specific style of image or transforming a photograph into a certain painting style. The workflow of such a single modality task is shown in Figure I.1. + +![Figure I.1: Workflow of a single modality task. The diagram shows a 'Unimodal model' block. On the left, 'Origin images' and 'Style images' are shown with a 'Train' arrow pointing to the model. Above the model, an 'Input' image is shown with a dashed arrow pointing to the model. On the right, a 'Generate' arrow points to a set of generated images, with the text 'F.748.56(25)' below them.](41a438d7e4adc17c3a4005e7c9500091_img.jpg) + +The diagram illustrates the workflow of a single modality task. It features a central 'Generative AI model' block labeled 'Unimodal model'. To the left, a dashed box contains 'Origin images' and 'Style images', with a 'Train' arrow pointing from this box to the model. Above the model, an 'Input' image is shown with a dashed arrow pointing to the model. To the right, a 'Generate' arrow points from the model to a set of generated images, with the text 'F.748.56(25)' below them. + +Figure I.1: Workflow of a single modality task. The diagram shows a 'Unimodal model' block. On the left, 'Origin images' and 'Style images' are shown with a 'Train' arrow pointing to the model. Above the model, an 'Input' image is shown with a dashed arrow pointing to the model. On the right, a 'Generate' arrow points to a set of generated images, with the text 'F.748.56(25)' below them. + +Figure I.1 – Workflow of a single modality task + +### I.2 Multi-modality task + +A multi-modality (multimodal) task generally refers to generating multimodal data with corresponding characteristics based on an input demand. Training data for the generative model contains each modality and its description. The inputs are prompts, such as text-generated images or text-generated audio prompts. Tasks could be generating images or audio based on the input text to generate a corresponding image or audio, or a description based on the image. The workflow of such a multi-modality task is shown in Figure I.2. + +![Figure I.2: Workflow of a multi-modality task. The diagram shows a 'Multimodal model' block. On the left, training inputs include a waveform labeled 'Description for the audio' and an image icon labeled 'Description for the image', with a 'Train' arrow pointing to the model. Below the model, an 'Input' section shows three prompts: A ('A white cat with two different colored eyes'), B ('A song about a cat'), and C (an image of a cat). Arrows labeled 'Text2Image Prompt', 'Text2Audio Prompt', and 'Text2Image Prompt' point from these inputs to the model. On the right, a 'Generate' arrow points to three outputs: A (an image of a cat), B (a waveform), and C (the text 'A cat sitting on the floor'). The text 'F.748.56(25)' is at the bottom right.](f57a881fcbcaf494a7dedc4f61224991_img.jpg) + +The diagram illustrates the workflow of a multi-modality task. It features a central 'Generative AI model' block labeled 'Multimodal model'. On the left, training inputs include a waveform labeled 'Description for the audio' and an image icon labeled 'Description for the image', with a 'Train' arrow pointing to the model. Below the model, an 'Input' section shows three prompts: A ('A white cat with two different colored eyes'), B ('A song about a cat'), and C (an image of a cat). Arrows labeled 'Text2Image Prompt', 'Text2Audio Prompt', and 'Text2Image Prompt' point from these inputs to the model. On the right, a 'Generate' arrow points to three outputs: A (an image of a cat), B (a waveform), and C (the text 'A cat sitting on the floor'). The text 'F.748.56(25)' is at the bottom right. + +Figure I.2: Workflow of a multi-modality task. The diagram shows a 'Multimodal model' block. On the left, training inputs include a waveform labeled 'Description for the audio' and an image icon labeled 'Description for the image', with a 'Train' arrow pointing to the model. Below the model, an 'Input' section shows three prompts: A ('A white cat with two different colored eyes'), B ('A song about a cat'), and C (an image of a cat). Arrows labeled 'Text2Image Prompt', 'Text2Audio Prompt', and 'Text2Image Prompt' point from these inputs to the model. On the right, a 'Generate' arrow points to three outputs: A (an image of a cat), B (a waveform), and C (the text 'A cat sitting on the floor'). The text 'F.748.56(25)' is at the bottom right. + +Figure I.2 – Workflow of a multi-modality task + +## Appendix II + +### Metrics calculation + +(This appendix does not form an integral part of this Recommendation.) + +This appendix describes qualitative and quantitative metrics for each type of AIGC, as described in clause 8.3. + +### II.1 Text generation metrics + +##### II.1.1 BLEU score + +Bilingual evaluation understudy (BLEU) score is a widely used metric for machine translation tasks. It calculates the precision of n-grams (up to a specified length) in the generated text compared to reference translations. The score ranges from 0 to 1, with higher values indicating better matches between the generated text and reference translations. + +$$BLEU = BP \times \exp(\sum_{n=1}^N w_n \log p_n) \quad (\text{II.1})$$ + +where: + +$p_n$ is the precision for n-grams (usually up to 4-grams) + +$w_n$ is the weight assigned to n-grams (often equal for all n) + +$BP$ is the brevity penalty, used to prevent very short translations from scoring well. + +##### II.1.2 Accuracy + +Accuracy is a straightforward metric used for tasks such as classification or question-answering. It measures the proportion of correctly generated or predicted words (or tokens) compared to the total number of words in the output. + +$$Accuracy = \frac{\text{Number of correct predictions}}{\text{Total predictions}} \quad (\text{II.2})$$ + +##### II.1.3 F1 score + +The F1 score is the harmonic mean of precision and recall, offering a balanced evaluation for tasks where both false positives and false negatives are important. It is commonly used in tasks such as text classification, named entity recognition (NER), and language generation. + +$$F1 \text{ Score} = 2 \times \frac{\text{Precision} \times \text{Recall}}{\text{Precision} + \text{Recall}} \quad (\text{II.3})$$ + +### II.2 Audio generation metrics + +##### II.2.1 MOS + +The mean opinion score (MOS) is the average of the ratings given by multiple listeners, providing an overall perception of the synthesized speech quality. + +$$MOS = \frac{1}{N} \sum_{i=1}^N r_i \quad (\text{II.4})$$ + +Where $r_i$ is the rating from listener $i$ , and $N$ is the total number of listeners. + +##### II.2.2 PESQ + +Perceptual evaluation of speech quality (PESQ) uses a model that simulates human auditory perception, and the specific calculation involves comparing the reference speech signal and the synthesized signal in the frequency domain. + +### II.3 Image generation metrics + +#### II.3.1 FID + +The Fréchet inception distance (FID) score is calculated as the Fréchet distance between two multivariate Gaussian distributions representing the real and generated image features. + +$$FID = \|\mu_r - \mu_g\|^2 + \text{Tr}(\Sigma_r + \Sigma_g - 2(\Sigma_r \Sigma_g)^{1/2}) \quad (\text{II.5})$$ + +$\mu_r, \mu_g$ are the mean feature vectors of real and generated images, respectively. + +$\Sigma_r, \Sigma_g$ are the covariance matrices of the real and generated image feature distributions. + +#### II.3.2 PSNR + +Peak signal-to-noise ratio (PSNR) is calculated by comparing the pixel-wise differences between the original and generated images. + +$$PSNR = 10 \cdot \log_{10} \left( \frac{(R_{\max})^2}{MSE} \right) \quad (\text{II.6})$$ + +$R_{\max}$ is the maximum possible pixel value (255 for 8-bit images). + +MSE is the mean squared error between the generated image and the reference (real) image. + +#### II.3.3 SSIM + +Structural similarity index measure (SSIM) measures the perceptual similarity between two images, considering structural information, luminance, and texture. It is designed to correlate better with human perception than PSNR, as it incorporates local patterns of pixel intensities and evaluates structural fidelity. + +The SSIM index ranges from -1 to 1, where 1 indicates perfect structural similarity between the generated and reference images. + +$$SSIM(x, y) = \frac{(2\mu_x\mu_y + C_1)(2\sigma_{xy} + C_2)}{(\mu_x^2 + \mu_y^2 + C_1)(\sigma_x^2 + \sigma_y^2 + C_2)} \quad (\text{II.7})$$ + +$\mu_x, \mu_y$ are the means of the images x and y + +$\sigma_x, \sigma_y$ are the variances + +$\sigma_{xy}$ is the covariance between x and y + +$C_1$ and $C_2$ are constants to avoid division by zero. + +### II.4 Video generation metrics + +#### II.4.1 GenVideo-RefVideo CLiP (corresponding frames) + +The GenVideo-RefVideo CLiP metric evaluates the temporal consistency and alignment between the frames of a generated video and the reference video. It compares each frame in the generated video to the corresponding frame in the reference video, typically using a feature-based similarity measure (e.g., cosine similarity, perceptual loss, or deep feature representations). + +$$\text{Cosine Similarity} = \frac{\text{Feature}_i \cdot \text{Feature}_j}{\|\text{Feature}_i\| \|\text{Feature}_j\|} \quad (\text{II.8})$$ + +where: + +$\text{Feature}_i$ are feature vectors of the i-th and j-th frames of the generated and reference videos. + +#### II.4.2 Video frame rate + +The frame rate of a video refers to the number of frames displayed per second (FPS). In video generation, this metric is essential for assessing the temporal smoothness and real-time performance of the generated video. Common frame rates for videos are 24 FPS, 30 FPS, or higher. + +$$\text{Frame Rate (FPS)} = \frac{\text{Total Number of Frames}}{\text{Video Duration in Seconds}} \quad (\text{II.9})$$ + +### II.5 3D generation metrics + +#### II.5.1 MDR + +Matching distance ratio (MDR) measures the geometric similarity between two 3D shapes, typically a generated 3D model and a reference model. It quantifies how well corresponding points or surfaces between the two 3D objects align. The matching distance is computed as the distance between corresponding points in the models, and the ratio compares these distances to some threshold to evaluate the degree of correspondence. + +$$\text{MDR} = \frac{\sum_{i=1}^N \|P_i - Q_i\|}{\sum_{i=1}^N \|P_i - P_i^*\|} \quad (\text{II.10})$$ + +$P_i$ and $Q_i$ are corresponding points on the generated and reference models, respectively + +$P_i^*$ is the closest point to $P_i$ in the reference model + +$N$ is the total number of points on the models. + +#### II.5.2 MPJPE + +Mean per joint positioning error (MPJPE) is commonly used for evaluating the accuracy of pose estimation or 3D human mesh generation. It measures the average Euclidean distance between corresponding joints in the generated 3D model and the reference ground truth. The lower the MPJPE, the more accurately the generated model replicates the pose or position of the reference model. + +$$\text{MPJPE} = \frac{1}{N} \sum_{i=1}^N \|P_i - Q_i\| \quad (\text{II.11})$$ + +$P_i$ and $Q_i$ are the positions of the $i$ -th joint in the generated and reference model, respectively. + +$N$ is the total number of joints. + +## Bibliography + +- [b-Transformer] Vaswani A., Shazeer N., Parmar N., Uszkoreit J., Jones L., Gomez A.N., Kaiser L., Polosukhin I. (2017), *Attention is all you need. Advances in Neural Information Processing Systems*. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.57-202510-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.748.57-202510-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..38bdd7bc1d3581fbdc3cf6b035f618693dbe3846 --- /dev/null +++ b/marked/F/T-REC-F.748.57-202510-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:75efa8e3513a799767c04a0bb7bcd87a550f2378f5880629068581808c1c9f7f +size 7353 diff --git a/marked/F/T-REC-F.748.57-202510-I_PDF-E/5860ad6bd2a2dd8d1ab12864b8f90f37_img.jpg b/marked/F/T-REC-F.748.57-202510-I_PDF-E/5860ad6bd2a2dd8d1ab12864b8f90f37_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7efdc4679db521f75285b6319554f2e15c4090d9 --- /dev/null +++ b/marked/F/T-REC-F.748.57-202510-I_PDF-E/5860ad6bd2a2dd8d1ab12864b8f90f37_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:95d2bab9d4a2266094beb19ed67881e526fda7435062a7d2cac0e6e66d3e61e5 +size 87439 diff --git a/marked/F/T-REC-F.748.57-202510-I_PDF-E/raw.md b/marked/F/T-REC-F.748.57-202510-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..14a784dd39c00f802c4b559b73f71634f4885be1 --- /dev/null +++ b/marked/F/T-REC-F.748.57-202510-I_PDF-E/raw.md @@ -0,0 +1,392 @@ + + +# Recommendation **ITU-T F.748.57 (10/2025)** + +SERIES F: Non-telephone telecommunication services +Multimedia services + +--- + +## **Technical requirements and assessment methods of generative artificial intelligence enabled multimedia applications** + +![ITU logo](390120de4fe440c42fea8154fcaad334_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.57 + +# Technical requirements and assessment methods of generative artificial intelligence enabled multimedia applications + +## Summary + +Recommendation ITU-T F.748.57 establishes a framework of technical requirements and assessment methods for generative artificial intelligence-enabled multimedia applications. This Recommendation provides users with a clearer understanding of generative artificial intelligence (AI) technology and support them in setting up generative AI applications more efficiently and appropriately. The Recommendation provides guidelines for users to assess the capabilities of generative artificial intelligence applications and promote their effectiveness. + +## History\* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.57 | 2025-10-17 | 21 | 11.1002/1000/16254 | + +## Keywords + +Assessment methods, generative artificial intelligence, technical requirements, testing metrics. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2025 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-------------------------------------------------------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 1 | +| 5 Conventions ..... | 2 | +| 6 Framework of technical capabilities for generative artificial intelligence enabled multimedia applications..... | 2 | +| 7 Technical requirements of generative artificial intelligence-enabled multimedia applications ..... | 3 | +| 7.1 Result quality..... | 3 | +| 7.2 Task diversity ..... | 3 | +| 7.3 Model capability..... | 3 | +| 7.4 Basic capability ..... | 4 | +| 7.5 Interaction capability ..... | 4 | +| 8 Assessment methods of generative artificial intelligence-enabled multimedia applications ..... | 4 | +| 8.1 Assessment objects..... | 4 | +| 8.2 Testing metrics ..... | 5 | +| 8.3 Testing methods..... | 6 | +| Appendix I – Use cases of generative artificial intelligence-enabled multimedia applications ..... | 7 | + + + +# Recommendation ITU-T F.748.57 + +## Technical requirements and assessment methods of generative artificial intelligence enabled multimedia applications + +# 1 Scope + +This Recommendation mainly focuses on generative artificial intelligence-enabled multimedia applications. It establishes technical requirements for these multimedia application, and sets up assessment methods to measure the performance of those applications. The general framework and requirements is provided in [ITU-T F.748.56]. + +The scope of this Recommendation includes three parts: + +- Framework of technical abilities and assessment methods for generative artificial intelligence enabled multimedia applications. +- Technical requirements of generative artificial intelligence-enabled multimedia applications. +- Assessment methods of generative artificial intelligence-enabled multimedia applications. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T F.748.56] Recommendation ITU-T F.748.56 (2025), *Artificial intelligence generated content – General framework and requirements*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +None. + +## 3.2 Terms defined in this Recommendation + +None. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-------|---------------------------------------------------| +| 3D | Three-dimensional | +| AI | Artificial Intelligence | +| API | Application Programming Interface | +| FID | Fréchet Inception Distance | +| IS | Inception Score | +| ROUGE | Recall-Oriented Understudy for Gisting Evaluation | + +# 5 Conventions + +In this Recommendation: + +- The keywords "**is recommended to**" indicate a requirement that is recommended but which is not absolutely required. Thus, this requirement does not need to be present to claim conformance. +- The keywords "**optional**" indicate an optional requirement that is permissible, without implying any sense of being recommended. These terms are not intended to imply that the vendor's implementation must provide the option, nor that the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. +- The keyword "**functions**" is defined as a collection of functionalities. + +# 6 Framework of technical capabilities for generative artificial intelligence enabled multimedia applications + +This Recommendation defines the technical requirements of generative artificial intelligence (AI) applications. Regarding result quality, generative AI applications are recommended to ensure high-quality output results. In terms of task diversity, generative AI applications are recommended to support various tasks from industries. For model capability, generative AI applications are recommended to possess a range of generative abilities across different modalities. In terms of interaction capability, generative AI applications are recommended to be able to interact with users and optimize performance based on user feedback. For basic capability, generative AI applications are recommended to have a clear input or output structure and support the configuration of parameters. + +Additionally, this Recommendation separately outlines evaluation methods, including specific assessment objects, testing metrics and testing methods, for each of the technical requirements, combined with generation modalities. + +![Diagram illustrating the framework of technical capabilities for generative artificial intelligence-enabled multimedia applications. The diagram shows a hierarchy of capabilities and their relationship to assessment methods.](5860ad6bd2a2dd8d1ab12864b8f90f37_img.jpg) + +The diagram illustrates the framework of technical capabilities for generative artificial intelligence-enabled multimedia applications. It is structured as follows: + +- Technical capabilities** (bottom level): + - Result quality + - Task diversity + - Interaction capability + - Model capability + - Basic capability +- Generative modalities** (second level): + - Image, video, speech, code, structured data, 3D models ... +- Generative AI enabled multimedia applications** (third level): + - Image generating platform, code assistant, 3D generating tools ... +- Various industries usage** (top level): + - Financial, automobile manufacturing, new media, biomedicine ... + +Arrows indicate a bottom-up flow from Technical capabilities to Generative modalities, then to Generative AI enabled multimedia applications, and finally to Various industries usage. A dashed box encloses the top three levels. To the right, a box labeled **Assessment methods** contains three sub-boxes: **Testing methods**, **Testing metrics**, and **Assessment objects**. Arrows point from the top three levels of the hierarchy to the Assessment methods box. The reference code F.748.57(25) is shown at the bottom right. + +Diagram illustrating the framework of technical capabilities for generative artificial intelligence-enabled multimedia applications. The diagram shows a hierarchy of capabilities and their relationship to assessment methods. + +**Figure 1 – Framework of technical capabilities for generative artificial intelligence-enabled multimedia applications** + +# **7 Technical requirements of generative artificial intelligence-enabled multimedia applications** + +## **7.1 Result quality** + +Generative AI applications are recommended to ensure the quality of results through inspection and archiving. + +Technical requirements: + +- a) Inspect users' suggestions: generative AI applications are recommended to set up mechanisms for collecting user feedback, including thumbs-up button, pop-up window, etc. +- b) Inspect information archiving: generative AI applications are recommended to support elements recording, including evaluation time, evaluation text, evaluation users, backtracking and finding of inspect information. The assessment of inspection information is recommended to occur within the storage cycle. + +## **7.2 Task diversity** + +Generative AI applications are recommended to support various tasks in different scenarios. + +Technical requirements: + +- a) For text generation, applications may optionally support conversion, dialogue, compression, generation and computing tasks. +- b) For image generation, applications may optionally support the creative generation and special effects processing tasks. +- c) For video generation, applications may optionally support video repairing, creative video generation, special effects animation generation tasks, and virtual image generation tasks. +- d) For audio generation, applications may optionally support text-to-audio tasks and audio generation tasks. +- e) For three-dimensional (3D) generation, applications may optionally support 3D model generation, modelling materials, and 3D model editing tasks. + +## **7.3 Model capability** + +For different generative modalities, the underlying model of generative AI applications is recommended to have a range of capabilities. + +Technical requirements: + +- a) For text-generated applications, they are recommended to support the capability of multi-language transformation. Besides, these applications are recommended to have task-based dialogue systems and chit-chat dialogue systems that support tasks like summary generation, poetry generation, story generation, search generation, machine translation, graph-text generation and mathematical operations. +- b) For image-generated applications, they are recommended to have capabilities such as size editing, colour tuning, adding filters, lens correction, view correction, free rotation, free cropping, curve tool, white balance correction, and lasso keying. In addition, these applications are recommended to support tasks like text-generated images, graphic-generated images, image restoration, image super-resolution, image re-illumination and font generation. +- c) For video-generated applications, they are recommended to have capabilities such as video cutting, video collage, inserting subtitles, adding watermarks, special effects processing, denoising and noise reduction, and face fabrication. Furthermore, these applications are + +recommended to support tasks like text-generated video, single-image-generated video, multi-image-generated video, video repair and video relighting. + +- d) For audio-generated applications, they are recommended to have capabilities such as frequency equalization, timbre settings, audio effects, and paragraph adjustment. Moreover, these applications are recommended to support tasks like melody generation, chord generation, style conversion, performance generation, audio synthesis and vocal synthesis. +- e) For 3D-generated applications, they are recommended to have capabilities such as 3D modelling, video texture, and displacement of an object. Additionally, these applications are recommended to support tasks like art design, animation generation, advertising production and object simulation. + +## 7.4 Basic capability + +Generative AI applications are recommended to have basic generative functionality and support configuration of parameters. + +Technical requirements: + +- a) Parameter configuration: applications based on generative AI are recommended to set up functions and resource import aspects. +- b) Input modality: applications of generative AI are recommended to ensure the diversity of input modes. Considering that the product has not yet reached a high level of maturity at this stage, the product only needs to meet one or more input modes. +- c) Generation function: applications based on generative AI are recommended to have different functions. These functions are recommended to be diverse and can be evaluated by selecting the corresponding modality in combination with the product type. +- d) Result evaluation: applications based on generative AI are recommended to support automatic optimization, which is divided into two types: manual optimization and automatic optimization. The diversity of each optimization modality needs to be evaluated. + +## 7.5 Interaction capability + +Generative AI applications are recommended to have abilities of system management, interaction mode, measurement mode, feedback mechanism and user experience. + +Technical requirements: + +- a) System management: the system management of generative AI is a comprehensive work, which is recommended to include the design of functional modules and user operation modules. +- b) Interaction mode: the service of generative AI is recommended to have more diverse interaction methods. +- c) Measurement mode: the service of generative AI is recommended to have the ability of customization and be able to achieve customized development and tuning. +- d) Feedback mechanism: the service of generative AI is recommended to have the ability to provide differentiated services for different types of users through a variety of methods. + +## 8 Assessment methods of generative artificial intelligence-enabled multimedia applications + +## 8.1 Assessment objects + +Specific definitions of test objects, which may include product-specific tools, application programming interfaces (APIs), or offline deployment packages. The current form of tools for + +generative artificial intelligence mainly includes the following types: online web pages, standalone apps, embedded systems, online interfaces, offline interfaces, and standalone plug-ins. + +All these applications are recommended to be assessment objects of generative artificial intelligence. Ordinary users can use generative artificial intelligence tools through online web pages and standalone apps. If users wish to use generative artificial intelligence developer mode, they can access it through an interface. Standalone plug-ins are developed based on specific needs, and large organizations use this approach to apply generative artificial intelligence. + +## 8.2 Testing metrics + +This part defines the test metrics in the context of specific generative modalities respectively. + +#### 8.2.1 Readability + +This metric is recommended to test the ease of understanding and acceptance of the generated text. Test methods are divided into two specific areas as follows. + +- a) Grammatical structure accuracy: focus on the accuracy of the generated text in terms of word collocation, punctuation and sentence structure. +- b) Generated text comprehensibility: the degree to which the generated text is comprehensible to the reader. + +#### 8.2.2 Expertise + +This metric is recommended to examine whether the generated text meets the specialization requirements of the domain or scenario. + +#### 8.2.3 User friendliness + +This metric is recommended to assess if the generated content1 is enjoyable and interesting to users. + +#### 8.2.4 Quality of rendering + +This metric is recommended to evaluate the rendering effects and the use of rendering techniques in the generated images. + +#### 8.2.5 Clarity + +This metric is recommended to evaluate the clarity of textures and edge contours in the generated video, as well as the stability and naturalness of the image. + +#### 8.2.6 Recall-oriented understudy for gisting evaluation (ROUGE) + +This metric is recommended to evaluate the similarity between the generated text and the reference text to assess the quality of text generation. + +In this formula, recall is calculated after splitting the results generated by the model and the standard results into n-grams. The appropriate n is selected, and Rouge-n is calculated as follows. + +$$Rouge-n = \frac{\sum_{S \in \{RefText\}} \sum_{gram_n \in S} Count_{match}(gram_n)}{\sum_{S \in \{RefText\}} \sum_{gram_n \in S} Count(gram_n)}$$ + +$Count(gram_n)$ : the number of n-grams in the reference text collection. + +$Count_{match}(gram_n)$ : the number of n-grams that the generated text shares with the reference text is considered. + +$RefText$ : the reference summary from text testing sets. + +--- + +1 Generation, creation and regulation of content is outside the scope of this Recommendation. + +#### 8.2.7 Fréchet inception distance (FID) + +This metric is recommended to focus on indicators that assess the difference between the generated model and the real data distribution. + +This formula calculates the metric between the feature vectors of the real image and the generated image. + +$$FID = \|\mu_r - \mu_g\|^2 + \text{Tr} \left( \Sigma_r + \Sigma_g - 2(\Sigma_r \Sigma_g)^{\frac{1}{2}} \right)$$ + +$d^2$ : FID distance. + +$(\mu_r, \Sigma_r)$ : mean and variance of the true distribution. + +$(\mu_g, \Sigma_g)$ : mean and variance of the generated distribution. + +$\text{Tr}$ : the trace of the representation matrix (sum of diagonal elements of the matrix). + +### 8.3 Testing methods + +This part defines the method of conducting the test according to the respective generative modalities. + +- a) For text generation, testing methods of text comparison and calibration are optionally used. First, set up test datasets, then compare the generated content with the correct answers, and calculate the accuracy of the generated content2. +- b) For image generation, testing methods are optionally used to compare the generated images with real images. First, input prompts into generative AI applications, then collect output results, compare generative images with ground truth, and finally evaluate performance according to testing metrics. +- c) For video generation, testing methods are optionally used to compare the generated videos with real videos. First, input prompts into generative AI applications, then collect output results and compare generative videos with ground truth, and finally evaluate performance according to testing metrics. + +--- + +2 Generation, creation and regulation of content are outside the scope of this Recommendation. + +# Appendix I + +### **Use cases of generative artificial intelligence-enabled multimedia applications** + +(This appendix does not form an integral part of this Recommendation.) + +At present, generative artificial intelligence (AI) has been widely used in a variety of industries in the real-world scenarios. Generative AI applications can add content to existing documents and rewrite parts or entire documents. For example, in the financial sector, some research reports can be completed with the help of generative AI assistants. In the insurance industry, some companies have used generative AI technology in actual model training and will train models with the help of generated image data to serve their business. + +Another example is the automobile manufacturing industry. The knowledge base that has been formed helps integrate the generative AI capability into an intelligent customer system, which enables automobile sales to provide a better customer experience. + +Besides, generative AI technology makes efforts towards the early stages of vehicle development and design. New car development in styling can further shorten the development cycle, reduce iteration, and ultimately accelerate of new car development. By using generative AI technology, protein sequences can be generated with predictable functions after being trained with 280 million sequences from known protein families. + +As can be seen, in a variety of industries, generative AI can effectively help staff in business, industry, and other fields fulfil their work and play a central role in improving overall efficiency. + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.58-202508-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/F/T-REC-F.748.58-202508-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c1629394c02f697f60ae954ad36bf274fa8082cb --- /dev/null +++ b/marked/F/T-REC-F.748.58-202508-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3763acd7a0ed56aa30f3f90468c1fb0cb54573eb9045bca3eddf8976b4f0f41d +size 7219 diff --git a/marked/F/T-REC-F.748.58-202508-I_PDF-E/1a85642ed2356d183ce598f2c8b3ee8b_img.jpg b/marked/F/T-REC-F.748.58-202508-I_PDF-E/1a85642ed2356d183ce598f2c8b3ee8b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..4b244f6334cf29db9ab745772884f9a22976eb01 --- /dev/null +++ b/marked/F/T-REC-F.748.58-202508-I_PDF-E/1a85642ed2356d183ce598f2c8b3ee8b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:24739dbe60a2255194c216248d5dd7c30f44ebaf529791b07d3c1a5c2e974697 +size 23229 diff --git a/marked/F/T-REC-F.748.58-202508-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg b/marked/F/T-REC-F.748.58-202508-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f7dc2e9f1118d6e1d5b505ddbd890519ca30d5a2 --- /dev/null +++ b/marked/F/T-REC-F.748.58-202508-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0fd2876a171d8aef149ac9f9d2012981007aabbdb1a13d47837bb3d077ffd376 +size 162226 diff --git a/marked/F/T-REC-F.748.58-202508-I_PDF-E/4cc7cdce3d498d8b0ba033a9be24ade5_img.jpg b/marked/F/T-REC-F.748.58-202508-I_PDF-E/4cc7cdce3d498d8b0ba033a9be24ade5_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..91d2ea2788e2766059ba790dedebbde6ed4f5034 --- /dev/null +++ b/marked/F/T-REC-F.748.58-202508-I_PDF-E/4cc7cdce3d498d8b0ba033a9be24ade5_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b63059b6638a282c4b268282b43157b2aa5e8bd8118ff470d20066556f2a436c +size 16086 diff --git a/marked/F/T-REC-F.748.58-202508-I_PDF-E/898fb89a50d9ec1dfb4e425c816976a7_img.jpg b/marked/F/T-REC-F.748.58-202508-I_PDF-E/898fb89a50d9ec1dfb4e425c816976a7_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..bf8bd17e8a43877722ddbd60812026665440b5a5 --- /dev/null +++ b/marked/F/T-REC-F.748.58-202508-I_PDF-E/898fb89a50d9ec1dfb4e425c816976a7_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f7175b34f3dbad45c102a37661f5201d7f3c531ae7e90244e40addb465760b76 +size 32960 diff --git a/marked/F/T-REC-F.748.58-202508-I_PDF-E/b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg b/marked/F/T-REC-F.748.58-202508-I_PDF-E/b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ac1b2bf364f93bb9d0f76028ac252d2a15378270 --- /dev/null +++ b/marked/F/T-REC-F.748.58-202508-I_PDF-E/b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c1c1e60d5ceb30bf57e9274696268c0b4a685cfcd26c402200d198adce4a514a +size 100959 diff --git a/marked/F/T-REC-F.748.58-202508-I_PDF-E/b6671cfafda3820aafe9a24fa7a4d8c7_img.jpg b/marked/F/T-REC-F.748.58-202508-I_PDF-E/b6671cfafda3820aafe9a24fa7a4d8c7_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a138a5a312341e3d2e9e2b6fbd1a517c8caebc14 --- /dev/null +++ b/marked/F/T-REC-F.748.58-202508-I_PDF-E/b6671cfafda3820aafe9a24fa7a4d8c7_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:650e0a105aa727f8dbee1e83f032a6e897ba544c27cc577bd5df6f87adb150bc +size 18965 diff --git a/marked/F/T-REC-F.748.58-202508-I_PDF-E/d9c0a780cd22626253dab4aa41699e2f_img.jpg b/marked/F/T-REC-F.748.58-202508-I_PDF-E/d9c0a780cd22626253dab4aa41699e2f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..0b4437262960a5dbe1b41ad9b6f9aa527f6d79b5 --- /dev/null +++ b/marked/F/T-REC-F.748.58-202508-I_PDF-E/d9c0a780cd22626253dab4aa41699e2f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f72b8009298392a1650f4caa838eda5078ef365da8934a014f4656dc6c2fa71f +size 65059 diff --git a/marked/F/T-REC-F.748.58-202508-I_PDF-E/e69b9188aa2c14ec6b21c83f711fef65_img.jpg b/marked/F/T-REC-F.748.58-202508-I_PDF-E/e69b9188aa2c14ec6b21c83f711fef65_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..6e46298be2a94276f37717806caf120187c6d033 --- /dev/null +++ b/marked/F/T-REC-F.748.58-202508-I_PDF-E/e69b9188aa2c14ec6b21c83f711fef65_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d11957fd26e22a9401990c41368f9787cd8922db35e5b42149b8ff1d4a12f341 +size 37781 diff --git a/marked/F/T-REC-F.748.58-202508-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.748.58-202508-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9c701a197524b86ff83ca9c4c2e3161e8ac20b9f --- /dev/null +++ b/marked/F/T-REC-F.748.58-202508-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:088af1ed4e5ac5ba37aa594bab6ae37f4fa41d06876872bc7017ce2ff326801c +size 37488 diff --git a/marked/F/T-REC-F.748.58-202508-I_PDF-E/raw.md b/marked/F/T-REC-F.748.58-202508-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..9337241db0f78391be7a7ab958e224e0edc0992c --- /dev/null +++ b/marked/F/T-REC-F.748.58-202508-I_PDF-E/raw.md @@ -0,0 +1,708 @@ + + +# Recommendation **ITU-T F.748.58 (08/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +**Requirements and architectures of multimedia +platform for digital human services using edge +cloud** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.58 + +# Requirements and architectures of multimedia platform for digital human services using edge cloud + +## Summary + +Media services utilizing digital human technology encompass diverse application scenarios and incorporate various essential technologies, including modelling, animation and rendering. To address the demands for high fidelity, real-time responsiveness and concurrent usage in digital human services, a new initiative aims to introduce additional functional capabilities based on multimedia and edge cloud technologies. These enhancements aim to enable a broad range of media services leveraging existing digital human technology while meeting the aforementioned requirements. + +Recommendation ITU-T F.748.58 specifies the use cases, requirements and architecture for a multimedia platform that facilitates processing, interaction and management in digital human services. The platform is built upon the existing technologies of edge cloud and digital human as a reference. + +## History\* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.58 | 2025-08-13 | 21 | 11.1002/1000/16431 | + +## Keywords + +Digital human, edge cloud, multimedia platform. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2025 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | | Page | +|--------------|--------------------------------------------------------------------------------------------|------| +| 1 | Scope..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 2 | +| 4 | Abbreviations and acronyms ..... | 2 | +| 5 | Conventions ..... | 2 | +| 6 | Overview..... | 2 | +| 7 | Requirements of multimedia platform for digital human services using the edge cloud ..... | 4 | +| 7.1 | Requirements for media processing ..... | 4 | +| 7.2 | Requirements for multimedia interaction..... | 7 | +| 7.3 | Requirements for content management ..... | 7 | +| 8 | Architectures of multimedia platform for digital human services using the edge cloud ..... | 7 | +| 8.1 | General architecture..... | 7 | +| 8.2 | Content management function..... | 8 | +| 8.3 | Media processing function ..... | 9 | +| 8.4 | Media interaction function ..... | 10 | +| 8.5 | Reference points ..... | 11 | +| Appendix I | – Use cases of DHMP using edge cloud..... | 13 | +| I.1 | Digital human call ..... | 13 | +| I.2 | Digital human multiparty call..... | 13 | +| I.3 | Digital human live streaming ..... | 14 | +| I.4 | Online digital human ..... | 15 | +| I.5 | Digital human socializing..... | 15 | +| Appendix II | – Key issues of DHMP using edge cloud..... | 17 | +| II.1 | Key issue No. 1: NeRF-based data compression..... | 17 | +| II.2 | Key issue No. 2: Differentiated function provision for multiple access ..... | 18 | +| Appendix III | – Correspondence between requirements and architectures ..... | 20 | +| III.1 | Correspondence between requirements and architectures..... | 20 | +| Bibliography | ..... | 21 | + + + +# Recommendation ITU-T F.748.58 + +## Requirements and architectures of multimedia platform for digital human services using edge cloud + +# 1 Scope + +This Recommendation specifies the requirements and architectures of multimedia platform for digital human services using edge cloud technology. + +The scope of this Recommendation includes: + +- Requirements of multimedia platform for digital human services; +- Architectures of multimedia platform for digital human services; and +- Use cases and key issues of multimedia platform for digital human services. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[[ITU-T F.700](#)] Recommendation ITU-T F.700 (2000), *Framework Recommendation for multimedia services*. + +[[ITU-T F.748.15](#)] Recommendation ITU-T F.748.15 (2022), *Framework and metrics for digital human application systems*. + +# 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 digital human** [ITU-T F.748.15]: A computer application that integrates the technologies of computer graphics, computer vision, intelligent speech and natural language processing. It can be used for digital content generation and human-computer interaction to help improve content production efficiency and user experience. + +**3.1.2 edge cloud** [b-ITU-T Y.3508]: A cloud computing deployed to the edge of the network accessed by cloud service customers (CSCs) with small capacity resources enabling cloud service. + +NOTE 1 – Enabled cloud service on the edge cloud is lightweight cloud service provided by a cloud service provider (CSP) depending on cloud service category. + +NOTE 2 – Lightweight cloud service refers to a portion of cloud service to reconfigure the functionality of cloud service to fit on edge cloud such as base station and gateway with small capacity resource. + +**3.1.3 platform** [b-ITU-T G.1081]: A hardware and/or software architecture that serves as a foundation or base for realizing a certain functionality. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following term: + +**3.2.1 digital human multimedia platform:** A hardware or software system that serves as a foundation for the identification and description of the multimedia aspects of digital human services. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|---------|-----------------------------------------------------------------------| +| 3DGS | Three Dimensional Gaussian Splatting | +| AR | Augmented Reality | +| CC | Core Cloud | +| CSP | Cloud Service Provider | +| DASH | Dynamic Adaptive Streaming over Hypertext transfer protocol | +| DHMP | Digital Human Multimedia Platform | +| DQN | Deep Q-Network | +| EC | Edge Cloud | +| HEIF | High efficiency video coding Image File | +| ISOBMFF | International Organization for Standardization Base Media File Format | +| MEC | Multi-access Edge Computing | +| MIHS | MPEG-I Haptic Stream | +| MMT | MPEG Media Transport | +| NeRF | Neural Radiance Field | +| QoE | Quality of Experience | +| QoS | Quality of Service | +| UE | User Equipment | +| VR | Virtual Reality | +| XR | Extended Reality | + +# 5 Conventions + +In this Recommendation: + +The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. + +The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement needs not be present to claim conformance. + +# 6 Overview + +The digital human multimedia platform (DHMP) focuses on the media component level of multimedia services reference model specified in [ITU-T F.700], shown in Figure 1, which specifies the three levels of service platform, namely service level, communication task level and media component level. For the media component level is made apparent through the identification and description of the multimedia aspects of the services. + +![Figure 1: Multimedia services reference model. A diagram showing a funnel-shaped model with four levels: Application level (top), Service level, Communication task level, and Media component level (bottom). Arrows point from the top level to a box labeled 'Multimedia applications'. A vertical double-headed arrow on the left is labeled 'Service platform'. The bottom right of the funnel is labeled 'F.748.58(25)'.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +Figure 1: Multimedia services reference model. A diagram showing a funnel-shaped model with four levels: Application level (top), Service level, Communication task level, and Media component level (bottom). Arrows point from the top level to a box labeled 'Multimedia applications'. A vertical double-headed arrow on the left is labeled 'Service platform'. The bottom right of the funnel is labeled 'F.748.58(25)'. + +**Figure 1 – Multimedia services reference model** + +Based on the digital human system specified in [ITU-T F.748.15], Figure 2 shows a media component of the DHMP as an example. The platform follows the modules of the digital human system in [ITU-T F.748.15], i.e., it includes an image module, speech module, animation module and interactive processing module; and because this platform uses cloud services, a new data transmission module is added as a media function integration for this platform. + +![Figure 2: A media component of digital human service platform. A detailed diagram of the 'Core/Edge cloud' platform. It contains three main modules: 'Digital human image management' (with Provision, Storage, Retrieval), 'Digital human speech and animation' (with Generation, Animation, Rendering), and 'Digital human interactive processing' (with Decision, Perception). A 'Data transmission' module on the right includes 'Encode decode' and 'Encapsulation decapsulation'. Arrows point from the platform to 'Multimodal output' (Audio, Video) and 'Multimodal input' (Audio, Video, Image, Text, Action). The bottom is labeled 'Reference media component of digital human service platform' and 'F.748.58(25)'.](b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg) + +Figure 2: A media component of digital human service platform. A detailed diagram of the 'Core/Edge cloud' platform. It contains three main modules: 'Digital human image management' (with Provision, Storage, Retrieval), 'Digital human speech and animation' (with Generation, Animation, Rendering), and 'Digital human interactive processing' (with Decision, Perception). A 'Data transmission' module on the right includes 'Encode decode' and 'Encapsulation decapsulation'. Arrows point from the platform to 'Multimodal output' (Audio, Video) and 'Multimodal input' (Audio, Video, Image, Text, Action). The bottom is labeled 'Reference media component of digital human service platform' and 'F.748.58(25)'. + +**Figure 2 – A media component of digital human service platform** + +- Digital human image management: A module that offers comprehensive management capabilities, facilitating the provision, storage and retrieval of digital human assets. It provides efficient content management, enabling easy search, categorization and modification of digital human elements, ensuring seamless integration into various projects. + +- Digital human speech and animation: A module that incorporates advanced modelling, animation and rendering technologies, allowing for the creation of lifelike and expressive digital humans. These digital representations can be integrated into virtual reality (VR), augmented reality (AR), gaming and other multimedia applications. +- Digital human interactive processing: A module that supports interactive functionalities, enabling users to interact with digital humans through natural user interfaces such as voice commands, gestures and facial expressions. This creates actions of a digital human to perceive, recognize, understand and make decisions. +- Data transmission: A module that enables digital human models to be distributed and enhances the overall immersion and engagement for users. It includes functions such as the encoding/decoding and encapsulation of data. + +The multimedia platform needs to be robust and versatile, revolutionizing the way human-centric content is processed, interacted with and managed. It serves as a centralized hub, integrating cutting-edge technologies and providing a seamless user experience across various applications. Therefore, to realize real-time and multi-access experience, the edge cloud can assist to offload media traffic and computing resources of cloud services, and the platform as a cloud service is executed on both edge and core clouds by interworking. Some of the media feature optimization based on edge cloud characteristics are as follows: + +- Multimedia data caching: Edge clouds provide caching of cloud service media. +- Multimedia data handling: Data handling on edge clouds reduces the interaction between clouds, filters redundant information of media from multiple end devices and batched data delivery to other clouds, respectively. +- Multi-access edge computing (MEC): It offers cloud-computing capabilities for media processing and an IT service environment at the edge of the network and provides ultra-low latency and high bandwidth as well as real-time access to radio network information that can be leveraged by multimedia applications. + +DHMP leverages edge cloud computing, enabling real-time processing and rendering of high-fidelity digital human content. By utilizing edge cloud resources, it ensures low latency and high responsiveness, meeting the demanding requirements of interactive digital human services. Its versatility and adaptability make it a powerful tool for content creators, developers and service providers in delivering captivating digital human experiences. + +With the core-edge-user nodes, the multimedia platform will clarify media requirements and architectures. In clause 7, requirements for the multimedia processing, interaction and management function are closely related to the existing digital human application system, and some supplement requirements based on the concept of core-edge-user architecture, such as requirements for data sharing among the user nodes, edge cloud nodes and core cloud nodes. Clause 8 specifies a multimedia platform architecture based on the requirements and existing technologies of the edge cloud and digital human, which can serve as a reference for media service providers. + +# **7 Requirements of multimedia platform for digital human services using the edge cloud** + +### **7.1 Requirements for media processing** + +#### **7.1.1 Generation and rendering** + +The high level of detail in modelling, materials, textures and lighting simulation for realistic physical rendering of a digital human will consume substantial computing resources and incur significant processing delays. Devices, such as lightweight tablets, mobile phones or HMDs, are often used as entry points for interacting with the digital human. When digital human generation is running on these devices, it is challenging to realize the required rendering and display performance, failing to meet + +the high-fidelity demands of the digital human. Therefore, it is essential to define the media processing requirements of a multimedia platform using edge cloud and explore offloading computationally intensive tasks to appropriate core/edge cloud nodes. + +The multimedia platform can perform the following media processing requirements to enhance digital human immersion and media processing latency. + +**GR-01:** It is required to perform real-time generation on the edge cloud nodes of DHMP. + +**GR-02:** It is recommended to support the generation and export of 3D models in commonly-used formats such as .fbx and .obj on the DHMP. + +**GR-03:** It is required to support the visualization of generated models with configurable rendering parameters, including lighting, texturing and camera angles that enable an assessment of model quality across multiple display devices with different capabilities. + +**GR-04:** It is required to support digital human model editing and re-rendering according to client instructions of user(s). + +**GR-05:** It is required to support 3D dynamic representation (3D models, video, and images) with audio output. + +**GR-06:** It is required to support real-time rendering on the edge cloud nodes and non-real-time rendering on the core cloud nodes. + +**GR-07:** It is required to support the adjustment of model parameters to optimize performance based on network conditions, device capabilities and application requirements. + +NOTE – Examples of model parameters include polygon count, texture resolution, level of detail (LOD), bone structure complexity and rendering complexity. + +#### 7.1.2 Animation and display + +Digital human services load a 3D model and extract facial expression or motion signals from images/videos received from a client. The data also contains the camera configuration information. The DHMP animates the 3D model in real-time based on expression or motion signals and renders the 3D model from the user's view to display. + +The DHMP can perform the following media processing requirements to achieve the animation and display of digital humans. + +**AD-01:** It is required to extract facial expression signals from audio or video sources. + +**AD-02:** It is required to support extracting motion data from users, including head movements and body movements from a user equipment (UE) client, and binding this data to 3D models. + +**AD-03:** It is required to support digital human animation on the edge cloud nodes with priority given to real-time interaction requirements, which includes ensuring a smooth visual experience by meeting performance standards for facial animation and body movements. + +**AD-04:** It is required to support real-time rendering with fixed mesh points and texture, and display the 3D model with the operations of users. + +**AD-05:** It is required to support the non-real-time rendering of an animated digital human on the cloud nodes. + +#### 7.1.3 Data compression + +Since digital human data are a combination of multimedia, including audio, video, graphics and metadata, the DHMP should support the integration of heterogeneous types of digital human coding formats with the following requirements. + +**DC-01:** It is recommended to support deep-learning-based encoding and processing to compress and enhance the quality of digital human data. + +**DC-02:** It is recommended to use public digital human datasets to pretrain digital human coding models. + +NOTE 1 – Examples of public digital human datasets include [b-THuman-1], [b-THuman-2] and [b-AGORA]. + +**DC-03:** It is recommended to use customized digital human datasets in different service scenarios to pretrain and optimize digital human coding models. + +**DC-04:** It is recommended to use at least one of deep learning methods as the baseline of digital human data coding models. + +**DC-05:** It is recommended to use reinforcement learning methods to support the generation and iterative improvement of deep-learning-based encoding models. + +NOTE 2 – Examples of reinforcement learning methods include actor-critic architectures and deep Q-networks (DQNs), while federated learning represents a distributed learning approach that preserves data privacy. + +**DC-06:** It is required to support adaptive selection and adjustment of digital human data encoding models based on factors of terminal capabilities and digital human data categories. + +**DC-07:** It is required to support adaptive dynamic compression and decompression based on the network conditions and user requirements to improve digital human data processing efficiency including dynamic encoding model selections, frame classification and prediction, and view-area-based streaming. + +**DC-08:** It is recommended to support joint optimization with computation capabilities, including computation resource scheduling within and between nodes, and task offloading based on different functions and contents. + +**DC-09:** It is required to support the coding of the digital human in 2D or 3D formats. + +**DC-10:** It is required to support the synchronization and integration of the different multimodal data with the coding formats of the digital human. + +**DC-11:** It is recommended to support semantic and signal representations of the digital human for the coding format, using a deep-learning-based algorithm for object detection, image segmentation, neural rendering and 3D representation techniques. + +NOTE 3 – Examples of applicable techniques include object detection algorithms (e.g., YOLO), image segmentation methods (e.g., SLIC), neural radiance fields (NeRFs), and 3D Gaussian Splatting (3DGS) for efficient 3D representation. These examples are provided for illustration and do not imply endorsement of specific implementations. + +**DC-12:** It is required to support both lossy and lossless coding. + +**DC-13:** It is recommended to support the deep-learning-based restoration of digital human. + +#### **7.1.4 Customization and trust processing** + +Digital human services face content protection against unauthorized access, leakage, destruction and tampering. During the lifecycle of multimedia applications, different media processing technologies can be employed to ensure the credibility of digital human content. + +**CTP-01:** It is required to support digital human customization that retains real-person core identity and portrait likeness while allowing for stylized adjustment (such as accessory updates) to meet requirements of user preferences for design style. + +**CTP-02:** It is recommended to support voice customization capabilities, including user-specific voice modelling and voice variation according to user-defined preferences. + +**CTP-03:** It is required to protect digital content for the digital human. + +### 7.2 Requirements for multimedia interaction + +Digital human services involve capturing objects, extracting a set of features and transmitting them to calibrate models. While a simple model can be constructed from a single image, more realistic models need a large amount of data and processing for training. The platform should enable the real-time exchange of animated information of digital human between the acquisition side and the display side. Additionally, it needs to handle the synchronization of multimedia information to ensure harmony between voice and performance. + +For real-time multimedia interaction among multiple points, the platform should meet the media transmission and support different formats of digital human media. At the same time, the platform should ensure quality of service (QoS) and QoE. + +**MI-01:** It is required to support multimodal data including audio, video, image, text and action signal. + +**MI-02:** It is required to identify and recognize media data including audio, video, image, text and action signals during upload to enable the corresponding animation engine. + +**MI-03:** It is required to support the synchronization and consistent representation of multimodal data. + +**MI-04:** It is required to support 3D media streaming using appropriate protocols. + +**MI-05:** It is recommended to support multimodal data inference to assist digital human models in perception and decision making. + +**MI-06:** It is recommended to support delay sensitive transmission of the corresponding animation streams. + +**MI-07:** It is recommended to handle QoE-based transmission. + +### 7.3 Requirements for content management + +Digital human services need unified content management for the digital human image, including the ingestion, distribution and hosting of digital human media content. When receiving a client request, the core cloud offers the flexibility to configure the nodes where digital human models can be stored and processed. + +The DHMP should enable the nearby invocation, sharing and management of digital human content. + +**CM-01:** It is required to support data structuring for 3D videos and 3D models to ingest. + +**CM-02:** It is required to support third-party platform as digital human content sources for the ingestion of 3D models. + +**CM-03:** It is required to support the provision of content (3D model, audio, video, image, graphics, and metadata) to other platforms or applications with authorization. + +**CM-04:** It is required to support the storage of 3D models on the core cloud node for the reuse of models. + +**CM-05:** It is required to support the caching of 3D model on edge cloud nodes to ensure timely reuse within the same session. + +# 8 Architectures of multimedia platform for digital human services using the edge cloud + +### 8.1 General architecture + +According to the analysis of media related requirements in clause 7, the above requirements can be mapped into three functions in the functional architecture of DHMP, which are deployed in the core cloud and edge cloud nodes respectively. The three functions, shown in Figure 3 are content management, media processing and media interaction. + +![Figure 3 – General architecture of DHMP. The diagram shows the architecture of the Digital Human Multimedia Platform (DHMP) and its interactions. At the top is the 'Service control for digital human application (Third party platform)'. Below it is the 'Digital human multimedia platform (DHMP)' which contains 'Core cloud node' and 'Edge cloud nodes'. The 'Core cloud node' has three functions: 'Content management function' (Data structuring, Model storage, Content provision), 'Media processing function' (Non-real-time generation, Non-real-time rendering, Customization, Trust processing, Data compression), and 'Media interaction function' (Data encapsulation, Data streaming). The 'Edge cloud nodes' have three functions: 'Content management function' (Data structuring, Model cache), 'Media processing function' (Real-time generation, Real-time rendering, Animation, Data compression), and 'Media interaction function' (Data ingest, Data inference, Data encapsulation). Interfaces are labeled M-1, M-2, M-3, M-4, P-1, P-2, I-0, I-1, and I-2. At the bottom are 'Cloud resources (Defined in [ITU-T Y.3508])'. On the right are 'Users' with 'Input', 'Output', 'Encapsulation', and 'Decapsulation' components. A legend at the bottom defines the arrow types: Control stream (dashed), Media stream based on internal interface (solid), and Media stream based on external interface (dotted).](4801720824e4b5e2361a5564f91cfb70_img.jpg) + +F.748.58(25) + +Figure 3 – General architecture of DHMP. The diagram shows the architecture of the Digital Human Multimedia Platform (DHMP) and its interactions. At the top is the 'Service control for digital human application (Third party platform)'. Below it is the 'Digital human multimedia platform (DHMP)' which contains 'Core cloud node' and 'Edge cloud nodes'. The 'Core cloud node' has three functions: 'Content management function' (Data structuring, Model storage, Content provision), 'Media processing function' (Non-real-time generation, Non-real-time rendering, Customization, Trust processing, Data compression), and 'Media interaction function' (Data encapsulation, Data streaming). The 'Edge cloud nodes' have three functions: 'Content management function' (Data structuring, Model cache), 'Media processing function' (Real-time generation, Real-time rendering, Animation, Data compression), and 'Media interaction function' (Data ingest, Data inference, Data encapsulation). Interfaces are labeled M-1, M-2, M-3, M-4, P-1, P-2, I-0, I-1, and I-2. At the bottom are 'Cloud resources (Defined in [ITU-T Y.3508])'. On the right are 'Users' with 'Input', 'Output', 'Encapsulation', and 'Decapsulation' components. A legend at the bottom defines the arrow types: Control stream (dashed), Media stream based on internal interface (solid), and Media stream based on external interface (dotted). + +**Figure 3 – General architecture of DHMP** + +### 8.2 Content management function + +This module facilitates platform-wide content management by standardizing data structures and overseeing content ingestion, provision and storage across the distributed nodes. + +#### a) Core cloud nodes + +To fulfil requirements CM-01, CM-02, CM-03 and CM-04, the core nodes on DHMP implement the following functional entities: + +- i) **Data structuring:** Supports ingestion of digital human models via a unified facial–body composite data structure (CM-01, and CM-02). Except for ingesting the digital human models generated from DHMP (CM-01), it also includes the integration of existing 3D models from different third-party platforms into the DHMP to realize the content management and application (CM-02). +- ii) **Model storage:** Maintains and stores digital human models in the unified facial body composite data structure (CM-04). +- iii) **Content provision:** Enables user to select of digital human models while embedding watermark verification metadata. Verification includes check algorithms, random sequence numbers and checksum values (CM-03). + +#### b) Edge cloud nodes + +To address requirements CM-01 and CM-05, the edge cloud nodes provide: + +- i) **Data structuring:** Unifies facial–body composite data structure for differentiated animation (CM-01). + +- ii) **Models cache:** Implements a facial–body composite indexing structure for efficient model and processed result retrieval. + +### 8.3 Media processing function + +This module provides media processing capabilities on DHMP, with non-real-time media processing on the core cloud nodes and real-time media processing on the edge cloud nodes. + +#### a) Core cloud nodes + +Compliant with the requirements GR-01, GR-02, GR-03, GR-04, GR-05, GR-06 and GR-07, functional entities include non-real-time generation, non-real-time rendering, customization and trust processing. + +- i) **Non-real-time generation:** Converts raw 3D data from acquisition devices into preliminary 3D models through preprocessing. This initial processing step lays the foundation for subsequent model refinement and customization. +- ii) **Non-real-time rendering:** Edits and renders preliminary 3D models into final assets. This process allows for detailed adjustments and enhancements to create a high-quality digital human representation. +- iii) **Customization:** For requirements of CTP-01, CTP-02, offers digital human image and voice customization and digital content protection. During the generation and editing of digital human models, the watermark information is generated according to the copyright and operational details of the model and embedded within it. The watermark information includes timestamp, random serial number, copyright, author and operational information. +- iv) **Trust processing:** For requirements of CTP-03, embeds/extracts watermarks in geometry/texture via pre-trained systems. Watermark information can be extracted from this model in geometry/texture by watermark processing systems. The watermark consistency would identify whether the digital human has been edited and attacked. According to the watermark information, the corresponding verification can be extracted in subsequent processing. And comparing with the verification in metadata, it would identify whether the digital human has been edited without authorization. +- v) **Data compression:** For requirements DC-09 and DC-10, 2D and 3D data coding is offered, including image/video codecs (H.264, H.265, H.266, AVS3), 3D point cloud codec, 3D dynamic mesh codec and haptics codec, while containing deep-learning-based methods. To address the requirements DC-01, DC-03, DC-04, DC-07 and DC-11, for NeRF-based or 3DGS-based 3D data, the data can be processed through semantic analysis, representation generation, feature transform and quantization followed by video coding, or neural network-based encoders or entropy coding. For the requirement DC-11, 2D/3D semantic information extraction and analysis functions are deployed, including semantic feature extraction, 2D/3D semantic segmentation, 2D/3D object detection, ROI detection, saliency detection and semantics-based bitrate adaptive optimization. For the mentioned-above requirements DC-12, multiple coding modes are supported, enabling both the lossy coding and lossless coding. For requirements DC-06 and DC-07, network conditions could be fed back to the encoder, allowing the adjustments of coding parameters/modes based on the network conditions. For the requirements DC-03, DC-06 and DC-07, the terminal (user side) could provide feedback to the encoding end based on the requirements of downstream tasks, thereby better selecting matched encoding parameters and adjusting bitrate control strategies. For the requirements DC-06, DC-13, after transmitting the encoded data to the decoding side, based on the quality feedback from the terminal side, the quality enhancement module can (but need not) be activated for enhancing the quality of rendered viewpoints or 3D models. For 2D videos, this involves frame interpolation, video super-resolution and motion deblurring and denoising methods. For 3D point clouds/mesh, upsampling / completion / denoising / frame rate up-conversion can be applied. + +#### b) Edge cloud nodes + +To meet the above requirements AD-01, AD-02, AD-03 and AD-04, real-time generation, animation and real-time rendering should be implemented. + +- i) **Real-time generation:** lightweight 3D model processing and generation, enabling quick model creation suitable for real-time applications at the edge. +- ii) **Real-time rendering:** Converts preliminary 3D model into final renderings, ensuring smooth and immediate visual output for users. +- iii) **Animation:** Based on the definition in [ITU-T F.748.15], dynamically animates models based on UE-input facial/body signals. The animation modes are capable of extracting facial and/or body animation signals and then animating the corresponding parts of the digital human model. These modes can be selected and switched according to input instructions from the UE. The platform animates a digital human model in animation mode and renders the 3D model with a user view to display after animation and distributes related animated and rendered digital human through different paths or ports. +- iv) **Data compression:** The same data compression functions as core cloud nodes are provided, ensuring consistent data handling and optimization across the platform. + +## 8.4 Media interaction function + +This module handles the sharing of digital human models and media data among different core/edge cloud nodes, and conducts inference operations before multimodal animation. The processing signals include video, image, 3D data (point cloud, mesh, implicit neural visual representation), text, haptics signal and audio, collectively referred as multimodal data. + +#### a) Core cloud nodes + +To meet requirements MI-01, MI-03 and MI-04, it should implement data encapsulation and data streaming. These functions entities enable encapsulation of multimodal data and distribution of multimodal digital human media. + +- i) **Data encapsulation:** When dealing with point clouds, the storage of point cloud compression data along with its associated metadata can adopt the ISO Base Media File Format (ISOBMFF) as specified in [b-ISO/IEC 14496-12]. For storage of non-timed geometry-based point cloud compression data and the associated metadata, the high efficiency video coding image file (HEIF) format specified in [b-ISO/IEC 23008-12] can be utilized. +- ii) **Data streaming:** The streaming of compression data relies on media streaming systems. Dynamic adaptive streaming over hypertext transfer protocol (DASH) as specified in [b-ISO/IEC 23009-1] or MPEG media transport (MMT) as specified in [b-ISO/IEC 23008-1] can be employed. In the case of haptic media, MIHS format for streaming haptic media specified in [b-ISO/IEC 23090-31] is applicable. + +#### b) Edge cloud nodes + +To meet the above requirements MI-01, MI-02 and MI-05, functional entities including data ingestion, data inference, data encapsulation should be deployed. + +- i) **Data inference:** The function enables the collection of multimodal media and the recognition of animation media, thereby facilitating corresponding inference processes and powering the animation engine. +- ii) **Data encapsulation:** After real-time animation and rendering, this function encapsulates multimodal data to enable media sharing. The media interaction module encapsulates and streams structural data if it needs to be transmitted (MI-04) and outputs the overall model according to the selected animation mode. + +- iii) **Data ingestion:** The function enables the implementation of multiple interactivity tasks and frameworks. Based on QoE related parameters obtained from the platform, network and UE such as latency, bandwidth and packet loss, interactivity tasks and frameworks can be established. This includes interaction modes and their respective levels between UE and DHMP. Different interaction modes correspond to different QoE levels of users with higher-level interaction modes generally providing a better user experience. To fulfil requirements MI-06 and MI-07, DHMP categorizes interaction modes into different levels, assigns specific interaction modes and levels to different interaction frameworks, and determines user QoE values. UE records the actual value of QoE for each interaction framework. By comparing these two QoE values, this function selects an interaction framework with the smallest difference for interaction and transmission. The function can work for block-based delivery and presentation of digital human models. This involves segmenting a frame of the digital human media into multiple blocks after encoding the digital human model. The segment information, including the spatial and temporal position relation between multiple blocks, as well as the importance and animation information of each block, should be recorded. The UE reads the segmented information, and provides feedback on the number of segments it can handle based on its capabilities. Subsequently, the DHMP decides on the number of segments to send to the UE taking into account client requests and network condition. + +### 8.5 Reference points + +The reference points between different modules are shown in Figure 3. + +The external reference points are: + +**Reference point I-0:** An external interface used for communication of input/output and media interaction module in client on user nodes. + +**Reference point I-1:** An external interface used for communication between the media interaction module on user nodes and media interaction module on edge cloud (EC) nodes to ingest user media data or forward digital human media from EC. + +**Reference point I-2:** An external interface used for communication between media interaction module on core cloud (CC) nodes and media interaction module on EC nodes to forward digital human media. + +The internal reference points are: + +**Reference point P-1:** An internal interface used for communication between the media interaction module and media processing module on EC nodes to forward digital human models. + +**Reference point P-2:** An internal interface used for communication between the media interaction module and media processing module on CC nodes to forward digital human models. + +**Reference point M-1:** An internal interface used for communication between the content management module and media processing module on CC nodes to provide digital human models for media processing. + +**Reference point M-2:** An internal interface used for communication between the content management module and media processing module on EC nodes to provide digital human models for media processing. + +**Reference point M-3:** An internal interface used for communication between the content management module and media interaction module on CC nodes to distribute multimodal digital human media. + +**Reference point M-4:** Used for communication between the content management module and media interaction module on EC nodes to distribute multimodal digital human media. + +DHMP can support various types of digital human applications based on the deployment of media functions on CC and EC nodes. Third-party service control for digital human application can be integrated into DHMP, and the platform can utilize edge cloud computing and other infrastructure resources. + +# Appendix I + +## Use cases of DHMP using edge cloud + +(This appendix does not form an integral part of the Recommendation.) + +### I.1 Digital human call + +A digital human call involves two users. User B views an animated digital human avatar of User A. The movements of User A are employed to animate his or her avatar. This will expand into a new type of calling experience. By introducing edge cloud technology in the digital human call, the real-time performance, reliability, bandwidth efficiency and security of calls can be significantly improved. + +![Diagram of a digital human call architecture showing the flow of data between two users (UE A and UE B) through edge and core cloud nodes within the DHMP, connected to a third-party service platform.](898fb89a50d9ec1dfb4e425c816976a7_img.jpg) + +The diagram illustrates the architecture of a digital human call. At the top, a 'Third party service platform' is connected to a 'Core cloud' node. The 'Core cloud' node is part of the 'DHMP' (Digital Human Media Platform) and is connected to two 'Edge cloud' nodes. On the left, 'UE A' (User Equipment A) is connected to the first 'Edge cloud' node. On the right, 'UE B' (User Equipment B) is connected to the second 'Edge cloud' node. Data flows from 'UE A' to the first 'Edge cloud', then to the 'Core cloud', then to the second 'Edge cloud', and finally to 'UE B'. The 'DHMP' is represented by a dashed box containing the 'Core cloud' and the two 'Edge cloud' nodes. The reference 'F.748.58(25)' is shown near 'UE B'. + +Diagram of a digital human call architecture showing the flow of data between two users (UE A and UE B) through edge and core cloud nodes within the DHMP, connected to a third-party service platform. + +Figure I.1 – A case of digital human call + +**Core cloud:** These nodes of DHMP are responsible for content management and non-real-time processing, such as digital human generation, non-real-time rendering and trust processing. The core cloud nodes should have content management capability. + +The nodes can access a third-party service platform, i.e., a digital human communication application server (such as Agora and Twilio). + +**Edge cloud:** These nodes of the DHMP are responsible for real-time processing, including motion signals creation, animation, uplink multimodal analysis, media encoding/decoding and distribution, and user interaction. + +**UE:** This supports images (e.g., RGB or RGB-D streams) and audio capture or text input. The UE can render and display digital human content. + +### I.2 Digital human multiparty call + +In a multiparty call scenario, the digital human avatars and audio of each participant are transmitted and spatially rendered in the direction of their geolocation. Users wear glasses-type devices and know the direction of the other participants. The digital human call described in clause I.1 only involves two parties, while a digital human multiparty call involves more than two parties. + +![Diagram of a digital human multiparty call architecture. At the top, a 'Third party service platform' connects to a 'Core cloud' node. The 'Core cloud' node is connected to an 'Edge cloud' node. These two cloud nodes are part of a 'DHMP' (Digital Human Multiparty) system, indicated by a dashed box. Below the DHMP system, a 'UE' (User Equipment) icon shows four people, representing multiple participants in a call. The label 'F.748.58(25)' is at the bottom right.](b6671cfafda3820aafe9a24fa7a4d8c7_img.jpg) + +Diagram of a digital human multiparty call architecture. At the top, a 'Third party service platform' connects to a 'Core cloud' node. The 'Core cloud' node is connected to an 'Edge cloud' node. These two cloud nodes are part of a 'DHMP' (Digital Human Multiparty) system, indicated by a dashed box. Below the DHMP system, a 'UE' (User Equipment) icon shows four people, representing multiple participants in a call. The label 'F.748.58(25)' is at the bottom right. + +**Figure I.2 – A case of digital human multiparty call** + +**Core cloud:** These nodes of the DHMP are responsible for content management and non-real-time processing, such as digital human generation, non-real-time rendering and trust processing. The core cloud nodes should have content management capability. The nodes can access a third-party service platform, i.e., a platform of conference application with functions of multiparty negotiation and stream mixing. + +**Edge cloud:** These nodes of DHMP are responsible for real-time processing, including motion signal creation, animation, uplink multimodal analysis, media encode/decode and distribution. + +**UE:** This supports images (e.g., RGB or RGB-D streams) and audio capture or text input. Users can render and display digital human content. + +### I.3 Digital human live streaming + +Digital human live streaming combines motion capture and image processing. It synthesizes video information or pictures of a real person into a virtual character in real-time. In digital human live streaming, communication is typically one-to-many and only the presenter possesses a digital human representation, and the audience participates as viewers without digital avatars. In contrast, a digital human multiparty call (I.2) involves many-to-many interactions, with each participant represented by their own digital human. + +In another scenario, a digital human can understand the semantics of user questions. This is made possible through the integration of advanced AI capabilities, including natural language processing (NLP) and large language models (LLMs). A digital human judge user emotions and make different interactive responses through user behaviour records and can generate live content in real-time. + +![Diagram of a digital human live streaming architecture. At the top, a 'Digital human live video application server' connects to a 'Core cloud' node. The 'Core cloud' node is connected to an 'Edge cloud' node. These two cloud nodes are part of a 'DHMP' (Digital Human Multiparty) system, indicated by a dashed box. Below the DHMP system, a 'UE' (User Equipment) icon shows a person with a camera, representing a viewer. To the right of the DHMP system, there is an icon of a globe with people, representing the audience. The label 'F.748.58(25)' is at the bottom right.](1a85642ed2356d183ce598f2c8b3ee8b_img.jpg) + +Diagram of a digital human live streaming architecture. At the top, a 'Digital human live video application server' connects to a 'Core cloud' node. The 'Core cloud' node is connected to an 'Edge cloud' node. These two cloud nodes are part of a 'DHMP' (Digital Human Multiparty) system, indicated by a dashed box. Below the DHMP system, a 'UE' (User Equipment) icon shows a person with a camera, representing a viewer. To the right of the DHMP system, there is an icon of a globe with people, representing the audience. The label 'F.748.58(25)' is at the bottom right. + +**Figure I.3 – An example of digital human live streaming** + +**Core cloud:** These nodes of DHMP are responsible for content management and non-real-time processing, such as digital human access, non-real-time rendering and trust processing. The cloud + +node access a third-party service platform, i.e., a live streaming platform, to implement the function of recording and auditing. + +Edge cloud: These nodes of DHMP are responsible for real-time processing, including motion signals creation, animation, uplink multimodal analysis and media distribution. + +NOTE – For local live streaming services, edge nodes can play the role of cloud nodes. + +UE: This supports images (e.g., RGB or RGB-D streams) and audio capture or text input. Users can render and display digital human content. + +### I.4 Online digital human + +Online digital human scenarios include the online generation of digital human, animation, intelligent and other processes using cloud service, while sharing a rich material library and content library. Compared with local digital human scenarios, online operations can reduce the application threshold and use cost for users. By introducing edge cloud technology in the online digital human scenarios, the real-time performance, reliability, bandwidth efficiency and security of scenarios can be significantly improved. + +![Diagram of a case of online digital human generation and editing. A dashed box labeled 'DHMP' contains a cloud icon with 'Core cloud' and 'Edge cloud' inside. Below the box is a 'UE' (User Equipment) icon showing a person at a laptop. A double-headed arrow connects the UE to the DHMP box. The label 'F.748.58(25)' is at the bottom right.](4cc7cdce3d498d8b0ba033a9be24ade5_img.jpg) + +The diagram illustrates a system architecture for online digital human generation and editing. At the top, a dashed rectangular box labeled 'DHMP' (Digital Human Media Platform) contains a cloud icon. Inside the cloud, there are two stacked rectangular boxes labeled 'Core cloud' and 'Edge cloud'. Below the DHMP box, there is a 'UE' (User Equipment) icon depicting a person sitting at a laptop. A vertical double-headed arrow connects the UE to the DHMP box, indicating bidirectional data flow. The label 'F.748.58(25)' is positioned at the bottom right of the diagram. + +Diagram of a case of online digital human generation and editing. A dashed box labeled 'DHMP' contains a cloud icon with 'Core cloud' and 'Edge cloud' inside. Below the box is a 'UE' (User Equipment) icon showing a person at a laptop. A double-headed arrow connects the UE to the DHMP box. The label 'F.748.58(25)' is at the bottom right. + +**Figure I.4 – A case of online digital human generation and editing** + +Core/edge cloud: + +- Servers on cloud/edge nodes support the generation of 3D models in various formats and conduct content management and non-real-time processing. +- The cloud has customization and editing functions for 3D models. +- The cloud provides non-real-time animation and rendering, including image and audio consistency, animation. +- The core/edge cloud nodes can review, output and save the image of a digital human model and can perform real-time recording of digital human with animation. +- The cloud provides online replacement of generated digital human voices, clothing and backgrounds. + +UE: This supports images (e.g., RGB or RGB-D streams) and audio capture or text input and provides visual editing interface. + +### I.5 Digital human socializing + +Based on 3D scene reconstruction, 3D object reconstruction and digital human technology, a virtual digital space is created with a rendering engine. Users can have immersive experience in the virtual space through extended reality (XR) devices in the image of digital people and interact in manners such as socializing and shopping together. As shown in Figure I.5, users are not necessarily limited to the same DHMP; nevertheless, they are able to interact socially and exchange data across different platforms. + +![Diagram illustrating a case of digital human socializing. At the top, a 'Digital human social application server' is connected to two 'DHMP' (Digital Human Metaverse Platform) units. Each DHMP unit contains a 'Core cloud' and an 'Edge cloud'. Below the DHMP units, a 'UE' (User Equipment) icon is shown, representing a user. Arrows indicate data flow from the DHMP units to the UE. The UE is surrounded by icons representing digital humans. The text 'F.748.58(25)' is visible in the bottom right corner of the diagram.](e69b9188aa2c14ec6b21c83f711fef65_img.jpg) + +Diagram illustrating a case of digital human socializing. At the top, a 'Digital human social application server' is connected to two 'DHMP' (Digital Human Metaverse Platform) units. Each DHMP unit contains a 'Core cloud' and an 'Edge cloud'. Below the DHMP units, a 'UE' (User Equipment) icon is shown, representing a user. Arrows indicate data flow from the DHMP units to the UE. The UE is surrounded by icons representing digital humans. The text 'F.748.58(25)' is visible in the bottom right corner of the diagram. + +**Figure I.5 – A case of digital human socializing** + +**Core cloud:** These nodes of DHMP are responsible for content management and non-real-time processing, such as digital human generation, non-real-time rendering and trust processing. The cloud offers the following: + +- It provides 3D model loading functions. +- It can simultaneously load different knowledge bases, such as customer service, assistants and teachers, through a third-party service platform. +- It provides a scene management function to ensure model consistency rationality in the same scene. +- The core cloud nodes should have digital human content management capabilities, and access to third-party service platform, i.e., a social application server, to implement the function of user management and scenario management. + +**Edge cloud:** These nodes of the DHMP are responsible for real-time processing, including motion signals creation, animation, uplink multimodal analysis and media distribution. + +**UE:** This supports images (e.g., RGB or RGB-D streams) and audio capture or text input. Users can render and display digital human social scenes. + +# Appendix II + +## Key issues of DHMP using edge cloud + +(This appendix does not form an integral part of the Recommendation.) + +### II.1 Key issue No. 1: NeRF-based data compression + +It is commonly known that the NeRF [b-Mildenhall] becomes an effective visual representation for 3D digital human and scenes. In view synthesis, NeRF models with explicit voxel grid representations exhibit good rendering performance in view synthesis. However, voxel grids require a large amount of data while constructing NeRF models. Therefore, the potential solution (DC-01, DC-04, DC-06, DC-07, DC-10, DC-12, DC-14) for NeRF-based data compression could be as shown in Figure II.1. + +![Flow chart of the NeRF-based data compression process.](d9c0a780cd22626253dab4aa41699e2f_img.jpg) + +The diagram illustrates the NeRF-based data compression pipeline. It is divided into two main sections: an Encoder and a Decoder, connected by a Network cloud. The Encoder takes an 'Input' and processes it through 'Semantic analysis', 'Representation generation', 'Video coding', 'Representation coding', and 'Metadata coding'. The output of the Encoder is sent to the Network. The Network provides feedback (Signal A) to the Encoder and receives data from the Encoder. The Decoder receives data from the Network and processes it through 'Video decoding', 'Representation decoding', and 'Metadata decoding'. The output of the Decoder is sent to a 'Terminal', which is connected to 'Quality enhancement'. The Terminal also provides feedback (Signal B) to the Encoder. The overall flow is: Input -> Encoder -> Network -> Decoder -> Terminal -> Quality enhancement. Signal A (Network to Encoder) and Signal B (Terminal to Encoder) are shown as feedback loops. + +Flow chart of the NeRF-based data compression process. + +Figure II.1 – The flow chart of the NeRF-based data compression + +At the encoding end, the input signal of the system may include a 3D digital human and corresponding 2D backgrounds/materials (note: backgrounds/materials are not mandatory and depend on the input source). Therefore, the input signal first undergoes semantic analysis to divide the digital human from the background/materials. Subsequently, the digital human signal enters the "representation generation" module to obtain a compact representation of the NeRF-based digital human; the basic compact representation then is further compressed by a representation encoder (e.g., video coding, deep-learning-based methods, feature coding). Meanwhile, the parameters of the NeRF model and other parameters (such as camera parameters, motion parameters) serve as metadata and enter a metadata encoder (lossless coding, e.g., entropy coding, dictionary coding) for encoding. Simultaneously, the background/materials are compressed by a video encoder. + +Afterward, the encoded data are encapsulated and distributed to the decoding end via network transmission. Network conditions (Signal A) are fed back to the encoder, allowing coding parameter adjustments based on the network conditions (this function can be enabled or disabled) (DC-06). + +At the decoding end, the received bitstream will be subject to initial parsing; the 2D background/material stream is decoded by a video decoder, while the representation stream is decoded by a representation decoder, and the metadata data stream is decoded by a metadata decoder. + +Subsequently, the data decoded from the three streams are input into the rendering module to synthesize the 3D view, thereby presenting the 3D digital human on the terminal. + +The terminal side could provide feedback to the encoding end based on the requirements of downstream tasks (signal B), thereby better selecting encoding parameters and adjusting bitrate control strategies, etc. (DC-06). Additionally, besides the aforementioned process, based on the quality feedback from the terminal side (Signal C), the decision to enable or disable the quality enhancement module is made, enhancing the quality of rendered viewpoints or 3D models (this function can be enabled or disabled) (DC-06, DC-14). This function could be run on the edge node. For example, for video, it may increase the frame rate, use the video super-resolution, motion deblurring, denoising methods and so on. For the point cloud, it may increase the frame rate, use the point cloud upsampling, point cloud completion, point cloud denoising, etc. + +NOTE – Regarding representation generation, there are many effective solutions that can be included in this proposed encoding and decoding structure. For example, there are some works on the compact representation generation of NeRF models that decompose the voxel grid representation into vectors or planes for accelerating the model inference and training. Based on this, three potential solutions for the representation encoder can be considered as an example: + +- Solution example No. 1: 3D grid representation can be decomposed into a series of tensor planes, vectors, MLP and various parameters. Among them, tensor planes occupy 99.2% of the size of the training model. Therefore, tensor planes can be converted into feature maps compatible with video encoding formats, and then compressed using video encoders. Specifically, feature packing is conducted, which converts tensor planes composed of multiple channels into a single feature map. Then, each value of the single feature map is quantized, converting it from a floating-point format to a 10-bit integer format compatible with the input format of the video encoder. Subsequently, the quantized sequence of feature maps is compressed using a video encoder, such as H.264, H.265, H.266, AVS3, AV1/2 or VP9. +- Solution example No. 2: 3D grid representation decomposition is similar to Example No. 1. Tensor planes would be compressed by a hybrid structure encoder. It consists of frequency domain transform (such as discrete cosine transform, wavelet transform), model pruning (transformation coefficient pruning), quantization and entropy coding. +- Solution example No. 3: 3D grid representation decomposition is similar to Example No. 1. Tensor planes would be compressed by a tuned neural network-based encoder, containing the implicit feature extraction, probability estimation (as auxiliary information for the entropy model), quantization and entropy coding. + +### II.2 Key issue No. 2: Differentiated function provision for multiple access + +In a digital human application with multiple access, the edge cloud based DHMP needs to support selecting and switching animation mode for expression and motion (AD-01, AD-02). These two kinds of animation modes are implemented in different ways technically, and the animation data that need to be extracted is also different. The platform can set two animation modes of expression and motion, such as 1:1 or 1:N, and distribute related animated and rendered digital humans according to different paths/ports (AD-04, AD-05). This will be conducive to the optimization of computing resource utilization and network transmission efficiency. + +Content management module unifies data structure for this differentiated animation (CM-01), and models basically form a face and body combination structure. It manages the content ingestion through M-1 and M-2, and sharing through M-0, M-3 and M-4. + +The media processing module on the edge node switches animation and rendering engine based on 1:1 or 1:N mode selection. For instance, 1:N mode will update the media interaction module settings (MI-02), and the motion information for motion animation can be generated and extracted through data inference from input I-1 (MI-01 and MI-05). Motion animation and rendering engine in the media + +processing module will be started through P-1 and synchronization and consistent representation will be handled in the media interaction module (MI-03). + +The media interaction module encapsulates and streams this structural data if they need to be transmitted (MI-04). Clients in the end device will display multimodal output in a consistent way. + +The differentiated function of this issue is to store and animate the digital human model according to the structure of the face and body, and one part of this structure contains the state before and the other the state after animation. The overall model is rendered and output according to the mode selection, which is more in tune with the experience of multiperson access applications. + +# Appendix III + +## Correspondence between requirements and architectures + +(This appendix does not form an integral part of the Recommendation.) + +### III.1 Correspondence between requirements and architectures + +The architectures of DHMP in clause 8 need to meet the requirements of clause 7. Table III.1 shows the correspondence between the architectures and the requirements. + +**Table III.1 – Correspondence between requirements and architectures** + +| Architectures | | Requirements | +|---------------------------------|---------------------|-----------------------------------------------------------------------------------------------------------------------------------------------| +| 8.2 Content management function | a) Core cloud nodes | CM-01, CM-02, CM-03, CM-04 | +| | b) Edge cloud nodes | CM-01, CM-05 | +| 8.3 Media processing function | a) Core cloud nodes | GR-01, GR-02, GR-03, GR-04, GR-05, GR-06, GR-07, CTP-01, CTP-02, CTP-03, DC-01, DC-03, DC-04, DC-06, DC-07, DC-08, DC-10, DC-11, DC-12, DC-13 | +| | b) Edge cloud nodes | AD-01, AD-02, AD-03, AD-04 | +| 8.4 Media interaction function | a) Core cloud nodes | MI-01, MI-03, MI-04 | +| | b) Edge cloud nodes | MI-01, MI-02, MI-04, MI-05, MI-06, MI-07 | + +# Bibliography + +- [b-ITU-T G.1081] Recommendation ITU-T G.1081 (2008), *Performance monitoring points for IPTV*. +- [b-ITU-T Y.3508] Recommendation ITU-T Y.3508 (2019), *Cloud computing - Overview and high-level requirements of distributed cloud*. +- [b-AGORA] AGORA – Avatars in Geography Optimized for Regression Analysis. Digital Human Dataset. +<> +- [b-ISO/IEC 14496-12] ISO/IEC 14496-12 (2022), *Information technology – Coding of audio-visual objects – Part 12: ISO base media file format*. +- [b-ISO/IEC 23008-1] ISO/IEC 23008-1:2023, *Information technology – High efficiency coding and media delivery in heterogeneous environments – Part 1: MPEG media transport (MMT)*. +- [b-ISO/IEC 23008-12] ISO/IEC 23008-12:2022, *Information technology – High efficiency coding and media delivery in heterogeneous environments – Part 12: Image file format for high efficiency image coding (HEIF)*. +- [b-ISO/IEC 23009-1] ISO/IEC 23009-1 (2022), *Information technology – Dynamic adaptive streaming over HTTP (DASH) – Part 1: Media presentation description and segment formats*. +- [b-ISO/IEC 23090-31] ISO/IEC 23090-31 (2025), *Information technology – Coded representation of immersive media – Part 31: Haptics coding*. +- [b-Mildenhall] Mildenhall, B., Srinivasan, P.P., Tancik, M., Barron, J.T., Ramamoorthi, R., & Ng, R. (2021). *Nerf: Representing scenes as neural radiance fields for view synthesis*, Communications of the ACM, Vol. 65, No. 1, pp. 99–106. +<> +- [b-THuman-1] THumanDataset 2.0. +<> +- [b-THuman-2] THumanDataset 4.0. +<> + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.59-202512-I_PDF-E/042733dc5e8e7f5f30b60adba3266cde_img.jpg b/marked/F/T-REC-F.748.59-202512-I_PDF-E/042733dc5e8e7f5f30b60adba3266cde_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ff8519d0398e07aeed8a0282a7f9c639bf27f6b3 --- /dev/null +++ b/marked/F/T-REC-F.748.59-202512-I_PDF-E/042733dc5e8e7f5f30b60adba3266cde_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b6c11c07ac401dab85ffe105387c7af326eecd93c0cb6e1aede002971ab8e544 +size 16458 diff --git a/marked/F/T-REC-F.748.59-202512-I_PDF-E/07b17a620c75522d53916a11e12d1bff_img.jpg b/marked/F/T-REC-F.748.59-202512-I_PDF-E/07b17a620c75522d53916a11e12d1bff_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..28de8dc1eff028d05b71e8e2948f70fdbd7ef1ae --- /dev/null +++ b/marked/F/T-REC-F.748.59-202512-I_PDF-E/07b17a620c75522d53916a11e12d1bff_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:95c04b161f1023a92c1ff2509e3fa0beef213120f8a3e56d94ef5f2eef0c6d0c +size 3375 diff --git a/marked/F/T-REC-F.748.59-202512-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg b/marked/F/T-REC-F.748.59-202512-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ac295b1b0c7f14b1c5a33ef6682b1b017c812477 --- /dev/null +++ b/marked/F/T-REC-F.748.59-202512-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37aed4fc114a39843438175326e0b7e1d09a9d6075723d82fea32040fd560ee4 +size 7192 diff --git a/marked/F/T-REC-F.748.59-202512-I_PDF-E/731f533b0599c8e42a063f06e4332045_img.jpg b/marked/F/T-REC-F.748.59-202512-I_PDF-E/731f533b0599c8e42a063f06e4332045_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..fde446e19fbc5b4329ecff0933e93c774deb54ed --- /dev/null +++ b/marked/F/T-REC-F.748.59-202512-I_PDF-E/731f533b0599c8e42a063f06e4332045_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:62fd167ecca62eaae95a39048774d31cd219f359ebf132acff324079f4fd1ccb +size 5306 diff --git a/marked/F/T-REC-F.748.59-202512-I_PDF-E/7e670a2b556b53ea9002dfff3a420e08_img.jpg b/marked/F/T-REC-F.748.59-202512-I_PDF-E/7e670a2b556b53ea9002dfff3a420e08_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ea07ac66b97c40070ed2dd0d6dca2952fd10fec9 --- /dev/null +++ b/marked/F/T-REC-F.748.59-202512-I_PDF-E/7e670a2b556b53ea9002dfff3a420e08_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cee10194eb905c0cf7403bcc500e4ad648dc27eaa706fa06a7e7e11387fa8738 +size 91834 diff --git a/marked/F/T-REC-F.748.59-202512-I_PDF-E/853f59c89931a666c07903b31d098277_img.jpg b/marked/F/T-REC-F.748.59-202512-I_PDF-E/853f59c89931a666c07903b31d098277_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..5cbce6919ff29ab72c246ffd4cd054e0ff8a38aa --- /dev/null +++ b/marked/F/T-REC-F.748.59-202512-I_PDF-E/853f59c89931a666c07903b31d098277_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a87429f5c490dc849019bcb5d12f7f274433e8e975ebb915005db9be1f04724d +size 53943 diff --git a/marked/F/T-REC-F.748.59-202512-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg b/marked/F/T-REC-F.748.59-202512-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ac93f2b87f1916461402064c83ca80dfe781521d --- /dev/null +++ b/marked/F/T-REC-F.748.59-202512-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:50348b20620a719d6a8ab0a33e484fa8086937073071e1b5c4e7dbdbe7c47b59 +size 185717 diff --git a/marked/F/T-REC-F.748.59-202512-I_PDF-E/b9ecbc3baefab13719e000faa6e0c7eb_img.jpg b/marked/F/T-REC-F.748.59-202512-I_PDF-E/b9ecbc3baefab13719e000faa6e0c7eb_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..667753428ee67c75de8a152a4ce4232a49d51820 --- /dev/null +++ b/marked/F/T-REC-F.748.59-202512-I_PDF-E/b9ecbc3baefab13719e000faa6e0c7eb_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:56ff90b2996ef26a7c3b0a73a2b700adacc9a8c47a7e767993f2ec9a076be9fd +size 80995 diff --git a/marked/F/T-REC-F.748.59-202512-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg b/marked/F/T-REC-F.748.59-202512-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..727dd3d7b5e22bb26a042bae0a0a5e42e2fab79f --- /dev/null +++ b/marked/F/T-REC-F.748.59-202512-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d6a496c50343aabbd077ff55f19da62a14d9a1da19bd50045c822bf6c863e570 +size 67226 diff --git a/marked/F/T-REC-F.748.59-202512-I_PDF-E/ff0952ef692c9d960ce5f6708bcc9711_img.jpg b/marked/F/T-REC-F.748.59-202512-I_PDF-E/ff0952ef692c9d960ce5f6708bcc9711_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..11d926a7d5bb683d295ee00a36ef379f75d8083f --- /dev/null +++ b/marked/F/T-REC-F.748.59-202512-I_PDF-E/ff0952ef692c9d960ce5f6708bcc9711_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:77444378aaba4305c0cb51ec4eaa1161acf4b433daf606c2193dc0be6e2f3e5d +size 70518 diff --git a/marked/F/T-REC-F.748.6-202503-I_PDF-E/20136850feb70fd71c7d41cdae203ebb_img.jpg b/marked/F/T-REC-F.748.6-202503-I_PDF-E/20136850feb70fd71c7d41cdae203ebb_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..49e91460932b2cead02ee18509cb3d0536e03bf6 --- /dev/null +++ b/marked/F/T-REC-F.748.6-202503-I_PDF-E/20136850feb70fd71c7d41cdae203ebb_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d65d59355e9e28a8e279a9975a771c16c46e2871e08ed1b1d8e868670308a3c1 +size 54764 diff --git a/marked/F/T-REC-F.748.6-202503-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.748.6-202503-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..6a574bf4c7fd169a72a97de0395e1970ee6f7092 --- /dev/null +++ b/marked/F/T-REC-F.748.6-202503-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5d863c71623a2b5a933b672e32a1463a7461603fd81d81e68543e5d90adf3b80 +size 7132 diff --git a/marked/F/T-REC-F.748.6-202503-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg b/marked/F/T-REC-F.748.6-202503-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..2a19915719b92c7c8a56d8be04e816324e0e193f --- /dev/null +++ b/marked/F/T-REC-F.748.6-202503-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8e8ac43d4705661d6fd74b152932d22b1389294a14f238a433259969468130ac +size 154303 diff --git a/marked/F/T-REC-F.748.6-202503-I_PDF-E/75e4b78ee25f885d73120e3066a5253e_img.jpg b/marked/F/T-REC-F.748.6-202503-I_PDF-E/75e4b78ee25f885d73120e3066a5253e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..80b6a42a4ce62540c1ef851238f1af01ecd4f1ef --- /dev/null +++ b/marked/F/T-REC-F.748.6-202503-I_PDF-E/75e4b78ee25f885d73120e3066a5253e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d23e4ad8baded6da42feb60da0e356ddf5af864c5c96d62f3a951b3f7d3f3efc +size 78983 diff --git a/marked/F/T-REC-F.748.6-202503-I_PDF-E/b6671cfafda3820aafe9a24fa7a4d8c7_img.jpg b/marked/F/T-REC-F.748.6-202503-I_PDF-E/b6671cfafda3820aafe9a24fa7a4d8c7_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9591044266092cf37b809e1fe000a641ae014fcf --- /dev/null +++ b/marked/F/T-REC-F.748.6-202503-I_PDF-E/b6671cfafda3820aafe9a24fa7a4d8c7_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e15b6468290eb5dfb85f1d917d622110c9109b1e76af62c9cc1220370359e736 +size 175475 diff --git a/marked/F/T-REC-F.748.6-202503-I_PDF-E/c5452f95f3b28f1bfe29e84fbc2e1267_img.jpg b/marked/F/T-REC-F.748.6-202503-I_PDF-E/c5452f95f3b28f1bfe29e84fbc2e1267_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1af97995ccceb7e234c28976c246efee9938187d --- /dev/null +++ b/marked/F/T-REC-F.748.6-202503-I_PDF-E/c5452f95f3b28f1bfe29e84fbc2e1267_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:171145396de81e765db5c7e697c1906b910613871ee585ea1f1a531ab27becf2 +size 142320 diff --git a/marked/F/T-REC-F.748.6-202503-I_PDF-E/c85ded401105f62f2d6ff26b3b5eb4af_img.jpg b/marked/F/T-REC-F.748.6-202503-I_PDF-E/c85ded401105f62f2d6ff26b3b5eb4af_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9a3dc24819250d34d6508509a521a884828449c3 --- /dev/null +++ b/marked/F/T-REC-F.748.6-202503-I_PDF-E/c85ded401105f62f2d6ff26b3b5eb4af_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f93837d8d0c12c0d369ee743a6d12d4522daf22916aa81bebe3656918497d7ff +size 146289 diff --git a/marked/F/T-REC-F.748.6-202503-I_PDF-E/c914f51f4427bc672dd0526cfc90ebe9_img.jpg b/marked/F/T-REC-F.748.6-202503-I_PDF-E/c914f51f4427bc672dd0526cfc90ebe9_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c5233935beca31d3d64323629284d06050ae86b2 --- /dev/null +++ b/marked/F/T-REC-F.748.6-202503-I_PDF-E/c914f51f4427bc672dd0526cfc90ebe9_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7a5e2c0a30e64a2c8434633f7d474b10e6a8af2a192ad0a9189ee807e3ad827e +size 109950 diff --git a/marked/F/T-REC-F.748.6-202503-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg b/marked/F/T-REC-F.748.6-202503-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d791e4efdf5308b269a8392e42ba6577468c2009 --- /dev/null +++ b/marked/F/T-REC-F.748.6-202503-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2766587f0cf1ecb58beaf7c6b317487b8d4704161707087eda0f105ad7719698 +size 48114 diff --git a/marked/F/T-REC-F.748.60-202512-I_PDF-E/raw.md b/marked/F/T-REC-F.748.60-202512-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..48db5123f0b6518b0bfad96dfa3495728a56aaa1 --- /dev/null +++ b/marked/F/T-REC-F.748.60-202512-I_PDF-E/raw.md @@ -0,0 +1,422 @@ + + +# Recommendation + +## **ITU-T F.748.60 (12/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +**General technical requirements and framework +for artificial intelligence cloud platform – High +availability requirements** + +![ITU logo](84a1d09fb489061482111515543b60dc_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.60 + +# General technical requirements and framework for artificial intelligence cloud platform – High availability requirements + +## Summary + +With the rapid development of its foundation model, the demand for high availability in the artificial intelligence (AI) cloud platform has significantly increased. The AI cloud platform not only delivers stable service capabilities but also enables real-time monitoring of resource utilization and operational status across all services, and promptly detects runtime failures while providing alerting and intelligent maintenance functions. + +Recommendation ITU-T F.748.60 provides general technical requirements for a high-availability AI cloud platform for foundation model development, including monitoring capabilities, alerting capabilities and operation capabilities. This Recommendation is intended to improve the application development support capabilities of the AI cloud platform to better assist implementers in developing the foundation model. + +## History\* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.60 | 2025-12-14 | 21 | 11.1002/1000/16585 | + +## Keywords + +Artificial intelligence, foundation model, high availability. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2026 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|---------------------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 1 | +| 5 Conventions ..... | 2 | +| 6 Framework high availability requirements of AI cloud platform..... | 2 | +| 6.1 Monitoring capabilities..... | 3 | +| 6.2 Alarm capabilities..... | 3 | +| 6.3 Operation capabilities ..... | 3 | +| 7 Functional requirements for monitoring, alarm and operation capabilities..... | 4 | +| 7.1 Monitoring capabilities..... | 4 | +| 7.2 Alarm capabilities..... | 5 | +| 7.3 Operation capabilities ..... | 6 | +| Bibliography..... | 9 | + + + +# Recommendation ITU-T F.748.60 + +## General technical requirements and framework for artificial intelligence cloud platform – High availability requirements + +## 1 Scope + +This Recommendation specifies the technical requirements for an artificial intelligence (AI) cloud platform. These are mainly focused on improving the high availability of the platform. The Recommendation addresses the following aspects: + +- Framework of high availability in AI cloud platform. +- Functional requirements for monitoring, alarming and operation capabilities. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[[ITU-T F.748.38](#)] Recommendation ITU-T F.748.38 (2024), *Technical specification for artificial intelligence cloud platform: General architecture*. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 checkpoint** [b-ITU-T T.62]: A checkpoint is a numbered mark inserted by the sender in the text stream to provide a reference point for error recovery. + +**3.1.2 node failure** [b-ITU-T X.1643]: An error where one or multiple network nodes could not be accessed in an orchestration system. + +### 3.2 Terms defined in this Recommendation + +None. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-----|-----------------------------------| +| AI | Artificial Intelligence | +| API | Application Programming Interface | +| CPU | Central Processing Unit | +| ECN | Explicit Congestion Notification | +| FCT | Flow Completion Time | +| FET | Flow Establishment Time | + +| | | +|------|------------------------------------| +| GPU | Graphics Processing Unit | +| I/O | Input/Output | +| IOPS | Input/Output Operations Per Second | +| NLP | Natural Language Processing | +| PFC | Priority Flow Control | +| RDMA | Remote Direct Memory Access | +| RoCE | RDMA over Converged Ethernet | +| SDK | Software Development Kit | + +## 5 Conventions + +In this Recommendation: + +- The keywords "**is required to**" indicate a requirement that must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "**is recommended to**" indicate a requirement that is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. + +# 6 Framework high availability requirements of AI cloud platform + +This Recommendation focuses on the monitoring, alarm and operation capabilities of AI cloud platforms. The framework AI cloud platform: General architecture [ITU-T F.748.38] includes the high availability part, which is composed of monitoring capabilities, alarm capabilities and operation capabilities, as shown in Figure 1. The framework established in this Recommendation includes the 14 functional blocks shown in the Figure 2. + +![Diagram of the framework of high availability for the AI cloud platform. It shows the 'AI cloud platform: General architecture (ITU-T F.748.38)' with components: Platform security, Performance, Model development, Model deployment, and Resource management. The 'High availability' component is highlighted with a red box and an arrow pointing to a detailed view. This view shows 'High availability of foundation model development in AI cloud platform' with three stacked boxes: Monitoring capabilities, Alarm capabilities, and Operation capabilities. Above the architecture diagram is a box for 'AI-based multimedia applications' listing various tasks like question answering, logical reasoning, etc. An upward arrow points from the architecture to the applications box.](5860ad6bd2a2dd8d1ab12864b8f90f37_img.jpg) + +The diagram illustrates the framework of high availability for the AI cloud platform. At the top, a box labeled "AI-based multimedia applications" lists various tasks: "Question answering, logical reasoning, mathematical ability, code generation, machine translation, text generation, sentiment analysis, multimodal understanding and generation capabilities ...". Below this, an upward arrow points to the "AI cloud platform: General architecture (ITU-T F.748.38)". This architecture is shown as a vertical stack of five boxes: "Platform security", "Performance", "Model development", "Model deployment", and "Resource management". To the right of this stack is a vertical box labeled "High availability", which is highlighted with a red border. An arrow points from this "High availability" box to a larger box on the right titled "High availability of foundation model development in AI cloud platform". This box contains three stacked boxes: "Monitoring capabilities", "Alarm capabilities", and "Operation capabilities", connected by downward arrows. The label "F.748.60(25)" is at the bottom right of the diagram. + +Diagram of the framework of high availability for the AI cloud platform. It shows the 'AI cloud platform: General architecture (ITU-T F.748.38)' with components: Platform security, Performance, Model development, Model deployment, and Resource management. The 'High availability' component is highlighted with a red box and an arrow pointing to a detailed view. This view shows 'High availability of foundation model development in AI cloud platform' with three stacked boxes: Monitoring capabilities, Alarm capabilities, and Operation capabilities. Above the architecture diagram is a box for 'AI-based multimedia applications' listing various tasks like question answering, logical reasoning, etc. An upward arrow points from the architecture to the applications box. + +**Figure1 – Framework of high availability for the AI cloud platform** + +![Figure 2 – Framework of high availability. A flow diagram showing three main columns: Monitoring capabilities, Alarm capabilities, and Operation capabilities. Monitoring capabilities are divided into Foundation model business (Running status, User experience, Resource occupancy) and AI cloud platform resource (AI compute (GPU/ASIC/FPGA), AI storage (SSD/HDD), AI network (IB/RoCE)). Alarm capabilities include Monitoring data mining, Intelligent alarm analysis, and Intelligent alarm threshold setting. Operation capabilities include Automatic scaling, Automatic deployment, Automatic schedule, Intelligent fault diagnosis, and Auto recovery. Arrows indicate a sequential flow from left to right.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +``` + +graph LR + subgraph Monitoring_capabilities [Monitoring capabilities] + subgraph FMB [Foundation model business] + RS[Running status] + UE[User experience] + RO[Resource occupancy] + end + subgraph ACP [AI cloud platform resource] + AC["AI compute (GPU/ASIC/FPGA)"] + AS["AI storage (SSD/HDD)"] + AN["AI network (IB/RoCE)"] + end + end + subgraph Alarm_capabilities [Alarm capabilities] + MD[Monitoring data mining] + IA[Intelligent alarm analysis] + IT[Intelligent alarm threshold setting] + end + subgraph Operation_capabilities [Operation capabilities] + AS1[Automatic scaling] + AD[Automatic deployment] + AS2[Automatic schedule] + IF[Intelligent fault diagnosis] + AR[Auto recovery] + end + Monitoring_capabilities --> Alarm_capabilities + Alarm_capabilities --> Operation_capabilities + +``` + +F.748.60(25) + +Figure 2 – Framework of high availability. A flow diagram showing three main columns: Monitoring capabilities, Alarm capabilities, and Operation capabilities. Monitoring capabilities are divided into Foundation model business (Running status, User experience, Resource occupancy) and AI cloud platform resource (AI compute (GPU/ASIC/FPGA), AI storage (SSD/HDD), AI network (IB/RoCE)). Alarm capabilities include Monitoring data mining, Intelligent alarm analysis, and Intelligent alarm threshold setting. Operation capabilities include Automatic scaling, Automatic deployment, Automatic schedule, Intelligent fault diagnosis, and Auto recovery. Arrows indicate a sequential flow from left to right. + +**Figure 2 – Framework of high availability** + +## 6.1 Monitoring capabilities + +The monitoring capabilities part includes foundation model business monitoring and AI cloud platform resource monitoring. Foundational model business monitoring encompasses running status, user experience and resource occupancy. Running status includes the operation records of model running status and running status fault detection. The user experience section specifies the monitoring of business response time and user experience. AI cloud platform resource monitoring includes AI compute, AI storage and AI network monitoring. AI compute includes health status of computing nodes and the length and processing speed of the task queue monitoring. AI storage includes AI training dataset loading and storage performance monitoring, and storage capacity management. AI network connectivity includes lossless indicators of AI networks, the path and latency of GPU interconnection and traffic collection and analysis. + +## 6.2 Alarm capabilities + +The alarm capabilities part includes monitoring data mining, intelligent alarm analysis and intelligent alarm threshold setting. This clause specifies the AI cloud platform's alarm and monitoring requirements, mandating intelligent tools for monitoring data mining, including anomaly detection, historical baseline modelling and data visualization; alarm analysis, covering event classification, root-cause diagnosis with actionable suggestions, priority sorting and recommended alert convergence to reduce redundancy; and threshold management, featuring dynamic auto-adjustment, customizable settings, periodic optimization, and application programming interface (API)/software development kit (SDK)-based alert aggregation/distribution. + +## 6.3 Operation capabilities + +The operation capabilities part includes automatic scaling, automatic deployment, automatic schedule, intelligent fault diagnosis and auto recovery. This clause outlines the operational capabilities required for an AI cloud platform, covering five key functions. The first is automatic scaling, including dynamic resource adjustment based on demand, template management and elastic scaling strategies, with recommended features like historical data analysis and failure recovery. The second is automatic deployment, supporting end-to-end lifecycle management of applications and AI models, with workflow automation and orchestration tools. The third is automatic scheduling, enabling intelligent resource allocation across pools, load balancing and cross-regional management, with optional user quota controls. The fourth is intelligent fault diagnosis, leveraging AI for fault detection, root-cause analysis and resolution suggestions, plus recommended fault databases and natural language processing (NLP)-based diagnostics. The fifth is auto-recovery, ensuring automated + +restoration of faults, data and services with backup mechanisms. These capabilities collectively enable resilient, self-optimizing AI cloud operations. + +## **7 Functional requirements for monitoring, alarm and operation capabilities** + +## **7.1 Monitoring capabilities** + +### **7.1.1 Foundation model business** + +##### **7.1.1.1 Running status** + +- a) The AI cloud platform is required to support monitoring and recording of the running status of tasks. + - i) It is required to monitor the running status of the base model, e.g., model loading status, inference request processing and model version updates. + - ii) It is required to support anomaly detection and alerting, establishing an anomaly detection mechanism that automatically triggers alerts when the model operation is abnormal and provide detailed exception information. + - iii) It is required to support log recording and analysis, comprehensively recording model operation logs for troubleshooting and performance optimization, e.g., successful requests, failed requests and their reasons, model version switching records. +- b) The AI cloud platform is recommended to support monitoring the training and inference processes and to conduct training optimization analysis. + - i) It is recommended to monitor the foundation model training progress, clearly displaying the completion ratio and estimated remaining time, which facilitates monitoring personnel to control the training process and plan subsequent tasks and deployment in advance. + - ii) It is recommended to support log queries during the inference process, e.g., the source of inference requests, input and output data and processing time. + +##### **7.1.1.2 User experience** + +- a) The AI cloud platform is recommended to support monitoring the user experience. + - i) It is recommended to support monitoring of response times, tracking of the response times, and to assist in judging the quality of user experience. + - ii) It is recommended to support monitoring of user success rates, establishing a mechanism to monitor the success rate of user requests to identify and address issues that may lead to user request failures in a timely manner. + +##### **7.1.1.3 Resource occupancy** + +- a) The AI cloud platform is required to support monitoring of business relationship topology and resource usage. + - i) It is required to support the monitoring of the status of training and inference instances and tasks, as well as the transition of states. + - ii) It is required to support the viewing of the business system relationship topology, displaying the dependencies between business systems. + - iii) It is required to support the monitoring of resource usage for foundation model training and inference instances and tasks, e.g., central processing unit (CPU)/graphics processing unit (GPU) usage rates, memory and video memory usage, disk input/output (I/O) and storage, and network bandwidth. + +#### **7.1.2 AI cloud platform resource** + +##### **7.1.2.1 AI compute** + +- a) The AI cloud platform is required to support monitoring of AI computing resources, including node status and task management. + - i) It is required to support monitoring of computing node status, implement comprehensive monitoring of the health status of computing nodes, including key indicators, e.g., CPU and GPU utilization, temperature and power consumption. + - ii) It is required to support task queue management, monitoring the length and processing speed of the task queue to ensure tasks are processed in a timely manner and to prevent task backlogs. +- b) The AI cloud platform is recommended to support defining monitoring metrics. + - i) It is recommended to support custom metric monitoring, e.g., expanding the monitoring metrics of pre-set resource types in the monitoring system through protocols. + +##### **7.1.2.2 AI storage** + +- a) The AI cloud platform is required to support monitoring of AI storage resources, e.g., storage capacity and storage performance. + - i) It is required to support AI training dataset loading and storage performance monitoring of foundation model check points, e.g., input/output operations per second (IOPS). + - ii) It is required to support storage capacity management, tracking of the usage of storage capacity by foundation model check point operations, setting up capacity warning mechanisms. + +##### **7.1.2.3 AI network connectivity** + +- a) The AI cloud platform is required to support monitoring of AI networks, e.g., lossless indicators, inter GPU communication paths and latency, link congestion and uneven load sharing information. + - i) It is required to support the monitoring of lossless indicators of AI networks, e.g., RDMA over converged ethernet (RoCE) network, explicit congestion notification (ECN), priority flow control (PFC) frames. + - ii) It is required to support monitoring the path and latency of GPU interconnection through the network, e.g., the devices, device ports, device forwarding latency, auxiliary network forwarding issues and network performance issues. + - iii) It is required to support traffic collection and analysis, e.g., flow completion time (FCT), flow establishment time (FET) and other information of GPU card communication in combination with end-to-end information. + - iv) It is required to support monitoring of network failures, e.g., node failure and optical module failure during AI training tasks. + +### **7.2 Alarm capabilities** + +### **7.2.1 Monitoring data mining** + +- a) The AI cloud platform is required to support analysis and visual presentation of monitoring data. + - i) It is required to support the intelligent recognition of status abnormalities of AI services and resource usage, e.g., service response delays, service interruptions, GPU resource exhaustion and storage exhaustion. + +- ii) It is required to support mining of AI business data and resource usage data and to establish a baseline model on historical monitoring data. +- iii) It is required to support AI business data and AI resource usage data visualization. + +#### **7.2.2 Intelligent alarm analysis** + +- a) The AI cloud platform is required to support the classification, diagnosis and sorting of alert events. + - i) It is required to support intelligent classification of alarm events, e.g., classifying alarm events intelligently based on factors. + - ii) It is required to support intelligent diagnosis of alarm causes and provide solutions or suggestions, e.g., by combining monitoring data and system logs and applying intelligent analysis techniques to identify the causes of alarms. + - iii) It is required to support intelligent sorting of alarm priorities, sorting them intelligently based on factors, e.g., the urgency and impact of the alarm events. +- b) The AI cloud platform is recommended to support the alert convergence. + - i) It is recommended to support alert convergence, which can analyse, merge and discard generated alert messages based on rules or intelligent algorithms. + +### **7.2.3 Intelligent alarm threshold setting** + +- a) The AI cloud platform is required to support automatic threshold setting and alerting capability. + - i) It is required to support dynamic threshold intelligent adjustment, which adjusts alarm thresholds intelligently based on factors, e.g., system operation status and load conditions. + - ii) It is required to support custom alarm threshold settings, e.g., providing custom alarm threshold settings. + - iii) It is required to support threshold verification and optimization mechanisms, e.g., regularly verifying and optimizing the set alarm thresholds. +- b) The AI cloud platform is recommended to support the aggregation and distribution of alert information. + - i) It is recommended to support the unified aggregation and distribution of alert information through application programming interface (API) or software development kit (SDK) interfaces. + +## **7.3 Operation capabilities** + +### **7.3.1 Automatic scaling** + +- a) The AI cloud platform is required to support automatic scaling, which is the automatic management of nodes in a cluster to adapt to demand. + - ii) It is required to support the creation and management of resource configuration templates. + - ii) It is required to support automatic resource scaling, dynamically increasing or decreasing computing resources based on business demands. + - iii) It is required to support elastic scaling strategies, providing various scaling strategies, e.g., pre-set strategies based on time periods and trigger strategies based on load thresholds. +- b) The AI cloud platform is recommended to support automatic scaling refers to the automatic management of nodes in a cluster to adapt to demand. + +- i) It is recommended to support the storage and analysis of historical resource usage data, e.g., providing trend forecasts and resource demand assessments through the analysis of historical data. +- ii) It is recommended to support automatic resource failure recovery, e.g., automatically restarting or reallocating resources when anomalies are detected. + +### **7.3.2 Automatic deployment** + +- a) The AI cloud platform is required to support automated deployment, including the automatic completion of application building, testing and deployment through scripts or tools, reducing errors and time consumption caused by manual intervention. + - i) It is required to support the business lifecycle management of, including deployment, release, operation, maintenance, version management and offline management of the foundation model. + - ii) It is required to support business automation orchestration, e.g., arranging automated job processes through file libraries, script libraries and API libraries, e.g., visual process orchestration, job node parameter passing, process redirection, process publishing and other features. + +### **7.3.3 Automatic schedule** + +- a) The AI cloud platforms is required to support automatic scheduling, which is the rational allocation and scheduling of resources among different types of resources to achieve efficient resource management and task execution. + - i) It is required to support unified management and scheduling of all resources, such as computing resources and storage resources, e.g., the allocation, release and adjustment of resources. + - ii) It is required to support the formation of multiple resource pools with computing, networking and storage, and carry out scheduling and allocation based on demand. + - iii) It is required to support automatic scheduling of resources, e.g., allocating resources to different containers or virtual machines. + - iv) It is required to support load balancing, e.g., distributing tasks evenly across computing nodes through intelligent load balancing algorithms. +- b) The AI cloud platform is recommended to support automatic scheduling, which is the rational allocation and scheduling of resources among different regions and the allocation of resources for different users. + - i) It is recommended to support cross-regional resource management, e.g., enabling the migration and scaling of resources between infrastructures in different geographical locations. + - ii) It is recommended to support resource quota management, e.g., limiting the maximum quota of resource usage for each user to prevent a single user from over-consuming resources. + - iii) It is recommended to support user whitelisting or priority sorting. + +### **7.3.4 Intelligent fault diagnosis** + +- a) The AI cloud platform is required to use advanced algorithms and technologies to monitor, analyse data, diagnose faults and predict potential problems for systems and devices on AI cloud platforms. + - i) It is required to support fault detection, e.g., using artificial intelligence methods to detect faults or abnormal behaviours in the system. + +- ii) It is required to support fault localization, e.g., identifying the cause and location of faults by combining system logs, monitoring data and other information. + - iii) It is required to support providing fault resolution suggestions, e.g., offering plans for faults to help operations and maintenance personnel restore the system to normal operation. +- b) The AI cloud platform is recommended to use advanced algorithms and technologies to build fault databases and diagnosis fault based on AI. + - i) It is recommended to support the monitoring and fault diagnosis of the foundation model training through AI natural language interaction methods, e.g., end-to-end network integration traffic analysis, congestion analysis and forwarding path optimization. + - ii) It is recommended to support the establishment of a database of equipment failures to update predictive models for equipment failures. + +#### **7.3.5 Auto recovery** + +- a) The AI cloud platform is required to implement automatic fault recovery, data recovery and service recovery. + - i) It is required to support automatic fault recovery, e.g., for certain predictable faults or abnormal behaviours, the system can automatically trigger recovery mechanisms to restore normal system operation without manual intervention. + - ii) It is required to support data recovery, offering data backup and restoration functions. + - iii) It is required to support service recovery, e.g., services automatically restart or switch to backup services. + +## Bibliography + +- [b-ITU-T T.62] Recommendation ITU-T T.62 (1993), *Control procedures for teletex and Group 4 facsimile services.* +- [b-ITU-T X.1643] Recommendation ITU-T X.1643 (2022), *Security requirements and guidelines for virtualization containers in cloud computing environments.* + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.61-202512-I_PDF-E/raw.md b/marked/F/T-REC-F.748.61-202512-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..017a66efeb5339db01394a419dcd70b9211afd4e --- /dev/null +++ b/marked/F/T-REC-F.748.61-202512-I_PDF-E/raw.md @@ -0,0 +1,487 @@ + + +# Recommendation **ITU-T F.748.61 (12/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +**General technical requirements and framework +for artificial intelligence cloud platform – +Inference optimization** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.61 + +# General technical requirements and framework for artificial intelligence cloud platform – Inference optimization + +## Summary + +With the rapid proliferation of foundation models, the importance of inference optimization is receiving increasing attention, as it enables foundation models to be deployed in scenarios such as cloud, edge, and endpoint, providing extensive services efficiently, and improving the efficiency of multimedia application development significantly. Therefore, the technical requirements for inference optimization are crucial. + +Recommendation ITU-T F.748.61 provides general technical requirements for inference optimization, including data processing, model processing, model deployment and model inference. This Recommendation is to specify framework and requirements that meets the needs of foundation model inference optimization and promotes the in-depth application and development of foundation models in various industries. + +## History\* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.61 | 2025-12-14 | 21 | 11.1002/1000/16586 | + +## Keywords + +Artificial intelligence, cloud platform, foundation model, inference optimization. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2026 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-------------------------------------------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 1 | +| 5 Conventions ..... | 2 | +| 6 Framework of inference optimization of AI cloud platform ..... | 2 | +| 6.1 General optimization techniques ..... | 3 | +| 6.2 Cloud-edge-end collaborative ..... | 3 | +| 7 General technical requirements for general optimization and cloud-edge-endpoint collaboration ..... | 3 | +| 7.1 General optimization techniques ..... | 3 | +| 7.2 Cloud-edge-endpoint collaboration ..... | 7 | +| Bibliography..... | 10 | + + + +# Recommendation ITU-T F.748.61 + +## General technical requirements and framework for artificial intelligence cloud platform – Inference optimization + +# 1 Scope + +This Recommendation specifies the general technical requirements for an artificial intelligence (AI) cloud platform, which are mainly focused on inference optimization of the platform. The Recommendation covers the following two aspects: + +- Framework of inference optimization for AI cloud platform. +- General technical requirements for general optimization and cloud-edge-endpoint collaboration. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[[ITU-T F.748.38](#)] Recommendation ITU-T F.748.38 (2024), *Technical specification for artificial intelligence cloud platform: General architecture*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 machine learning** [b-ITU-T Y.3172]: Processes that enable computational systems to understand data and gain knowledge from it without necessarily being explicitly programmed. + +**3.1.2 quantization** [b-ITU-T F.748.53]: A technique to change the bit width in the original high-precision complex model from a high-bit representation to a low-bit width. + +**3.1.3 model inference** [b-ITU-T Y.3179]: Process by which a deployed machine learning model generates a result. + +**3.1.4 inference engine** [b-ITU-T Y.3179]: Functionality that provides runtime environment for a machine learning model and exposes corresponding machine learning model inference capability. + +**3.1.5 inference services** [b-ITU-T Y.3115]: Services that provide authorized consumers with the capabilities of inference based on the artificial intelligence/machine learning models and data within a single domain and/or across domains. + +## 3.2 Terms defined in this Recommendation + +None. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|-----------------------------------| +| AI | Artificial Intelligence | +| API | Application Programming Interface | +| CPU | Central Processing Unit | +| DP | Data Parallelism | +| EP | Expert Parallelism | +| GEMM | General Matrix Multiplication | +| GPU | Graphics Processing Unit | +| HTTP | Hypertext Transfer Protocol | +| I/O | Input/Output | +| KV | Key-Value | +| LLM | Large Language Model | +| MLA | Multi-head Latent Attention | +| MoE | Mixture of Experts | +| MTP | Multi-Token Prediction | +| PD | Prefill-Decode | +| PP | Pipeline Parallelism | +| PTQ | Post-Training Quantization | +| QAT | Quantization-Aware Training | +| QoS | Quality of Service | +| SDK | Software Development Kit | +| SLO | Service Level Objectives | +| TP | Tensor Parallelism | + +# 5 Conventions + +In this Recommendation: + +- The keywords "**is required to**" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The keywords "**is recommended**" indicate a requirement that is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. + +# 6 Framework of inference optimization of AI cloud platform + +[ITU-T F.748.38] for artificial intelligence cloud platforms provides a framework for artificial intelligence cloud platforms. This Recommendation mainly focuses on the optimization requirements of deployment and inference module. Inference optimization includes the optimization techniques required for model deployment, inference and service steps, enabling AI cloud platforms to provide low-latency, high-throughput and more reliable services, to enhance the convenience and efficiency of developing AI multimedia applications. This Recommendation focuses on two aspects for foundation model inference scenarios in artificial intelligence cloud platforms: general optimization techniques and cloud-edge-end collaborative optimization, as shown in Figure 1. + +![Figure 1 – Framework of inference optimization for AI cloud platform. The diagram shows a hierarchical structure. At the top is 'AI-based multimedia applications' with examples like question answering and code generation. Below this is 'Inference optimization of AI cloud platform', which contains 'Cloud-edge-endpoint collaboration' (Data transmission, Model deployment, Collaborative inference) and 'General optimization techniques' (Data processing, Model processing, Model deployment, Model inference). At the bottom are 'Multimedia data, business data, attributes' and 'Foundation model (Text, image, audio, multimodal)'.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +The diagram illustrates the framework of inference optimization for an AI cloud platform. It is structured as follows: + +- AI-based multimedia applications** (Top layer): Includes question answering, logical reasoning, mathematical ability, code generation, machine translation, text generation, multimodal understanding and generation capabilities ... +- Inference optimization of AI cloud platform** (Middle layer): + - Cloud-edge-endpoint collaboration**: Data transmission collaboration, Model deployment collaboration, Collaborative inference. + - General optimization techniques**: + - Data processing**: Distributed processing, Acceleration technology. + - Model processing**: Model compression, Model loading cache. + - Model deployment**: Automated deployment, Distributed deployment. + - Model inference**: Memory optimization, Computation optimization, Scheduling optimization, Service cache optimization, Load balancing, Adaptive inference optimization, Model orchestration. +- Multimedia data, business data, attributes** and **Foundation model (Text, image, audio, multimodal)** (Bottom layer). + +Figure 1 – Framework of inference optimization for AI cloud platform. The diagram shows a hierarchical structure. At the top is 'AI-based multimedia applications' with examples like question answering and code generation. Below this is 'Inference optimization of AI cloud platform', which contains 'Cloud-edge-endpoint collaboration' (Data transmission, Model deployment, Collaborative inference) and 'General optimization techniques' (Data processing, Model processing, Model deployment, Model inference). At the bottom are 'Multimedia data, business data, attributes' and 'Foundation model (Text, image, audio, multimodal)'. + +F.748.61(25) + +**Figure 1 – Framework of inference optimization for AI cloud platform** + +## 6.1 General optimization techniques + +This clause specifies a comprehensive framework of general optimization techniques for foundation model inference applications, spanning the entire lifecycle from data processing, model processing to deployment to model inference. The framework aims to enhance system throughput, reduce end-to-end latency, improve resource utilization efficiency and ensure service stability and scalability. Core components include distributed data processing and acceleration techniques for efficient input data preparation; model compression and loading cache strategies to enable model light weighting and rapid initialization; support for automated and distributed deployment configurations, integrated with parallel acceleration strategies and disaggregated architectures to ensure reliable deployment of large-scale models. During inference, optimization is achieved through memory management, operator fusion, scheduling, service caching, load balancing and adaptive inference control, enabling dynamic adaptation to traffic fluctuations and improved computational efficiency. Furthermore, the standard emphasizes model orchestration capabilities to support flexible integration across models, components and systems. + +## 6.2 Cloud-edge-end collaborative + +This clause specifies a collaborative framework for foundation model inference across cloud, edge and endpoint tiers, enabling end-to-end orchestration of data transmission, model deployment, and inference execution. A unified, policy-driven control loop facilitates dynamic adaptation and ensures efficient, coherent propagation of system state, task context and operational policies across all layers, supporting high-performance and scalable inference services in distributed environments. + +# 7 General technical requirements for general optimization and cloud-edge-endpoint collaboration + +## 7.1 General optimization techniques + +From the perspective of the entire process of multimedia application development, general optimization techniques for foundation model inference scenarios includes inference data processing, model processing, model deployment and model inference. + +### **7.1.1 Data processing** + +Data processing focuses on the handling of inference input data, including storage, access, pre-processing and retrieval stages. + +#### **7.1.1.1 Distributed processing** + +Distributed processing focuses on the distributed processing of inference data, achieving parallel acceleration. + +- a) It is required to support distributed data processing by enabling compatibility with distributed data processing frameworks and providing scalable processing capabilities across clustered environments. +- b) It is recommended to support data parallel access through scalable distributed storage architectures and concurrent data access mechanisms across multiple computing nodes. + +#### **7.1.1.2 Acceleration technology** + +Data processing acceleration technology focuses on optimization techniques to enhance data location, processing and read/write performance. + +- a) It is required to support asynchronous data processing to reduce pipeline waiting time, improve speed and throughput, e.g., asynchronous tokenization and detokenization; overlapping distributed communication and computation through asynchronous communication techniques to lower data transmission latency. +- b) It is required to support data compression and efficient data encoding, aimed at reducing storage requirement and improving data input/output (I/O) efficiency. +- c) It is recommended to support dynamic data sharding and tiered storage allocation, to enable efficient management and accelerated access for frequently accessed data. +- d) It is recommended to support efficient data retrieval through specific retrieval strategies to enhance data matching, location and reading, e.g., vector similarity search, hierarchical retrieval, asynchronous parallel retrieval. + +### **7.1.2 Model processing** + +Model processing focuses on the preparations before model deployment to achieve availability and efficiency in different environments and devices, including model compression and model loading cache. + +#### **7.1.2.1 Model compression** + +Model compression focuses on various model compression methods to achieve model light weighting and improve inference efficiency while ensuring inference accuracy. + +- a) It is required to support basic model compression methods including quantization, pruning and knowledge distillation, e.g., post-training quantization (PTQ), model quantization. +- b) It is recommended to support compression methods that leverage model and data characteristics, e.g., mixed-precision quantization. +- c) It is recommended to support compression methods tailored to specific model architectures, e.g., expert quantization and expert pruning for mixture of experts (MoE) models. +- d) It is recommended to support combinations of multiple compression methods, e.g., the combination of quantization and pruning. +- e) It is recommended to support dynamic adaptive compression methods based on models and data, e.g., quantization-aware training (QAT), dynamic-aware quantization. + +#### **7.1.2.2 Model loading cache** + +Model loading cache focuses on cache optimization strategies to improve deployment and inference efficiency, e.g., cache optimization strategies for the initial deployment of foundation models. + +- a) It is recommended to support multi-level caching mechanisms and parallelized preloading of sharded model components, to improve initialization speed and reduce latency during the first inference request. +- b) It is recommended to support ultra-fast loading through specific model formats, e.g., converting models to specialized formats for rapid loading. + +### **7.1.3 Model deployment** + +Model deployment focuses on optimization techniques during the model deployment process, including automated deployment and distributed deployment. + +#### **7.1.3.1 Automated deployment** + +Automated deployment focuses on the automation of the model deployment process, including deployment configuration and adaptive deployment adjustment. + +- a) It is recommended to support automated deployment configurations that integrate with container orchestration tools to enable automated container deployment, scaling and management across multiple nodes. +- b) It is recommended to support dynamic adjustment of deployment strategy based on model and service characteristics, enabling automated deployment, e.g., dynamically allocating deployment resources based on service traffic fluctuations, or automatically selecting prefill-decode (PD) disaggregated inference deployment based on request concurrency and model structure. + +#### **7.1.3.2 Distributed deployment** + +Distributed deployment refers to achieving the deployment of large-sized models and the integration of various sub-models through distributed methods. + +- a) It is required to support distributed deployment configurations across single-node multi-graphics processing unit (GPU) and multi-node multi-GPU environment. +- b) It is required to support canary deployment to ensure stable and reliable services during distributed deployment process. +- c) It is required to support general distributed parallel acceleration strategies, e.g., data parallelism (DP), pipeline parallelism (PP) and tensor parallelism (TP). +- d) It is recommended to support distributed parallel acceleration strategies tailored to specific model architectures, e.g., expert parallelism (EP). +- e) It is recommended to support deployment optimization through high-performance interconnect technologies to reduce communication latency and improve data transfer efficiency in distributed inference environments. +- f) It is recommended to PD disaggregated deployments, including distributed key-value (KV) cache management with cross-node sharing and eviction control. + +### **7.1.4 Model inference** + +Model inference focuses on optimization techniques during the model inference process to adapt to service traffic fluctuations and reduce resource requirements, including memory optimization, computation optimization, etc. + +#### **7.1.4.1 Memory optimization** + +Memory optimization focuses on memory optimization during model inference, including memory sharing strategy and memory optimization techniques. + +- a) It is recommended to support memory sharing strategies through shared model parameter storage and request parameter reuse of inference context data. +- b) It is recommended to support multi-dimensional GPU memory optimization techniques that dynamically migrate intermediate activations or model weights between GPU and central processing unit (CPU) memory based on access frequency and execution schedule. + +#### **7.1.4.2 Computation optimization** + +Computation optimization focuses on optimization at the operator and model levels, including operator fusion and model-level computation optimization. + +- a) It is required to optimize operators by operator fusion to reduce time consumption during memory access and kernel launch, e.g., general matrix multiplication (GEMM) operator's optimization, fusion of attention operator. +- b) It is recommended to support model-level computation optimization, e.g., multi-token prediction (MTP), speculative decoding to improve inference parallelism and multi-head latent attention (MLA) model structure optimization to reduce inference GPU memory (KV cache) usage. + +#### **7.1.4.3 Scheduling optimization** + +Scheduling optimization focuses on the optimization of inference tasks and service requests, improving system performance, including batch scheduling optimization, service scheduling optimization, etc. + +- a) It is required to support batch scheduling optimization to handle queries of varying lengths in concurrent scenarios, to improve system throughput and compute resource utilization, e.g., continuous batching and dynamic batching. +- b) It is required to support optimization of service scheduling strategies through dynamic request dispatching, e.g., according to service priority, resource loading and cache hit rate to allocate requests to the optimal queue in order to reduce latency and maintain system load balance. +- c) It is required to support cache-aware request routing that directs inference tasks to processing queues with higher model or context reuse potential, based on cache residency status and access patterns, to reduce redundant computation and end-to-end latency. + +#### **7.1.4.4 Service cache optimization** + +Service cache optimization focuses on cache optimization for specific tasks during the service process, such as optimization for hotpot request caching. + +- a) It is required to support multi-level caching strategies that allocate frequently accessed data across hierarchical storage tiers, e.g., storing frequent data in GPU memory to reduce redundant computation and I/O operations. +- b) It is recommended to support cache prefetching strategies, e.g., prefetch context-critical content such as shared prompts, prefix embeddings, and KV cache blocks. + +#### **7.1.4.5 Load balancing** + +Load balancing focuses on stability assurance optimizations during the model service process due to traffic fluctuations, including rate limiting, elastic scaling, etc. + +- a) It is required to support user-defined elastic resource policies, e.g., setting elasticity thresholds and elasticity rules. + +- b) It is required to support elastic resource policies based on pre-set rules. +- c) It is recommended to support adaptive elastic scaling to automatically and timely ensure system stability, e.g., predicting required resources based on concurrency, resource loading and computational capacity to achieve automatic scaling. + +#### **7.1.4.6 Adaptive inference optimization** + +Adaptive inference optimization focuses on dynamically adjusting models, resources and service parameters based on real-time traffic and service requirements to meet service demands. This includes parameter adaptive adjustment, deployment strategy adaptive adjustment. + +- a) It is recommended to support dynamic inference configuration adjustment according to traffic, model and service level objectives (SLO), e.g., dynamically adjusting parallelism and batch-size according to service latency constraints. +- b) It is recommended to support dynamic selection of optimal deployment strategies according to traffic, model and SLO, e.g., for large language model (LLM) inference scenarios, selecting whether to adopt PD disaggregated inference strategy and dynamically adjust PD resource allocation based on concurrency and computational capacity. + +#### **7.1.4.7 Model orchestration** + +Model orchestration focuses on the orchestration and combination capabilities of service construction, improving scalability, including interface support, orchestration and collaboration, etc. + +- a) It is recommended to support standardized interfaces to facilitate unified service invocation, e.g., software development kit (SDK), hypertext transfer protocol (HTTP) application programming interface (API). +- b) It is recommended to support configuring service parameters and model parameters through standardized interfaces. +- c) It is recommended for orchestration and collaboration to flexibly handle complex logic, e.g., interaction and orchestration between models, between models and components (e.g., vector databases) and between models and systems (e.g., public model repositories, business systems and public search engine). + +## **7.2 Cloud-edge-endpoint collaboration** + +From the perspective of the entire process of multimedia application development, cloud-edge-endpoint collaboration optimization for foundation model inference scenarios covers the entire process: data transmission collaboration, model deployment collaboration and collaborative inference. This clause defines the technical requirements for cloud-edge-endpoint collaboration optimization part of artificial intelligence cloud platforms. + +### **7.2.1 Data transmission collaboration** + +Data transmission collaboration focuses on the collaboration of cloud-edge-end data transmission, including data types, transmission strategies, collaboration mechanism, etc. + +##### **7.2.1.1 Data types** + +It is recommended to support data types including text, audio, image, video, 3-D point cloud and multi-modal data. + +#### **7.2.1.2 Transmission strategies** + +Cloud-side technical requirements include: + +- a) It is recommended to support encrypted tunnels from the edge and enforce tenant-isolated forwarding paths for each user workload. It is recommended to support a load-balancing mechanism. It can optionally employ a software-defined path for stateless flows. + +- b) It is recommended to support issues and adjust transmission policies in real time via the cloud-side controller or policy engine according to flow characteristics. + +Edge-side technical requirements include: + +- a) It is required to establish encrypted tunnels to the cloud and to enforce tenant-isolated virtual networks inside the edge cluster. +- b) It is recommended to support receive and enforce policy updates from the cloud. +- c) It is recommended to provide software-defined flow identification and scheduling on edge nodes. +- d) It is recommended to support activating a software-defined reliable-connection channel for stateful flows to guarantee data ordering and reliability. + +Endpoint-side technical requirements include: + +- a) It is recommended to leverage access-layer encryption for authentication. +- b) It is recommended to support embedding a policy agent in endpoint that automatically selects direct-to-cloud, edge-relay, or peer-to-peer paths according to distributed policy tags. + +#### **7.2.1.3 Collaboration mechanism** + +The data transmission collaboration mechanism is required to orchestrate end-to-end policies via a unified software-defined control plane of the distributed cloud, enabling runtime adaptation of routing, quality of service (QoS) and security parameters in response to network quality, load and data-type characteristics. + +### **7.2.2 Model deployment collaboration** + +Model deployment collaboration focuses on cloud-edge-end model deployment collaboration strategies, which can be reasonably deployed according to different resource scenarios of cloud, edge and endpoint, including distributed deployment strategies and model deployment collaboration mechanism, etc. + +#### **7.2.2.1 Distributed deployment strategies** + +Cloud-side technical requirements include: + +- a) It is recommended to host a multi-AZ (available zone) repository that guarantees download speed, and load-balanced failover. +- b) It is recommended to maintain a unified model-version registry with a metadata centre indexing model type, dependency graphs and compatibility matrices. +- c) It is recommended to establish continuous integration pipelines that support iterative updates and parameter tuning, utilizing the data gathered from edges and endpoints. +- d) It is recommended to orchestrate grey-scale and phased rollout strategies. +- e) It is recommended to support monitor edge and endpoint workloads in real time and to trigger dynamic redistribution policies. + +Edge-side technical requirements include: + +- a) It is recommended to support model status checks and functional validations. +- b) It is recommended to issue failure alert to cloud in case of deployment failure and guarantee atomic deployment with an automated rollback path to the last stable revision. +- c) It is recommended to support cache models locally and to perform incremental upgrades. +- d) It is recommended to expose health telemetry to the cloud for global scheduling decisions. +- e) It is recommended to use the inter-edge network to support fetching cached models from other edges to accelerate transmission. + +Endpoint-side technical requirements include: + +- a) It is recommended to store only the minimal model artifacts needed for local inference. +- b) It is recommended to support conflict scans and to signal incompatibility prior to any update. +- c) It is recommended to support atomic switch-over and instantaneous rollback initiated by either cloud or edge. + +#### **7.2.2.2 Collaboration mechanism** + +The model deployment collaboration mechanism is required to facilitate the deployment of the new model version –which has been trained, tested, packaged and hosted on the cloud – to edges and endpoints. It is recommended to establish a single control-plane protocol that synchronizes version states, deployment intents and rollback signals across cloud, edge and endpoints, to guarantee that every tier performs consensus-based conflict detection and maintains a consistent view of the global model fleet. + +### **7.2.3 Collaborative inference** + +Collaborative inference focuses on cloud-edge-end model collaborative inference strategies, improving response speed while ensuring the accuracy of inference, including service collaborative scheduling, etc. + +##### **7.2.3.1 Service collaborative scheduling** + +Cloud-side technical requirements include: + +- a) It is required to serve as the ultimate backstop that ingests intermediate results from edges, performs format conversion, data cleaning and multi-source fusion, and emits the final inference response. +- b) It is recommended to monitor the model-instance status of every edge. +- c) It is recommended to classify incoming inference tasks by latency/throughput profiles and to compute optimal task-decomposition graphs dynamically. + +Edge-side technical requirements include: + +- a) It is recommended to accept task bundles from the cloud and to perform secondary decomposition by weighing task type, local workload and endpoint capabilities. +- b) It is recommended to collect partial or pre-processed results from endpoints, to execute secondary inference and fusion, and to forward residual workloads or aggregated intermediates to the cloud. +- c) It is recommended to dispatch subtasks to endpoints and to trigger inter-edge collaboration when local capacity is exceeded. + +Endpoint-side technical requirements include: + +- a) It is recommended to execute assigned inference or pre-processing workloads independently. +- b) It is recommended to return results directly to the user upon successful local completion or to upload pre-processed data to the edge otherwise. +- c) It is required to detect and immediately report run-time failures to the supervising edge. + +##### **7.2.3.2 Collaboration mechanism** + +The inference collaboration mechanism is required to maintain a unified, policy-driven control loop that propagates resource snapshots, task graphs and error signals among cloud, edges and endpoints in real time to guarantee atomic task hand-offs and ordered result aggregation across all tiers. + +## Bibliography + +- [b-ITU-T F.748.53] Recommendation ITU-T F.748.53 (2025), *Representation and compression methods of artificial intelligence models.* +- [b-ITU-T Y.3115] Recommendation ITU-T Y.3115 (2022), *AI enabled cross-domain network architectural requirements and framework for future networks including IMT-2020.* +- [b-ITU-T Y.3172] Recommendation ITU-T Y.3172 (2019), *Architectural framework for machine learning in future networks including IMT-2020.* +- [b-ITU-T Y.3179] Recommendation ITU-T Y.3179 (2021), *Architectural framework for machine learning model serving in future networks including IMT-2020.* + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.66-202512-I_PDF-E/raw.md b/marked/F/T-REC-F.748.66-202512-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..d8945d955ad38f2c1ccb8acd6357e05bab8e1b73 --- /dev/null +++ b/marked/F/T-REC-F.748.66-202512-I_PDF-E/raw.md @@ -0,0 +1,547 @@ + + +# Recommendation **ITU-T F.748.66 (12/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +# --- **Requirements and framework for embodied artificial intelligence systems** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.66 + +# Requirements and framework for embodied artificial intelligence systems + +## Summary + +Embodied artificial intelligence (EAI) extends artificial intelligence into the physical world. It integrates artificial intelligence into a physical body that interacts autonomously with and adapts to the physical world. Embodied intelligence systems enable machines to perceive, decision making, execute, and autonomously adapt to complex real-world environments. + +Recommendation ITU-T F.748.66 provides the definition of EAI and introduces the framework and technical requirements for embodied artificial intelligence systems. It provides a unified framework for system capability building and performance requirements that guide the design, development, evaluation, and interoperability of embodied artificial intelligence (AI) systems. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.66 | 2025-12-14 | 21 | 11.1002/1000/16591 | + +## Keywords + +Decision-making, embodied artificial intelligence, embodied execution, embodied interaction, embodied perception. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2026 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|---------------------------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 1 | +| 5 Conventions ..... | 2 | +| 6 Overview of EAI..... | 2 | +| 6.1 Introduction of EAI ..... | 2 | +| 6.2 Features of EAI..... | 2 | +| 7 Framework of the EAI system..... | 3 | +| 7.1 Basic layer ..... | 3 | +| 7.2 Functional layer ..... | 4 | +| 7.3 Application layer ..... | 5 | +| 8 Technical requirements for EAI system ..... | 5 | +| 8.1 Perception capability ..... | 5 | +| 8.2 Decision-making capability..... | 5 | +| 8.3 Execution capability ..... | 6 | +| 8.4 Interaction capability ..... | 6 | +| 8.5 Learning capability ..... | 6 | +| Appendix I – Use cases with Embodied AI ..... | 7 | +| Appendix II – Use cases of EAI system capabilities in domain-specific scenarios ..... | 8 | +| II.1 Automotive manufacturing..... | 8 | +| II.2 Electronics manufacturing..... | 8 | +| II.3 Logistics and warehousing ..... | 8 | +| II.4 Retail services..... | 9 | +| II.5 Home services ..... | 9 | +| II.6 Hospitality service ..... | 9 | +| II.7 Search, rescue and security..... | 10 | +| II.8 Healthcare..... | 10 | +| II.9 Transportation..... | 10 | +| II.10 Education and research..... | 11 | +| Bibliography..... | 12 | + + + +# Requirements and framework for embodied artificial intelligence systems + +# 1 Scope + +The Recommendation provides the framework and technical requirements for embodied artificial intelligence (EAI) systems, including the system workflow and the requirements for perception, decision-making, execution, interaction, and learning capabilities. It addresses the following subjects: + +- a) The framework of embodied artificial intelligence systems. +- b) The technical requirements for embodied artificial intelligence systems. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ISO 10218-2] ISO 10218-2 (2025), *Robotics – Safety requirements – Part 2: Industrial robot applications and robot cells*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following term defined elsewhere: + +**3.1.1 artificial intelligence (AI)** [b-ITU-T F.749.13]: An interdisciplinary field, usually regarded as a branch of computer science, dealing with models and systems for the performance of functions. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following term: + +**3.2.1 embodied artificial intelligence:** Artificial intelligence integrated into physical systems that interact autonomously with and adapt to the physical world. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations: + +| | | +|------|----------------------------------| +| AI | Artificial Intelligence | +| EAI | Embodied Artificial Intelligence | +| PCB | Printed Circuit Board | +| RGBD | Red Green Blue Depth | +| VFM | Vision Foundation Model | +| VLA | Vision Language Action Model | +| VLM | Vision Language Model | + +# 5 Conventions + +The following conventions are used in this Recommendation: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted, if conformance to this Recommendation is to be claimed. +- The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. + +# 6 Overview of EAI + +## 6.1 Introduction of EAI + +Embodied AI (EAI) is an advanced and emerging branch of artificial intelligence (see [b-IEEE Systems] and [b-Zhang]). It extends artificial intelligence into the physical world, further reducing human cognitive burden in physical tasks, and enables systems to improve through direct data collection and interaction with the environment. + +The concept of EAI and its three key features are illustrated in Figure 1. "obtain information" indicates how the EAI perceives the physical world through three modes: human-machine interaction, interaction with the surrounding environment, and multi-machine interaction. "Task execution" shows how EAI acts back on that world to accomplish tasks. EAI plays a key role in helping embodied bodies such as robots evolve into more intelligent and capable forms. However, its applications are not limited to robots; EAI can also appear in other smart devices and systems that interact with the physical world. The foundation model, whose intelligence is instantiated in the body, shows cognitive, learning, and collaborative features. EAI system use case can be found in Appendix I. + +![Conceptual diagram and key features of EAI](07b17a620c75522d53916a11e12d1bff_img.jpg) + +The diagram illustrates the conceptual framework of Embodied AI (EAI). It consists of two main components: the 'Physical world' and 'Embodied AI'. The 'Physical world' box on the left contains three sub-sections: 'Human-machine' (represented by a person icon), 'Surroundings' (represented by a globe icon), and 'Multi-machine' (represented by a robot icon). The 'Embodied AI' box on the right contains a 'Physical body' section with icons of a humanoid robot, a quadruped robot, a robotic arm, and a drone. To the right of the physical body is a 'Foundation model' section with a cube icon. Arrows indicate the flow of information: 'Obtain information' from the Physical world to the Physical body, and 'Task execution' from the Foundation model back to the Physical world. Within the Physical world, three double-headed arrows labeled 'Cognition', 'Learning', and 'Collaborative' connect the Human-machine, Surroundings, and Multi-machine sections to the Physical body. + +Conceptual diagram and key features of EAI + +Figure 1 – Conceptual diagram and key features of EAI + +## 6.2 Features of EAI + +EAI systems have three main features (In the narrow sense, an EAI system ideally possesses all three; in the broad sense, it can be considered as an EAI system if it has at least one of the following features): + +- a) Cognition ability to form an internal model of the world through physical perception, decision-making, and human-machine interaction. + +- b) Collaborative ability to interact and cooperate with other embodied systems and to operate in a cloud-edge-device collaborative architecture. +- c) Learning ability to learn and improved by interacting with the environment, gaining new skills, and adapting to changes over time. + +# 7 Framework of the EAI system + +The general framework of the EAI system can be divided into three layers, including the basic layer, functional layer, and application layer, shown in Figure 2. + +![Figure 2 – General framework of EAI systems. The diagram shows three layers: Application layer, Functional layer, and Basic layer. The Application layer contains 'Application scenarios' and 'Service interfaces, user/task-oriented services'. The Functional layer contains five modules: 'Embodied perception' (Multimodal perception, Environment perception, Dynamic perception), 'Embodied decision-making' (Task planning, Decision optimization), 'Embodied execution' (Locomotion, Manipulation), 'Embodied interaction' (Human-machine interaction, Multi-machine collaboration), and 'Embodied learning' (Learning from experience). The Basic layer contains 'Foundation model', 'Computing platform', and 'Physical body'. An arrow points from the Functional layer to the Application layer. A small code 'F.748.66(25)' is in the bottom right corner.](b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg) + +Figure 2 – General framework of EAI systems. The diagram shows three layers: Application layer, Functional layer, and Basic layer. The Application layer contains 'Application scenarios' and 'Service interfaces, user/task-oriented services'. The Functional layer contains five modules: 'Embodied perception' (Multimodal perception, Environment perception, Dynamic perception), 'Embodied decision-making' (Task planning, Decision optimization), 'Embodied execution' (Locomotion, Manipulation), 'Embodied interaction' (Human-machine interaction, Multi-machine collaboration), and 'Embodied learning' (Learning from experience). The Basic layer contains 'Foundation model', 'Computing platform', and 'Physical body'. An arrow points from the Functional layer to the Application layer. A small code 'F.748.66(25)' is in the bottom right corner. + +Figure 2 – General framework of EAI systems + +## 7.1 Basic layer + +### 7.1.1 Foundation model + +The foundation models include vision foundation models (VFMs) (see [b-Park]), vision language models (VLMs) (see [b-Sarch]), and vision language action models (VLAs) (see [b-Black]), which are obtained through pre-training or equivalent methods and possess semantic understanding, scene reasoning, and vision-language fusion capabilities. They support the system in forming abstract representations and perceptual reasoning about the environment, roles, and tasks, serving as the cognitive foundation for perception, decision-making, and interaction functionalities. + +### 7.1.2 Computing platform + +The computing platform integrates cloud, edge, and on-premises environments, enabling heterogeneous resource orchestration for training and inference. It ensures low-latency, scalable execution of perception, decision-making, and control pipelines, with embedded mechanisms for security, privacy, and real-time performance. + +### 7.1.3 Physical body + +The physical body of the EAI system is a carrier of embodied AI – humanoid robots, quadrupeds, robot arms, autonomous vehicles, and drones all serve this role. It encompasses the perception system, computing unit, execution system, drive system, and energy system. + +- a) The perception system acquires environmental data through sensors, enabling situational awareness. +- b) The computing unit provides both base-level and AI-driven computation capabilities. + +- c) The execution system, composed of mechanical parts, end-effectors, and drive mechanisms, converts decision-making instructions into actions to accomplish task goals. +- d) The drive system, utilizing motors, hydraulics, or pneumatics, delivers the necessary force and control to ensure action precision and stability. +- e) The energy system, typically in the form of high-density batteries, supplies continuous and stable power to the entire physical body. + +## 7.2 Functional layer + +### 7.2.1 Embodied perception + +The embodied perception module is responsible for acquiring and processing information, establishing awareness and understanding of the external environment, which supports reliable decision-making and successful task execution. + +- a) Multi-modal perception: processes and fuses sensory inputs from different modalities to form coherent representations of the environment. +- b) Environment perception: detects, identifies, and maps static and dynamic elements in the surrounding physical space. +- c) Dynamic perception: tracks temporal changes in the environment and predicts evolving states to support real-time decision-making. + +### 7.2.2 Embodied decision-making + +The embodied decision-making module receives environmental perception information, conducts inference analysis and task planning, and generates step-by-step decision instructions to guide the embodied execution module. + +- a) Task planning: generates executable plans by decomposing tasks into goal-directed action sequences. +- b) Decision optimization: selects optimal strategies and parameters based on real-time context, constraints, and performance objectives. + +### 7.2.3 Embodied execution + +The embodied execution module receives commands from the decision-making module and performs actions. It carries out tasks including locomotion and manipulation. + +- a) Locomotion: controls physical movement and navigation of the embodied system across varying terrains and spaces. +- b) Manipulation: regulates motion trajectories and force application for object-level interaction and tool use. + +### 7.2.4 Embodied interaction + +The embodied interaction module continuously provides feedback on action outcomes and receives instructions through human-machine or multi-machine interaction. This enables the system to quickly identify effective versus ineffective strategies and make more intelligent decisions. + +- a) Human-machine interaction: perceives human states and intentions, and delivers responsive behaviours to support effective collaboration. +- b) Multi-machine collaboration: coordinates actions and shares state information among multiple embodied systems to achieve joint objectives. + +### **7.2.5 Embodied learning** + +The embodied learning module updates internal models, behaviours, and skills through interaction with the environment, real-time feedback, and historical experience. It supports imitation learning (see [b-Hussein]) and autonomous exploration lifelong learning (see [b-Zhu]). By leveraging event memory, role context, and spatial mapping, it refines system representations and behaviours. It adjusts to task changes and corrects erroneous decisions through autonomous reflection. + +## **7.3 Application layer** + +The application layer is responsible for providing service interfaces for embodied AI, as well as user- or task-oriented services of embodied AI systems. It operates above functionality layers, translating system capabilities into domain-specific scenarios applications. Representative applications can be found in Appendix II. + +# **8 Technical requirements for EAI system** + +## **8.1 Perception capability** + +### **8.1.1 Multi-modal perception** + +- a) The system is required to process multi-modal inputs from the environment, including text, audio, image, tactile, and force signals. +- b) The system is recommended to accept humidity, gas concentration, light intensity, and motion data as inputs. + +### **8.1.2 Environment perception** + +The system is required to perceive its surroundings, including spatial structure, object relations, and potential changes. + +### **8.1.3 Dynamic perception** + +The system is required to perceive its own physical state and internal state, including position, velocity, acceleration, battery level, temperature, and computational resources. + +## **8.2 Decision-making capability** + +### **8.2.1 Task planning** + +- a) The system is required to generate control commands by itself or other control modules to control actuators to perform locomotion, manipulation, and obstacle avoidance based on its current motion state, environment perception, and intended task objectives. +- b) The system is recommended to decompose tasks into steps, each of which contains atomic actions. + +NOTE – Atomic actions are fundamental movements and represent the basic building blocks for more complex actions (see [b-Akkaladevi]), which cannot be further decomposed into smaller actions (see [b-Aein]). + +- c) The system is recommended to access the environmental context and task objectives. + +### **8.2.2 Decision optimization** + +- a) The system is recommended to dynamically optimize its control strategy to improve the accuracy, efficiency and safety of the chosen actions. This optimization may include the selection of control strategies, parameter adjustment, or model tuning. +- b) The system is recommended to adjust decisions in real time in response to the changing environments and situations. + +## **8.3 Execution capability** + +### **8.3.1 Locomotion** + +- a) The system is required to adapt to different terrains and environmental changes during locomotion. +- b) The system is required to ensure stable and controllable locomotion, including controlled acceleration, steering, and payload management. +- c) The system is required to detect and avoid both static and dynamic obstacles to ensure safe navigation. + +### **8.3.2 Manipulation** + +- a) The system is required to support diverse manipulation tasks, including pick-and-place, push, in-hand manipulation, and sequenced tasks. +- b) The system is required to adapt to variations in load, material textures, object pose, shape, and colour, regardless of the background, lighting conditions, and different camera parameters. +- c) The system is required to establish robust safety protocols to protect operators and maintenance personnel from physical systems-related hazards. See [ISO 10218-2]. + +## **8.4 Interaction capability** + +### **8.4.1 Human-machine interaction** + +The system is required to support an advanced level of human-machine interaction, including one or more of the natural language processing, gesture recognition, facial expression detection, and body posture analysis. + +### **8.4.2 Multi-machine collaboration** + +The system is recommended to support multi-machine interaction, based on inter-machine and machine-to-infrastructure communication, synchronized coordination of tasks and state, and adaptive and decentralized control strategies. + +## **8.5 Learning capability** + +- a) The system is required to acquire knowledge and skills through a data-driven learning process. This process may be implemented during the design phase, pre-deployment training, or operation. + +NOTE – Training can be conducted with data from third-party, simulation episodes, or real-world interaction. After deployment, some systems may operate with fixed models, while others may continue to update and refine their knowledge through online or incremental learning. + +- b) The system is recommended to autonomously engage in trial-and-error learning through experience. +- c) The system is recommended to learn skills and knowledge through experience, improving its performance in tasks. + +# Appendix I + +## Use cases with Embodied AI + +(This appendix does not form an integral part of this Recommendation.) + +Figure I.1 shows an embodied AI (EAI) system deployed in a retail store that assists in monitoring stock, interacting with customers, and managing goods. Robots act as embodied agents and integrate with cloud-edge-device computing to ensure efficient store operations. + +![Diagram of a retail use case for Embodied AI. A customer (stick figure) interacts with a robot (stick figure with arms). The robot manipulates goods on a shelf and navigates. A camera (security camera icon) perceives the scene and sends data transmission to an EDGE device (cloud icon on a monitor). The EDGE device sends micro-services to a robot icon. The robot icon sends count services to a database (DB w/r). The database is labeled F.748.66(25).](cab0834804fb031b43865554cc8d06ab_img.jpg) + +``` +graph TD + Customer[Customer] -- Interact --> Robot1[Robot] + Robot1 -- Manipulate --> Shelf[Shelf] + Shelf -- Navigate --> Robot1 + Camera[Camera] -- Perceive --> DataTransmission[Data transmission] + DataTransmission --> EDGE[EDGE] + EDGE -- Micro-service --> Robot2[Robot] + Robot2 -- Count service --> DB[(DB w/r)] + DB -- F.748.66(25) +``` + +Diagram of a retail use case for Embodied AI. A customer (stick figure) interacts with a robot (stick figure with arms). The robot manipulates goods on a shelf and navigates. A camera (security camera icon) perceives the scene and sends data transmission to an EDGE device (cloud icon on a monitor). The EDGE device sends micro-services to a robot icon. The robot icon sends count services to a database (DB w/r). The database is labeled F.748.66(25). + +Figure I.1 – Retail use case + +Primary actor: Robots equipped with red green blue depth (RGBD) camera, microphone, tactile sensors, embodied AI + +Secondary actors: Customers + +- Embodied perception: Robots and store sensors collect data on stock levels, store environment, and customer activities. Sensor data are fused and transmitted to the appropriate computing platform (device, edge server, or cloud platform) according to the workload and urgency. +- Embodied decision-making: The EAI system analyses perception data to generate control decisions. Decisions are passed to the robots, specifying navigation routes, manipulation tasks, or customer interaction actions. +- Embodied execution: Robots navigate autonomously to target locations in the store. Robots manipulate goods, such as restocking shelves or moving products. +- Embodied interaction: Robots interact with customers and staff using natural interfaces (speech, gesture, or display). Interaction is guided by embodied AI models to provide personalized assistance and answer queries. +- Embodied learning: Data collected during store operations are stored in server-side databases. Embodied AI models are re-trained at an interval with the latest data to acquire new knowledge. Pre-collected datasets (from simulations, real scenarios, or online sources) are also used for model updates to improve performance. + +# Appendix II + +## Use cases of EAI system capabilities in domain-specific scenarios + +(This appendix does not form an integral part of this Recommendation.) + +This appendix illustrates how embodied intelligence systems can enhance perception, decision-making, execution, interaction, and learning capacities across scenarios including automotive manufacturing, electronics manufacturing, logistics and warehousing, retail services, home services, hospitality services, search, rescue and security, healthcare, transportation, education and research. + +### II.1 Automotive manufacturing + +EAI systems could perform part installation, welding, painting, detect surface cracks, and battery assembly, enabling consistency and speed in assembly. The functions involve: + +- a) Perception: The system supports identifying parts and detecting surface cracks or dimensional deviations. +- b) Decision-making: The system supports adjusting operational parameters in real time, ensuring precision and responsiveness in dynamic factory conditions. +- c) Execution: The system supports executing dexterous operations via integrated vision-force control, enabling small-component installation, bolt fastening, and battery module assembly. +- d) Interaction: The system supports working safely alongside humans. +- e) Learning: The system supports to learn multi-modal production data and machine behaviour to enable optimization of factory layout, workflow design, and assembly strategies. + +### II.2 Electronics manufacturing + +EAI systems could perform component fabrication, printed circuit board (PCB) assembly, adhesive application, inspections, packing, speeding up the coating and assembly, and finishing products. The functions involve: + +- a) Perception: The system supports identifying and locating micro parts. +- b) Decision-making: The system supports adjusting placement trajectories and force parameters based on visual and force feedback. +- c) Execution: The system supports handling cable assembly, the screwing and fastening of components, or picking and placing tiny components. +- d) Interaction: The system supports safe collaboration alongside human operators, enhancing flexible deployment. +- e) Learning: The system supports to learn skills according to the needs of the production line, helping improve product quality and resource use. + +### II.3 Logistics and warehousing + +EAI systems could perform order picking, packing, sorting, labelling, and transporting goods on the factory floor, adapting to changing layouts and order demands. The functions involve: + +- a) Perception: The system supports identifying inventory, detecting spills or misplaced items, and perceiving warehouse layout changes in real time. +- b) Decision-making: The system supports planning and adjusting sub-task sequences, adapting to changing layouts and order demands. + +- c) Execution: The system supports navigating freely among warehouse racks, pick boxes, and placing them precisely to assemble orders on the factory floor. +- d) Interaction: The system supports coordinating multiple robots to collaboratively execute tasks. +- e) Learning: The system supports to learn from operational feedback derived during runtime to optimize mobile navigation routes and order-picking and delivery strategies. + +### **II.4 Retail services** + +EAI systems could guide customers to products, respond to inquiries, and detect spills or misplaced items in store aisles, providing customer-facing services. The functions involve: + +- a) Perception: The system supports detecting shelf inventory levels, recognizing misplaced or out-of-stock items, and perceiving customer presence and behaviour in store aisles. +- b) Decision-making: The system supports determining replenishment tasks, navigation paths, and in-store promotions or guidance. +- c) Execution: The system supports restocking shelves, guiding shopping carts, and operating automated checkouts. +- d) Interaction: The system supports interaction with customers through natural language prompts, touchscreen interfaces, or gesture-based guidance to assist in locating products, answering questions, or facilitating cashier-less checkout experiences. +- e) Learning: The system supports to learn from customer behaviour, inventory turnover, and restocking patterns to optimize store layout, shelf assignment, and personalized product suggestions. + +### **II.5 Home services** + +EAI systems could assist with different household chores and tasks, providing companionship and assistance to elderly or disabled individuals at home. The functions involve: + +- a) Perception: The system supports the perception of humans, objects, and environmental context using multi-modal perception. +- b) Decision-making: The system supports reasoning over perceived environmental context to determine actions or assistance tasks. +- c) Execution: The system supports performing household chores and tasks, including fetching objects, cleaning floors, cleaning windows, mowing lawns, or food preparation. +- d) Interaction: The system supports communication with users via speech, gesture, or touchscreen interfaces. +- e) Learning: The system supports to learn from user interactions and environmental experiences to personalize behaviour, task execution, and decision support over time. + +### **II.6 Hospitality service** + +EAI systems could provide luggage delivery, room service routing, and guest inquiry response, enhancing guest experience. The functions involve: + +- a) Perception: The system supports the perception of guest presence, identity, departmental locations, and obstacles using multimodal sensing. +- b) Decision-making: The system supports reasoning over perceived environmental context to determine actions or assistance tasks. +- c) Execution: The system supports operating mobile manipulators to enter elevators, navigate corridors, and deliver items across multiple floors. + +- d) Interaction: The system supports interaction with guests using natural language, touch interface, or multilingual communication to assist, provide information, or handle check-in/out processes. +- e) Learning: The system supports to learn from guest interactions and service performance to enhance task accuracy, response strategies, and guest satisfaction. + +### **II.7 Search, rescue and security** + +EAI systems could identify individuals in the case of a disaster or emergency. The functions involve: + +- a) Perception: The system supports recognizing targets or environmental threats. +- b) Decision-making: The system supports updating environmental maps and plan search, navigation, or patrol strategies based on real-time environmental information. +- c) Execution: The system supports navigating debris-laden remote terrain and detecting anomalies, avoiding obstacles. +- d) Interaction: The system supports working with human teams to improve search and rescue efficiency. +- e) Learning: The system supports to learn from operational feedback and environmental interaction to optimize search efficiency. + +## **II.8 Healthcare** + +EAI systems could provide medical support and patient care. The functions involve: + +- a) Perception: The system supports identifying medicines and medical equipment, and monitoring patient status. +- b) Decision-making: The system supports planning delivery routes and analysing and monitoring patient conditions. +- c) Execution: The system supports delivering medicines and medical equipment from pharmacies or warehouses and assists elderly or mobility-impaired patients. +- d) Interaction: The system supports interaction with healthcare staff and patients via natural interfaces, to coordinate deliveries and support caregiving. +- e) Learning: The system supports to learn from operational feedback and patient data to optimize delivery efficiency and personalize caregiving routines. + +### **II.9 Transportation** + +EAI systems could reduce traffic congestion, smooth traffic flow, and implement smarter traffic management. The functions involve: + +- a) Perception: The system supports the perception of surrounding objects, traffic signs, lanes, and pedestrians. +- b) Decision-making: The system supports adjusting traffic signal timing, routing, resource allocation, and vehicle state. +- c) Execution: The system supports implementing adaptive vehicle control and coordinated vehicle routing. +- d) Interaction: The system supports integrating and communicating with infrastructure elements to prioritize flow and response. +- e) Learning: The system supports forecasting traffic patterns and continuously optimizes traffic control strategies. + +## **II.10 Education and research** + +EAI systems could automate repetitive lab tasks and provide classroom assistance. The functions involve: + +- a) Perception: The system supports identifying laboratory objects, equipment, and student actions. +- b) Decision-making: The system supports selecting and scheduling tasks to execute based on real-time context in lab workflows or classroom activities. +- c) Execution: The system supports performing laboratory operations or instructional guidance tasks. +- d) Interaction: The system supports interaction with users – students, educators, or researchers – using instructional dialogue. +- e) Learning: The system supports refining task execution capabilities and strategies based on feedback from user interaction and experimental results. + +# Bibliography + +- [b-ITU-T F.749.13] Recommendation ITU-T F.749.13 (2021), *Framework and requirements for civilian unmanned aerial vehicle flight control using artificial intelligence*. +- [b-Aein] Ain, M.J., Aksoy, E.E., Wörgötter, F. (2019), *Library of actions: Implementing a generic robot execution framework by using manipulation action semantics*. The International Journal of Robotics Research. Volume 38, Issue 8. +<> +- [b-Akkaladevi] Akkaladevi, S.C., Plasch, M., Maddukuri, S., Eitzinger, C., Pichler, A., Rinner, B. (2018), *Toward an Interactive Reinforcement Based Learning Framework for Human Robot Collaborative Assembly Processes*. Front. Robot. AI. +<> +- [b-Black] Black, K., Brown, N., Driess, D., Esmail, A., Equi, M., Finn, C., Fusai, N., Groom, L., Hausman, K., Ichter, B., Jakubczak, S., Jones, T., Ke, L., Levine, S., Li-Bell, A., Mothukuri, M., Nair, S., Pertsch, K., Shi, L.X., Tanner, J., Vuong, Q., Walling, A., Wang, H., and Zhilinsky, U. (2024), *$\pi_0$ : A Vision-Language-Action Flow Model for General Robot Control*. CoRR. +<> +- [b-Hussein] Hussein, A., Gaber, M.M., Elyan, E., and Jayne, C. (2017), *Imitation learning: A survey of learning methods*. ACM Computing Surveys (CSUR), 50(2), 1-35. +<> +- [b-IEEE Systems] *IEEE Systems, Man, and Cybernetics Society Announcement: Establishment of the Technical Committee on Embodied AI Systems*. (2024). +<> +- [b-Park] Park, N., Kim, W., Heo, B., Kim, T., and Yun, S. (2023), *What do self-supervised vision transformers learn?* arXiv preprint arXiv:2305.00729. +<> +- [b-Sarch] Sarch, G., Jang, L., Tarr, M.J., Cohen, W.W., Marino, K., and Fragkiadaki, K. (2024), *VLM agents generate their own memories: Distilling experience into embodied programs of thought*. arXiv:2406.145696. +<> +- [b-Zhang] Zhang, H., Zhu, C., Wang, X., Zhou, Z., Hu, S., Zhang, L.Y. (2024), *BadRobot: Jailbreaking LLM-based embodied AI in the physical world*. arXiv preprint arXiv:2407.20242v1. +<[https://www.rivista.ai/wp-content/uploads/2024/10/BadRobot\\_Jailbreaking\\_LLM-based\\_Embodied\\_AI\\_in\\_the.pdf](https://www.rivista.ai/wp-content/uploads/2024/10/BadRobot_Jailbreaking_LLM-based_Embodied_AI_in_the.pdf)> +- [b-Zhu] Zhu, D., Bu, Q., Zhu, Z., Zhang, Y., and Wang, Z. (2024), *Advancing autonomy through lifelong learning: a survey of autonomous intelligent systems*. Frontiers in neurorobotics, Vol.18:1385778. +<> + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.67-202512-I_PDF-E/raw.md b/marked/F/T-REC-F.748.67-202512-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..191a1d66c875440a0f024dc418d255123a0e2c08 --- /dev/null +++ b/marked/F/T-REC-F.748.67-202512-I_PDF-E/raw.md @@ -0,0 +1,777 @@ + + +# Recommendation **ITU-T F.748.67 (12/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +## --- **Requirements and framework for advanced intelligent customer service system** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with white grid lines and the letters 'ITU' in a bold, blue, sans-serif font. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.67 + +# Requirements and framework for advanced intelligent customer service system + +## Summary + +Recommendation ITU-T F.748.67 specifies the requirements and framework of the advanced intelligent customer service system (AICSS). Typical use cases of AICSS are provided in the appendix. + +Advanced intelligent customer service system (AICSS) is a customer service system that integrates artificial intelligence (e.g., large language model, multimodal large model, NLP, ASR and TTS) technology, omnichannel access, omni media interaction and multimodal interaction capabilities. It aims to provide customers with reliable, efficient, unified customer service experiences, and improve the work efficiency of agents while reducing the customer service costs for enterprises. + +This Recommendation can be used to guide relevant parties in the design, development, selection, and evaluation of the construction of the advanced intelligent customer service system. + +## History\* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.67 | 2025-12-14 | 21 | 11.1002/1000/16592 | + +## Keywords + +Advanced intelligent customer service system. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2026 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|------------------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 3 | +| 6 Overview..... | 3 | +| 6.1 Introduction of AICSS..... | 3 | +| 6.2 Classification of AICSS ..... | 4 | +| 7 Design principles and framework for AICSS ..... | 5 | +| 7.1 Design principles for AICSS ..... | 5 | +| 7.2 Framework of AICSS ..... | 5 | +| 8 Requirements for AICSS ..... | 8 | +| 8.1 Functional requirements ..... | 8 | +| 8.2 Performance requirements ..... | 10 | +| 8.3 Openness requirements..... | 11 | +| 8.4 Reliability requirements ..... | 12 | +| 8.5 Security requirements ..... | 12 | +| Appendix I – Use cases of advanced intelligent customer service system ..... | 14 | +| I.1 Omni-channel and omni-media customer service scenario ..... | 14 | +| I.2 Video + digital human customer service scenario..... | 15 | +| I.3 Intelligent inspection service scenario..... | 16 | +| I.4 Intelligent agent assistance service scenario ..... | 18 | +| I.5 Intelligent outbound call scenario ..... | 19 | +| Bibliography..... | 22 | + + + +# Recommendation ITU-T F.748.67 + +## Requirements and framework for advanced intelligent customer service system + +# 1 Scope + +This Recommendation specifies the requirements and framework of the advanced intelligent customer service system (AICSS). In particular, this Recommendation covers the following aspects for advanced intelligent customer service system. + +- Overview of AICSS including the introduction and classification; +- Design principles and framework of AICSS; +- Requirements of the AICSS. + +Typical use cases of AICSS are provided in an appendix. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +- [ITU-T G.711] Recommendation ITU-T G.711 (1988), *Pulse code modulation (PCM) of voice frequencies*. +- [ITU-T G.729] Recommendation ITU-T G.729 (2012), *Coding of speech at 8 kbit/s using conjugate-structure algebraic-code-excited linear prediction (CS-ACELP)*. +- [ITU-T H.264] Recommendation ITU-T H.264 (V15) (2024), *Advanced video coding for generic audiovisual services*. +- [ITU-T H.265] Recommendation ITU-T H.265 (V11) (2026), *High efficiency video coding*. +- [ISO/IEC 11172-3] ISO/IEC 11172-3:1993, *Information technology – Coding of moving pictures and associated audio for digital storage media at up to about 1.5 Mbit/s – Part 3: Audio*. +- [ISO/IEC 13818-7] ISO/IEC 13818-7:2006, *Information technology – Generic coding of moving pictures and associated audio information Part 7: Advanced Audio Coding (AAC)*. + +# 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 application** [b-ITU-T Y.2091]: A structured set of capabilities, which provide value-added functionality supported by one or more services, which may be supported by an API interface. + +**3.1.2 artificial intelligence (AI)** [b-ITU-T F.749.13]: An interdisciplinary field, usually regarded as a branch of computer science, dealing with models and systems for the performance of functions generally associated with human intelligence, such as reasoning and learning. + +**3.1.3 automatic speech recognition (ASR)** [b-ITU-T H.625]: A system that can recognize continuous speech, often having phoneme-sized references, using lexical, syntactic, semantic, and pragmatic knowledge, and reacts appropriately (therefore having interpreted the message and found the corresponding action to be taken). + +**3.1.4 digital human** [b-ITU-T F.748.15]: A computer application that integrates the technologies of computer graphics, computer vision, intelligent speech, natural language processing. It can be used for digital content generation and human-computer interaction to help improve content production efficiency and user experience. + +**3.1.5 intelligent customer service system** [b-ITU-T F.746.16]: A customer service system based on natural language processing, speech recognition, text to speech and other artificial intelligence technologies. After converging the user's questions through single or multiple rounds of conversation, the intelligent customer service system relies on a maintained background knowledge base and interface components to realize intelligent dialogue. + +**3.1.6 knowledge base** [b-ITU-T F.746.3]: A collection of knowledge resources that consist of structured and unstructured data. The knowledge base is used to provide information to the various applications that are related to information provision such as the QA system and search system. + +**3.1.7 natural language processing** [b-ITU-T F.746.3]: A method that analyses text in natural languages through several processes such as part-of-speech recognition, syntactic analysis and semantic analysis. + +**3.1.8 text-to-speech synthesis (TTS)** [b-ITU-T P.10]: A TTS process generates a speech signal from text codes. It is usually composed of the two parts: + +- a language-dependent text processing part (the high-level processing part), which generates from the character string (by reading rules, vocabulary, and semantic analysis) a set of phonetic, prosodic, etc., parameters which are used by: + - an acoustical signal generating part, the synthesizer itself, which generates the audible speech. + +### **3.2 Terms defined in this Recommendation** + +This Recommendation defines the following term: + +**3.2.1 advanced intelligent customer service system (AICSS)**: A customer service system that integrates artificial intelligence (with foundation models as a core component), omnichannel access, omni-media processing, and multimodal interaction aims to deliver reliable, efficient, and unified customer service experiences, improve the agents' work efficiency, and reduce enterprises' customer service costs. + +# **4 Abbreviations and acronyms** + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-------|----------------------------------------------| +| ACD | Automatic Call Distribution | +| AI | Artificial Intelligence | +| AICSS | Advanced Intelligent Customer Service System | +| ASR | Automatic Speech Recognition | +| CMS | Call Monitoring Service | +| CS | Customer Service | +| CTI | Computer Telephony Integration | +| DTMF | Dual-Tone Multi-Frequency | + +| | | +|--------|--------------------------------------| +| IVR | Interactive Voice Response | +| IVVR | Interactive Voice and Video Response | +| LLM | Large Language Model | +| MGM | Media Gateway Module | +| NAT | Network Address Translation | +| NLP | Natural Language Processing | +| PII | Personally Identifiable Information | +| SBC | Session Border Controller | +| TTS | Text-to-Speech | +| VoIP | Voice over Internet Protocol | +| VoLTE | Voice over Long-Term Evolution | +| WAV | Waveform Audio File Format | +| WebRTC | Web Real-Time Communication | + +## 5 Conventions + +In this Recommendation, the following conventions apply: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted, if conformance to this Recommendation is to be claimed. +- The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. + +# 6 Overview + +### 6.1 Introduction of AICSS + +Customer service (CS), originally called call centre, originated in the 1930s and routed users' call requests to the department of customer service professionals who handle inbound calls, and manually answer calls to solve customer problems. With the natural language processing (NLP) technologies, call centres gradually become intelligent customer service system (sometimes also called contact centre) and are an important bridge for communication between enterprises and customers, widely used in telecommunication operators, banks, insurance, and securities industries. + +With the application and development of foundation model technologies, intelligent customer service system has gradually developed and entered the stage of next-generation customer service stage with the change of integrating multi-modal user interaction (not limited to voice, the video interaction gradually becomes mainstream), large language model (LLM)-based voice agents and intelligent agent assistance, HD video interactions, providing customers with seamless and consistent omnichannel communication experience. At this stage companies not only attach importance to improvement of customer service experience but also emphasize the use of artificial intelligence (AI) to enhance the work efficiency of seat staff. + +Advanced intelligent customer service system (AICSS) is a system that integrates artificial intelligence and omni media interactive communication capabilities, aiming to provide enterprises with efficient and intelligent customer service solutions. The AICSS platform is mainly composed of omni-channel accessing module, omni media processing module, routing and calling controlling module (including automatic call distribution (ACD), computer telephony integration (CTI), + +interactive voice response (IVR), interactive voice and video response (IVVR)), AI service engine module, and operation management module. The application layer is mainly composed of the most common customer service applications, such as voice customer service (CS), video CS, chat CS, digital human CS, intelligent inspection for conservations, intelligent agent assistant, voice bot, and chatbot. Nowadays the outbound call marketing function is added to AICSS and plays an increasingly important role in enterprise operation. + +### 6.2 Classification of AICSS + +#### 6.2.1 Classification based on deployment + +- Public cloud AICSS: A public cloud-based customer service system relies on the public cloud to deploy the platform and business to provide customer service cloud services, which is characterized by fast launch, low cost, and strong scalability. However, there are several disadvantages for public cloud AICSS, including high data security risks (shared infrastructure may expose theoretical attack surfaces), vendor lock-in, inability to customize infrastructure, and performance variability (latency dependent on internet quality of public cloud). +- In-house AICSS: All hardware and software of the customer service are deployed on-premises, and the infrastructure (computing, storage, and network) is provided by a private cloud deployed in the enterprise's local, which is characterized by highly secure, highly scalable, and high flexibility. +- Managed AICSS: The customer service system service provider delivers all infrastructure and software systems. Enterprises rent resources and deploy agents to launch operations rapidly, with short setup time, low upfront risk, and controlled costs. However, the operation cost is high, and both operational and service data are managed entirely by third-party, customization is limited by third-party. +- Out-sourcing AICSS: Enterprises outsource all related matters of the call centre, including equipment, personnel and management, to another customer service system operation enterprise as a whole, this model offers distinct advantages such as zero operational burden, instant scalability, and cost efficiency. However, it also entails challenges like uncontrollable service quality, data opacity, and security vulnerabilities. + +#### 6.2.2 Classification based on business type + +- Inbound AICSS: The inbound customer service system mainly aims to provide high-quality customer service, support and solve customer problems and needs, these services include customer inquiries, technical support, order processing and complaint handling account inquiries. The defining characteristic of this model is the unpredictable nature of inbound traffic, which is entirely driven by real-time customer demand. +- Outbound AICSS: The outbound customer service system is strategically oriented toward sales enhancement, revenue growth, customer loyalty cultivation, and customer data collection, achieved through proactive customer engagement by enterprises. This proactive communication model enables targeted outreach to nurture leads, strengthen client relationships, and gather actionable insights. +- Hybrid AICSS: The hybrid customer service system that can handle both inbound and outbound customer requests simultaneously, comprehensively meeting enterprises' communication needs with customers, further realizing the integration of agent resources, and improving work efficiency. + +# 7 Design principles and framework for AICSS + +### 7.1 Design principles for AICSS + +AICSS design principles are as follows: + +- **Technical advancement:** The system design is recommended to adopt forward-looking technologies (e.g., LLM, cloud computing) and integrate best practices into architectural design to maintain industry leadership over its lifecycle. +- **System stability and reliability:** The system design is recommended to ensure high availability and business continuity and be capable of running stably and reliably in the long-term. It is recommended to have high throughput, powerful processing capabilities, and error-free transmission capabilities to meet the processing requirements during peak business periods. Meanwhile, the system is required to withstand the pressure brought by various abnormal situations and ensure uninterrupted $7 \times 24$ -hour operation. For a large-sized AICSS system with the human agent capacity more than 1 000, a reliability standard of 99.999% is recommended to be considered to guarantee the stable and reliable operation of the system under high call-access volumes. For evidence and references, please see clause 8.4. +- **System security:** The system design is recommended to ensure the security of the whole system across multiple aspects, including information database, management system, and data transmission. Build systems on mature and stable hardware and software. The system is recommended to implement comprehensive backup and recovery plans, strict security controls, standardized operation management, and swift fault-response measures to safeguard overall system security and stable operation. +- **System scalability:** The system design is recommended to offer excellent scalability in functionality, business, and capacity, supporting seamless component-level expansion, adapting to increased user traffic without architecture changes, and allows flexible addition of new features. The platform is recommended to use mainstream virtualization and containerization architectures, enabling distributed microservice deployment. Each microservice can be deployed independently, communicating via lightweight protocols. Hardware requirements are device-agnostic, leveraging generic cloud services or virtual machines for scalable capacity. +- **System openness:** The system design is recommended to support open APIs for the platform, enabling enterprise to perform secondary applications development on top of it. The system is required to provide diverse business scenario APIs and open-source tools to reduce the development cost of upper-layer applications, enhance development efficiency, support integration and verification with multiple third-party systems, and shorten the time-to-market for new services. +- **System maintainability:** All system elements are recommended to support remote, centralized management with a visual topology map for intuitive system-wide monitoring. This enables unified resource scheduling, rapid issue detection/resolution, and proactive maintenance through comprehensive performance analytics, ensuring optimal system stability, security, and efficiency. + +### 7.2 Framework of AICSS + +AICSS is composed of an infrastructure layer, a platform layer, and an application layer. These three layers realize different functions: + +- **Infrastructure layer:** this layer includes network, computing, and storage, providing operational support for upper-layer platforms and applications. + +- Platform layer: this layer realizes the core functional module for the AICSS system, including omni-channel accessing, omni media processing, routing and calling controlling, AI service engine, the customer service processing monitoring, inspections and operations, as well as the centralized interfaces enabling the upper applications developing. +- Application layer: the layer includes the most common customer services, such as video/voice agent, agent copilot, workorder management, knowledge manager copilot, operation analysis insight, and intelligent inspection (refer to clause I.3). + +NOTE 1 – Agent copilot is an AI-powered assistant, powered by large language models (LLM), that supports human agents in AICSS by providing real-time intent recognition, sentiment analysis, response words recommendations, and automated post-call summarization to improve efficiency and reduce handling time. + +NOTE 2 – Knowledge manager copilot is an AI-powered tool that automates the extraction and generation of Q&A pairs from multi-format and multi-lingual business documents, streamlining the creation and maintenance of knowledge bases for AICSS. + +Other support systems are shown in Figure 7-1 including knowledge base, the CRM system, product system, and billing system. + +![Figure 7-1 – Framework of AICSS. This is a layered architecture diagram showing the flow from Customer to Infrastructure layer. The layers are: Customer (top), Omni channel (Voice/SMS/Video/SMS/Fax and H5/Web/WebRTC/VoIP/Email/APP), Network, Application layer (Video/Voice agent, Agent copilot, Workorder management, Knowledge manager copilot, Operation analysis insight, Intelligent inspection, ...), Platform layer (Open API, Media access module, Real-time audio and video service, Instant message service, Call and routing control module, Operation management module, AI capability service), and Infrastructure layer (Storage, Computing, Network). External systems on the right include Human agent, Knowledge base, CRM, and Products and billing. A legend at the bottom defines the box types: solid for AICSS components, dashed for user access channels, and dash-dot for external system components.](1439cb942d9e363bbb3161b5540dd8c6_img.jpg) + +Legend: + +- A component of AICSS +- ⋯ User and the access channel from user +- ⋯ A component from other systems + +F.748.67(25) + +Figure 7-1 – Framework of AICSS. This is a layered architecture diagram showing the flow from Customer to Infrastructure layer. The layers are: Customer (top), Omni channel (Voice/SMS/Video/SMS/Fax and H5/Web/WebRTC/VoIP/Email/APP), Network, Application layer (Video/Voice agent, Agent copilot, Workorder management, Knowledge manager copilot, Operation analysis insight, Intelligent inspection, ...), Platform layer (Open API, Media access module, Real-time audio and video service, Instant message service, Call and routing control module, Operation management module, AI capability service), and Infrastructure layer (Storage, Computing, Network). External systems on the right include Human agent, Knowledge base, CRM, and Products and billing. A legend at the bottom defines the box types: solid for AICSS components, dashed for user access channels, and dash-dot for external system components. + +**Figure 7-1 – Framework of AICSS** + +Figure 7-2 shows the composition architecture for AICC platform layer. + +![Figure 7-2 – Function composition of AICSS platform layer. The diagram shows a hierarchical structure of the AICSS platform. At the top is the 'Open API' layer with 9 components: Outbound call, IVR/IVVR orchestration, Call handling API, Video/Audio API, Management API, CMS API, AI service API, Calling report API, and File operation API. Below this is the 'Media access service module' with 7 components: SBC, Social media gateway, 5G SMS gateway, Web gateway, MGM gateway, Email gateway, and WebRTC gateway. Below that are two modules: 'Call and routing control module' (CTI, IVR/IVVR, CMS, ACD) and 'Real-time audio and video service module' (Audio/Video conference, Audio/Video recording, Signalling processing unit, Whiteboard, Desktop sharing). Below these are 'Operation and management module' (Operation and management toolset, Configuration toolset, Deployment toolset) and 'AI capability service module' (ASR, NLP, TTS, ML, LLM, MLM, Digital human). At the bottom right is the 'Instant message service module' (Conversation management, Group management, Message transmission, Conversation status).](af7916c89a458fdab6c3f443217388ae_img.jpg) + +Figure 7-2 – Function composition of AICSS platform layer. The diagram shows a hierarchical structure of the AICSS platform. At the top is the 'Open API' layer with 9 components: Outbound call, IVR/IVVR orchestration, Call handling API, Video/Audio API, Management API, CMS API, AI service API, Calling report API, and File operation API. Below this is the 'Media access service module' with 7 components: SBC, Social media gateway, 5G SMS gateway, Web gateway, MGM gateway, Email gateway, and WebRTC gateway. Below that are two modules: 'Call and routing control module' (CTI, IVR/IVVR, CMS, ACD) and 'Real-time audio and video service module' (Audio/Video conference, Audio/Video recording, Signalling processing unit, Whiteboard, Desktop sharing). Below these are 'Operation and management module' (Operation and management toolset, Configuration toolset, Deployment toolset) and 'AI capability service module' (ASR, NLP, TTS, ML, LLM, MLM, Digital human). At the bottom right is the 'Instant message service module' (Conversation management, Group management, Message transmission, Conversation status). + +F.748.67(25) + +**Figure 7-2 – Function composition of AICSS platform layer** + +The AICSS platform components are described as follows: + +**Media access service module:** Enables audio, video, and multimedia access, and supports unified broadband and narrowband access. + +- a) Media gateway module (MGM): A media gateway connects with carrier E1 lines to enable voice and signalling access, converts them into SIP protocol, and interfaces with the soft switch. +- b) WebRTC gateway: Used for web real-time communication (WebRTC) audio/video access in H5 mini-programs or webpages. +- c) Social media gateway: Docks with social media channels. +- d) Email gateway: Implements email filtering, monitoring, and encryption management to improve the management efficiency of the customer service email system. +- e) Session border controller (SBC): Provides network isolation capabilities superior to firewalls and achieves network address translation (NAT) traversal through a full-proxy approach. +- f) IMT-2020 message gateway: Used for IMT-2020 messaging access and adaptation. + +**Call and routing control module:** Provide omnichannel call routing, assignment, and scheduling capabilities for voice, video, and digital media, supporting both inbound and outbound business processes across all channels. Main subcomponents are as follows: + +- a) CTI: A core telephony control middleware in AICSS solutions, supporting unified routing across multiple customer channels and enabling telephony functions such as call management, voice playback, recording, dual-tone multi-frequency (DTMF) collection, call transfer, conference calls, text-to-speech (TTS), and automatic speech recognition (ASR). +- b) IVR: A self-service voice platform. By integrating TTS, ASR, semantic understanding, and multi-turn dialogue engines, IVR provides voice bot and chatbot capabilities to enable intelligent voice interactions with customers. +- c) IVVR: An interactive voice and video application technology that allows users to listen to voice explanations from customer service agents while watching instructional or operational videos. + +- d) **Call monitoring service (CMS):** A real-time monitoring service for AICSS that tracks call statuses and provides real-time monitoring APIs for upper-layer applications to develop AICSS monitoring applications. +- e) **ACD:** An auto calling distribution components enables audio and video media access and supports unified broadband and narrowband integration. + +**Realtime audio and video service module:** Provide audio processing, video processing, and multimedia processing with capabilities including video/voice recording, voice over long-term evolution / voice over Internet protocol (VoLTE/VoIP) video access and processing, desktop sharing and three-way video conference, and whiteboard. + +**Operation and management module:** Provide operation and management tools including O&M tools, configuration tools, and deployment tools for AICSS operation. + +**Instant message service module:** Provide multimedia text dialogue (including text, images, emojis, voice, and video). + +**AI capability service module:** An AI platform component that provides engines including NLP, ASR, TTS, large language model (LMM), and multimodal large model (MLM), and offers AI capability interfaces for upper-layer intelligent applications to support their development. + +**Open API:** The functional interfaces provided by the platform to upper-layer applications are used to support the development of enterprise-level AICSS business applications, including call handling, IVR orchestration, CMS, management, outbound call, AI service, real video and audio service, calling report and file operation interfaces. For the detailed introduction for APIs, refer to clause 8.3. + +# **8 Requirements for AICSS** + +### **8.1 Functional requirements** + +#### **8.1.1 Self-service voice function requirements** + +SVR-001: AICSS is required to support audio encoding format including [ITU-T G.711] A-law, [ITU-T G.711] u-law and Annex A of [ITU-T G.729]. + +SVR-002: AICSS is required to support TTS play. + +SVR-003: AICSS is recommended to support playing combined voice prompts to users, including phone numbers, numerical values, prices, dates, and times. + +SVR-004: AICSS is required to support for audio playback control, including the ability to immediately stop the current voice playback. + +SVR-005: AICSS is recommended to support multi-language functions. + +#### **8.1.2 Routing and queuing function requirements** + +RQR-001: AICSS is required to support transferring the customer to matching idle human agents based on the customer profile information, agent status, and other information according to routing and queuing strategies. + +RQR-002: AICSS is required to support overflow routing, including routing between skill groups, cross-platform overflow routing, and cross-regional overflow routing. + +RQR-003: The system's routing rules is required to support configuration based on multiple business types, such as customer level, customer waiting time, agent idle time, and agent skills. + +RQR-004: AICSS is required to support unified routing and queuing for voice, video, multimedia, and social media from different channels. + +RQR-005: AICSS is required to support flexible configuration of routing priorities and processing strategies for different access channels to adapt to changing business needs. + +RQR-006: AICSS is recommended to support automatic outbound calling including predictive outbound, preview outbound, pre-occupy outbound and IVR outbound. + +NOTE 1 – Predictive outbound is an outbound mode that uses algorithms to predict call answer rates and dials in advance, aiming to improve the work efficiency of human agents. + +NOTE 2 – Preview outbound is an outbound mode where agents first review customer information, and the system initiates the call only after the agent confirms. + +NOTE 3 – Pre-occupy outbound is an outbound mode that occupies communication resources in advance to ensure a fast connection once the call is initiated. + +NOTE 4 – IVR outbound calls automatically initiated through an IVR system, which are applied in scenarios such as voice notifications and verification code broadcasts. + +#### **8.1.3 Human agent management requirements** + +AMR-001: AICSS is required to support agent management functions, including issuing, modifying, revoking, deleting agents and their skills, and batch operations supporting. + +AMR-002: AICSS is required to support agent skill management, including adding, deleting, modifying, querying skills, and configuring skill-based routing strategies. + +AMR-003: AICSS is required to support agent transfers, including these following functions: + +- Transferring to the automatic process for continued service after agent release. +- Transferring to the appropriate agent after digit collection is completed in the automatic process. +- Transferring to skilled queues, agent queues, or designated agents for processing. + +#### **8.1.4 Calling reports requirements** + +CRR-001: AICSS is required to support real-time reports by dimensions including agent group, agent, and routing (skill category). + +CRR-002: AICSS is required to provide real-time reports for online text conversations, audio calls, and video calls. + +CRR-003: AICSS is recommended to support real-time call statistics reports at the whole system level, including current incoming call volume, answered calls, queued calls, abandoned calls, short-abandoned calls, abandoned calls during ringing, average queue duration, and average talk (service) duration. + +CRR-004: AICSS is recommended to support configurable refresh frequency for reports, with configuration changes taking effect in real time. + +CRR-005: AICSS is recommended to support historical report queries by period (e.g., minutes, hourly, daily, monthly, and annual). + +CRR-006: AICSS is recommended to support generation of outbound traffic statistics reports. + +#### **8.1.5 Agent conversation recording requirements** + +ACR-001: Encoding formats of conversation recording is required to support [ITU-T G.711] A-law, [ITU-T G.711] u-law and [ITU-T G.729] Annex A. + +ACR-002: File storage formats of recording are required to support at least one of waveform audio file format (WAV), [ISO/IEC 11172-3]/MP3 and [ISO/IEC 13818-7]/AAC. + +NOTE – WAV is an audio file format standard for storing an audio bitstream on personal computers, and it is an application of the resource interchange file format (RIFF) bitstream format method for storing data in chunks. + +ACR-003: Recording files are required to support encrypted storage. + +#### 8.1.6 AI capabilities requirements + +AIR-001: The AICSS is required to have integration capabilities with intelligent engines, including third-party intelligent components such as ASR, TTS, and NLP. + +AIR-002: The AICSS is recommended to support intelligent functions including virtual agent (e.g., voice bot, chatbot), conversation automatic summary, conversation script recommending, customer data analytics, intelligent work order filling, digital human customer service, intelligent quality inspection, and intelligent FAQ extraction. + +NOTE 1 – Conversation automatic summary is an intelligent function that generates work order summaries based on agent-customer communication content, reviewed and confirmed by human agents, providing a convenient operation experience. + +NOTE 2 – Intelligent work order filling is an intelligent function that enables automatically assisted filling of work orders, supports work order dispatch tracking, and automatically recommends work order types, and generating work order summary. + +#### 8.1.7 Requirements of omnichannel access and multimedia interaction + +OMR-001: AICSS is required to support multiple channel accesses, such as mobile websites, apps, mobile messages, web portals, VoLTE video access, telephony, and email. + +OMR-002: AICSS is required to support multimedia access, including audio, video, IVR, and IVVR. + +OMR-003: AICSS is required to support rich media services (e.g., text, images, emojis, voice and video). + +OMR-004: AICSS is required to support audio-video services including VoIP, VoLTE, WebRTC and other audio-video calling capabilities, multi-party collaborative video calls, and seamless switching between audio and video calls. + +OMR-005: AICSS is recommended to support one-way video outbound calling, with the ability to switch to two-way video with customer consent without call interruption. + +OMR-006: AICSS is recommended to support remote collaboration including video interaction, desktop sharing, whiteboarding, document sharing, and customer online signature capture. + +OMR-007: AICSS is required to ensure uniform processing and response experiences across all access channels. + +OMR-008: AICSS is required to enable real-time synchronization of customer conversation records across multi-channel interactions, ensuring all service personnel/system modules access the latest session history. + +OMR-009: AICSS is required to support at least one of the video coding formats: [ITU-T H.264] and [ITU-T H.265]. + +OMR-010: AICSS is required to support handover from video IVVR to human agent calls. + +OMR-011: AICSS is recommended to support visual IVVR workflow development, providing rich controls, algorithms, and data structures to support complex IVVR video self-service flows and custom routing logic. + +### 8.2 Performance requirements + +PR-001: Maximum capacity performance metrics: for the maximum concurrent call volume of AICSS, it is recommended to follow Table 8-1's calculation method, if the human agents capacity is no larger than 1 000, the maximum capacity (concurrent call volume) is required to be no less than $2 \times$ human agents capacity. If the human agents capacity number is larger than 1 000 and no larger than 2 000, the maximum capacity is required to be no less than 4 000 concurrent calls, if the human + +agents capacity number is larger than 2 000, the CTI/IVR maximum capacity is required to be no less than the maximum concurrent calls capacity, where $\text{maximum} = (R_v \times N_v + R_d \times N_d) + B$ (for detail definition, please refer to Table 8-1 below). + +**Table 8-1 – Maximum capacity performance metrics** + +| Human agents capacity | AICSS maximum capacity (concurrent call volume) | +|----------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Capacity $\leq 1\ 000$ | $\geq 2 \times \text{Capacity of human agents}$ | +| $1\ 000 < \text{Capacity} \leq 2\ 000$ | $\geq 4\ 000$ | +| Capacity $> 2\ 000$ | $\geq (R_v \times N_v + R_d \times N_d) + B$
$R_v$ – Voice agent concurrency ratio, with a recommended value range of 3 to 4;
$N_v$ – Number of voice human agents;
$R_d$ – Digital agent concurrency ratio, with a recommended value range of 1.2 to 1.8;
$N_d$ – Number of digital channel human agents;
$B$ – Redundancy and peak buffer, recommended to be 15% to 20% of the calculated value. | + +NOTE 1 – For AICSS with fewer than 1 000 human agents, the CTI/IVR must support concurrent calls exceedingly twice the agent count. This requirement aligns with the Erlang C model – the industry standard for customer service capacity planning – which scientifically determines resources needed to achieve the basic customer service levels (e.g., 80% of calls answered within 20 seconds). The model inherently mandates that concurrent capacity significantly surpasses agent numbers to accommodate peak queueing and IVR loads, with a $2 \times$ multiplier representing a fundamental baseline. For over 1 000 agents, systems must handle larger traffic bursts with higher availability. The 4 000 concurrent call requirement provides a 2.67:1 ratio for 1 500-agent systems, ensuring strong redundancy and burst capacity. + +NOTE 2 – For AICSS with 2 000 + human agents, the Erlang model must be expanded to omnichannel operations (such as email, chat and social media), and according to the omnichannel mixed traffic model (the scientific foundation and core basis for capacity planning and technical architecture design in modern, large-scale customer service), A maximum capacity value match $(R_v \times N_v + R_d \times N_d) + B$ – concurrency capacity calibrated for mixed workloads – aligns with Gartner/DMG industry benchmarks and leading vendor specifications. + +### 8.3 Openness requirements + +OR-001: The platform is recommended to provide configuration interfaces enabling secondary developers to create, query, modify, and delete customer service system resources (such as skills, human agent IDs and IVR workflows). + +OR-002: The platform is recommended to provide call connection interfaces enabling secondary developers to create customer service system agent workbenches, handle audio, video and multimedia call connections for agents, controlling agents' statuses. + +OR-003: The platform is recommended to provide IVR orchestration interfaces enabling secondary developers to design and implement IVR workflows. + +OR-004: The platform is recommended to provide management and reporting interfaces enabling secondary developers to develop applications enabling visual management and statistical analysis of real-time resource statuses and real-time and historical call statistics data. + +OR-005: The platform is recommended to provide call detail recording file interfaces enabling secondary developers to retrieve raw call detail records from customer service systems. + +OR-006: The platform is recommended to provide monitoring capability interfaces enabling secondary developers to implement real-time and historical monitoring functions of customer service system through open APIs. + +OR-007: The platform is recommended to provide automatic outbound calling capability interfaces enabling secondary developers to implement automatic outbound business functions via open APIs. + +OR-008: The platform is recommended to provide intelligent capability interfaces enabling developers to implement business functions via open APIs, including virtual agent (voice bot, chatbot), conversation automatic summary, conversation script recommending, intent recognition, emotion recognition, customer data analytics, intelligent work order filling, digital human customer service, intelligent quality inspection and intelligent FAQ extraction. + +OR-009: The platform is recommended to provide monitoring interfaces enabling secondary developers to implement real-time and historical monitoring functions of customer service systems. + +### 8.4 Reliability requirements + +RR-001: The AICSS is recommended to achieve a minimum overall availability of 99.99%. + +NOTE 1 – The 99.99% availability requirement represents the enterprise-grade benchmark for critical business systems which is specified in clause 8.7.1, *Service availability management*, of [b-ISO/IEC 20000-1]. + +RR-002: If human agents capacity exceeds 1 000 and the system handles real-time financial transactions, emergency response (such as government hotlines), or daily consultation volumes of no less than 500 000 cases, the system reliability rate is recommended to meet 99.999% availability. + +NOTE 2 – According to the business impact analysis (BIA) requirements specified in clause 8.2.2 of [b-ISO 22301], this customer service system described in RR-002 has been identified as a critical business process with a very short maximum tolerable period of disruption (MTPD). To meet this business requirement, and in adherence to the service availability management processes outlined in [b-ISO/IEC 20000-1] and the 'carrier-grade' service best practices (similar definition can be found according to [b-ITU-T E.800]), the architectural design target for this system is set at 99.999% availability. This target ensures alignment between the technical design and business continuity requirements. + +### 8.5 Security requirements + +#### 8.5.1 Data encryption requirements + +DR-001: It is required not to print communication contents and sensitive information by any services within the system. + +DR-002: Authentication credentials and keys is required not be stored in plaintext in the system. + +DR-003: It is required to encrypt authentication credentials and keys before storage. + +DR-004: It is required not to use proprietary cryptographic algorithms, non-standard cryptographic algorithms and known insecure cryptographic algorithms. + +DR-005: It is required not to print passwords and communication contents in any logs or call detail records is prohibited. + +DR-006: It is required to support TLS/HTTPS access for channels transmitting personally identifiable information (PII) and highly sensitive personal data. + +DR-007: It is required to support TLS encryption for signalling and SRTP encryption for media transmission to ensure secure data communication. + +#### 8.5.2 Access control and identity authentication security + +AUR-001: It is required to establish secure access control to prevent unauthorized access by anonymous users. + +AUR-002: It is required to prohibit directory listing browsing to prevent site information leakage and unauthorized file downloads. + +AUR-003: It is required to verify user permissions for important operations to prevent horizontal and vertical privilege escalation. + +AUR-004: It is required to support identity authentication mechanisms to manage and maintain users and platform devices, ensuring the security of their usage and access. + +AUR-005: It is required to support decentralized and zonal management, achieving identity recognition and authentication through unique identity identifiers and account passwords, and enabling different permissions for different identities. + +#### **8.5.3 Integrity security requirements** + +IR-001: It is required to support least privilege principle for all processes, least privilege principle for all user accounts, and network traffic isolation capabilities. + +#### **8.5.4 Log and audit** + +LA-001: It is required to support logging and auditing functions, and log key operations and PII data operations for security auditing. + +#### **8.5.5 Deployment security** + +DS-001: It is required to support secure deployment capabilities and support the division of security domains and networks for each module of the system to ensure network security after deployment. + +## Appendix I + +### Use cases of advanced intelligent customer service system + +(This appendix does not form an integral part of this draft Recommendation.) + +### I.1 Omni-channel and omni-media customer service scenario + +Figure I.1 shows the omni-channel and omni-media customer service scenario. In this scenario, customers can access the advanced intelligent customer service system (AICSS) through multiple channels and media forms, such as telephone, email, social media, online service portals on official websites, and video customer service. This multi-channel access allows customers to choose the most suitable communication method according to their preferences and convenience, improving the flexibility and convenience of customer service. + +![Diagram of the omni-channel and omni-media customer service scenario. On the left, a dashed box contains icons for 'Hard phone', 'SMS', 'APP', 'Video', 'Email', and 'Other'. An arrow points to a second dashed box containing 'Voice', 'Video', and 'Multimedia' (text, image, form). A 'Unified integrated access' icon connects this to a vertical 'Service engine' bar. From the engine, a 'Manual service' path leads to a red 'Intelligent routing' bar, which then points to an 'All-media unified work platform' containing 'Phone', 'Online', and 'Video' options. An 'Intelligent service' path from the engine leads to a robot icon. The label 'F.748.67(25)' is at the bottom right.](3da1a07cb87051bf616c9876db958cf0_img.jpg) + +The diagram illustrates the omni-channel and omni-media customer service scenario. It shows a flow from various customer channels (Hard phone, SMS, APP, Video, Email, Other) through a unified integrated access point to a Service engine. The Service engine then routes to either a Manual service (via Intelligent routing to Phone, Online, Video) or an Intelligent service (via a robot icon). The label F.748.67(25) is present at the bottom right. + +Diagram of the omni-channel and omni-media customer service scenario. On the left, a dashed box contains icons for 'Hard phone', 'SMS', 'APP', 'Video', 'Email', and 'Other'. An arrow points to a second dashed box containing 'Voice', 'Video', and 'Multimedia' (text, image, form). A 'Unified integrated access' icon connects this to a vertical 'Service engine' bar. From the engine, a 'Manual service' path leads to a red 'Intelligent routing' bar, which then points to an 'All-media unified work platform' containing 'Phone', 'Online', and 'Video' options. An 'Intelligent service' path from the engine leads to a robot icon. The label 'F.748.67(25)' is at the bottom right. + +**Figure I.1 – Omni-channel and omni-media customer service scenario** + +When customers access AICSS through these channels, the enterprise can first provide intelligent robot services to solve standardized consultations or provide personalized consultations based on customer profile information to respond to customer needs as quickly as possible, improve customer satisfaction, and relieve the service pressure of manual agents during peak periods. + +If the intelligent robot fails to meet customer needs, manual service support is available. At this time, AICSS can quickly and accurately locate the customer's consultation intent based on the customer profiles, customer journey information, selected service types, and historical interaction records (such as human-machine interactions, work orders, and outbound calls), assign suitable skilled agents to customers, and provide the optimal path for business processing. + +The main functions of the omni-channel multimedia customer service are as follows: + +- a) **Unified access for all channels:** Supporting unified access to channels including phone, SMS, email, and other social media applications and website, leveraging the complementary strengths of each channel to form a service matrix. +- b) **Integrated work platform:** Providing a multi-dimensional agent workspace that integrates various service modes such as voice, online chat, and video services, enabling seamless operation without switching between systems. +- c) **Enterprise-wide service support:** Enabling unified access for customer service representatives, marketing staff, back-office personnel, etc., offering an enterprise-wide service solution. + +- d) Unified data management: Implements traceable records of customer service interactions across all channels and centrally manages data generated during business processes. +- e) Intelligent application support: Supporting intelligent features for customer-facing services as needed for business development, such as intelligent voice navigation and AI chatbots. + +The functional requirements generated from the omni-channel and omni-media customer service are as follows: + +- a) Multi-channel integration support: Capable of omni-media customer engagement (such as telephone, video, APP, SMS, email and H5); integrating robot and human agent resources, and supporting API-based open access to various business lines. +- b) NLP, ASR, and TTS capabilities: Supporting speech recognition, semantic understanding, multi-turn conversations, and the ability to identify specific instructions and business keywords; with multi-language support. +- c) Knowledge base construction capability: Supporting knowledge base management, construction, and updates. +- d) Multimedia unified routing capability: Supporting omni-channel resource integration and analysis, as well as the construction and optimization of multimedia routing strategies. + +### I.2 Video + digital human customer service scenario + +Figure I.2 shows the video + digital human customer service scenario. This video + digital human customer service is built on IMT-2020/VoLTE technology. When customers call the enterprise's hotline, they can watch digital human demonstrations or promotional videos before connection. After connection, the traditional IVR is upgraded to an interactive voice and video response (IVVR) interface. Customers can tap virtual menus on their mobile phones to trigger services. For example, changing the service password requires identity verification via a virtual keyboard first. Once verified, customers can interact with digital humans via voice, which will provide visual broadcasts accompanied by text, enhancing the user experience. + +Digital humans are connected to the business knowledge base and embedded in various service channels in a unified H5 format to meet diverse scenario needs. When digital customer service fails to resolve issues, calls are automatically transferred to human agents. Video customer service enables real-time observation of customers' expressions and tone, allowing for optimized service delivery and more reliable identity verification. It also extends services to the elderly and customers in remote areas, reducing the burden on physical branches. + +![Diagram of the video + digital human customer service scenario. The flow starts with 'Customers call XXX hotline', leading to 'Play video before connection' and 'Enter the video interactive self-service after connecting'. A 'Voice interaction' loop connects the customer to a 'Digital human' (using NLP, NLU, ASR, etc.). The digital human can 'Initiate real customer service to take over when the virtual digital person cannot solve the problem', transferring to a 'Human' agent. The human agent uses 'Voice driven, using timbre conversion capabilities' to interact with the customer.](4cc7cdce3d498d8b0ba033a9be24ade5_img.jpg) + +The diagram illustrates the video + digital human customer service scenario. It shows a flow from a customer calling a hotline to a digital human interface, with options for video playback and self-service. The digital human uses NLP, NLU, and ASR for interaction. If the digital human cannot solve the problem, it initiates real customer service, transferring to a human agent. The human agent uses timbre conversion capabilities for voice interaction. + +``` + +graph LR + C[Customers call XXX hotline] --> V[Play video before connection] + V --> I[Enter the video interactive self-service after connecting] + I --> DH[Digital human] + DH -- "Using NLP, NLU, ASR, etc." --> I + DH -- "Initiate real customer service to take over when the virtual digital person cannot solve the problem" --> H[Human] + H -- "Voice driven, using timbre conversion capabilities" --> I + C -- "Voice interaction" --> DH + +``` + +Diagram of the video + digital human customer service scenario. The flow starts with 'Customers call XXX hotline', leading to 'Play video before connection' and 'Enter the video interactive self-service after connecting'. A 'Voice interaction' loop connects the customer to a 'Digital human' (using NLP, NLU, ASR, etc.). The digital human can 'Initiate real customer service to take over when the virtual digital person cannot solve the problem', transferring to a 'Human' agent. The human agent uses 'Voice driven, using timbre conversion capabilities' to interact with the customer. + +Figure I.2 – Video + digital human customer service scenario + +The main featured functions of the video + digital human customer service application are as follows: + +- a) It supports customers to access videos via IMT-2020/VoLTE hotlines, providing visualized voice navigation and video-digitized services. +- b) Based on core AI capabilities such as virtual digital humans, speech recognition, semantic understanding, and speech synthesis, digital human customer service provides users with real-time voice interactions similar to those of real people online. For example, it offers services like information inquiry and business handling. + +The functional requirements for the AICSS platform are as follows: + +- a) Support for multi-channel video access, including IMT-2020 messaging, video apps, VoLTE and video. +- b) Provision of high-definition video calls and conferences, with support for audio-video switching, dual-video recording, and three-party recording. +- c) Enablement of remote collaboration features, such as video interaction, desktop sharing, and digital signature. +- d) One-way video outbound calling capability, with seamless switching to two-way video upon user consent. +- e) Support for IVVR functionality and transfer to human agents. + +### **I.3 Intelligent inspection service scenario** + +Figure I.3 shows the intelligent inspection service scenario. Voice call and text conversation data between customer service system agents and customers can be automatically quality-inspected and analysed by an intelligent quality inspection application form the AICSS. This not only significantly reduces the workload of manual quality inspection and addresses the low coverage issue of manual inspection, but also transcribes all recordings and conducts full-volume quality inspection, thereby reducing business risks. Additionally, it enables retrieval and analysis based on multi-dimensional data. Building on traditional business model maintenance, intelligent quality inspection, and hot-word analysis functions, it continuously optimizes and innovates features like creating customer tags, improving multi-channel and multi-system access, expands analysis dimensions, and enriches analysis methods. It provides data integration and support for enterprises in aspects such as reducing customer complaint rates, enhancing service and marketing skills, and seizing market opportunities, helping to identify customer pain points and business patterns, and feeding back into the optimization of products and business processes to enable the customer service system to achieve greater value. + +![Figure 1.3 – Intelligent inspection service scenario. A flow diagram showing the process from Customer service centre to Call data, then to Quality inspection and analysis system, and finally to Application scenarios. The system includes modules like Risk model, Role recognition, Speech-to-text, Full coverage, Risk monitoring, Recognition engine, Quality inspection, Business analysis, Trend analysis, Comparative analysis, Cause digging, Quality inspection monitoring, Manual review, and Quality inspection model (Semantic, Speech, Rule, Process detection). Application scenarios include Risk identification and control, Service quality inspection, and Hot public opinion monitoring.](c914f51f4427bc672dd0526cfc90ebe9_img.jpg) + +The diagram illustrates the intelligent inspection service scenario, organized into four main columns: + +- Customer service centre:** Includes Incoming calls, Outbound calls, and Online customer service. +- Call data:** Includes Recording files and Conversation texts. +- Quality inspection and analysis system:** + - Top section: Risk model, Complaint early-warning, Dissatisfaction analysis, Risk monitoring, Recognition engine, Quality inspection, Business analysis, Trend analysis, Comparative analysis, Cause digging. + - Middle section: Quality inspection monitoring, Manual review. + - Bottom section: Quality inspection model, 01 Semantic detection, 02 Speech detection, 03 Rule detection, 04 Process detection. +- Application scenarios:** + - Risk identification and control:** Customer complaints and reports, Identification of major and urgent incidents, Reminders for uncivilized service. + - Service quality inspection:** 100% quality inspection coverage, Customizable quality inspection scenario models, Automatic team service quality assessment. + - Hot public opinion monitoring:** Monitoring and trend analysis of group complaints, Customizable monitoring scenarios for operations, In-depth clustering and analysis of problems in hot scenarios. + +F.748.67(25) + +Figure 1.3 – Intelligent inspection service scenario. A flow diagram showing the process from Customer service centre to Call data, then to Quality inspection and analysis system, and finally to Application scenarios. The system includes modules like Risk model, Role recognition, Speech-to-text, Full coverage, Risk monitoring, Recognition engine, Quality inspection, Business analysis, Trend analysis, Comparative analysis, Cause digging, Quality inspection monitoring, Manual review, and Quality inspection model (Semantic, Speech, Rule, Process detection). Application scenarios include Risk identification and control, Service quality inspection, and Hot public opinion monitoring. + +**Figure 1.3 – Intelligent inspection service scenario** + +In terms of agent quality inspection, the intelligent quality inspection system achieves 100% coverage, mainly in the application scenarios of risk identification and control, and service quality inspection. For call recordings, the system uses technologies such as speech recognition, and dialogue analysis to restructure quality inspection management capabilities, model and analyse quality inspection standards, and establish models for customer emotions and complaint risks. It converts all agent-customer calls into text and sets quality inspection rules based on multiple dimensions including agent keywords, taboo words, polite expressions, emotions, speaking speed, and mute duration, automatically scoring the service quality of agents. + +In terms of business operations, the main application scenario is hot-topic public opinion analysis. By applying intelligent speech analysis technology, it mines and analyses valuable information from recording files and conversation data, such as group-complaint incidents, attention to hot-topic issues, and problem clustering analysis, providing support for customer service, operations, and marketing. + +The main special functions of intelligent quality inspection are as follows: + +- Recognition engine:** Covers all voice and text data and supports speech-to-text conversion and role recognition. +- Risk monitoring:** Based on compliance and risk management requirements, build models to achieve complaint early-warning and dissatisfaction analysis. +- Business analysis:** Standardize processes, mine hotspots, analyse trends of hot words, discover new words and display business focus points. +- Automatic scoring:** The system automatically scores call recordings. +- Manual review:** Human agents can conduct secondary review of quality inspection results. The review results can be fed back to the quality inspection platform for independent model training, and can also be used for label classification to form a multi-dimensional service case library including excellent cases, warning cases, etc. +- Quality inspection rules:** Supporting not only keyword detection, but also speech rate detection, silence detection, overlapping speech detection, volume detection, emotion detection, as well as specification and process detection. + +The intelligent quality inspection has the following requirements for the capabilities of the customer service system platform: + +- a) Support for semantic recognition and keyword detection. +- b) Support for sound recognition, emotion monitoring, and intent recognition. +- c) Support for multi-scenario quality inspection, covering machine inspection, manual sampling inspection, manual review, and quality inspection appeals, to increase inspection coverage, improve inspection efficiency, and effectively reduce labour costs for quality inspection. +- d) Connectivity with multiple systems, including recording, work order, CRM, and alarm systems. +- e) Support for automatic scoring of call recordings. + +### I.4 Intelligent agent assistance service scenario + +Figure I.4 shows the intelligent agent assistance service scenario. The intelligent agent assistant, built on artificial intelligence technologies such as speech recognition, natural language understanding, and large language models, provides real-time business guidance and monitoring for agents through intelligent means like on-screen transcription, knowledge-at-hand, and risk control. This helps improve customer service skills, service levels, and operational efficiency. + +![Diagram of the intelligent agent assistance service scenario showing interactions between Human agents, Intelligent assistants, the Intelligent assistant platform, and Administrators.](4b87467ad9642943235f48f7d4b59449_img.jpg) + +The diagram illustrates the intelligent agent assistance service scenario. It features four main entities: Human agent, Intelligent assistant, Intelligent assistant platform, and Administrators, along with a Customer. The interactions are as follows: + +- Customer** interacts with the **Intelligent assistant platform** via **Online text interaction** (1) and **Phone voice interaction in progress** (1). The platform uses **ASR Real-time speech-to-text** for voice interactions. +- The **Intelligent assistant platform** interacts with the **Human agent** and **Intelligent assistant** via **Synchronize to agent interface** (2). This provides them with **Notify agents of dialogue violations**, **Knowledge retrieval recommendations**, **Case sharing recommendations**, and **Autofill/Intelligent summary**. +- Both the **Human agent** and **Intelligent assistant** provide **Human judgement of text reply** (3) and **Human judgement voice reply** (3) to the platform. +- The **Intelligent assistant platform** also interacts with **Administrators** via **Sync to administrator interface** (2), which triggers an alert (**Alert agents for violations and notify supervisors**) to the administrators. + +Diagram of the intelligent agent assistance service scenario showing interactions between Human agents, Intelligent assistants, the Intelligent assistant platform, and Administrators. + +F.748.67(25) + +**Figure I.4 – Intelligent agent assistance service scenario** + +During the voice interaction between agents and customers, the intelligent assistant can assist agents to enhance service efficiency. Through continuous monitoring and real-time transcription, on one hand, it conducts real-time analysis of conversations, providing agents with associated knowledge prompts and recommendations, and reminding agents to correct and guide in a timely manner. On the other hand, it monitors the call content in real-time, avoiding risks by intercepting prohibited words, issuing violation warnings, and notifying supervisors to intervene and manage. + +In addition, the intelligent assistant can reduce operational risks and improve full-process quality inspection and control. If it triggers the configured return visit quality inspection process rules, it will display the corresponding process and prompt the completion status of process nodes for agents. During the communication between agents and customers, when the intelligent assistant detects situations such as sensitive words, significant emotional fluctuations, or pauses exceeding the set value in real-time, the system automatically triggers an early warning for managers to supervise and + +take over. After the communication between agents and customers ends, the intelligent assistant will assist in automatically generating a work order summary and automatically filling this work order for this service, helping agents improve work efficiency. + +The characteristic functions of the intelligent agent assistance are as follows: + +- a) Conversation script recommendation: Automatically searches knowledge base points and recommends them to customer service agents and recommends appropriate scripts and response methods according to the customer identities. +- b) Data analysis: Forms a customer profile by analysing customer transaction records, consultation records, and behaviour preferences, and provides the customer profile data to agents as a reference. +- c) Process quality inspection prompt: Sets return visit quality inspection compliance rules, detects whether agents handle business in accordance with the rules, and prompts agents in real-time for process omissions or violations so that agents can correct business processes in a timely manner. +- d) Real-time monitoring and early warning: Monitors business compliance in real-time and gives timely early warnings for service quality. For example, if prohibited words are used or problems escalate, a pop-up prompt and in-station message notification will be issued. +- e) Agent call process monitoring: The real-time transcription results will be synchronously notified or displayed in a pop-up window for supervisors, who can provide assistance or intervention through the dialogue box. +- f) Intelligent work order summary: Automatically generates a work order summary based on the communication content between agents and customers, which is reviewed and confirmed by agents, providing a convenient operation experience. + +The functional requirements of the intelligent agent assistance for the AICSS platform are as follows: + +- a) Support ASR and TTS to achieve real-time speech-to-text conversion within seconds. +- b) Support knowledge retrieval and push: Be able to quickly retrieve the knowledge base based on the information input by agents and push relevant results to agents. +- c) Script guidance and recommendation: Recommend appropriate scripts and response methods according to the customer's identity. +- d) Data analysis: Analyse customer transaction records, consultation records, behaviour preferences, etc. to form customer profiles, and provide agents with customer profile data as a reference. +- e) Intelligent work order filling: Automatically achieve rapid work order filling or assist in filling, support work order dispatch tracking, and be able to automatically recommend work order types. + +### **I.5 Intelligent outbound call scenario** + +Figure I.5 shows the intelligent outbound call scenarios. The intelligent outbound calling system integrates AI technologies such as speech recognition, synthesis, and natural language understanding. It adopts multiple modes like real-time transfer to human agents and user-initiated transfer, optimizing scripts and strategies based on the business scenarios and user preferences. The system aims to create standardized, intelligent, and human-like outbound services, reducing repetitive manual work and enhancing large-scale business application efficiency. + +![Figure I.5 – Intelligent outbound call scenario. This diagram illustrates the flow of an intelligent outbound call system. On the left, a funnel-shaped model shows the hierarchy of call tasks: 'All outbound call tasks' at the top, followed by 'Intelligent voice outbound', 'Pre-occupied/Predictive outbound calls', and 'Manual outbound calls', leading to 'Unified outbound call reports'. Above this, the 'Robot - Human funnel - Style outbound call task execution model' is mentioned. The central 'Intelligent outbound call platform' contains three main components: 'Outbound call task list', 'Outbound call process design', and 'Outbound call dialing strategy'. Below the dialing strategy, a 'Robot agent' is shown with 'TTS' (Text-to-Speech), 'NLP' (Natural Language Processing), and 'ASR' (Automatic Speech Recognition) modules. The process starts with 'Managers' providing input to the platform. The platform then interacts with a 'Customer' through a three-step process: 1. 'Successfully call the customer', 2. 'IVR plays a questionnaire or notification', and 3. 'Robot conducts human-like speech interaction'. If 'Speech recognition fails', the call is transferred to a 'Human agent' via 'Request to transfer to a human agent'. The 'Human agent' can 'Obtain outbound call tasks' from the platform and 'Conduct manual outbound calls based on outbound call tasks'. The platform also provides 'Unified outbound call tasks and outbound call result reports' to the human agent. A 'Pre-occupied outbound call intelligent number screening' step is also shown. The diagram is labeled 'F.748.67(25)' in the bottom right corner.](cbdfdade780e677eb1c1aef3081ce9ef_img.jpg) + +Figure I.5 – Intelligent outbound call scenario. This diagram illustrates the flow of an intelligent outbound call system. On the left, a funnel-shaped model shows the hierarchy of call tasks: 'All outbound call tasks' at the top, followed by 'Intelligent voice outbound', 'Pre-occupied/Predictive outbound calls', and 'Manual outbound calls', leading to 'Unified outbound call reports'. Above this, the 'Robot - Human funnel - Style outbound call task execution model' is mentioned. The central 'Intelligent outbound call platform' contains three main components: 'Outbound call task list', 'Outbound call process design', and 'Outbound call dialing strategy'. Below the dialing strategy, a 'Robot agent' is shown with 'TTS' (Text-to-Speech), 'NLP' (Natural Language Processing), and 'ASR' (Automatic Speech Recognition) modules. The process starts with 'Managers' providing input to the platform. The platform then interacts with a 'Customer' through a three-step process: 1. 'Successfully call the customer', 2. 'IVR plays a questionnaire or notification', and 3. 'Robot conducts human-like speech interaction'. If 'Speech recognition fails', the call is transferred to a 'Human agent' via 'Request to transfer to a human agent'. The 'Human agent' can 'Obtain outbound call tasks' from the platform and 'Conduct manual outbound calls based on outbound call tasks'. The platform also provides 'Unified outbound call tasks and outbound call result reports' to the human agent. A 'Pre-occupied outbound call intelligent number screening' step is also shown. The diagram is labeled 'F.748.67(25)' in the bottom right corner. + +**Figure I.5 – Intelligent outbound call scenario** + +Managers pre-configure calling tasks, processes, and strategies (e.g., call times, redial rules) and upload contact lists. The system then automatically dials the customers according to the settings. The robot's voice can be pre-recorded templates or synthesized by the text-to-speech (TTS) software. After connection, the robot interacts via IVR. When encountering recognition failures, unresolved issues, or user requests, it automatically transfers the call to human agents, who can access interaction data to continue service. The system monitors the entire outbound process and records the results. + +The system uses a funnel-based execution model: robots handle initial calls, and unsuccessful contacts are passed to human agents. Human agents employ a pre-reservation mode, screening invalid numbers in advance to improve efficiency. + +Key features of the automatic outbound calling system are as follows: + +- a) Automated batch dialing: Standardized scripts for automatic task generation and execution. +- b) Intelligent voice interaction: Scenario-specific scripts enable smart human-machine dialogue. +- c) Pre-reservation outbound: Reserve idle agents before calling and transfer immediately upon connection. +- d) Predictive dialing: Estimate agent availability to initiate calls and efficiently route successful connections. +- e) Preview outbound: Suitable for complex tasks; agents review customer profile information before initiating calls and control call frequency. + +The functional requirements of the intelligent outbound call business for the AICSS platform are as follows: + +- a) Support automatic outbound dialling and multiple outbound strategies, including predictive, preview, pre-occupation, and IVR outbound call. +- b) Support the capability of IVVR outbound calls. +- c) Support speech recognition and understanding, accurately recognize customer speech and understand customer semantics through (natural language processing (NLP) technology. +- d) Support the capability of voice chatbots to enhance the interactive experience. +- e) Support intelligent routing, routing customers to the most suitable marketing representatives or agents based on their intentions. + +- f) Support data analysis, record interaction content throughout the process, analyse customer semantics to achieve accurate intention classification, and display data in multiple dimensions. + +## Bibliography + +- [b-ITU-T E.800] Recommendation ITU-T E.800 (2008), *Definitions of terms related to quality of service*. +- [b-ITU-T F.746.3] Recommendation ITU-T F.746.3 (2015), *Intelligent question answering service framework*. +- [b-ITU-T F.746.16] Recommendation ITU-T F.746.16 (2022), *Technical requirements and evaluation methods of intelligence levels of intelligent customer service systems*. +- [b-ITU-T F.748.15] Recommendation ITU-T F.748.15 (2022), *Framework and metrics for digital human application systems*. +- [b-ITU-T F.749.13] Recommendation ITU-T F.749.13 (2021), *Framework and requirements for civilian unmanned aerial vehicle flight control using artificial intelligence*. +- [b-ITU-T H.625] Recommendation ITU-T H.625 (2017), *Architecture for network-based speech-to-speech translation services*. +- [b-ITU-T P.10] Recommendation ITU-T P.10/G.100 (2017), *Vocabulary for performance, quality of service and quality of experience*. +- [b-ITU-T Y.2091] Recommendation ITU-T Y.2091 (2011), *Terms and definitions for next generation networks*. +- [b-ISO/IEC 20000-1] ISO/IEC 20000-1:2018, *Information technology – Service management – Part 1: Service management system requirements*. +<> +- [b-ISO 22301] ISO 22301:2019, *Security and resilience – Business continuity management systems – Requirements*. +<> + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.748.72-202512-I_PDF-E/raw.md b/marked/F/T-REC-F.748.72-202512-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..e688d7f3d918919e2758140dbf9d01daa44f0996 --- /dev/null +++ b/marked/F/T-REC-F.748.72-202512-I_PDF-E/raw.md @@ -0,0 +1,404 @@ + + +# Recommendation **ITU-T F.748.72 (12/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +## **Requirements and framework of multimodal generative artificial intelligence (AI)-enabled multi-view transformation** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.748.72 + +# Requirements and framework of multimodal generative artificial intelligence (AI)-enabled multi-view transformation + +## Summary + +Multimodal generative artificial intelligence (AI) can help to deal with different modalities (e.g., images, videos, laser points, map) and solve the problems of traditional AI-based multi-view transformation (MVT) systems, such as lack of capability to produce obstructed or blocked views and information, and deficient in prior information which is necessary to comprehend spatial relations. Recommendation ITU-T F.748.72 specifies the framework, use cases and requirements of a multimodal generative AI-enabled multi-view transformation (MEMVT) system, including information compensation, multi-view fusion and completion, multimodal temporal fusion and application functions. This Recommendation is used to augment the generation capability of multi-view transformation systems, thereby enhancing its precision and robustness, and improving the transformation quality and application effectiveness. + +## History\* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.748.72 | 2025-12-14 | 21 | 11.1002/1000/16595 | + +## Keywords + +Framework, multi-view transformation, multimodal generative AI. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2026 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|--------------------------------------------------------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 1 | +| 5 Conventions ..... | 2 | +| 6 Introduction..... | 2 | +| 7 Functional framework..... | 3 | +| 8 Functional requirements of MEMVT ..... | 4 | +| 8.1 Information compensation..... | 4 | +| 8.2 Multi-view fusion and completion ..... | 4 | +| 8.3 Multimodal temporal fusion ..... | 4 | +| 8.4 Tracking and decision-making assistant..... | 5 | +| 8.5 Visualization..... | 5 | +| 8.6 Prediction..... | 5 | +| 8.7 Target analysis enhancement..... | 5 | +| 8.8 Simulation scenario library generation..... | 5 | +| 8.9 Real-time vehicular collaborative control optimization ..... | 5 | +| 8.10 Evaluation..... | 6 | +| 8.11 Management and services..... | 6 | +| Appendix I – Use cases of multimodal generative AI enabled multi-view transformation..... | 7 | +| I.1 MEMVT system for port cargo tracking ..... | 7 | +| I.2 MEMVT system for traffic awareness ..... | 7 | +| I.3 MEMVT system for personal safety distance measurement during the inspection of a substation power system ..... | 8 | +| I.4 MEMVT system for simulation scenario library generation..... | 8 | +| I.5 MEMVT system for real-time vehicular collaborative control optimization..... | 9 | +| Bibliography..... | 10 | + + + +# Recommendation ITU-T F.748.72 + +# Requirements and framework of multimodal generative artificial intelligence (AI)-enabled multi-view transformation + +## 1 Scope + +This Recommendation specifies requirements and framework of a multimodal generative AI enabled multi-view transformation (MEMVT) system, including: + +- Framework of the MEMVT system. +- Requirements of the MEMVT system, including information compensation, multi-view fusion and completion, multimodal temporal fusion, visualization, prediction, tracking and decision-making assistant, target analysis enhancement, simulation scenario library, real time vehicular collaborative control optimization, evaluation, management and service. + +Use cases of the MEMVT system are provided in Appendix I. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following term defined elsewhere: + +**3.1.1 artificial intelligence** [b-ITU-T F.749.13]: An interdisciplinary field, usually regarded as a branch of computer science, dealing with models and systems for the performance of functions generally associated with human intelligence, such as reasoning and learning. + +### 3.2 Terms defined in this Recommendation + +None. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-------|--------------------------------------| +| 3DGS | Three-Dimensional Gaussian Splashing | +| AI | Artificial Intelligence | +| BEV | Birds' eye view | +| CNN | Convolutional Neural Network | +| LiDAR | Light Laser Detection and Ranging | +| LLM | Large Language Model | + +MEMVT Multimodal Generative AI Enabled Multi-view Transformation + +MoE Mixture of Experts + +RADAR Radio Detection and Ranging + +## 5 Conventions + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement needs not be present to claim conformance. +- The keywords "can optionally" indicate an optional requirement which is permissible, without implying any sense of being recommended. These terms are not intended to imply that the vender's implementation must provide the option, and the feature can be optionally enabled by the network operator/service provider. Rather, it means that the vender may optionally provide the feature and still claim conformance with the specification. + +## 6 Introduction + +Multi-view transformation system based on traditional AI (e.g., using convolutional neural network (CNN) architecture) is widely used in the real world. For example, Figure 1 shows a roadside bird's eye view (BEV) transformation system by using CNN, converting multimodal information (e.g., camera and light laser detection and ranging (LiDAR) information from roadside sensors) to BEV space for spatial perception and understanding. However, certain limitations persist as outlined below: + +- Lack of capability to produce obstructed or blocked views and information. This includes instances of partially obstructed objects, missing segments due to sparse sensing, etc. +- Deficient in prior information necessary to comprehend spatial relations. This deficit results in limited perception, accuracy and robustness. + +These limitations will cause potential safety hazards when used in applications such as cooperative collision avoidance, as the approaching vehicle can be dismissed by road-side detectors due to obstruction. These limitations will also cause significant efficiency drop when used in applications such as port cargo planning, as occupancy may be misjudged due to sight-block. + +![Figure 1: Example of traditional multi-view transformation system (CNN architecture). The diagram shows three parallel input streams, each consisting of a camera image and a LiDAR point cloud. Each stream is processed by a CNN to produce a 'Results' output. These three 'Results' outputs are then combined in a 'Cross-view late fusion' block, which produces a final Bird's Eye View (BEV) image. The BEV image shows a road with vehicles and a red line indicating a detected boundary. The text 'F.748.72(25)' is visible at the bottom right of the BEV image.](731f533b0599c8e42a063f06e4332045_img.jpg) + +``` +graph LR; subgraph Stream1; Input1[Camera + LiDAR] --> CNN1[CNNs] --> Result1[Results]; end; subgraph Stream2; Input2[Camera + LiDAR] --> CNN2[CNNs] --> Result2[Results]; end; subgraph Stream3; Input3[Camera + LiDAR] --> CNN3[CNNs] --> Result3[Results]; end; Result1 --> Fusion[Cross-view late fusion]; Result2 --> Fusion; Result3 --> Fusion; Fusion --> BEV[BEV Image]; +``` + +Figure 1: Example of traditional multi-view transformation system (CNN architecture). The diagram shows three parallel input streams, each consisting of a camera image and a LiDAR point cloud. Each stream is processed by a CNN to produce a 'Results' output. These three 'Results' outputs are then combined in a 'Cross-view late fusion' block, which produces a final Bird's Eye View (BEV) image. The BEV image shows a road with vehicles and a red line indicating a detected boundary. The text 'F.748.72(25)' is visible at the bottom right of the BEV image. + +**Figure 1 – Example of traditional multi-view transformation system (CNN architecture)** + +Multimodal generative AI technology can generate unknown multimodal information with the input of massive multimodal information from multiple sensors, such as image, video, laser points, map, etc. It can help to solve the above problems of traditional CNN based MVT system, that is, the limited view transformation range and insufficient information completion ability. Moreover, multimodal + +generative AI enabled multi-view transformation (MEMVT) can deal with different modalities such as images, videos, laser point clouds, etc. + +By pre-training with massive information derived from various sensing data, multimodal generative AI can transfer or fuse the original sensing angle-of-view into a more desired view. For example, from horizontal view to bird's eye view. Additionally, it can support multi-view information complementarity in complex situations such as spatial occlusion, object overlaps, etc. Therefore, the MEMVT system can solve these obstruction problems and make the detection more accurate and robust. + +## 7 Functional framework + +Figure 2 illustrates the framework of a multimodal generative AI-enabled multi-view transformation (MEMVT) system. The MEMVT system generates specific multimodal view output with input of single-view sources. The types of single-view sources are image, video, LiDAR and radio detection and ranging (RADAR) points and map. The output of the MEMVT system is a multimodal view including information of the image, video, laser points and map. + +![Figure 2 – Framework of the MEMVT system. The diagram shows the MEMVT system as a large box containing three main functional blocks: Information compensation, Multi-view fusion and completion, and Multimodal temporal fusion. Below these, a row of six boxes lists specific application functions: Tracking and decision-making assistant, Visualization, Prediction, Target analysis enhancement, Simulation scenario library, and Real time vehicular collaborative control optimization. At the bottom, two more boxes represent Evaluation and Management and service. The system takes 'Single-view sources (Image, video, laser points, map, etc.)' as input and produces a 'Specific view output (Multimodal)' as output. A small code 'F.748.72(25)' is in the bottom right corner.](7a0db9703b68b3d06cdaeefc084c0006_img.jpg) + +| | | | | | | +|----------------------------------------|---------------|----------------------------------|-----------------------------|-----------------------------|--------------------------------------------------------| +| MEMVT system | | | | | | +| Information compensation | | Multi-view fusion and completion | | Multimodal temporal fusion | | +| Tracking and decision-making assistant | Visualization | Prediction | Target analysis enhancement | Simulation scenario library | Real time vehicular collaborative control optimization | +| Evaluation | | | Management and service | | | + +Single-view sources (Image, video, laser points, map, etc.) → MEMVT system → Specific view output (Multimodal) + +F.748.72(25) + +Figure 2 – Framework of the MEMVT system. The diagram shows the MEMVT system as a large box containing three main functional blocks: Information compensation, Multi-view fusion and completion, and Multimodal temporal fusion. Below these, a row of six boxes lists specific application functions: Tracking and decision-making assistant, Visualization, Prediction, Target analysis enhancement, Simulation scenario library, and Real time vehicular collaborative control optimization. At the bottom, two more boxes represent Evaluation and Management and service. The system takes 'Single-view sources (Image, video, laser points, map, etc.)' as input and produces a 'Specific view output (Multimodal)' as output. A small code 'F.748.72(25)' is in the bottom right corner. + +**Figure 2 – Framework of the MEMVT system** + +All the functions supported by the MEMVT system in Figure 2 are realized by using multi-view source encoder, view transform encoder and multi-task decoder. The workflow is shown in Figure 3. + +As shown in Figure 3, during the training period of the MEMVT system, features from massive single-view sources (general content, empirical content and reflow content) can be extracted by the encoder as the standard single-view tokens of transformer architecture, and they can then be mapped into a uniform view as multiple X-view tokens. This workflow will be circled for each time slide to realize the basic functions, including information compensation, multi-view fusion and completion. + +As shown in Figure 3, during the inference period of the MEMVT system, the system will output the specific view to realize the application functions based on the above basic functions, including tracking and decision-making assistant, visualization, prediction, target analysis enhancement, simulation scenario library generation and real time vehicular collaborative control optimization. + +Additionally, the MEMVT system also supports the evaluation, management and service functions during the whole period. + +Use cases of the MEMVT system are shown in Appendix I. + +![Figure 3: Typical workflow of the MEMVT system. The diagram is divided into two main sections: 'Training period of multimodal generative AI' and 'Inference period of multimodal generative AI'. In the training period, 'Single-view sources (Multimodal)' including Image, Video, Map, and LiDAR/RADAR points are processed by an 'Encoder' to produce 'Single-view Tokens'. These tokens are then processed by a 'View transform encoder' which incorporates 'Temporal fusion' (indicated by a circular arrow). The output is an 'X-view' (represented by a grid of colored blocks) and 'Tokens'. In the inference period, these 'X-view' and 'Tokens' are processed by a 'Multi-task Decoder' to produce the final 'X-view output'. A reference code 'F.748.72(25)' is shown in the bottom right corner.](a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg) + +Figure 3: Typical workflow of the MEMVT system. The diagram is divided into two main sections: 'Training period of multimodal generative AI' and 'Inference period of multimodal generative AI'. In the training period, 'Single-view sources (Multimodal)' including Image, Video, Map, and LiDAR/RADAR points are processed by an 'Encoder' to produce 'Single-view Tokens'. These tokens are then processed by a 'View transform encoder' which incorporates 'Temporal fusion' (indicated by a circular arrow). The output is an 'X-view' (represented by a grid of colored blocks) and 'Tokens'. In the inference period, these 'X-view' and 'Tokens' are processed by a 'Multi-task Decoder' to produce the final 'X-view output'. A reference code 'F.748.72(25)' is shown in the bottom right corner. + +**Figure 3 – Typical workflow of the MEMVT system** + +## 8 Functional requirements of MEMVT + +The requirements for each function in Figure 2 are listed in the following clauses. An individual and any personal information in single-view sources (image, video, LiDAR and RADAR points and map) used for MEMVT system are anonymized. The application functions (related to clauses 8.4 to 8.9) are realized by the three basic functions (related to clauses 8.1 to 8.3). All the application functions are illustrated in the use cases of Appendix I. + +### 8.1 Information compensation + +- The MEMVT system is required to support information compensation for partially obstructed objects (for example, information compensation of the obstructed back view position of inspectors in clause I.3). +- The MEMVT system is recommended to generate compensation knowledge by multimodal and multi-view fusion (for example, compensation knowledge of the distance between inspectors and substation equipment in clause I.3). +- Model fine-tuning is recommended to optimize the compensation results (for example, optimize the distance between inspectors and substation equipment in clause I.3). +- The input formats of single-view sources can optionally be text, voice, single image, multiple images and videos, to repair the partially available information from multiple sources. + +### 8.2 Multi-view fusion and completion + +- The MEMVT system is required to support multi-view fusion and compensation, by transforming single-view sources into one fused view and extracting features from the fused view. +- The MEMVT system is required to support completion of missing segments within the fused view, due to sparse sensing or broken sensors. + +### 8.3 Multimodal temporal fusion + +- The MEMVT system is required to support the temporal information fusion of many temporal slide view sources. +- The MEMVT system is recommended to leverage memory units or feature fusing models to translate temporal-adjacent frames from multiple sources, during the training and inference procedure. +- The temporal fusion is required to compensate for missing information based on historic data, as well as enhances the trajectory prediction ability, especially under occlusions. + +### **8.4 Tracking and decision-making assistant** + +- a) The MEMVT system is recommended to support trajectory tracking and decision-making assistant for connected vehicles by transferring the useful multi-view information to the vehicle's decision-making system. +- b) The MEMVT system is recommended to build up cross-view object association and 2D-3D interaction to produce correct trajectories and for further prediction. +- c) Feature-guided position embedding can optionally be used to aggregate 2D image features to 3D detection queries. + +### **8.5 Visualization** + +- a) The MEMVT system is recommended to support visualization by using multi-view fusion and transformation. For multi-view scenarios, a visualization under a common, unified protocol is required. +- b) The MEMVT system is recommended to support graceful visualization other than legacy multi-view concatenation, by transforming multi-view inputs into BEV space. +- c) The 3D view is recommended to be realized in the BEV space. + +### **8.6 Prediction** + +- a) The MEMVT system is recommended to support prediction of spatio-temporal information of the target objects under various scenarios. +- b) The dynamics and occlusions of objects are recommended to be considered. + +### **8.7 Target analysis enhancement** + +The MEMVT system is recommended to support target analysis enhancement, including enhancement of target retrieval and scenario understanding by transforming useful multi-view information, including trajectories and categories of passing vehicles, high-definition map, and occupation units. + +### **8.8 Simulation scenario library generation** + +- a) The MEMVT system is recommended to support simulation scenario library generation based on roadside-view traffic data and three-dimensional Gaussian splashing (3DGS) technology. +- b) The MEMVT system is recommended to support the fusion of simulation software and multimodal sensor models to generate vehicle-view data and simulation scenario library. + +### **8.9 Real-time vehicular collaborative control optimization** + +- a) The MEMVT system is recommended to provide optimized real-time vehicular collaborative control instructions of vehicular acceleration and steering angle. +- b) The MEMVT system is recommended to provide vehicular trajectory predictions with pictures, videos and intersection text information (e.g., signal light status, road sign semantics). +- c) The MEMVT system is recommended to provide inter-section semantic information analysis of traffic signal timing and conflict point distribution. +- d) The MEMVT system is recommended to provide dynamically calling domain knowledge library of traffic rules and vehicle dynamics knowledge. +- e) The MEMVT system is recommended to provide real-time evaluation of collision probability and traffic efficiency. + +### **8.10 Evaluation** + +- a) The MEMVT system is recommended to support the model evaluation of multi-view transformation, including object detection, tracking and trajectory prediction. +- b) The object detection is recommended to evaluate model outputs of object spatial location, category, orientation and other attributes. +- c) The tracking is recommended to evaluate the model outputs of object trajectory. +- d) Trajectory prediction is recommended to evaluate model outputs of future routes of objects. + +### **8.11 Management and services** + +- a) The MEMVT is recommended to support access control management of multi-view transformation inference interface. +- b) The MEMVT is recommended to support operation management, including single-view resource management and model deployment. +- c) The MEMVT is recommended to support the plug-in service, including multi-view transformation models addition, replacement and fine-tuning and interoperability with other services. + +## Appendix I + +### Use cases of multimodal generative AI enabled multi-view transformation + +(This appendix does not form an integral part of this Recommendation.) + +### I.1 MEMVT system for port cargo tracking + +Figure I.1 shows the MEMVT system for port cargo tracking (related to clause 8.4). According to the accuracy and stability of the MEMVT system, port cargo tracking can be realized anywhere, anytime. Additionally, it also generates the blind area view and information on the containers and realizes accurate matching of container loading within the specific position in port. + +![Figure I.1: MEMVT system for port cargo tracking. The image consists of two side-by-side panels. The left panel shows a night view of a port with several shipping containers and a truck, with pink rectangular bounding boxes overlaid on the containers to indicate tracking. The right panel shows a daytime view of a large red container being lifted by a container crane at a port facility.](eb22a8740f7c6a0f6ee98f16d99ed8b9_img.jpg) + +Figure I.1: MEMVT system for port cargo tracking. The image consists of two side-by-side panels. The left panel shows a night view of a port with several shipping containers and a truck, with pink rectangular bounding boxes overlaid on the containers to indicate tracking. The right panel shows a daytime view of a large red container being lifted by a container crane at a port facility. + +Figure I.1 – MEMVT system for port cargo tracking + +### I.2 MEMVT system for traffic awareness + +Figure I.2 shows the MEMVT system for traffic awareness (related to clauses 8.5 and 8.6). Compared to the traditional multi-view transformation system, MEMVT system can provide blind area view transformation, visualization of the whole traffic environment and prediction of traffic jams, so that it can improve sensor robustness, traffic management and efficiency. + +![Figure I.2: MEMVT system for traffic awareness. The image is a composite of several views of a multi-lane highway. At the top, there are three video frames showing traffic from different perspectives, with 'Video loss' text in the gaps. Below these are two frames showing radar data overlaid on a road map, with 'Radar loss' text in the gaps. The bottom section is a large, wide-angle view of the highway with numerous small car icons representing detected vehicles. Two rectangular areas on the road are highlighted with orange and blue outlines. The text 'F.748.72(25)' is visible in the bottom right corner.](6b7b3f3d6f9341906163682cf12d1ea1_img.jpg) + +Figure I.2: MEMVT system for traffic awareness. The image is a composite of several views of a multi-lane highway. At the top, there are three video frames showing traffic from different perspectives, with 'Video loss' text in the gaps. Below these are two frames showing radar data overlaid on a road map, with 'Radar loss' text in the gaps. The bottom section is a large, wide-angle view of the highway with numerous small car icons representing detected vehicles. Two rectangular areas on the road are highlighted with orange and blue outlines. The text 'F.748.72(25)' is visible in the bottom right corner. + +Figure I.2 – MEMVT system for traffic awareness + +### I.3 MEMVT system for personal safety distance measurement during the inspection of a substation power system + +Figure I.3 shows the MEMVT system for personal safety distance measurement (related to clause 8.7). According to the information compensation capability of MEMVT, the back view position of inspectors can be generated, and then the accurate distance between inspectors and substation equipment is inferred and calculated, so as to judge the safe distance between inspectors and substation equipment. + +![Figure I.3: MEMVT system for personal safety distance measurement. A worker in a yellow safety vest and white hard hat is shown with a blue bounding box. A red line indicates the personal safety distance to a piece of equipment, labeled 'Personal safety distance 1043 mm'. The image is labeled F.748.72(25) in the bottom right corner.](7832324609ad3cc688064e0341612b32_img.jpg) + +Figure I.3: MEMVT system for personal safety distance measurement. A worker in a yellow safety vest and white hard hat is shown with a blue bounding box. A red line indicates the personal safety distance to a piece of equipment, labeled 'Personal safety distance 1043 mm'. The image is labeled F.748.72(25) in the bottom right corner. + +Figure I.3 – MEMVT system for personal safety distance measurement during an inspection of a substation power system + +### I.4 MEMVT system for simulation scenario library generation + +Figure I.4 shows the MEMVT system for simulation scenario library (related to clause 8.8). Compared to the traditional multi-view transformation system, the MEMVT system can provide transformation from roadside-view to vehicle-view and generate simulation scenario library, providing efficient and diverse scenario data for end-to-end training of autonomous driving. + +![Figure I.4: MEMVT system for test scenario library. The interface displays a multi-view simulation environment with a 3D road model and multiple camera feeds. A sidebar on the right lists various vehicle types and actions. A bottom panel shows icons for different vehicle models and colors, including 'car_grey', 'van_black', 'bus_blue', 'car_red', 'truck_yellow', and 'car_blue'.](3293245c6893d9d49c2c878828423ecd_img.jpg) + +Figure I.4: MEMVT system for test scenario library. The interface displays a multi-view simulation environment with a 3D road model and multiple camera feeds. A sidebar on the right lists various vehicle types and actions. A bottom panel shows icons for different vehicle models and colors, including 'car\_grey', 'van\_black', 'bus\_blue', 'car\_red', 'truck\_yellow', and 'car\_blue'. + +Figure I.4 –MEMVT system for test scenario library + +### **I.5 MEMVT system for real-time vehicular collaborative control optimization** + +To meet the real-time requirements of multi-vehicle collaborative control for complex interactive scenarios (related to clause 8.9), a large language model (LLM) is used to analyse semantic information of intersections (such as signal timing and conflict point distribution) based on trajectory prediction results. A hybrid expert system (Mixture of Experts (MoE)) is designed to dynamically call domain knowledge bases (traffic rules, vehicle dynamics). Combining real-time risk perception of collision probability and traffic efficiency, the MEMVT system can break through the adaptability bottleneck of traditional rule-based control, dynamically optimize vehicle control instructions such as acceleration and steering angle, and improve the response speed. + +## Bibliography + +[b-ITU-T F.749.13] Recommendation ITU-T F.749.13 (2021), *Framework and requirements for civilian unmanned aerial vehicle flight control using artificial intelligence*. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.749.10-201905-I_PDF-E/raw.md b/marked/F/T-REC-F.749.10-201905-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..a2eedaa66457e7532919ca886bfcc9e5489db045 --- /dev/null +++ b/marked/F/T-REC-F.749.10-201905-I_PDF-E/raw.md @@ -0,0 +1,527 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.749.10** + +(05/2019) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Requirements for communication services of +civilian unmanned aerial vehicles** + +Recommendation ITU-T F.749.10 + +ITU-T + +![ITU logo](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) features a stylized globe with a red lightning bolt striking through it, symbolizing global connectivity and telecommunications. The letters 'ITU' are prominently displayed in blue within the globe. + +ITU logo + +International +Telecommunication +Union + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.749.10 + +## Requirements for communication services of civilian unmanned aerial vehicles + +## Summary + +Recommendation ITU-T F.749.10 provides the requirements for communication services of civilian unmanned aerial vehicles (CUAVs), as well as the use cases of CUAV in industry and consumer application areas. This Recommendation includes a general communication service framework, communication system requirements, requirements for flight control communication and flight data transport and requirements for mission payload communication service (such as audio / video / images transport, sensor data transport and communication signal relay). + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.749.10 | 2019-05-14 | 16 | 11.1002/1000/13900 | + +## Keywords + +Civilian unmanned aerial vehicle, communication service requirements, ground control station. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2019 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | | Page | +|--------------|-----------------------------------------------------------------------------------------|------| +| 1 | Scope..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 1 | +| 4 | Abbreviations and acronyms ..... | 2 | +| 5 | Conventions ..... | 2 | +| 6 | Introduction..... | 2 | +| 7 | Communication system framework of CUAV ..... | 2 | +| 7.1 | Overview of communication system ..... | 2 | +| 7.2 | General requirements for communication system ..... | 4 | +| 8 | Requirements for flight control and flight data transport ..... | 6 | +| 8.1 | Requirements for communication of flight control ..... | 6 | +| 8.2 | Requirements for CUAV flight data transport ..... | 6 | +| 9 | Requirements for mission payload service communication ..... | 7 | +| 9.1 | Requirements for mission payload data –transport related ..... | 7 | +| 9.2 | Requirements for mission payload video / images – transport related..... | 7 | +| 9.3 | Requirements for mission payload communication signal relay..... | 8 | +| Appendix I | – Use cases of civilian unmanned aerial vehicle communication services..... | 9 | +| I.1 | Agriculture and plant protection related services ..... | 9 | +| I.2 | Power line and petroleum pipeline inspection related services..... | 9 | +| I.3 | Police and traffic security surveillance related services..... | 9 | +| I.4 | Natural disaster monitoring related services ..... | 9 | +| I.5 | Aerial photography and video related services ..... | 10 | +| I.6 | Logistics express delivery related services..... | 10 | +| I.7 | Meteorological, territorial resources and scientific research related
services ..... | 10 | +| I.8 | Network communication relay related services ..... | 10 | +| I.9 | Bridge monitoring related services..... | 10 | +| I.10 | Industrial disaster monitoring related services ..... | 11 | +| Bibliography | ..... | 12 | + + + +# Recommendation ITU-T F.749.10 + +## Requirements for communication services of civilian unmanned aerial vehicles + +# 1 Scope + +This Recommendation specifies the requirements for communication services of civilian unmanned aerial vehicles (CUAVs), including communication system requirements, flight control communication requirements, flight data transport requirements and mission payload communication service requirements. The regulations and supervision of CUAV flights are out the scope of this Recommendation. + +The scope of this Recommendation includes: + +- overview of communication service framework; +- communication service requirements. + +# 2 References + +None. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following term defined elsewhere: + +**3.1.1 device** [b-ITU-T Y.4000]: With regard to the Internet of things, this is a piece of equipment with the mandatory capabilities of communication and the optional capabilities of sensing, actuation, data capture, data storage and data processing. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 civilian unmanned aerial vehicle:** An unmanned flying device controlled by a ground control station or telecontroller via various wireless communication means. It usually consists of an aeroplane body, a power device, aviation electrical and electronic equipment and mission payload equipment, etc. and is used in non-military application areas such as industrial and consumer areas to complete the specific operation and transportation of data including audio, video and image. + +**3.2.2 flight control system:** This is the sum of all components and driving devices of instruction transferring, rudder motion. It is mainly composed of airborne and ground control terminals. The one that is airborne includes three parts: airborne sensors, steering engine and flight control units. + +**3.2.3 ground control station:** A ground control station is a device which is used to realize the functions of mission planning, flight control, payload control, flight path display, parameter display, image and video display and mission information displaying, recording and distributing. + +**3.2.4 ISM bands:** The industrial, scientific and medical radio bands are radio bands (portions of the radio spectrum) reserved internationally for industrial, scientific and medical purposes other than telecommunications. + +**3.2.5 mission payload equipment:** The mission payload equipment consists of an audio / video / image acquisition device, signal relay device, remote electronic detection /sense device and other auxiliary devices. + +**3.2.6 telecontroller:** A piece of equipment used by human beings to control an unmanned aerial vehicle. It is usually composed of an operating device, coding device, transmitting device, receiving device, decoding device and executing mechanism. + +# **4 Abbreviations and acronyms** + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|------------------------------------| +| 4G | Fourth Generation | +| 5G | Fifth Generation | +| AP | Access Point | +| CUAV | Civilian Unmanned Aerial Vehicle | +| GCS | Ground Control Station | +| GNSS | Global Navigation Satellite System | +| IoT | Internet of Things | +| ISM | Industrial Scientific Medical | +| M2M | Machine-to-Machine | +| SIM | Subscriber Identity Module | +| UAV | Unmanned Aerial Vehicle | + +# **5 Conventions** + +In this Recommendation: + +The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. + +The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus this requirement needs not be present to claim conformance. + +The keywords "can optionally" and "may" indicate an optional requirement which is permissible, without implying any sense of being recommended. These terms are not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +# **6 Introduction** + +Unmanned aerial vehicles were originally developed and used in a military context. However, as the industry and consumer UAV market grows, industry and consumer UAV technologies have developed rapidly and matured enough to support a wide range of industry and civilian consumer areas. As civilian applications of UAVs are expected to grow rapidly, requirements for the communication service of civilian UAVs need to be addressed. + +# **7 Communication system framework of CUAV** + +## **7.1 Overview of communication system** + +The communication system of civilian unmanned aerial vehicles includes airborne wireless communication modules and a positioning device, a telecontroller, a ground control station (GCS), communication networks as well as a civil aviation flight management server. For different + +application areas, a telecontroller or ground control station can be used to provide flight control and communication services. In order to acquire the position information and the flight data of CUAVs for regulatory monitoring and check-ups, cellular networks (such as 4G/5G) can be used to transport the CUAV identification, position, altitude, speed, flight status to the civil aviation flight management server which will be used by the relevant departments. Also, the CUAV flight management server can receive other flight-prohibition zone data and send this data to the GCS or telecontroller to control the CUAV flight. In addition, if there are multiple CUAVs in a cooperative flight, it is required to realize communication between CUAVs. Figure 7-1 gives a communication reference framework of CUAVs. + +![Figure 7-1: Communication reference framework of civilian unmanned aerial vehicles. The diagram shows a central CUAV flight management server connected to a Ground control station, Cellular networks, and an Other flight-prohibition zone database. Two CUAVs (drones) are shown. Communication case 1 is between a Telecontroller and a CUAV. Communication case 2 is between a Ground control station and a CUAV. Communication case 3 is between a CUAV and Cellular networks. Communication case 4 is between a CUAV and a Small/micro base station. Communication case 5 is between a CUAV and a CUAV.](acfc53eca625d62b38aa2563efa95c3e_img.jpg) + +The diagram illustrates the communication reference framework for civilian unmanned aerial vehicles (CUAVs). At the center is a **CUAV flight management server**. To its left, a **Ground control station** (consisting of a laptop and a monitor) is connected to it. To its right, it is connected to **Cellular networks** (represented by a tower) and an **Other flight-prohibition zone database** (represented by a cylinder). Above the server, two CUAVs (drones) are shown. **Communication case 1** (magenta line) is between a **Telecontroller** (a handheld device) and a CUAV. **Communication case 2** (blue line) is between the **Ground control station** and a CUAV. **Communication case 3** (yellow line) is between a CUAV and **Cellular networks**. **Communication case 4** (cyan line) is between a CUAV and a **Small/micro base station** (a small tower). **Communication case 5** (green line) is between the two CUAVs. A small text label **F.749.10(19)\_F7-1** is located at the bottom right of the diagram. + +Figure 7-1: Communication reference framework of civilian unmanned aerial vehicles. The diagram shows a central CUAV flight management server connected to a Ground control station, Cellular networks, and an Other flight-prohibition zone database. Two CUAVs (drones) are shown. Communication case 1 is between a Telecontroller and a CUAV. Communication case 2 is between a Ground control station and a CUAV. Communication case 3 is between a CUAV and Cellular networks. Communication case 4 is between a CUAV and a Small/micro base station. Communication case 5 is between a CUAV and a CUAV. + +**Figure 7-1 – Communication reference framework of civilian unmanned aerial vehicles** + +Figure 7-1 illustrates five cases of communication service requirements. + +#### **Case 1: communication requirements between a CUAV and a telecontroller** + +Two ways of communication are recommended to realize the communication between a CUAV and a telecontroller. One way is based on the ISM bands to realize the communication between the CUAV and a telecontroller. The other is based on the 4G/5G communication network to realize communication between the telecontroller and CUAV under the cellular base station controlling the wireless resource allocation. + +### **Case 2: communication requirements between a CUAV and ground control station** + +The 4G/5G communication networks and ISM bands are recommended. One way is based on the ISM bands to realize the communication between the ground control station and CUAV. It is a very promising way to use 5G communication networks to real-time control the CUAV flight. + +#### **Case 3: communication requirements between CUAVs and cellular networks** + +Cellular networks (e.g., 4G/5G networks) are recommended to transport the CUAV flight data and mission payload data (such as application data, video and image, etc.). + +#### **Case 4: communication requirements between CUAVs** + +The 4G/5G communication mode or the Wi-Fi communication mode is recommended. E.g., a wireless AP (access point) can be set up on one CUAV, through which to command or to coordinate other nearby CUAVs flight. + +#### **Case 5: communication requirements between a CUAV and small / micro-base stations** + +With the aid of CUAVs, the communication signal of the public communication network can be transported to a place (via small / micro-base stations) where the public communication network cannot cover areas. Also, when the public communication network is damaged or failed, CUAVs act as the relay sources to transport data from one small / micro-base station (or IoT equipment with access point) to another small / micro-base station (or IoT equipment with access point). In addition, CUAVs can be used as a temporary collector to collect data from small / micro-base stations (or IoT equipment) in a certain area only when the public communication network is damaged or invalid. + +## **7.2 General requirements for communication system** + +The communication system shall have a communication module, telecontroller, ground control system, communication networks and management devices to finish functions such as numbering for CUAVs, identification authentication for CUAV, etc. Furthermore, there are communication interfaces between the CUAV and other devices. + +### **7.2.1 Numbering requirement for CUAV** + +- A CUAV is required to have a unique number. The number is optionally to include the country code, industry domain code, sustained flight time code, energy-used mode code, enterprise name code and CUAV type code. It is required to paste the code slice (e.g., two-dimensional code) to the CUAV body. + +### **7.2.2 Identification authentication requirements for CUAV** + +- A CUAV and the CUAV owner are required to have a real name registration and send the registration information to the CUAV flight management server controlled by the civil aviation management department. +- A CUAV is required to have a unique digital identifier or device identification code. The assigned unique identifier or device identification code is the identity of the CUAV used for the CUAV identification, authentication. This unique digital identifier or device identification code can optionally be an International Mobile Subscriber Identification Number (IMSI), or other M2M device identifiers. +- A CUAV is required to have one or more communication modules associated with the unique digital identifier. The communication module with the unique digital identifier shall support functions that prevent non-authorized access from the external and internal networks, complying with one or several authentication methods along with authorization management, and should support one or several encryption modes. + +### **7.2.3 The functional requirements for telecontroller** + +The telecontroller shall have functions such as display function, control function and communication function. + +- The telecontroller is recommended to have the display function to show its remaining energy, as well as other information. +- The telecontroller is recommended to have the control function to control the flight direction, flight speed, take-off and landing, etc. + +- The telecontroller is recommended to have the communication function to communicate with the CUAV. The communication module shall realize the encoding and decoding functions of the sending and receiving signals to the CUAV. + +### **7.2.4 The functional requirements for ground control system** + +The ground control system refers to the ground control station which fulfils CUAV flight control and management, integrated data display, map and flight path display, mission planning, etc. The specific configuration can be determined by the mission. + +##### **7.2.4.1 Flight control and management** + +Flight control and management is required to fulfil the CUAV functions' check-ups before take-off (launch), and to operate and control the CUAV in the process of cruise flight, mission execution and landing (return). Also, it is required to coordinate various related equipment with a unified system clock, as well as to fulfil flight state monitoring, operation mode switch and fault diagnosis and treatment, etc. + +##### **7.2.4.2 Integrated display system** + +The integrated display system is required to display the flight parameter and mission parameter, which will be used as the analysis and judgement basis for the operator to monitor the state of the CUAV flight and the mission payload equipment. The information displayed in the integrated display system includes, but is not limited to, the following: + +- The flight parameter is recommended to include the height, speed, course, flight path coordinates, flight attitude, residual power, flight time, satellite navigation parameter, etc. +- The mission planning is recommended to generate CUAV flight routes, automatically or manually. The mission payload parameter can optionally display the working status information of the mission payload equipment. +- It can optionally require the integrated display system to be a graphical and digital display. For flight failure status and / or a mission payload equipment fault state, it is recommended to note with sound, light or red colour to attract operator attention. + +#### **7.2.5 Interface requirements with external devices** + +There are six communication interfaces between the CUAV and external devices. + +- A CUAV is recommended to have a communication interface between the CUAV and telecontroller to control the CUAV flight. +- A CUAV is recommended to have a communication interface between the CUAV and ground control station. CUAV flight control and data can be transported through this interface. +- A CUAV is recommended to have a communication interface between the CUAV and cellular communication networks. The CUAV flight data and the mission payload information (such as data, video and image, etc.) can be transported through this interface. +- A CUAV is recommended to have a communication interface between a CUAV and another CUAV. Controlling and / or coordinating data between CUAVs is transported through this interface. +- A CUAV is recommended to have a communication interface between a CUAV and small / micro-base stations. Signal relay and data collection and exchange can be transported through this interface. +- A GCS is recommended to have a communication interface between the GCS and CUAV flight management server. The CUAV flight data can be uploaded to a CUAV flight management server and flight-prohibition data can be downloaded through this interface. + +# **8 Requirements for flight control and flight data transport** + +To complete the mission of the CUAV, it is required to control the CUAV and adjust the flight status of a CUAV according to the flight data of the CUAV. Therefore, the transport of flight control and flight data is required. + +## **8.1 Requirements for communication of flight control** + +A communication link between a CUAV and a telecontroller or a GCS is required to guarantee flight control of the CUAV. A communication link for flight control is required and optionally based on cellular communication networks (4G/5G networks) to support direct and indirect connection environments. Flight control with a direct connection is recommended to support a single-hop communication mode over cellular communication networks or ISM bands. Flight control with an indirect connection is recommended to support the multi-hop communication mode among multiple CUAVs using cellular communication modes. In addition, a CUAV is required to receive a flight control signal under a handover process among multiple base stations in order to guarantee a robust connectivity. + +- It is required to keep uninterrupted real-time reliable communication of flight control between a CUAV and a GCS or telecontroller with a low latency. +- It is required that a GCS or telecontroller remotely controls the direction of flight, speed, height, etc. of a CUAV, and the CUAV sends its status information to a GCS through the established communication link. +- A CUAV is recommended to combine the global navigation satellite system (GNSS) with sensor units (e.g., optionally supporting 6 freedom degree inertial measurement units, magnetic field gauge, barometric altimeter) in order to provide high stability and reliability of the flight control means. +- It is recommended that multiple flight control modes should be realized for different environmental operations. +- A CUAV is required to hover automatically when it loses its remote control signals. After exceeding a given time, the CUAV flight control system shall calculate the best back route and return it safely. Otherwise, the GCS or telecontroller may send an automatic returning order or pressing "one key" returning command. +- It is recommended that the communication of flight control between a CUAV and GCS or telecontroller can optionally choose cellular networks (e.g., 4G/5G), ISM bands. +- Before flight, a CUAV is required to be activated (based on the location and user information) online to ensure the CUAV operators use "correct geospatial information and flight capabilities". If not, then the flight distance and function of the CUAV will be limited. +- A CUAV is recommended to support various technologies to realize intelligent flight control, e.g., using artificial intelligence to fulfil the flight control and multi-CUAVs coordinated flight control. + +## **8.2 Requirements for CUAV flight data transport** + +- It is required to have different flight data transport modes when it has different mission targets. It can be divided into real-time transport and non-real-time transport. +- A CUAV is required to receive real-time flight notifications and the flight-prohibition zone data from the CUAV flight management server. +- A CUAV is required to transport the flight data (e.g., height, speed, etc.) to the GCS or CUAV flight management server. The following table lists some type of flight data. + +**Table 1 – Flight data of a CUAV and the related components or meaning** + +| | The type of flight data | Including components (not limited) | +|---|-----------------------------------------------------------|---------------------------------------------------------------------------------------------------------| +| 1 | Speed data | Maximum flat speed, cruise speed, minimum speed, etc. | +| 2 | Height data | Height ceiling, the maximum usage height and minimum flight altitude | +| 3 | Continuation time and maximum continuation time of flight | | +| 4 | Flight radius | The flight radius refers to the distance from taking off to the far most point (returning point) | +| 5 | Manoeuvre flight performance | Minimum turning radius, the maximum climbing rate and the maximum descent rate | +| 6 | Weight data | Maximum take-off weight; maximum mission payload equipment weight | +| 7 | Flight posture stability | Pitch and tilt angle stationarity, the tilt angle stationary degree and the yaw angle stationary degree | +| 8 | Positioning accuracy | Positioning precision and the target location precision | + +- It is recommended that data communication between the CUAV and CUAV flight management server can optionally choose cellular networks (e.g., 4G/5G), ISM bands and satellite communication. + +# **9 Requirements for mission payload service communication** + +Mission payload equipment includes audio / video / image acquisition equipment, remote electronic detection /sense equipment, signal relay equipment and other auxiliary equipment etc. mounted on the CUAV or integrated in the CUAV. + +## **9.1 Requirements for mission payload data –transport related** + +Mission payload data may come from remote electronic sense equipment (such as temperature sensor/ humidity sensor) in the CUAV. + +- A CUAV is recommended to support the collection and storage of mission payload data. This kind of data services have lower requirements for the delay and bandwidth and higher data integrity and reliability, and all the collected data is temporarily stored in the CUAV to be processed after coming back, e.g., data from micro-stations, sensors or IoT devices can be collected by a CUAV and exported to the third-party application server. +- A CUAV is optionally to support to transport real-time mission payload data (e.g., industrial disaster data and territorial resources and science research data, etc.) to the GCS or mobile terminal. +- It is required a secure data transmission channel be established before the beginning of the communication and that two-way authentication is performed between the CUAV and the GCS or CUAV flight management server. Networks that can be applied to data transmission services include cellular networks (e.g., 4G/5G), ISM bands and satellite. + +## **9.2 Requirements for mission payload video / images – transport related** + +Mission payload video / images can be acquired by high-resolution cameras, infrared cameras, low light reconnaissance cameras, etc. + +- It is required that secure and high bandwidth video / images transmission channels shall be established before the beginning of the communication and that two-way authentication is + +performed between the CUAV and the GCS or CUAV flight management server. Networks that can be applied to video / pictures transmission services include cellular networks (e.g., 4G/5G), ISM bands and satellite. + +- It is recommended that the delay is less than 100 milliseconds when using 4G/5G networks or ISM bands to control the CUAV. +- It is recommended that a communication system is capable to adjust transmission power, based on the environmental conditions and requirements, as well as the wireless transmission distance. +- It is required that the communication system supports the transmission of standard definition (SD) and high-definition (HD) video and images. As to HD, video and pictures shall at least reach 2 Mbit/s above with the delay less than 500 milliseconds. The system has the ability to dynamically adjust encoding speed and have an error tolerance and error correcting coding ability to meet the requirements of the transmission of pictures / video. +- A CUAV is optionally to support to transport real-time video / images to the GCS or mobile terminal (e.g., video / images related with power line and petroleum pipeline inspection, police and traffic security surveillance, disaster monitoring and aerial photography and video, etc.) + +## **9.3 Requirements for mission payload communication signal relay** + +Mission payload communication signal relay is a process to amplify and retransmit / relay the signal from one place to other place(s). + +- A CUAV is recommended to act as the relay sources of cellular networks to provide flexible and specific time interval public network access services for remote areas and mountain areas (e.g., when signals are cut off by high mountains). It is required that the signals sent by the CUAV are stable and do not interfere with other radio signals. +- When the cellular network base stations or other IoT network base stations are damaged or broken down, CUAVs can optionally be used as temporary base stations for the residents of the affected region and devices to provide emergency network access services (such as audio and video services, distance education, remote electronic medical as well as specific IoT devices access service). +- A CUAV can optionally act as a signal transmitter and data receiver for IoT devices (such as geological monitoring devices, farmland and infrastructure monitoring equipment etc.) in remote areas and mountainous areas to send control information (such as status control, start up and shut down) and periodically collect data from these monitoring devices. + +# Appendix I + +## Use cases of civilian unmanned aerial vehicle communication services + +(This appendix does not form an integral part of this Recommendation.) + +### I.1 Agriculture and plant protection related services + +#### – Service requirements + +It is required that CUAVs fulfil agricultural insecticide spraying, agricultural fertilizer dispensing, crop monitoring, etc. + +#### – Communication requirements + +A CUAV can be controlled by a telecontroller within the range of human vision or a GCS. The GCS monitors and sets up flight control modes for the CUAV. Sometimes, it is required to transport high-definition images and video to a GCS. + +### I.2 Power line and petroleum pipeline inspection related services + +#### – Service requirements + +It is required that CUAVs fulfil power line inspection, oil and gas pipeline monitoring, survey of drilling sites for the oil and gas industry, oil spill and oil slick detection and monitoring, etc. + +#### – Communication requirements + +A CUAV can be controlled by the GCS to monitor and cruise along the power line or oil pipeline to find whether there are any hazards or faults. If any, the location of hazards and fault information will be transported back to the GCS. A CUAV may save locally high-definition images or video or transport them in real-time to a GCS depending on the application requirements. + +### I.3 Police and traffic security surveillance related services + +#### – Service requirements + +It is required that CUAVs fulfil information gathering, aerial crime scene inspection, special and anti-terrorist operations, perimeter surveillance, surveillance of large public gatherings, road traffic surveillance, smuggling surveillance, etc. + +#### – Communication requirements + +A CUAV can be controlled by the GCS or a telecontroller and high-definition images or video are transported to a GCS or mobile terminal. + +### I.4 Natural disaster monitoring related services + +#### – Service requirements + +It is required that CUAVs fulfil natural disaster analysis, avalanche survivor search, forest fire prevention, the monitoring and investigation of disasters such as drought, fire, flood, landslide, earthquake and volcano eruption, etc. + +#### – Communication requirements + +A CUAV can be controlled by the GCS or a telecontroller and high-definition images and video and data are transported to a GCS or mobile terminal. Sometimes a CUAV can use satellites to transport images or video. + +### **I.5 Aerial photography and video related services** + +#### **– Service requirements** + +It is required that CUAVs fulfil news / disaster reporting, photography / film (capture the beautiful picture and shooting angle) for tourism, city publicity and extreme sports (taken from every angle tracking shots, downhill skiing athletes etc.), etc. + +#### **– Communication requirements** + +A CUAV can be controlled by the GCS or a telecontroller and high-definition images or video are transported to the GCS or a mobile terminal. + +### **I.6 Logistics express delivery related services** + +#### **– Service requirements** + +It is required that CUAVs fulfil transporting cargoes to remote areas or emergency medical / food supplies, or point-to-point transportation of small items, etc. + +#### **– Communication requirements** + +A CUAV can be controlled by the GCS or a telecontroller to reach a given spot or a place. + +### **I.7 Meteorological, territorial resources and scientific research related services** + +#### **– Service requirements** + +It is required that CUAVs fulfil weather assessments and measurements, atmospheric measurements, cloud seeding, typhoon tracking and measurement, land resource planning and geological exploration, marine mammal monitoring, iceberg surveillance and tracking, national and regional pollution monitoring, fishery monitoring, wildlife monitoring and protection, wildlife researching such as inventory, migration, conservation, etc. + +#### **– Communication requirements** + +A CUAV can be controlled by the GCS or a telecontroller and high-definition images and video and data are transported to a GCS or a mobile terminal. Sometimes, a CUAV can use satellites to transport images or video. + +### **I.8 Network communication relay related services** + +#### **– Service requirements** + +It is required that CUAVs provide flexible, low cost Wi-Fi network access, or act as an emergency communications network base station and provide network access services to residents in remote and mountain areas, etc. + +#### **– Communication requirements** + +A CUAV can be controlled by the GCS and acts as a temporary base station to relay communication signals for the residents of the affected region and devices when the cellular network base stations are damaged or have broken down, or acts as a signal transmitter and data receiver for IoT devices in remote areas and mountainous areas. + +### **I.9 Bridge monitoring related services** + +#### **– Service requirements** + +It is required for CUAVs to fulfil water level monitoring, crack detection of bridge structure such as decks, piers, abutments, etc. + +#### **– Communication requirements** + +A CUAV can be controlled by the GCS or a telecontroller and high-definition images and video and data are transported to a GCS or mobile terminal. Sometimes, a CUAV can use satellites to transport images or video. + +### **I.10 Industrial disaster monitoring related services** + +#### **– Service requirements** + +It is required for CUAVs to fulfil fire detection, fall accident detection, explosion accident detection of facilities such as factories, thermal power plants, nuclear power plants, construction sites, etc. + +#### **– Communication requirements** + +A CUAV can be controlled by the GCS or a telecontroller and high-definition images and video and data are transported to a GCS or servers or mobile terminal. Sometimes, a CUAV can use satellites to transport images or video. + +## Bibliography + +[b-ITU-T Y.4000] Recommendation ITU-T Y.4000/Y.2060 (2012), *Overview of the Internet of things*. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound-programme and television-transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.749.11-201911-I_PDF-E/1439cb942d9e363bbb3161b5540dd8c6_img.jpg b/marked/F/T-REC-F.749.11-201911-I_PDF-E/1439cb942d9e363bbb3161b5540dd8c6_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ac224d8368b85724879d5f114633b82c0faf5107 --- /dev/null +++ b/marked/F/T-REC-F.749.11-201911-I_PDF-E/1439cb942d9e363bbb3161b5540dd8c6_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:25f1e76d6a13ed2cbe07c24af16a7aecc2876a7d9abddb71b93dceff884f571c +size 49911 diff --git a/marked/F/T-REC-F.749.11-201911-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.749.11-201911-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9002ee68e2b9ebf63d27d39ba8c73bdafb0e6be6 --- /dev/null +++ b/marked/F/T-REC-F.749.11-201911-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8df701536c25ab8246d2868de3c20c8a4e25168bb52c5eb9078278ab1b5b4fd8 +size 6859 diff --git a/marked/F/T-REC-F.749.11-201911-I_PDF-E/b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg b/marked/F/T-REC-F.749.11-201911-I_PDF-E/b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c080195126408b3bb52419466162cbe569ffc71d --- /dev/null +++ b/marked/F/T-REC-F.749.11-201911-I_PDF-E/b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d2b6a8cc0437e4289c4ca6aed9d042e6268541bc1f4662f18a54f54ea0a22b16 +size 69704 diff --git a/marked/F/T-REC-F.749.11-201911-I_PDF-E/raw.md b/marked/F/T-REC-F.749.11-201911-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..f0287a568f4caf3784d9f24566e37861bb5db25b --- /dev/null +++ b/marked/F/T-REC-F.749.11-201911-I_PDF-E/raw.md @@ -0,0 +1,392 @@ + + +International Telecommunication Union + +**ITU-T** + +**F.749.11** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +(11/2019) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Requirements for civilian unmanned aerial +vehicles enabled mobile edge computing** + +Recommendation ITU-T F.749.11 + +![ITU logo](84a1d09fb489061482111515543b60dc_img.jpg) + +The logo of the International Telecommunication Union (ITU) features a globe with a red lightning bolt striking through it, symbolizing global connectivity and telecommunications. To the right of the globe, the text "International Telecommunication Union" is written in a blue, sans-serif font, with "ITU" in a larger, bold font above it. + +ITU logo + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.749.11 + +## Requirements for civilian unmanned aerial vehicles enabled mobile edge computing + +## Summary + +Civilian unmanned aerial vehicle (CUAV) enabled mobile edge computing (MEC) utilizes CUAV as an MEC platform to realize a flexible, efficient and on-demand computing service that can be rapidly deployed and move according to the practical service needs of devices. Recommendation ITU-T F.749.11 describes the framework and specifies requirements for a CUAV-MEC system, including functional, service and security requirements. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.749.11 | 2019-11-29 | 16 | 11.1002/1000/14104 | + +## Keywords + +Civilian unmanned aerial vehicle, mobile edge computing. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2020 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|---------------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Overview..... | 2 | +| 7 System framework of the CUAV-MEC ..... | 2 | +| 8 Function requirements for CUAV-MEC ..... | 3 | +| 8.1 Requirements for location placement for CUAV-MEC..... | 3 | +| 8.2 Requirements for communication services for CUAV-MEC ..... | 4 | +| 8.3 Requirements for task offloading for CUAV-MEC ..... | 4 | +| 8.4 Requirements for resource allocation for CUAV-MEC..... | 4 | +| 8.5 Requirements for CUAV collaboration for CUAV-MEC..... | 4 | +| 9 Service requirements for CUAV-MEC ..... | 4 | +| 10 Security requirements for CUAV-MEC ..... | 5 | +| Appendix I - Typical scenarios of CUAV-enabled mobile edge computing..... | 6 | +| I.1 Emergency scenarios deployment ..... | 6 | +| I.2 Temporary hot-spot scenarios deployment ..... | 6 | +| I.3 High mobility scenarios deployment..... | 6 | +| I.4 Rural or remote areas scenarios deployment..... | 7 | +| Bibliography..... | 8 | + + + +# Recommendation ITU-T F.749.11 + +## Requirements for civilian unmanned aerial vehicles enabled mobile edge computing + +## 1 Scope + +This Recommendation specifies the system framework and requirements for civilian unmanned aerial vehicle-enabled mobile edge computing (CUAV-MEC). + +The scope of this Recommendation includes: + +- system framework of CUAV-MEC; +- functional requirements for CUAV-MEC; +- service requirements for CUAV-MEC; +- security requirements for CUAV-MEC. + +## 2 References + +None. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 application** [b-ITU-T Y.101]: A structured set of capabilities, which provide value-added functionality supported by one or more services. + +**3.1.2 civilian unmanned aerial vehicle** [b-ITU-T F.749.10]: An unmanned flying device controlled by a ground control station or telecontroller via various wireless communication means. It usually consists of an aeroplane body, a power device, aviation electrical and electronic equipment and mission payload equipment, etc. and is used in non-military application areas such as industrial and consumer areas to complete the specific operation and transportation of data including audio, video and image. + +**3.1.3 device** [b-ITU-T Y.4000]: With regard to the Internet of things, this is a piece of equipment with the mandatory capabilities of communication and the optional capabilities of sensing, actuation, data capture, data storage and data processing. + +**3.1.4 service** [b-ITU-T Y.101]: A structure set of capabilities intended to support applications. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 civilian unmanned aerial vehicle collaboration:** Utilization of the ability of multiple unmanned aerial vehicles to accomplish specific tasks through network communication. + +**3.2.2 civilian unmanned aerial vehicle-enabled mobile edge computing:** Utilization of civilian unmanned aerial vehicles as a mobile edge computing platform to accomplish computation-intensive and latency-critical applications through task offloading. + +**3.2.3 communication service:** Service that controls information transmission, including uplink transmission of input data for the application (offloading from the devices to civilian unmanned aerial vehicles (CUAVs)) and downlink transmission of the results (from the CUAVs to the devices). + +**3.2.4 mobile edge computing:** System which provides an information technology (IT) service environment and cloud-computing capabilities at the edge of an access network which contains one or more type of access technology, and in close proximity to devices. + +NOTE – Based on the definition of multi-access edge computing in [b-ETSI GS MEC 001]. + +**3.2.5 resource allocation:** Allocation that controls resource usage, including communication, computing and cache resources within civilian unmanned aerial vehicles according to the task offloading decisions. + +**3.2.6 task offloading:** Offloading determining where to execute tasks and how to allocate tasks according to various factors, such as the computation demands of devices and existing resources of civilian unmanned aerial vehicles. + +**3.2.7 task requirements:** The required communication and computing resources and the executed latency constraints for applications generated by devices. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|----------|----------------------------------------------------------------| +| 4G | Fourth Generation | +| 5G | Fifth Generation | +| CUAV | Civilian Unmanned Aerial Vehicle | +| CUAV-MEC | Civilian Unmanned Aerial Vehicle-enabled Mobile Edge Computing | +| MEC | Mobile Edge Computing | +| WiFi | Wireless Fidelity | + +## 5 Conventions + +In this Recommendation: + +The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. + +The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement needs not be present to claim conformance. + +## 6 Overview + +Mobile edge computing (MEC) brings computation and storage to the edge of a mobile network and promises dramatic reduction in latency and mobile energy consumption. However, in many realistic-scenarios, such as emergency and public safety situations, temporary hotspot areas, and military scenarios, the static ground base stations are damaged or it is difficult to meet the requirements of the devices. The civilian unmanned aerial vehicle (CUAV) is a promising solution to extend edge service scope and provide offloading opportunities for devices in such scenarios owing to its mobility, autonomy and flexibility. Because some typical scenarios for CUAV-MEC are appearing and growing rapidly, requirements for CUAV-MEC need to be addressed. + +## 7 System framework of the CUAV-MEC + +Figure 1 shows a reference framework for the CUAV-MEC system, which has three layers: a device interface, CUAV on-board servers and a central management platform. + +The device interface layer is used by various kinds of devices, such as cars, mobile phones and sensors. A device application locates directly in the device that exploits the CUAV-MEC service. + +Data generated by devices can be uploaded to CUAV on-board servers supported by the communication service. + +The CUAV on-board server layer offers four main function modules for location placement, communication service, task offloading and resource allocation. Location placement determines the locations of CUAVs according to service demands and devices. As a communication module in the middle layer, the communication service module allows CUAV on-board servers to communicate with lower devices and the upper central platform to transmit related data. The task-offloading module aims to make suitable offloading decisions for devices. The resource allocation module controls the usage of communication, computing and cache resources within CUAVs. + +The central management platform layer provides multiple management services and controls the overall MEC function. The device management module manages devices and CUAVs hosted on lower layers. The service management module profiles the CUAV server services available. The status monitoring module monitors CUAV computation resource status and communication channel status. + +![Diagram of the CUAV-MEC system framework showing three layers: Central management platform, CUAV onboard servers, and Device interface.](b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg) + +The diagram illustrates the system framework of CUAV-MEC, organized into three horizontal layers connected by double-headed blue arrows indicating bidirectional communication. + +- Central management platform (Top Layer):** Contains three modules: **Service management**, **Device management**, and **Status monitoring**. It is accompanied by an icon of a cloud server. +- CUAV onboard servers (Middle Layer):** Contains four modules: **Location placement**, **Communication service**, **Task offloading**, and **Resource allocation**. It is accompanied by icons of two drones and servers. +- Device interface (Bottom Layer):** Contains two modules: **Device application** and **Communication service**. It is accompanied by icons of a heart, a smartphone, a car, and a person. + +F.749.11(19)\_F01 + +Diagram of the CUAV-MEC system framework showing three layers: Central management platform, CUAV onboard servers, and Device interface. + +**Figure 1 – The system framework of CUAV-MEC** + +## **8 Function requirements for CUAV-MEC** + +### **8.1 Requirements for location placement for CUAV-MEC** + +FUN-001: The location information of devices includes the location of all devices currently associated with the CUAV. + +FUN-002: The CUAV-MEC system is required to feasibly move the CUAV toward devices and provide MEC service for connected devices. It is of critical importance to determine the CUAV + +parameters affecting service performance, such as altitude, location and trajectory according to changes in device location information. + +### **8.2 Requirements for communication services for CUAV-MEC** + +FUN-003: The CUAV-MEC system is required to establish a stable and reliable communication link between a CUAV and devices so as to transmit input data for task execution and output results. + +FUN-004: The CUAV-MEC system is required to establish a stable and reliable communication link between a CUAV and the central management platform so as to interactively share control and management data. + +FUN-005: The CUAV-MEC system is recommended to establish a stable and reliable communication link between CUAVs so as to collaboratively serve the devices. + +FUN-006: The CUAV-MEC system is recommended to establish the communication link by utilizing fourth generation (4G), fifth generation (5G) or wireless fidelity (WiFi) access methods. The CUAV-MEC system is recommended to adjust the data transmission and network mode in response to changes in bandwidth status. + +### **8.3 Requirements for task offloading for CUAV-MEC** + +FUN-007: The CUAV-MEC system is required to make suitable decisions on task offloading aiming to determine whether the offloading is profitable for devices and can reduce energy consumption and execution delay. + +FUN-008: The CUAV-MEC system is required to make suitable decisions to satisfy the task requirements of devices. + +FUN-009: The CUAV-MEC system is recommended to design different task-offloading strategies according to the various applications requirements and consider various parameters that play critical roles for the performance of task offloading, including latency, bandwidth utilization, context awareness and scalability. + +FUN-010: The CUAV-MEC system is recommended to make adaptive task-offloading decisions based on changes in communication, storage and computation resources. + +### **8.4 Requirements for resource allocation for CUAV-MEC** + +FUN-011: The CUAV-MEC system is required to allocate communication resources for devices. + +FUN-012: The CUAV-MEC system is required to allocate computing resources for devices. + +FUN-013: The CUAV-MEC system is recommended to provide a resource allocation strategy during the whole edge computing stage in order to significantly improve system performance and network resource utilization. The resource allocation strategy includes joint communication and computation resource allocation, service scheduling and CUAV selections. + +### **8.5 Requirements for CUAV collaboration for CUAV-MEC** + +FUN-014: The CUAV-MEC system is recommended to serve devices in a cooperative way according to the mobility of devices and changes in service demand. + +FUN-015: The CUAV-MEC system is recommended to utilize the CUAV collaboration to decrease handover, optimize resource allocation and improve performance. + +## **9 Service requirements for CUAV-MEC** + +SRV-001: The CUAV-MEC system is required to support data transmission and data processing in multiple formats, such as image, audio and video. + +SRV-002: The CUAV-MEC system is required to support real-time data transmission from devices to CUAVs. + +SRV-003: The CUAV-MEC system is required to support real-time data processing on CUAVs for multiple devices. + +SRV-004: The CUAV-MEC system is required to support collection of the communication and computation resource status of CUAVs. + +SRV-005: The CUAV-MEC system is required to support collection of the task requirements and environmental status of devices. + +SRV-006: The CUAV-MEC system is required to provide task execution information for ground devices. + +SRV-007: The CUAV-MEC system is recommended to support task management. The task requirement and task execution mechanisms are required to be efficient and flexible. + +SRV-008: The CUAV-MEC system is recommended to utilize CUAV collaboration to jointly serve all associated devices. + +SRV-009: The CUAV-MEC system is recommended to provide different services according to task priority. + +SRV-010: The CUAV-MEC system is recommended to have the ability to store corresponding contents so as to rapidly provide services. + +## **10 Security requirements for CUAV-MEC** + +SEC-001: The CUAV-MEC system is required to permit authorized devices to access the system and obtain edge computing services. The CUAV-MEC system is required to forbid an unauthorized device to utilize any resource of the system and perform incorrect operations in CUAV-MEC. + +SEC-002: The CUAV-MEC system is required to ensure the security of data transmission. It is required to protect privacy of data communication and provide network transmission with high availability and high reliability. + +SEC-003: The CUAV-MEC system is required to ensure security of data processing. It is required to provide appropriate mechanisms to control and manage security of data collection, storage and computation. The required security mechanisms include device privacy protection, network attack avoidance, data security isolation and service authorization. + +## Appendix I + +### Typical scenarios of CUAV-enabled mobile edge computing + +(This appendix does not form an integral part of this Recommendation.) + +This appendix describes four typical scenarios for CUAV-MEC. See Figure I.1. + +![Diagram illustrating four typical scenarios of CUAV-MEC: Emergency, High mobility, Temporary hot-spots, and Rural or remote. Each scenario is represented by a small image within a large oval, with a CUAV (drone) and MEC (Mobile Edge Computing) unit above it, connected by a lightning bolt symbol.](1439cb942d9e363bbb3161b5540dd8c6_img.jpg) + +The diagram shows four distinct scenarios for CUAV-MEC deployment, each enclosed in a large oval. Above each scenario is a CUAV (drone) icon connected to an MEC (Mobile Edge Computing) unit, which is further connected to the scenario by a lightning bolt symbol. The scenarios are: 1. Emergency: Aerial view of a disaster area with water and buildings. 2. High mobility: A busy highway with many cars. 3. Temporary hot-spots: A modern urban area with large buildings and people. 4. Rural or remote: A view of a mountainous or rural landscape. Each MEC unit is represented by a grey box with a blue and green bar, labeled 'MEC'. + +Diagram illustrating four typical scenarios of CUAV-MEC: Emergency, High mobility, Temporary hot-spots, and Rural or remote. Each scenario is represented by a small image within a large oval, with a CUAV (drone) and MEC (Mobile Edge Computing) unit above it, connected by a lightning bolt symbol. + +F.749.11(19)\_FI.1 + +Figure I.1 – Typical scenarios of CUAV-MEC + +### I.1 Emergency scenarios deployment + +CUAV can play an important role in disaster or emergency relief scenarios, as ground devices with limited processing capabilities can benefit from CUAV-aided execution of data analytics application, such as assessing victim location, hazardous terrain and structures. In addition, if the ground communication infrastructure is damaged or inadequate, the CUAV can simultaneously provide connectivity and computation service for ground devices. + +### I.2 Temporary hot-spot scenarios deployment + +The temporary hot-spot scenarios usually create hard-to-predict inhomogeneous service demands. Meeting the requirements of this scenario is challenging, especially for delay-sensitive and compute-intensive applications. The existing facilities cannot meet the needs of devices, while establishing the temporary infrastructures can easily lead to cost increase and waste of resources. CUAV is appropriate to apply in this scenario, because of its high flexibility, scalability and mobility, enabling dynamic adaptation to device movement and changes in service demands. + +### I.3 High mobility scenarios deployment + +The mobility of CUAVs enables devices with high mobility and data execution demand, e.g., autonomous driving and road navigation, to be served. CUAV-MEC can provide vehicle monitoring and address various analytics, which can reduce traffic congestion, increase road safety and enhance the overall driving experience. + +### **I.4 Rural or remote areas scenarios deployment** + +A rural or remote edge network often does not offer a large computation service due to its sparsely spread populations. Moreover, the service is not always reliable, which leads to an inadequate device experience in rural or remote areas. CUAV-MEC is considered to be a flexible, efficient, low-cost and low-power infrastructure to provide various functionalities, including communication, computing, caching and transportation, for rural or remote areas. + +## Bibliography + +- [b-ITU-T F.749.10] Recommendation ITU-T F.749.10 (2019), *Requirements for communication services of civilian unmanned aerial vehicle*. +- [b-ITU-T Y.101] Recommendation ITU-T Y.101 (2000), *Global Information Infrastructure terminology: Terms and definitions*. +- [b-ITU-T Y.4000] Recommendation ITU-T Y.4000/Y.2060 (2012), *Overview of the Internet of things*. +- [b-ETSI GS MEC 001] ETSI GS MEC 001 V2.1.1 (2019), *Multi-access Edge Computing (MEC); Terminology*. Available [viewed 2019-12-27] at: +[https://www.etsi.org/deliver/etsi\\_gs/MEC/001\\_099/001/02.01.01\\_60/gs\\_MEC001v020101p.pdf](https://www.etsi.org/deliver/etsi_gs/MEC/001_099/001/02.01.01_60/gs_MEC001v020101p.pdf) + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.749.12-202008-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.749.12-202008-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9002ee68e2b9ebf63d27d39ba8c73bdafb0e6be6 --- /dev/null +++ b/marked/F/T-REC-F.749.12-202008-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8df701536c25ab8246d2868de3c20c8a4e25168bb52c5eb9078278ab1b5b4fd8 +size 6859 diff --git a/marked/F/T-REC-F.749.12-202008-I_PDF-E/33a8f3f01dfa8bce75d23017855a13c5_img.jpg b/marked/F/T-REC-F.749.12-202008-I_PDF-E/33a8f3f01dfa8bce75d23017855a13c5_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ec491683d8a4adee32f5450b0782c11f5ad6f3d7 --- /dev/null +++ b/marked/F/T-REC-F.749.12-202008-I_PDF-E/33a8f3f01dfa8bce75d23017855a13c5_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f62d7cd3388e603646aca9e92e9493a674f37f9b682cad907b720f8b0dadc09d +size 115091 diff --git a/marked/F/T-REC-F.749.12-202008-I_PDF-E/acfc53eca625d62b38aa2563efa95c3e_img.jpg b/marked/F/T-REC-F.749.12-202008-I_PDF-E/acfc53eca625d62b38aa2563efa95c3e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9d5e79ebfab07f1c4454316d0c93025a73e849ab --- /dev/null +++ b/marked/F/T-REC-F.749.12-202008-I_PDF-E/acfc53eca625d62b38aa2563efa95c3e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2e64fa76394acce5e40818f963922332d86187a5febbfb1a8fd72fd7fb77f202 +size 142524 diff --git a/marked/F/T-REC-F.749.12-202008-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg b/marked/F/T-REC-F.749.12-202008-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..84d472e83dbb5623c7b65fd43109fa2a6cbd6996 --- /dev/null +++ b/marked/F/T-REC-F.749.12-202008-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8199a5db75269456ff984ea681baee015db734699e2750ebf69db96bc9cde544 +size 129375 diff --git a/marked/F/T-REC-F.749.12-202008-I_PDF-E/raw.md b/marked/F/T-REC-F.749.12-202008-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..c09ef632df4aea4a0f6d0473707791ea7289b43f --- /dev/null +++ b/marked/F/T-REC-F.749.12-202008-I_PDF-E/raw.md @@ -0,0 +1,566 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.749.12** + +(08/2020) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Framework for communication application of +civilian unmanned aerial vehicle** + +Recommendation ITU-T F.749.12 + +ITU-T + +![ITU logo](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) features a stylized globe with a red lightning bolt striking across it. To the right of the globe, the text "International Telecommunication Union" is written in a blue, sans-serif font, with "ITU" in a larger, bold font above it. + +ITU logo + +International +Telecommunication +Union + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +## Recommendation ITU-T F.749.12 + +# Framework for communication application of civilian unmanned aerial vehicle + +## Summary + +The civilian unmanned aerial vehicles are widely used in industry and consumer areas such as agriculture and plant protection, power line and petroleum pipeline inspection, police and traffic security surveillance, disaster monitoring, aerial photography and videography, express delivery, forestry and forest fire monitoring, meteorological, resource and scientific research, etc. + +Recommendation ITU-T F.749.12 presents the general framework for communication application of civilian unmanned aerial vehicle (CUAV) and its functional entities, reference points, etc. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.749.12 | 2020-08-13 | 16 | 11.1002/1000/14331 | + +## Keywords + +Civilian unmanned aerial vehicle, communication application, framework. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2020 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-----------------------------------------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Introduction ..... | 2 | +| 7 High-level communication application frameworks ..... | 2 | +| 7.1 Application layer ..... | 3 | +| 7.2 Service support and application support layer..... | 3 | +| 7.3 Network layer ..... | 4 | +| 7.4 Device layer..... | 4 | +| 7.5 Management ..... | 4 | +| 8 Functional entity of communication application ..... | 4 | +| 8.1 Functional entity framework and reference points..... | 4 | +| 8.2 Application ..... | 6 | +| 8.3 Services..... | 6 | +| 8.4 Service functionalities ..... | 7 | +| 8.5 Network functions ..... | 8 | +| 8.6 Device functions..... | 8 | +| 8.7 Management ..... | 9 | +| Appendix I – Example use case of CUAV security flight monitoring based on cellular networks (4G/5G) ..... | 11 | +| Bibliography..... | 13 | + + + +# Recommendation ITU-T F.749.12 + +## Framework for communication application of civilian unmanned aerial vehicle + +# 1 Scope + +This Recommendation defines the general framework for communication application of civilian unmanned aerial vehicle (CUAV) and its functional entities, reference points, etc. The communication applications include flight control, flight data transportation, mission payload data services and video/images services, etc. The regulations and supervision of civilian unmanned aerial vehicle flight are out scope of this Recommendation. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T Y.4000] Recommendation ITU-T Y.4000/Y.2060 (2012), *Overview of the Internet of things*. + +# 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 civilian unmanned aerial vehicle** [b-ITU-T F.749.10]: An unmanned flying device controlled by a ground control station or telecontroller via various wireless communication means. It usually consists of an aeroplane body, a power device, aviation electrical and electronic equipment and mission payload equipment, etc., and is used in non-military application areas such as industrial and consumer areas to complete the specific operation and transportation of data including audio, video and image. + +**3.1.2 flight control system** [b-ITU-T F.749.10]: This is the sum of all components and driving devices of instruction transferring, rudder motion. It is mainly composed of airborne and ground control terminals. The one that is airborne includes three parts: airborne sensors, steering engine and flight control units. + +**3.1.3 ground control station** [b-ITU-T F.749.10]: A ground control station is a device which is used to realize the functions of mission planning, flight control, payload control, flight path display, parameter display, image and video display and mission information displaying, recording and distributing. + +**3.1.4 ISM bands** [b-ITU-T F.749.10]: The industrial, scientific and medical radio bands are radio bands (portions of the radio spectrum) reserved internationally for industrial, scientific and medical purposes other than telecommunications. + +**3.1.5 mission payload equipment** [b-ITU-T F.749.10]: The mission payload equipment consists of an audio / video / image acquisition device, signal relay device, remote electronic detection / sense device and other auxiliary devices. + +**3.1.6 telecontroller** [b-ITU-T F.749.10]: A piece of equipment used by human beings to control an unmanned aerial vehicle. It is usually composed of an operating device, coding device, transmitting device, receiving device, decoding device and executing mechanism. + +### **3.2 Terms defined in this Recommendation** + +None. + +## **4 Abbreviations and acronyms** + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|------------------------------------------------| +| CUAV | Civilian Unmanned Aerial Vehicle | +| eSIM | Embedded Subscriber Identification Module | +| GCS | Ground Control Station | +| IoT | Internet of Things | +| IMSI | International Mobile Subscriber Identification | +| ISM | Industrial Scientific Medical | +| M2M | Machine-to-Machine | +| USIM | Universal Subscriber Identification Module | +| VR | Virtual reality | +| 4G | Fourth Generation | +| 5G | Fifth Generation | + +# **5 Conventions** + +In this Recommendation: + +- The keywords "shall" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The keywords "should" indicate an optional requirement which is permissible. This term is not intended to imply that the vendor's implementation must provide the option, and the feature can be optionally enabled by the vendor. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +# **6 Introduction** + +Civilian unmanned aerial vehicles are widely used in industry, personal and consumer and other third party application fields. These applications have some common service and application features (such as flight service and control, application service and control) and management functions (such as CUAV management and flight database management), etc. This application framework is summarized and described from application layer, service support and application support layer, network layer and device layer. + +# **7 High-level communication application frameworks** + +Figure 1 shows the overview of a CUAV communication application framework based on the Internet of things (IoT) reference model [ITU-T Y.4000]. It is composed of four layers as well as management, as illustrated in Figure 1. + +The four layers are as follows: + +- Application layer, +- Service support and application support layer, +- Network layer, +- Device layer. + +![Figure 1 – Overview of CUAV communication application high level framework. The diagram shows four layers: Application layer, Service support and application support layer, Network layer, and Device layer. The Application layer includes Industry application (Plant protection, Monitoring, Power line and pipeline inspection, Visual surveillance, ...), Personal and consumer application (e.g., aerial photography), and Other third party application. The Service support and application support layer includes Flight data service, Application data service (including audio/video, etc.), Flight control session management, and Mission payload application management. The Network layer includes Networking and transport, Non-cellular network access (Wireless LAN, ISM band, Satellite), and Cellular network access (4G, 5G). The Device layer includes CUAV, Telecontroller, Mobilephone, and GCS. A Management block on the right contains CUAV management, Flight database management, Wireless access management, and Management. Arrows indicate bidirectional communication between layers and between the Management block and the Service support and application support layer and Network layer.](acfc53eca625d62b38aa2563efa95c3e_img.jpg) + +The diagram illustrates the high-level framework for CUAV communication, organized into four layers: + +- Application layer:** Contains three categories of applications: + - Industry application:** Plant protection, Monitoring, Power line and pipeline inspection, Visual surveillance, ... + - Personal and consumer application (e.g., aerial photography)** + - Other third party application** +- Service support and application support layer:** Contains four functional blocks: + - Flight data service + - Application data service (including audio/video, etc.) + - Flight control session management + - Mission payload application management +- Network layer:** Contains three functional blocks: + - Networking and transport + - Non-cellular network access:** Wireless LAN, ISM band, Satellite + - Cellular network access:** 4G, 5G +- Device layer:** Contains four functional blocks: + - CUAV + - Telecontroller + - Mobilephone + - GCS + +Vertical double-headed arrows connect the layers. A separate **Management** block on the right contains: + +- CUAV management +- Flight database management +- Wireless access management +- Management + +Double-headed arrows also connect the **Service support and application support layer** and the **Network layer** to the **Management** block. A jagged line connects the **Network layer** to the **CUAV** in the **Device layer**. + +Figure 1 – Overview of CUAV communication application high level framework. The diagram shows four layers: Application layer, Service support and application support layer, Network layer, and Device layer. The Application layer includes Industry application (Plant protection, Monitoring, Power line and pipeline inspection, Visual surveillance, ...), Personal and consumer application (e.g., aerial photography), and Other third party application. The Service support and application support layer includes Flight data service, Application data service (including audio/video, etc.), Flight control session management, and Mission payload application management. The Network layer includes Networking and transport, Non-cellular network access (Wireless LAN, ISM band, Satellite), and Cellular network access (4G, 5G). The Device layer includes CUAV, Telecontroller, Mobilephone, and GCS. A Management block on the right contains CUAV management, Flight database management, Wireless access management, and Management. Arrows indicate bidirectional communication between layers and between the Management block and the Service support and application support layer and Network layer. + +F.749.12(20)\_F01 + +**Figure 1 – Overview of CUAV communication application high level framework** + +### 7.1 Application layer + +The application layer comprises three kinds of CUAV applications. They are industry application (e.g., plant protection, monitoring, power line and pipeline inspection, visual surveillance, etc.), personal and consumer application (e.g., aerial photography) as well as other third-party application. + +### 7.2 Service support and application support layer + +The service support and application support layer consists of the following functionality groupings: + +- **Flight control and flight data support functionalities:** These support functionalities fulfil the CUAV real-time flight control, collecting the flight data of CUAV (including position, altitude and speed) to realize the real-time monitoring of CUAV flight and keeping away from flight-prohibition zone, +- **Application support functionality,** such as audio/video (including pictures) service and application data service can be provided to fulfil service as well as the related service management. + +### **7.3 Network layer** + +The network layer consists of the following capabilities: + +- Networking and transport capabilities: provide relevant control functions of network connectivity and transport resource control and connectivity for the transport of CUAV service and application specific data information, as well as the transport of CUAV-related control and management information. +- Network access capabilities: provide relevant control functions of network access, mobility managements or authentication, authorization, etc. +- Satellite communication is an optional communication method when Internet protocol networking is not available. + +### **7.4 Device layer** + +The device layer includes the following four kinds of devices: + +- CUAV +- Mobile phone +- Telecontroller +- GCS (ground control station) + +The mobile phone, telecontroller and GCS, as controlling devices, control CUAVs to fulfil the payload mission. + +### **7.5 Management** + +The management capabilities can be categorized into CUAV management and flight database management. + +- CUAV management is based on the requirements of civil aviation management department. CUAV management completes the registration and management of CUAV users and CUAVs. +- Flight database management entails receiving real-time notification and release of flight-prohibition zone data from civil aviation management department. The flight database can be used by flight control and application. + +# **8 Functional entity of communication application** + +### **8.1 Functional entity framework and reference points** + +The functional entity framework of the CUAV communication application is shown in Figure 2. All independent functional blocks are connected to each other through the reference points. There are three types of reference points in the CUAV communication application functional framework: flight control and flight data reference points, application and service reference points and management reference points. + +![Functional entity framework diagram showing layers: Cloud server, Service, Service functionalities, Network functions, and Device functions. It includes reference points A_sa, A_af, A_sm, A_mn, A_nd, A_un, C_fc, C_fm, C_fn, C_un, C_nd and management interfaces M_us, M_fs, M_an. A legend at the bottom defines arrow colors: black for Control and flight data, red for Management, and blue for Application service.](cfef993dcc8fb513de79eb1f93cf26ae_img.jpg) + +The diagram illustrates the functional entity framework for a cloud server and its associated functions. It is organized into five horizontal layers, each containing specific functional blocks. The layers and their components are as follows: + +- Cloud server:** Contains the 'Applications' block. +- Service:** Contains 'Application data service (including audio/video, etc.)' and 'Flight data service'. +- Service functionalities:** Contains 'Mission payload application management' and 'Flight control session management'. +- Network functions:** Contains 'Networking and transport function', 'Non-cellular network access function', and 'Cellular network access function'. +- Device functions:** Contains 'CUAV function' and 'Flight controller function'. + +Reference points (interfaces) are defined between adjacent layers and components: + +- $A_{sa}$ : Between Cloud server and Service. +- $A_{af}$ : Between Application data service and Flight data service. +- $A_{sm}$ : Between Service and Service functionalities. +- $A_{mn}$ : Between Service functionalities and Network functions. +- $A_{nd}$ : Between Network functions and Device functions. +- $A_{un}$ : Between CUAV function and Network functions. +- $C_{fc}$ : Between Flight data service and Flight control session management. +- $C_{fm}$ : Between Flight control session management and Mission payload application management. +- $C_{fn}$ : Between Flight control session management and Network functions. +- $C_{un}$ : Between CUAV function and Network functions. +- $C_{nd}$ : Between Device functions and Network functions. + +Management interfaces are shown on the right side, connecting to the Management layer: + +- $M_{us}$ : Between Service and Management. +- $M_{fs}$ : Between Service functionalities and Management. +- $M_{an}$ : Between Network functions and Management. + +Management layer components include: Flight database management, CUAV management, Wireless access management, and Management. + +Legend: + +- Black arrow: Control and flight data +- Red arrow: Management +- Blue arrow: Application service + +F.749.12(20)\_F02 + +Functional entity framework diagram showing layers: Cloud server, Service, Service functionalities, Network functions, and Device functions. It includes reference points A\_sa, A\_af, A\_sm, A\_mn, A\_nd, A\_un, C\_fc, C\_fm, C\_fn, C\_un, C\_nd and management interfaces M\_us, M\_fs, M\_an. A legend at the bottom defines arrow colors: black for Control and flight data, red for Management, and blue for Application service. + +**Figure 2 – Functional entity framework and the reference points** + +#### 8.1.1 Control and flight data interface reference points + +- $C_{nd}$ . The reference point $C_{nd}$ is located between the device functions and network functions. It is used for the interaction and transport of flight control command and flight data between the networks and flight controllers. +- $C_{un}$ . The reference point $C_{un}$ is located between the CUAV device functions and network functions. It is used for the interaction and transport of flight control command and flight data between the networks and CUAVs. +- $C_{fn}$ . The reference point $C_{fn}$ is located between the flight control session management functionality and network functions. It is used for the management of flight control session and access control. +- $C_{fc}$ . The reference point $C_{fc}$ is located between the flight data service and flight control session management functionality. It is used for the transport and processing of flight data. +- $C_{fm}$ . The reference point $C_{fm}$ is located between the flight control session management functionality and mission payload application management functionality. It is used for the transport of the specific flight control (including tracking, or along a regular direction or specific area) based on the specific application requirements. + +#### 8.1.2 Application and service interface reference points + +- $A_{un}$ . The reference point $A_{un}$ is located between the CUAV device functions and network functions. It is used for the interaction and transport of application control command and application (including audio and video) data between the networks and CUAVs. + +- And. The reference point And is located between the flight controller function and network functions. It is used for the interaction and transport of application control command and application data between the networks and flight controllers. +- Amn. The reference point Amn is located between the mission payload application management functionality and network functions. It is used for the management of application control session and application data transport. +- Asm. The reference point Asm is located between the application data service and mission payload application management functionality. It is used for the transport and processing of application data. +- Aaf. The reference point Aaf is located between the application data service and flight data service. It is used for the transport of flight data to some application to be used when necessary. +- Asa. The reference point Asa is located between the applications and application data service. It is used for the application data presentation, application instruction processing and application logic execution. + +#### 8.1.3 Management interface reference points + +- Mus. The reference point Mus is located between the management and service. It is used for the transport and management of CUAV devices and flight data. +- Mfs. The reference point Mfs is located between the management and service functionalities. It is used for the management of CUAV flight controlling and mission payload application controlling. +- Man. The reference point Man is located between the management and network functions. It is used for the management of CUAVs' access controlling, authentication and authorization. + +### 8.2 Application + +The main function of the application is to provide interfaces and send requests for communication application procedure and data transport. Application realizes interface operations and display, such as CUAV flight status and location display, flight path, mission payload high-definition video and images presentation through virtual reality (VR) helmet, high-definition display screen for users to watch and use. + +### 8.3 Services + +Services includes flight data service and application data service. + +#### 8.3.1 Flight data service + +Flight data service is a functional module to support the function of interacting with flight control session management and flight database management. Flight data service should complete the following functions: + +- Flight data service shall provide flight-prohibition zone data query for flight control session management, such as electronic fence maps of flight-prohibition zone, densely populated areas, and airports as well as its viewing functions. +- Flight data service shall receive the latest real-time flight notification and the flight-prohibition zone data from the CUAV flight database management server and save the current flight data to the database management server. +- Flight data service shall provide flight data query and application capabilities for application data service, such as analysing user's behaviour, reasonable flight speed, flight altitude and route through "big data" method, so as to provide users and regulatory departments with more safe, feasible and efficient flight mode and supervision reference. + +#### 8.3.2 Application data service + +Application data service supports the function of interacting with applications from application cloud server, and can be used by different applications (e.g., video monitoring, inspection and surveillance, etc.) based on the instructions from mission payload application management. + +Application data service should provide the following functions: + +- Application data service shall support the collection and storage of mission payload data. Cloud server can be used to manage, classify, store, and read application data. +- Application data service should support media processing, such as decomposition and synthesis of standard definition, high-definition videos and images to form a 3D or VR video for application display and playback, and support the storage and distribution of these images and videos. + +### 8.4 Service functionalities + +Service functionalities include flight control session management and mission payload application management. + +#### 8.4.1 Flight control session management + +Flight control session management is a functional module to support the function of interacting with flight data service and CUAV management. Flight control session management shall complete CUAV flight process monitoring and flight control session management. + +- Flight control session management shall review and record the flight plan (e.g., CUAV information, flight time schedule, flight mission, flight area, take-off place, maximum flight altitude, etc.) submitted by the user before each flight, and evaluate the user's flight credit grade to determine whether the flight is permitted (Appendix I gives an example use case of CUAV security flight monitoring based on cellular network). +- Flight control session management shall support real-time monitoring each CUAV and its owner's information, flight position, altitude and speed, flight mission, flight time, etc., and it shall display this information properly, as well as upload them to the flight database management server. +- Flight control session management shall support real-time flight warning in flight-prohibition zone and densely populated area (which can be made into electronic fence map). If CUAVs fly close to or enter electronic fence areas, then the immediate warning information shall be sent to users (for example, notifying or sending messages to users to inform illegal flights with time and locations). +- Flight control session management shall support control of the CUAV flying along specific routes/boundaries at specific speeds and altitudes according to the specific requirements of the application. +- Flight control session management shall support the CUAV login. After registration, the CUAV needs to initiate login request to the flight control session management system before interacting with it. Flight control session management system authenticates the information and login password of the CUAV, then establishes session data link with the CUAV. If there is no data interaction with the management system for a long time, the CUAV needs to send heartbeat information to the management system to maintain the login status. +- Flight control session management shall support the CUAV exit. When the CUAV ends the session with the management system, an exit request needs to be initiated by the CUAV, and the management system updates the CUAV status as an exit. + +#### **8.4.2 Mission payload application management** + +Mission payload application management is a functional module to support the function of interacting with applications data service and flight control session management. + +Mission payload application management should fulfil the following functions: + +- Mission payload application management shall fulfil session management and data synchronization functions for different types of applications, and provide unified query functions for users (including CUAV owners, users and supervisors) and their applications, such as query user information, flight weather, etc. +- Mission payload application management should support the control of mission payload task mode and operation mode according to application requirements. +- Mission payload application management should support the application of network resources with different bandwidth and service levels according to the application requirements. +- If the application needs to fly along the specific route and flight mode, mission payload application management shall support sending specific flight requirements to the flight control session management and obtain timely flight data. +- Mission payload application management should support the ability to dynamically adjust encoding speed and have error tolerance and error correcting coding ability to meet the requirements of transmission of pictures/video. + +### **8.5 Network functions** + +Network functions include networking and transport function as well as the network access function. These functions provide relevant control functions of network connectivity, such as access and transport resource control functions, mobility management or authentication, authorization, and accounting. Transport capabilities focus on providing connectivity for the transport of IoT service and application specific data, as well as the transport of IoT-related control and management information. + +#### **8.5.1 Networking and transport function** + +Networking and transport function shall support flight related data and application related data transport. It can be divided into real-time transport and non-real-time transport. + +- Networking and transport function shall support transporting real-time reliable flight control command and flight data. +- Networking and transport function shall support different transport mode when it has different mission targets, including to transport real-time high definition video/images and non-real-time application data. +- Transport function should support the transport of application data and video/images through satellite. + +#### **8.5.2 Network access function** + +The network access includes cellular 4G/5G and non-cellular access networks, such as industrial scientific medical (ISM) band and WLAN. They shall operate and conform to the regulatory limits and parameters defined in regional and national regulations. + +### **8.6 Device functions** + +CUAV can be controlled by GCS, telecontroller and mobile phones. Correspondingly, the device functions include CUAV function and CUAV flight controller function. These functions should control CUAV to complete the mission payload tasks according to the application requirements. + +#### **8.6.1 CUAV function** + +The following are application requirements of the CUAV function: + +- CUAV functions shall include identification and authentication function (e.g., international mobile subscriber identification (IMSI) number, or other machine-to-machine (M2M) device identifiers). +- Before flight, CUAV function shall support to be activated (based on the location and user information) online to ensure the CUAV operators use "correct geospatial information and flight capabilities". If not, then the flight distance and function of the CUAV will be limited. +- CUAV functions should support to act as the relay sources of cellular network to provide flexible and specific time interval public network access service for remote areas and mountain areas (e.g., when signals are cut off by high mountains). +- When the cellular network base stations or other IoT network base stations are damaged or broken-down, CUAV functions shall support to be used as the operator's temporary base stations for the residents of the affected region and devices to provide emergency network access services. +- When a CUAV is used as an operators' temporary base stations to provide emergency network access services, it shall obtain the location information of the IoT devices/terminals to be served, then inquire whether the devices/terminals are the pre-set service objects before the event, establish a wireless connection with the devices/terminals and provide services after confirmation. +- CUAV functions should support to be acted as a signal transmitter and data receiver for IoT devices (such as geological monitoring devices, farmland and infrastructure monitoring equipment etc.) in remote areas and mountainous areas to send control information (such as status control, start up and shut down) and periodically collect data from these monitoring devices. + +#### **8.6.2 Flight controller function** + +This function includes the following: + +- Flight controllers include telecontroller, mobile phone and GCS. Flight controller functions shall provide support to operate and control the CUAV in the process of cruise flight (for example, flying along the contour line under specific application, etc.), mission execution and landing (return), and to coordinate various related equipment with a unified system clock, as well as to fulfil flight state monitoring, operation mode switch and fault diagnosis and treatment, etc. +- Flight controller functions shall support that multiple flight control modes should be realized for different environmental operations. When a CUAV loses its remote control signals, flight controller functions shall support to hover automatically, and calculate the best back route returning after exceeding a given time. Flight controller functions shall support pressing "one key" returning command. +- Flight controller functions shall support various technologies to realize intelligent flight control, e.g., using artificial intelligence to fulfil the flight control and multi-CUAVs coordinated flight control. + +### **8.7 Management** + +#### **8.7.1 CUAV management** + +CUAV management is a functional module to support the function of interacting with flight control session management and flight database management to realize the management of user and CUAV + +access, including registration, login, exit and cancellation. CUAV management system shall complete the following functions: + +- CUAV management shall provide a function to support user registration. CUAV management system shall provide function to authenticate users and enterprise information (including personal identity, business license, address, contact information, etc.) when they login management system. +- CUAV management shall provide function to support the CUAV registration. CUAV management system shall also provide function to register with the CUAV information (including the number of a CUAV, universal subscriber identification module (USIM) card number, communication module ID, CUAV model, flight control equipment number, category and related technical parameters, etc.) before the first flight. The national registration number is unique and traceable. After the management system checks the relevant information correctly, the CUAV status is recorded as registered. At the same time, the management system generates login password for the CUAV and sends it to the CUAV. Appendix I gives an example of the registration authentication process. + +#### **8.7.2 Flight database management** + +Flight database management is a functional module of CUAV flight management system. It interacts with CUAV management module and flight data service module. Flight database management module shall provide the following functions: + +- The flight database management module shall receive the electronic map data of flight-prohibition zone, densely populated area and airports from the management department, and send the updated notice and data to CUAV flight control system (telecontroller, mobile phone or GCS). +- The flight database management module shall receive and store flight data from the flight data service module. + +#### **8.7.3 Wireless access management** + +The following are functions of the wireless access management: + +- The wireless access management should support management of multiple wireless access network and keep uninterrupted real-time reliable communication of flight control between a CUAV and a GCS or telecontroller with low latency. +- The wireless access management should support management single-hop communication link between a CUAV and a telecontroller or a GCS and multi-hop communication mode among multiple CUAVs using cellular communication mode. + +# Appendix I + +### Example use case of CUAV security flight monitoring based on cellular networks (4G/5G) + +(This appendix does not form an integral part of this Recommendation.) + +Based on the authorized spectrum, the 4G/5G cellular network can provide large bandwidth, low delay, and high reliability communication services, and support the efficient and orderly air traffic management of CUAV in the future. Each CUAV have a unique digital identifier and a serial number which is composed of country code, industry domain code, enterprise name code and enterprise's own definition code. A CUAV shall be registered to the civil aviation management department before flight. Each CUAV may have one or more communication modules associated with the digital identifier. The communication module can optionally use the universal subscriber identification module (USIM) / embedded subscriber identification module (eSIM) card with an IMSI number. + +Figure I.1 illustrates a use case of CUAV security flight monitoring based on cellular network (4G/5G). + +![Sequence diagram of CUAV security flight monitoring based on cellular network (4G/5G).](33a8f3f01dfa8bce75d23017855a13c5_img.jpg) + +The diagram illustrates a sequence of interactions between four entities: Users, CUAV flight control management, CUAV, and Civil aviation management department. The sequence is as follows: + +- [1] Users send a "Real name registration" message to the Civil aviation management department. +- [2] The Civil aviation management department sends "Forward users and CUAVs information" to the CUAV flight control management. +- [3] The Civil aviation management department sends "Transmit real-time notification and flight-prohibition data" to the CUAV flight control management. +- [4] The CUAV flight control management sends "Receive flight-prohibition data" to the Users, who then perform "Take off operation". Simultaneously, the CUAV performs "Take off ready". +- [5] The CUAV initiates an "online authorization request" to the CUAV flight control management, which responds with "Verification (User info., CUAV info. check)". +- [6] The CUAV sends a "Flight data report (High reliability, low delay, wide coverage communication connection)" to the CUAV flight control management, which responds with "Flight monitoring (Avoiding flight-prohibition zone)". +- [7] Both the Users and the CUAV perform "Return/landing". + +Icons representing each entity are shown above their respective lifelines: three people for Users, a cloud with a tower and server for CUAV flight control management, a helicopter for CUAV, and a cloud with a server for the Civil aviation management department. + +Sequence diagram of CUAV security flight monitoring based on cellular network (4G/5G). + +F.749.12(20)\_FI.1 + +Figure I.1 – CUAV security flight monitoring based on cellular network (4G/5G) + +- 1) Users register to the civil aviation management department and send the information of user ID and CUAVs to the civil aviation management department. +- 2) Civil aviation management department save users and CUAVs information and send down this information to the CUAV flight control management. The CUAV flight control management binds user information, communication module IMSI and CUAV information. +- 3) Civil aviation management department transmits real-time flight control notification and flight-prohibition data to the CUAV flight control management. +- 4) Users initiate taking off operation, the CUAVs are ready to communicate with the CUAV flight control management before take-off. +- 5) Before take-off, CUAV initiates online authorization request to the CUAV flight control management. The CUAV flight control management verifies users and CUAVs information to determine whether to take off and sends the verification result to the CUAV. +- 6) If permitted, the CUAV takes off and online reports flight data to the CUAV flight control management via 4G/5G network. The CUAV flight control management continuously monitors the flight of CUAV. If the CUAV approaches the flight-prohibition zone, it will alert the CUAV or the operator. +- 7) If the CUAV enters the flight-prohibition zone, it will force the CUAV to leave or return/landing. + +# Bibliography + +- [b-ITU-T F.749.10] Recommendation ITU-T F.749.10 (2019), *Requirements for communication services of civilian unmanned aerial vehicles*. + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.749.13-202106-I_PDF-E/raw.md b/marked/F/T-REC-F.749.13-202106-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..94173d64e57a7e845962d63cfa50ade8c160282f --- /dev/null +++ b/marked/F/T-REC-F.749.13-202106-I_PDF-E/raw.md @@ -0,0 +1,478 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.749.13** + +(06/2021) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Framework and requirements for civilian +unmanned aerial vehicle flight control using +artificial intelligence** + +Recommendation ITU-T F.749.13 + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +## **TELEGRAPH SERVICE** + +| | | +|-----------------------------------------------------------------|-------------| +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | + +## **MOBILE SERVICE** + +| | | +|--------------------------------------------------------|-------------| +| Mobile services and multdestination satellite services | F.110–F.159 | +|--------------------------------------------------------|-------------| + +## **TELEMATIC SERVICES** + +| | | +|-------------------------------------------|-------------| +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | + +## **MESSAGE HANDLING SERVICES** + +| | | +|--|-------------| +| | F.400–F.499 | +|--|-------------| + +## **DIRECTORY SERVICES** + +| | | +|--|-------------| +| | F.500–F.549 | +|--|-------------| + +## **DOCUMENT COMMUNICATION** + +| | | +|--------------------------------------|-------------| +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | + +## **DATA TRANSMISSION SERVICES** + +| | | +|--|-------------| +| | F.600–F.699 | +|--|-------------| + +## **MULTIMEDIA SERVICES** + +| | | +|--|--------------------| +| | F.700–F.799 | +|--|--------------------| + +## **ISDN SERVICES** + +| | | +|--|-------------| +| | F.800–F.849 | +|--|-------------| + +## **UNIVERSAL PERSONAL TELECOMMUNICATION** + +| | | +|--|-------------| +| | F.850–F.899 | +|--|-------------| + +## **ACCESSIBILITY AND HUMAN FACTORS** + +| | | +|--|-------------| +| | F.900–F.999 | +|--|-------------| + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.749.13 + +# Framework and requirements for civilian unmanned aerial vehicle flight control using artificial intelligence + +## Summary + +Recommendation ITU-T F.749.13 provides a framework of civilian unmanned aerial vehicle flight control using artificial intelligence, including the flight navigation control of a civilian unmanned aerial vehicle (CUAV) itself and the specific flight control based on the vertical industry application requirements. Although CUAVs have been widely used in industry and consumption areas, there are also problems in the development of CUAVs. In addition to the policy and legal supervision, the other problem is how CUAVs avoid obstacles during the flight, and how the CUAVs applied in the vertical industry can automatically navigate, track or fly along a specific area according to the mission requirements, to achieve partial or complete autonomous flight and control. + +This Recommendation also provides a flight control framework and functional requirements for each specific implementation. This way the product and system integrators can design and produce specific products and systems according to this framework. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.749.13 | 2021-06-13 | 16 | 11.1002/1000/14684 | + +## Keywords + +Artificial intelligence, civilian unmanned aerial vehicle, flight control. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2021 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-------------------------------------------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Introduction ..... | 2 | +| 7 Framework of CUAV flight control using AI ..... | 3 | +| 7.1 Flight and application data provision ..... | 4 | +| 7.2 Platform for training model and inference of AI capability ..... | 4 | +| 7.3 Integration of flight control system ..... | 5 | +| 8 Functional requirements for CUAV flight control using AI ..... | 5 | +| 8.1 Requirements for flight and application data provision ..... | 5 | +| 8.2 Requirements for platform for training, modelling and inference of AI capabilities..... | 6 | +| 8.3 Requirements for integration of flight control system ..... | 8 | +| Appendix I – Use case of civilian unmanned aerial vehicle flight control using artificial intelligence..... | 10 | +| Bibliography..... | 12 | + + + +# Recommendation ITU-T F.749.13 + +## Framework and requirements for civilian unmanned aerial vehicle flight control using artificial intelligence + +# 1 Scope + +This Recommendation provides a framework of civilian unmanned aerial vehicle flight control using artificial intelligence (AI), including the flight navigation control of a civilian unmanned aerial vehicle (CUAV) and the specific flight control according to the vertical industry application requirements. The regulations and supervision of civilian unmanned aerial vehicle flights are out of the scope of this Recommendation. + +The scope of this Recommendation includes: + +- Framework of CUAV flight control using AI, +- Functional requirements for CUAV flight control using AI. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 civilian unmanned aerial vehicle** [b-ITU-T F.749.10]: An unmanned flying device controlled by a ground control station or telecontroller via various wireless communication means. It usually consists of an aeroplane body, a power device, aviation electrical and electronic equipment and mission payload equipment, etc. and is used in non-military application areas such as industrial and consumer areas to complete the specific operation and transportation of data including audio, video and image. + +**3.1.2 flight control system** [b-ITU-T F.749.10]: This is the sum of all components and driving devices of instruction transferring, rudder motion. It is mainly composed of airborne and ground control terminals. The one that is airborne includes three parts: airborne sensors, steering engine and flight control units. + +**3.1.3 ground control station** [b-ITU-T F.749.10]: A ground control station is a device which is used to realize the functions of mission planning, flight control, payload control, flight path display, parameter display, image and video display and mission information displaying, recording, and distributing. + +**3.1.4 mission payload equipment** [b-ITU-T F.749.10]: The mission payload equipment consists of an audio/video/image acquisition device, signal relay device, remote electronic detection/sense device, and other auxiliary devices. + +**3.1.5 telecontroller** [b-ITU-T F.749.10]: A piece of equipment used by human beings to control an unmanned aerial vehicle. It is usually composed of an operating device, coding device, transmitting device, receiving device, decoding device, and executing mechanism. + +**3.1.6 artificial intelligence (AI)** [b-ISO/IEC 2382]: An interdisciplinary field, usually regarded as a branch of computer science, dealing with models and systems for the performance of functions generally associated with human intelligence, such as reasoning and learning. + +**3.1.7 capability** [b-ITU-R M.1224-1]: The ability of an item to meet a service demand of given quantitative characteristics under given internal conditions. + +## **3.2 Terms defined in this Recommendation** + +This Recommendation defines the following term: + +**3.2.1 flight control:** It is the whole flight process to realize CUAV gesture stability and control, navigation and path planning, avoiding obstacles and emergency control, etc., including taking off, flying in the air, executing the mission and returning to the field. + +# **4 Abbreviations and acronyms** + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|----------------------------------------------------| +| 4G | The 4th Generation Mobile Communication Technology | +| 5G | The 5th Generation Mobile Communication Technology | +| AI | Artificial Intelligence | +| CPU | Central Processing Unit | +| CUAV | Civilian Unmanned Aerial Vehicle | +| GCS | Ground Control Station | +| GNSS | Global Navigation Satellite System | +| GPU | Graphics Processing Unit | +| HDD | High Definition Display | +| NPU | Neural-network Processing Units | + +# **5 Conventions** + +In this Recommendation: + +- The keywords "is required to" indicate a requirement that must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "is recommended" indicate a requirement that is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. + +# **6 Introduction** + +Currently, with the rapid development of civilian unmanned aerial vehicles (CUAV), these unmanned aerial vehicles are widely used in industry and consumption areas such as agriculture and plant protection, power line and petroleum pipeline inspection, police & traffic security surveillance, disaster monitoring, express delivery, forestry and forest fire monitoring, etc. Although the manual control of CUAV can be realized by telecontroller and ground control station (GCS) in these applications; due to the complexity and uncontrollability of the external environment, or the specific mission requirements, it is still hoped that CUAVs have partial or whole automatic ability of take-off, flight control and landing. Thus, CUAVs need some degree of intelligence, to avoid obstacles, + +prevent collisions, and complete industrial application flight missions. For example, using CUAV to inspect the power line or oil pipeline requires CUAVs to fly along a specific route (capable of automatically navigating and avoiding obstacles) and to automatically adjust the flight height (capable of precise control of the flight state) and speed according to the shape of power line/oil line so that the high definition display (HDD) cameras in CUAV can easily check and diagnose power lines/pipelines. + +At present, artificial intelligence (AI) technology has developed rapidly and has achieved great success in many fields, such as computer vision systems, pattern recognition, image recognition and so on. With the deep application of cloud computing and edge computing, all these provide a strong support platform for the training and modelling of AI. + +This Recommendation provides a framework and functional requirements for each specific implementation, so the product and system integrators can design and produce specific products and systems according to this framework. + +# **7 Framework of CUAV flight control using AI** + +The flight control system of CUAV is a resource-limited system. Therefore, the provision of algorithm and model of flight control need the help of the external system and platform. + +Flight control using AI includes three aspects of data collection, processing and modelling, and application/execution process. Figure 1 presents a framework diagram of CUAV flight control using AI. + +![Figure 1 – Overview of framework diagram of flight control using AI. The diagram is a hierarchical flowchart showing four levels of the system. Level 1 (top): 'Flight and application data provision' with boxes for 'Agriculture plant protection', 'Logistics express delivery', 'Power line and petroleum pipeline inspection', 'Police and traffic security surveillance', 'Aerial photography and video', and '...'. Level 2: 'Platform for training, modelling and inference of AI capability' containing 'Data preservation, classification and management' at the top, followed by two columns of modules. The left column has 'Gesture stabilization and control', 'Common positioning and navigation', 'Flight management', 'Emergency control', and 'Common flight control AI components'. The right column has 'Mission avoiding obstacles', 'Target/boundary recognition', 'Mission path planning', 'Mission track control', and 'Mission specific control AI components'. A double-headed arrow connects the two columns. Level 3: 'Integration of flight control system' with 'Main control module', 'Inertial measurement unit', 'Positioning and navigation module', 'Flight control sensors', 'Specific mission equipment and control software/hardware', and 'Communication and other auxiliary modules'. Level 4 (bottom): 'Deployment in CUAV/GCS/servers' with images of a power line inspection drone, a ground control station laptop, a delivery drone, and a remote control unit. The label 'F.749.13(21)_F01' is in the bottom right corner.](a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg) + +Figure 1 – Overview of framework diagram of flight control using AI. The diagram is a hierarchical flowchart showing four levels of the system. Level 1 (top): 'Flight and application data provision' with boxes for 'Agriculture plant protection', 'Logistics express delivery', 'Power line and petroleum pipeline inspection', 'Police and traffic security surveillance', 'Aerial photography and video', and '...'. Level 2: 'Platform for training, modelling and inference of AI capability' containing 'Data preservation, classification and management' at the top, followed by two columns of modules. The left column has 'Gesture stabilization and control', 'Common positioning and navigation', 'Flight management', 'Emergency control', and 'Common flight control AI components'. The right column has 'Mission avoiding obstacles', 'Target/boundary recognition', 'Mission path planning', 'Mission track control', and 'Mission specific control AI components'. A double-headed arrow connects the two columns. Level 3: 'Integration of flight control system' with 'Main control module', 'Inertial measurement unit', 'Positioning and navigation module', 'Flight control sensors', 'Specific mission equipment and control software/hardware', and 'Communication and other auxiliary modules'. Level 4 (bottom): 'Deployment in CUAV/GCS/servers' with images of a power line inspection drone, a ground control station laptop, a delivery drone, and a remote control unit. The label 'F.749.13(21)\_F01' is in the bottom right corner. + +**Figure 1 – Overview of framework diagram of flight control using AI** + +## 7.1 Flight and application data provision + +AI algorithms, reasoning and modelling require a lot of flight data and vertical industry application flight control data collected by CUAV flight sensors and applications. The sensors include gyroscopes, accelerometers, barometers, global navigation satellite system (GNSS) modules, infrared sensors, laser sensors, ultrasonic sensors, visual sensors, and audio and video equipment carried by the mission payload. Application data from agriculture and plant protection, power line and petroleum pipeline inspection, police and traffic security surveillance, disaster monitoring, express delivery, forestry and forest fire monitoring, etc. will greatly promote the training and modelling of AI in a flight control system. + +## 7.2 Platform for training model and inference of AI capability + +The platform shall collect all kinds of flight control and application data, and the data will be classified, stored and managed. It is necessary to authorize access to this data before using them to provide data security and privacy and rights protection. The application of artificial intelligence in flight control systems requires the help of AI training frameworks (e.g., TensorFlow, etc.) and inference frameworks. They are usually deployed in a platform with central processing unit (CPU), graphics processing unit (GPU) and neural-network processing units (NPU) with required hardware resources for scheduling and using, for the implementation of massive training. The framework can also be a lightweight software framework running on a CUAV side, which can improve the operational efficiency and improve real-time and accuracy of flight control. To respond quickly and + +timely to the flight environment, complete automatic navigation and obstacle avoidance, the dedicated flight control chips and hardware can be deployed on a CUAV to provide basic computational capabilities for automatic navigation inference for vertical industry application scenarios. + +Third-party software/middleware service providers, using an open AI platform, provide modular encapsulation of application-specific flight control algorithms based on large numbers of flight and application data provisions, or provide algorithm tools or middleware for flight control system integrators and application developers. + +Common flight control AI capability components include but are not limited to the following: + +- Gesture stabilization and control +- Common positioning and navigation +- Flight management +- Emergency control + +Mission specific flight control AI capability components include but are not limited to the following: + +- Mission avoiding obstacles +- Target/boundary recognition +- Mission path planning +- Mission track control + +Common flight control capabilities, such as positioning and navigation, and flight management functions will relate to obstacle avoidance and path planning. Flight control for specific missions requires flight gesture stability and navigation. Hence, the classification of common flight control AI capability and mission specific flight control AI capability is not absolute. They interact and cooperate with each other to complete the entire flight process. + +## 7.3 Integration of flight control system + +CUAV manufacturers or flight control system integrators apply the above-mentioned common flight control and mission specific flight control AI capability components, including AI software and algorithms, hardware and chips, to the main control module, inertial measurement unit, positioning and navigation module, flight control sensors, specific mission equipment and control software and hardware, etc. to realize the partial or full-automatic flight control of CUAV. + +# 8 Functional requirements for CUAV flight control using AI + +The flight control in this Recommendation consists of two aspects, one is the common flight navigation control of a CUAV (including gesture stabilization and control, positioning and navigation, normal take-off and landing, etc.); the other is the specific mission flight control (including mission avoiding obstacles, tracking, or along a direction or specific area, etc.) based on the vertical industry application requirements. + +## 8.1 Requirements for flight and application data provision + +The data required by an AI algorithm, reasoning and modelling include but are not limited to the following: + +- **Flight data.** Examples include maximum flat speed, cruise speed, minimum speed, height ceiling, the maximum usage height and minimum flight altitude, flight radius (referring to the distance from taking off to the far most point), pitch and tilt angle, the tilt angle degree, and the yaw angle degree, etc. + +- **Vertical industry application flight data.** Examples include continuation time and maximum continuation time of flight, minimum turning radius, the maximum climbing rate and the maximum descent rate, maximum take-off weight, maximum mission payload equipment weight, positioning and the target location, mission path planning, and track control data, etc. +- **Power system and flight environment data.** Examples include motor speed, temperature, humidity, external wind speed, ambient temperature, power supply, image and video of the working area and environment obtained by mission payload equipment, recognition of specific target and environment, avoidance operation process, alarm operation data, etc. + +## 8.2 Requirements for platform for training, modelling and inference of AI capabilities + +Flight control using AI includes common flight control and mission specific flight control. The platform needs to complete the functions of training, modelling and inference of AI capabilities. + +### 8.2.1 Requirements for common flight control AI components + +Common flight control includes gesture stabilization and control, positioning and navigation, flight management, and emergency control. The AI training framework and platform use all kinds of flight data, combined with geographical fence and flight-prohibition zones data provided by the flight management department, to fulfil the training, modelling, and inference process for common flight control. The final products are the flight control AI software/algorithm, hardware/chips, which can be used by CUAV manufacturers or third-party flight control system integrators. The functional requirements for these AI components, such as software/algorithm, along with hardware/chips integrated with AI software and algorithm, include but are not limited to the following items: + +##### 8.2.1.1 Gesture stabilization and control + +GS-01: CUAV is required to use an inertial measurement unit (flight attitude sensing equipment), air pressure sensor, ultrasonic sensor, etc. to realize CUAV gesture stability and control without GNSS positioning. + +GS-02: CUAV is recommended to have online situational awareness and variable autonomy control capabilities to help achieve semi-automatic/automatic flight functions. + +GS-03: When the autonomous flight of CUAV cannot be controlled by itself or needs to be taken over, it is required to switch to the telecontroller or GCS to take over, to restore the correct posture and state. + +##### 8.2.1.2 Common positioning and navigation + +PN-01: CUAV is required to create two dimensions or three-dimension maps of the environment online or offline according to the environmental information obtained by various airborne sensors. + +PN-02: CUAV navigation system is recommended to realize autonomous flight of designated route according to environment map, position, speed, and flight attitude of reference coordinate system. + +PN-03: CUAV navigation system is recommended to realize navigation through GNSS, inertial guidance, or binocular vision, or use the combination of inertial guidance plus binocular vision plus GNSS. + +PN-04: For a vertical industry application or indoor application, the GNSS is either not available or unable to meet the requirements. For this purpose, multiple small base stations (signal sources) are required to be used. The distances from the three base stations closest to the CUAV selected, along with the identifier and the receiving signal strength of the CUAV and the identifiers of small base stations (including their coordinates) are used to determine the CUAV coordinates. + +##### **8.2.1.3 Flight management** + +FM-01: Flight management function is required to complete CUAV flight process monitoring and flight control session management. + +FM-02: Flight management function is required to record the mission path and its track. + +FM-03: Flight management function is required to support the receiving track information transmitted by CUAV in real-time. The longitude, latitude and altitude data extracted and the corresponding position are then displayed in the map window, so that the operator can see the CUAV position information directly. + +FM-04: Flight management function is required to support data playback mode. After receiving the flight data and saving the data, the whole process of the flight path can be reproduced through the data playback mode, including the position, attitude, and other important data of the CUAV, which is convenient for the analysis of the flight status and mission execution status of the CUAV. + +##### **8.2.1.4 Emergency control** + +EC-01: The flight control function is required to automatically record the flight path and calculate the return route when the control signal is disturbed, or the wireless control link is interrupted and realize the automatic return and landing. + +EC-02: The flight control function is required to support the lock course to control the direction of CUAV when the direction cannot be identified. + +EC-03: The flight control function is required to support the avoidance of the flight-prohibition zone and emergency management flight areas designated by the aviation management department. + +### **8.2.2 Requirements for mission specific control AI components** + +The flight control process of a specific task is based on the common flight control. Therefore, the functional requirements of the two complement each other. The flight control process of the task load needs to be based on positioning and navigation to complete the mission path planning and track control, and at the same time, it is also required to realize the mission avoiding obstacles and target/boundary recognition. So, the functional requirements for these AI components, such as software/algorithm, along with hardware/chips integrated with AI software and algorithm, include but are not limited to the following items: + +##### **8.2.2.1 Mission avoiding obstacles** + +AO-01: CUAV is required to use one or several of infrared sensors, ultrasonic sensors, laser sensors, and visual sensors to sense environmental information and measure the distance of obstacles to avoid obstacles. + +AO-02: CUAV is recommended to use the map model of flight area (e.g., two and three-dimensional map) to reasonably plan the route to bypass obstacles, so as to realize autonomous flight. + +##### **8.2.2.2 Target/boundary recognition** + +The identifications can be done on CUAVs or sent back to ground control stations (GCS) and cloud servers. + +TR-01: For consumption applications, such as aerial photography, CUAV is required to distinguish the target object (such as people) from the environment, and constantly track and identify the moving target. + +TR-02: For plant protection applications, CUAV is required to identify the crops and boundary of the field, e.g., trees, haystacks, electric poles and wires, as well as distinguish the boundary and narrow path between crops and farmlands. + +TR-03: For applications related to the inspection of power lines and oil pipelines, CUAV is required to identify oil pipeline/power lines to monitor and cruise along the power line or oil pipeline and find whether there are any hazards or faults. + +TR-04: For natural disaster monitoring related applications, CUAV is recommended to distinguish avalanche survivors, forest fires, disasters such as drought, flood, landslide as well as aquatic plants from floating garbage, and harmful liquids, etc. + +##### **8.2.2.3 Mission path planning** + +PP-01: CUAV is required to plan flight routes to fly a larger range and accomplish more tasks in a shorter time and cost. + +PP-02: For vertical industry applications, CUAV is required to fly along specific directions and paths. For example, using CUAV to inspect the power line or oil pipeline requires CUAVs to fly along a specific route (capable of automatically navigating and avoiding obstacles) and to automatically adjust the flight height (capable of precise control of the flight state) and speed according to the shape of power line/oil line so that the HDD cameras in CUAV can easily check and diagnose power lines/pipelines. + +PP-03: When multiple CUAVs are flying in a certain formation, the flight control system is required to ensure that the CUAVs in the formation avoid collision and can change the formation according to the defined mission requirements. + +##### **8.2.2.4 Mission track control** + +TC-01: CUAV is recommended to support the use of vector maps to control the track and realize a three-dimensional display with the current track control angle. The estimated total range of the CUAV can be displayed in the information prompt window, to make the track control and operation more intuitive. + +TC-02: Mission track control function is required to support the mouse movements on the digital map. The latitude and longitude information of the point can be displayed at the mouse position, and new navigation points can be added by clicking the mouse to improve the efficiency of route planning. + +## **8.3 Requirements for integration of flight control system** + +Flight control system includes the main control module, inertial measurement unit, positioning and navigation module, flight control sensors, specific mission equipment, control software and hardware, communication, and other auxiliary modules. These modules shall use some or all AI functions to meet the application requirements of different vertical industries in flight gesture control, positioning and navigation, obstacle avoidance, target boundary recognition, mission path planning and track control. The AI functions, algorithms, software/hardware should be implemented in the way of components, from which one or more components can be selected and utilized when integrating into vertical industry application systems. These will improve the integration efficiency of the system and shorten the development time. + +FC-01: The integrated flight control system is required to be deployed to CUAV and GCS, or on the cloud servers. When deployed to the cloud server, it can make full use of the low delay, high bandwidth, and reliability of the 4th generation mobile communication technology / the 5th generation mobile communication technology (4G/5G) network and beyond to realize the data and commands of CUAV download and upload. + +FC-02: The real-time data and images/videos acquired by CUAV sensors are recommended to be transported to the cloud servers through the 4G/5G network and beyond. The cloud shall use these data to complete the training, modelling, and inferencing of AI components. Sometimes, these data can be processed in real-time by the network edge computing system, and the processing results will be sent to the CUAV to control the flight of CUAV. + +FC-03: For a vertical industry application or indoor application, the global navigation satellite system (GNSS) is either not available or unable to meet the requirements. Flight control system is required to interact with small base stations and select a small base station that is closest to the CUAV (the distance reaches a specified value). Then the small base station sends flight control instructions to the CUAV to complete the mission tasks. + +## Appendix I + +## Use case of civilian unmanned aerial vehicle flight control using artificial intelligence + +(This appendix does not form an integral part of this Recommendation.) + +In the process of low altitude plant protection, CUAVs may encounter obstacles such as cables, trees, buildings, etc. Thus, CUAVs must have the function of autonomous obstacle avoidance equipment to identify trees, haystacks, electric poles and wires. So, it is necessary to have an obstacle avoidance equipment such as laser radar, millimetre wave radar, binocular stereo vision, etc. Before spraying pesticides, CUAVs also need to identify crops and field narrow roads in the field. Figure I.1 presents some examples about the specific shape of the task area, path planning, and so on. + +![Figure I.1: Examples of sloping fields, binocular vision and path planning. The figure consists of three parts: top-left shows a CUAV spraying a sloping field; bottom-left shows a CUAV in flight over a field; right shows a software interface with a map and path planning lines.](9d5d3616a843e8063fdc3613055aaef8_img.jpg) + +The figure is composed of three distinct visual elements. The top-left portion is a photograph of a multi-rotor drone in mid-air, positioned over a steep, terraced hillside covered in green crops. The drone is actively spraying a fine mist of liquid. The bottom-left portion is another photograph of a similar drone, viewed from a slightly different angle, flying over a flat, lush green field. The right portion of the figure is a screenshot of a computer-based flight control software. It displays a top-down map view of a field with a complex, pre-planned flight path overlaid. The path is marked by a series of yellow and green lines and numbered waypoints. A sidebar on the right of the map shows various telemetry data, including altitude, speed, and battery level. + +Figure I.1: Examples of sloping fields, binocular vision and path planning. The figure consists of three parts: top-left shows a CUAV spraying a sloping field; bottom-left shows a CUAV in flight over a field; right shows a software interface with a map and path planning lines. + +**Figure I.1 – Examples of sloping fields, binocular vision and path planning** + +At present, many plant protection CUAVs have one or more environmental sensing equipment, such as infrared sensors, ultrasonic sensors, laser sensors, and binocular vision sensors which can measure the distance of obstacles and identify the environment. Many plant protection CUAV manufacturers or other flight control system integrators can use the existing flight control historical data, combined with the data of various sensors collected by CUAV on the fields to practise online (based on cloud computing technology or edge computing, etc.) or offline training, modelling and reasoning of the flight control system. The trained software and hardware can be used to form real-time flight action instructions with partial intelligence of achieving obstacle avoidance and completing mission flight process, as well as to shorten the time for route planning, flying a larger range and completing more tasks with less cost, and realize track control based on high-precision map. For example, through the collection of three-dimensional space XYZ coordinates, six degrees of freedom (6DOF), yaw angle (yaw), roll angle (roll), pitch angle (pitch), Euler angle, and other data, it can provide a coherent and fluent action data basis for CUAV's gesture control and motion planning. When the flight target area is modelled as an accurate three-dimensional map, the CUAV can go skimming flight based on the ultra-low altitude automatic terrain matching and realize automatic obstacle recognition and avoidance, position and gesture control. + +Flight control system using AI can fly along a specific route (capable of automatically navigating and avoiding obstacles) and to automatically adjust the flight height (capable of precise control of the flight state) and speed, according to the shape of terraced fields or other shapes. The CUAV can sprinkle pesticide along a specific shape and at an appropriate height above the specific area, thus avoiding wastage of the pesticide. + +In addition, many CUAV flight control systems have the following functions: + +- Friendly user interface. Flight path and track of CUAV can be displayed graphically. +- In the route planning mode, with the mouse moving on the digital map of the target area, the longitude and latitude data of the point on the map can be displayed at the same time. At this time, click the mouse to automatically add points as a new waypoint in the route. +- After the actual flight process of a CUAV, the data is obtained and saved, the whole flight process of the CUAV can be reproduced through data playback mode. + +# Bibliography + +- [b-ITU-T F.749.10] Recommendation ITU-T F.749.10 (2019), *Requirements for communication services of civilian unmanned aerial vehicles.* +- [b-ISO/IEC 2382] ISO/IEC 2382:2015, *Information technology – Vocabulary.* +- [b-ITU-R M.1224-1] Recommendation ITU-R M.1224-1 (2012), *Vocabulary of terms for International Mobile Telecommunications (IMT.)* + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.749.14-202106-I_PDF-E/raw.md b/marked/F/T-REC-F.749.14-202106-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..3f4f95fc524a595469bebaea91372638076eea2c --- /dev/null +++ b/marked/F/T-REC-F.749.14-202106-I_PDF-E/raw.md @@ -0,0 +1,417 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.749.14** + +(06/2021) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Requirements of coordination for civilian +unmanned aerial vehicles** + +Recommendation ITU-T F.749.14 + +ITU-T + +![ITU logo](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with a white grid pattern, overlaid by a stylized 'ITU' text in blue. + +ITU logo + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +### **TELEGRAPH SERVICE** + +| | | +|-----------------------------------------------------------------|-------------| +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | + +### **MOBILE SERVICE** + +| | | +|--------------------------------------------------------|-------------| +| Mobile services and multdestination satellite services | F.110–F.159 | +|--------------------------------------------------------|-------------| + +## **TELEMATIC SERVICES** + +| | | +|-------------------------------------------|-------------| +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | + +## **MESSAGE HANDLING SERVICES** + +F.400–F.499 + +### **DIRECTORY SERVICES** + +F.500–F.549 + +### **DOCUMENT COMMUNICATION** + +| | | +|--------------------------------------|-------------| +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | + +## **DATA TRANSMISSION SERVICES** + +F.600–F.699 + +## **MULTIMEDIA SERVICES** + +**F.700–F.799** + +## **ISDN SERVICES** + +F.800–F.849 + +## **UNIVERSAL PERSONAL TELECOMMUNICATION** + +F.850–F.899 + +## **ACCESSIBILITY AND HUMAN FACTORS** + +F.900–F.999 + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.749.14 + +# Requirements of coordination for civilian unmanned aerial vehicles + +## Summary + +Civilian unmanned aerial vehicle (CUAV) coordination refers to utilizing the coordination of multiple CUAVs to accomplish specific tasks through network communication. To enable CUAVs to cooperate well and accomplish tasks successfully, three important parts need to be fulfilled. First, a stable and reliable network connection is required to be established among CUAVs to ensure that every CUAV can connect to the others; second, efficient and reliable data transmission among CUAVs is required so as to ensure that the perception information, control information, coordination information and other required information is exchanged among CUAVs; finally, a CUAV collaborative task execution strategy is required to be reasonably worked out during the coordination process in order to improve the efficiency and quality of collaborative task execution. Recommendation ITU-T F.749.14 specifies the requirements of CUAV coordination, including the requirements of network connection among CUAVs, the requirements of data transmission among CUAVs, and the requirements of CUAV collaborative task execution. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.749.14 | 2021-06-13 | 16 | 11.1002/1000/14685 | + +## Keywords + +Civilian unmanned aerial vehicle, coordination, CUAV. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2021 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|---------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Overview of the framework of CUAV coordination..... | 2 | +| 7 Requirements of the network connection ..... | 3 | +| 7.1 Requirements of topology control ..... | 3 | +| 7.2 Requirements of channel resource management ..... | 3 | +| 8 Requirements of data transmission among CUAVs ..... | 3 | +| 8.1 Requirements of channel-coding data ..... | 3 | +| 8.2 Requirements of routing data ..... | 3 | +| 9 Requirements of CUAV collaborative task execution..... | 4 | +| 9.1 Requirements of data sensing and information fusion ..... | 4 | +| 9.2 Requirements of task assignment ..... | 4 | +| 9.3 Requirements of cloud-end collaborative computing..... | 4 | +| Appendix I – Typical scenarios of CUAV coordination..... | 5 | +| I.1 Disaster search and rescue..... | 5 | +| I.2 Environment monitoring ..... | 6 | +| I.3 Network coverage..... | 6 | +| Bibliography..... | 7 | + + + +# Requirements of coordination for civilian unmanned aerial vehicles + +## 1 Scope + +This Recommendation specifies requirements of coordination for civilian unmanned aerial vehicles, including the framework, network connection, data transmission and task processing requirements during the coordination stage. The regulations and supervision of civilian unmanned aerial vehicle flight are out scope of this Recommendation. + +The scope of this Recommendation includes: + +- Requirements of the network connection among civilian unmanned aerial vehicles (CUAVs) +- Requirements of data transmission among CUAVs +- Requirement of CUAV collaborative task execution + +## 2 References + +None. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following term defined elsewhere: + +**3.1.1 civilian unmanned aerial vehicle** [b-ITU-T F.749.10]: An unmanned flying device controlled by a ground control station or telecontroller via various wireless communication means. It usually consists of an aeroplane body, a power device, aviation electrical and electronic equipment and mission payload equipment, etc. and is used in non-military application areas such as industrial and consumer areas to complete the specific operation and transportation of data including audio, video and image. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 civilian unmanned aerial vehicle (CUAV) coordination:** CUAV coordination refers to utilizing the ability of multiple unmanned aerial vehicles (UAVs) to accomplish specific tasks through network communication. + +**3.2.2 position-based routing:** A routing method that does not need to know the topology of the network but makes each forwarder transmit data to its neighbour closest to the destination, according to the position of its neighbours. It is applicable for the scenario where the nodes have high mobility, and the network topology changes drastically. + +**3.2.3 rateless codes:** Channel codes that continuously encode and transmit data at a high rate regardless of the link states. When the receiver accumulates adequate data packets, it can decode the original data successfully. + +**3.2.4 topology-based routing:** A routing method that detects the topology information of the network and selects the optimal routing path from the source to the destination according to this topology information. It is applicable for the scenario where nodes have low mobility and the network topology does not change drastically. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|----------------------------------| +| CUAV | Civilian Unmanned Aerial Vehicle | +| GPS | Global Positioning System | +| QoS | Quality of Service | +| UAV | Unmanned Aerial Vehicle | + +## 5 Conventions + +In this Recommendation: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement needs not be present to claim conformance. + +## 6 Overview of the framework of CUAV coordination + +The framework of CUAV coordination includes three parts of requirements: requirements of network connection, requirements of data transmission and requirements of collaborative task execution. First of all, the reliable network connection among CUAVs is required. This is the basis of CUAV coordination that ensures that every CUAV can connect with each other. In this part, a topology control mechanism and a channel resource management mechanism are required. Second, efficient data transmission is required. This ensures that all kinds of information (control information, coordination information, perception information and other required information) can be exchanged efficiently. In this part, channel-coding data and a routing data are required. Finally, the collaborative task execution is required. This ensures that collaborative tasks can be completed efficiently. In this part, data sensing, information fusion, task assignment and cloud-end collaborative computing are required. The overall framework of CUAV coordination is shown in Figure 6-1. + +![Diagram of the overall framework of CUAV coordination showing three main components: Node connection, Data transmission, and Task execution, centered around a CUAV and connected to a Cloud and other CUAVs.](e1a0d046fbe7f28f5e93a47091851747_img.jpg) + +The diagram illustrates the overall framework of CUAV coordination. At the center is a CUAV (Civilian Unmanned Aerial Vehicle) icon. Above it is a Cloud icon with three small squares representing nodes. Below the central CUAV is another CUAV icon. To the left of the central CUAV is a box labeled 'Node connection' containing 'Topology control' and 'Channel resource management'. To the right is a box labeled 'Task execution' containing 'Data sensing and information fusion', 'Task assignment', and 'Cloud-end collaborative computing'. Below the central CUAV is a box labeled 'Data transmission' containing 'Channel-coding data' and 'Routing data'. Dashed lines connect the central CUAV to the Cloud, the left 'Node connection' box, the right 'Task execution' box, and the bottom 'Data transmission' box. The left and right CUAV icons are also connected to the central CUAV by dashed lines. + +| Node connection | +|-----------------------------| +| Topology control | +| Channel resource management | + +| Task execution | +|-------------------------------------| +| Data sensing and information fusion | +| Task assignment | +| Cloud-end collaborative computing | + +| Data transmission | +|---------------------| +| Channel-coding data | +| Routing data | + +Diagram of the overall framework of CUAV coordination showing three main components: Node connection, Data transmission, and Task execution, centered around a CUAV and connected to a Cloud and other CUAVs. + +F.749.14(21)\_F6-1 + +Figure 6-1 – The overall framework of CUAV coordination + +## **7 Requirements of the network connection** + +### **7.1 Requirements of topology control** + +TC-01: CUAVs are required to know their own information such as residual energy, moving speed, maximum movable distance, maximum load, etc. + +TC-02: CUAVs are required to be aware of the surrounding environment and obtain external environmental conditions such as terrain, wind direction, hydrology, obstacle distribution, etc. + +TC-03: CUAVs are required to use their own information (such as residual energy, moving speed, maximum movable distance, maximum load, etc.) and surrounding environment information (such as terrain, wind direction, hydrology, obstacle distribution, etc.) to actively control topology, and actively optimize the performance of network connection. + +### **7.2 Requirements of channel resource management** + +CRM-01: A channel resource management mechanism is required, which supports channel resources (such as spectrum resources) management, and collision avoidance when multiple CUAVs communicate with each other using 4G/5G, wireless LAN or other wireless communication technologies. + +CRM-02: The channel resource management mechanism is recommended to be distributed, which has low delay and low overhead. + +CRM-03: It is recommended that the control channel only transmits control information and the data channel only transmits data information. This can ensure the efficiency and reliability of the control information. + +CRM-04: It is recommended to group multiple CUAVs according to their types, locations, coverage, and wireless access methods. Different groups of CUAVs are required to use different control channels to reduce mutual interference and improve collaboration efficiency and security. + +## **8 Requirements of data transmission among CUAVs** + +### **8.1 Requirements of channel-coding data** + +CCD-01: In a one-to-one transmission scenario, it is recommended to use rateless codes as channel codes. This is because rateless codes can adapt the dynamic change of link qualities with time, guarantee the reliability of data transmission and maximize the efficiency of data transmission. + +CCD-02: In a one-to-many transmission scenario, it is recommended to use rateless codes as channel codes. This is because rateless codes can adapt different qualities of multiple links and make a single CUAV transmitter meet different data receiving requirements of multiple CUAV receivers without changing rateless coding parameters. + +CCD-03: In a one-to-many transmission scenario, it is recommended to set different rateless codes parameters for different CUAV receivers with different communication capacity or computing capacity, which can meet different QoS requirements (throughput, reliability, delay time, etc.) of different CUAV receivers and different application services. + +### **8.2 Requirements of routing data** + +RD-01: In some application scenarios (such as disaster search and rescue, target tracking, etc.), CUAVs usually have high mobility, and as a result, the topology of the CUAV network may change drastically. So, it is recommended to use position-based routing in these application scenarios as position-based routing does not need to detect and maintain the topology of the network, but makes each forwarder transmit data to its neighbour closest to the destination according to the position (acquired by GPS services) of its neighbours. + +RD-02: In some application scenarios (such as environmental monitoring, network coverage, etc.), CUAVs have low mobility and the topology of the CUAV network in these application scenarios does not change drastically. So, it is recommended to use topology-based routing. The topology-based routing selects the optimal routing path from the source to the destination according to the network topology information. + +RD-03: It is recommended to introduce multiple parameters (such as position change, energy surplus, computing capacity, communication capacity, etc.) of CUAV as evaluation indicators for selecting the path from the source to the destination. + +## **9 Requirements of CUAV collaborative task execution** + +### **9.1 Requirements of data sensing and information fusion** + +DSIF-01: To eliminate various defects in a single type of sensor, it is recommended that CUAVs use multiple kinds of sensors (including vision sensors, lidar, ultrasonic radar, infrared radar, millimetre-wave radar, etc.) to sense the various types of information of the environment and tasks. + +DSIF-02: To enable CUAVs to analyse and recognize various changes in the dynamic environment and task requirements accurately, it is recommended to perform unified and efficient information fusion of various types of heterogeneous information sensed by multiple sensors on multiple CUAVs. + +DSIF-03: It is recommended to adopt a variety of dynamic and adjustable information fusion methods according to the dynamic external environment, various task types, different dynamic task requirements, and heterogeneous node resources. + +### **9.2 Requirements of task assignment** + +TA-01: Task assignment requires considering the dynamicity and the uncertainty of the external environment, task requirement change, the heterogeneity of CUAV resources, task assignment efficiency, task assignment delay and other constraints. + +TA-02: Before tasks start, it is recommended to perform offline task assignments for each CUAV according to the known task requirements and CUAV resources. + +TA-03: During the execution of tasks, it is recommended to perform the real-time online evaluation of the external environment, task requirements and the CUAV status. + +TA-04: According to the evaluation results, it is recommended to perform real-time online task adjustments for each CUAV. + +TA-05: For simple task scenarios, it is recommended to use a centralized task assignment method to obtain global optimal solutions. + +TA-06: For complex task scenarios with high uncertainty, it is recommended to use a distributed task assignment method to improve the efficiency of the task assignment. + +### **9.3 Requirements of cloud-end collaborative computing** + +CECC-01: For simple computing tasks, it is recommended that computation be done locally on the CUAV to reduce the computing delay. + +CECC-02: For complex computing tasks, it is recommended that computation be done on the cloud platform to reduce the computing and storage overhead of CUAVs. + +CECC-03: The computing task offloading from CUAV to the cloud platform will introduce communication overhead and corresponding delays. "Cloud-end" collaborative computing requires balancing the calculation cost, communication cost, communication delay and other performance indicators. + +## Appendix I + +### Typical scenarios of CUAV coordination + +(This appendix does not form an integral part of this Recommendation.) + +With the tremendous progress in the research of civilian unmanned aerial vehicles (CUAVs) technology, CUAVs are not only widely used in military, but also widely used in civilian and commercial fields. In commercial or civilian applications, CUAVs can be used for disaster search and rescue, environment monitoring, target tracking, network coverage, etc. As the workload and complexity of tasks in these scenarios continue to increase, using a single CUAV to perform tasks will increase the burden of this single CUAV in terms of load, communication, computing, storage, and endurance. In contrast, using multiple CUAVs in a cooperative way to accomplish tasks can not only improve coverage and range, but also reduce the risk of mission failure. In addition, CUAV coordination can effectively improve the system stability and the task execution efficiency. This clause describes various scenarios for CUAV coordination. + +### I.1 Disaster search and rescue + +In this scenario, multiple unmanned aerial vehicles are used to search and detect single or multiple targets (stationary or mobile) and keep track of the state of the target(s) to facilitate rescue victims in efficient time. See Figure I.1 and Figure I.2. Due to the time critical nature and the mobility of targets, mission response time is the most important. Thus, searching for different targets in different areas using multiple CUAVs in a collaborative manner is required, and can greatly shorten the search time. In this process, multiple CUAVs require network communication to exchange information with each other. Therefore, a long range communication and multi-hop connectivity are required in such scenarios. + +![Figure I.1: CUAVs are used to search and detect targets in disasters. The image shows two UAVs flying over a body of water, with one UAV pointing a red laser at a small boat labeled 'USVs' and the other pointing a green laser at a person in the water. A tree is visible in the foreground on the right.](2de51a31da9b0ffd9d017f035d498997_img.jpg) + +Figure I.1: CUAVs are used to search and detect targets in disasters. The image shows two UAVs flying over a body of water, with one UAV pointing a red laser at a small boat labeled 'USVs' and the other pointing a green laser at a person in the water. A tree is visible in the foreground on the right. + +Figure I.1 – CUAVs are used to search and detect targets in disasters + +![Figure I.2: CUAVs keep track of the target(s) until victims are rescued. The image shows a single UAV flying over a body of water at sunset, with a red laser pointing at a person in the water. A city skyline is visible in the background.](8c49beafe7c55ea578abcaee4bdca0ba_img.jpg) + +Figure I.2: CUAVs keep track of the target(s) until victims are rescued. The image shows a single UAV flying over a body of water at sunset, with a red laser pointing at a person in the water. A city skyline is visible in the background. + +Figure I.2 – CUAVs keep track of the target(s) until victims are rescued + +### I.2 Environment monitoring + +In this scenario, multiple CUAVs sense and collect environmental information (such as temperature information and humidity information). CUAVs should carry these data directly to the sink node or transmit these data to the sink node through multiple hops. If the detected area is large, communication range may be limited. Thus, multiple CUAVs are required to work in a cooperative way, and multi-hop communication is required to enable continuous connectivity between these CUAVs. See Figure I.3. + +![Three CUAVs flying over a mountainous landscape, each with a yellow cone representing its sensor coverage area.](e3b8510f6a2194e250205ab7bc38076d_img.jpg) + +A photograph showing three black quadcopter drones flying in a triangular formation over a lush green mountainous landscape under a blue sky with white clouds. Each drone has a yellow semi-transparent cone extending downwards, representing its field of view or sensor coverage area over the terrain. + +Three CUAVs flying over a mountainous landscape, each with a yellow cone representing its sensor coverage area. + +**Figure I.3 – CUAVs monitor the environment** + +### I.3 Network coverage + +In this scenario, CUAVs are used as air base stations to provide network coverage to users in disasters where ground network infrastructures are seriously destroyed, or in big events with a large number of audiences (such as a football match or an outdoor concert). These CUAVs connect the base station to the users' device in a multi-hop manner. See Figure I.4. + +![A CUAV flying over a city, with yellow cones representing its network coverage area.](7b4218684dabfd7cc15685f82dbe5faf_img.jpg) + +A photograph of a white and black drone flying over a densely built urban area. The drone is positioned centrally, with three yellow semi-transparent cones extending downwards to different parts of the city, illustrating its network coverage area. Other smaller drones are visible in the background. + +A CUAV flying over a city, with yellow cones representing its network coverage area. + +**Figure I.4 – CUAVs provide network coverage** + +## Bibliography + +- [b-ITU-T F.749.10] Recommendation ITU-T F.749.10 (2019), *Requirements for communication services of civilian unmanned aerial vehicle.* + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.749.15-202203-I_PDF-E/raw.md b/marked/F/T-REC-F.749.15-202203-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..a2d3393a2a8c5c742e36e883be6553af0cbcb8ac --- /dev/null +++ b/marked/F/T-REC-F.749.15-202203-I_PDF-E/raw.md @@ -0,0 +1,518 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.749.15** + +(03/2022) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Requirements for inspection and examination +services using civilian unmanned aerial vehicles** + +Recommendation ITU-T F.749.15 + +ITU-T + +![ITU logo](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with a white grid pattern, overlaid by a stylized 'ITU' text in blue. + +ITU logo + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multidestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +## Recommendation ITU-T F.749.15 + +# Requirements for inspection and examination services using civilian unmanned aerial vehicles + +## Summary + +Recommendation ITU-T F.749.15 specifies requirements for high-definition (HD) and virtual reality inspection and examination (IaE) services by human beings using civilian unmanned aerial vehicles (CUAVs), including those for fore-end devices capturing HD images and videos of objects and surroundings, network communication, service and application support, as well as service presentation and playback. There are many requirements for careful IaE of specific objects or facilities and surroundings, such as power lines, oil pipelines, bridges and viaducts on water and land, rivers and lakes or inspection of areas subject to emergency or disaster. With the help of HD cameras, panoramic cameras and other sensors (such as infrared sensors), a CUAV can easily accomplish IaE tasks quickly and efficiently with high quality. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.749.15 | 2022-03-16 | 16 | 11.1002/1000/14964 | + +## Keywords + +Civilian unmanned aerial vehicle, inspection and examination, requirements. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2022 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|------------------------------------------------------------------------------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Introduction ..... | 2 | +| 7 Functional diagram of inspection and examination services ..... | 3 | +| 8 Requirements for inspection and examination services..... | 4 | +| 8.1 Requirements for fore-end devices..... | 4 | +| 8.2 Requirements for network communication ..... | 6 | +| 8.3 Requirements for service and application support ..... | 7 | +| 8.4 Requirements for service presentation and playback ..... | 8 | +| Appendix I – Examples of high-definition and virtual reality inspection and examination services using civilian unmanned aerial vehicles ..... | 10 | +| Bibliography..... | 13 | + + + +# Requirements for inspection and examination services using civilian unmanned aerial vehicles + +# 1 Scope + +This Recommendation specifies requirements for high-definition (HD) and virtual reality (VR) inspection and examination (IaE) services by human beings using civilian unmanned aerial vehicles (CUAVs), including those for fore-end devices capturing HD images and videos of objects and surroundings, network communication, service and application support, and service presentation requirements. IaE tasks completed by machine vision lie outside the scope of this Recommendation. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T F.749.12] Recommendation ITU-T F.749.12 (2020), *Framework for communication application of civilian unmanned aerial vehicles*. + +[ITU-T H.264] Recommendation ITU-T H.264 (2021), *Advanced video coding for generic audiovisual services*. + +[ITU-T H.265] Recommendation ITU-T H.265 (2021), *High efficiency video coding*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 civilian unmanned aerial vehicle** [b-ITU-T F.749.10]: An unmanned flying device controlled by a ground control station or telecontroller via various wireless communication means. It usually consists of an aeroplane body, a power device, aviation electrical and electronic equipment and mission payload equipment, etc. and is used in non-military application areas such as industrial and consumer areas to complete the specific operation and transportation of data including audio, video and image. + +**3.1.2 flight control system** [b-ITU-T F.749.10]: This is the sum of all components and driving devices of instruction transferring, rudder motion. It is mainly composed of airborne and ground control terminals. The one that is airborne includes three parts: airborne sensors, steering engine and flight control units. + +**3.1.3 ground control station** [b-ITU-T F.749.10]: A ground control station is a device which is used to realize the functions of mission planning, flight control, payload control, flight path display, parameter display, image and video display and mission information displaying, recording and distributing. + +**3.1.4 mission payload equipment** [b-ITU-T F.749.10]: The mission payload equipment consists of an audio/video/image acquisition device, signal relay device, remote electronic detection/sense device and other auxiliary devices. + +**3.1.5 telecontroller** [b-ITU-T F.749.10]: A piece of equipment used by human beings to control an unmanned aerial vehicle. It is usually composed of an operating device, coding device, transmitting device, receiving device, decoding device and executing mechanism. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following term: + +**3.2.1 inspection and examination (IaE)**: A process to carefully check and look over specific objects / facilities and surroundings to determine whether these objects and surroundings are in a normal, damaged, dangerous or defective state. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|----------|------------------------------------------| +| CUAV | Civilian Unmanned Aerial Vehicle | +| DASH | Dynamic Adaptive Streaming over HTTP | +| GCS | Ground Control Station | +| HD | High Definition | +| HTTP | Hypertext Transfer Protocol | +| IaE | Inspection and Examination | +| IMT-2020 | International Mobile Telecom System-2020 | +| ISM | Industrial Scientific Medical | +| TV | Television | +| VR | Virtual Reality | + +# 5 Conventions + +In this Recommendation: + +- The phrase "is required to" indicates a requirement that must be strictly followed and from which no deviation is permitted if conformity to this Recommendation is to be claimed. +- The phrase "is recommended" indicates a requirement that is recommended but which is not absolutely required. Thus, this requirement needs not be present to claim conformity. +- The phrases "can optionally" and "may" indicate an optional requirement that is permissible, without implying any sense of being recommended. These terms are not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator or service provider. Rather, it means the vendor may optionally provide the feature and still claim conformity with the specification. + +# 6 Introduction + +There are many requirements for careful IaE of specific objects or facilities and surroundings to determine whether they are in a normal, damaged, dangerous or defective state. For example, IaE of hundreds of kilometres power lines and oil pipelines are needed to see whether they are aging or damaged, corroded or leaking, etc. IaE of bridges and viaducts over water and land are also needed regularly to check their appearance and structure, whether cracks exist and bridge decks, piers, + +abutments are damaged. The workload in these processes is not only time-consuming and laborious with a long cycle, but also very unsafe in some situations or not suitable for human beings to complete. With the help of HD cameras and other sensors (such as infrared sensors), objects or facilities can be carefully observed over the long term, or historical images or data can be compared with current images and data. CUAVs can easily accomplish inspection work quickly and efficiently with high quality. It is also necessary to observe whether any water pollution, illegal closure and occupation is affecting wide rivers, lakes and seas over long distances. This can be realized using a panoramic camera or a group of cameras deployed at certain positions carried by CUAVs. The collected data and video or images are transported back to an IaE service management system through a network and displayed or presented on an HD display screen or VR terminal. + +# 7 Functional diagram of inspection and examination services + +Based on the system requirements for IaE service in [ITU-T F.749.12], Figure 1 is a functional diagram of IaE service based on CUAV use. + +![Functional diagram of inspection and examination service using a civilian unmanned aerial vehicle. The diagram shows four layers: Application layer (Presentation and playback) with VR terminal, HD TV, and Computer; Service and application support layer with Audio and video processing, Inspection and examination service management, and CUAV management; Network layer with Networking and transport, Non-cellular network access (Satellite, ISM band), and Cellular network access (IMT-2020); and Fore-end device layer with CUAV, Panoramic cameras, Telecontroller/GCS, and Other mission payload and auxiliary equipment. Double-headed arrows connect the layers, and zigzag lines connect the Network layer to the Fore-end device layer.](b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg) + +The diagram illustrates the functional architecture of an IaE service based on CUAV use, organized into four layers: + +- Application layer (Presentation and playback):** This top layer includes three user interface options: VR terminal, HD TV, and Computer. +- Service and application support layer:** This layer contains three main components: Audio and video processing, Inspection and examination service management, and CUAV management. +- Network layer:** This layer is responsible for Networking and transport. It is divided into two access methods: Non-cellular network access (which includes Satellite and ISM band) and Cellular network access (which includes IMT-2020). +- Fore-end device layer:** This bottom layer includes the CUAV, Panoramic cameras, Telecontroller/GCS, and Other mission payload and auxiliary equipment. + +The layers are interconnected by double-headed vertical arrows, indicating bidirectional communication. The Network layer and the Fore-end device layer are connected by zigzag lines, representing a wireless link. + +Functional diagram of inspection and examination service using a civilian unmanned aerial vehicle. The diagram shows four layers: Application layer (Presentation and playback) with VR terminal, HD TV, and Computer; Service and application support layer with Audio and video processing, Inspection and examination service management, and CUAV management; Network layer with Networking and transport, Non-cellular network access (Satellite, ISM band), and Cellular network access (IMT-2020); and Fore-end device layer with CUAV, Panoramic cameras, Telecontroller/GCS, and Other mission payload and auxiliary equipment. Double-headed arrows connect the layers, and zigzag lines connect the Network layer to the Fore-end device layer. + +F.749.15(22) + +**Figure 1 – Functional diagram of inspection and examination service using a civilian unmanned aerial vehicle** + +CUAV-based IaE service can be functionally divided into four levels: fore-end device layer; network layer; service and application support layer; and application layer. In terms of implementation methods, the whole service can be processed based on the cloud platform, including audio and video coding and decoding; alternatively, multiple sub images are spliced, projected, rotated, etc. to become a complete video or image. Finally, presentation and playback of IaE objects are completed by HD display equipment and a VR terminal. + +# **8 Requirements for inspection and examination services** + +## **8.1 Requirements for fore-end devices** + +Fore-end devices include a CUAV and telecontroller/ground control station (GCS), panoramic camera as well as other mission loads and auxiliary equipment. + +### **8.1.1 Requirements for the civilian unmanned aerial vehicle** + +#### **8.1.1.1 Flight test capability** + +FT-01: Protection of a CUAV is required not to be lower than a certain level. + +FT-02: At a higher altitude, the hovering time of a CUAV during normal operation is required not to be less than 20 min. + +FT-03: A CUAV is required to fly stably under light rain conditions, and the flight time to be not less than a certain value. + +FT-04: A CUAV is required to work and hover normally under the environmental conditions of instantaneous wind speed not greater than 10 m/s (level 5 wind). + +FT-05: During a self-inspection test, a CUAV is required to give alarm prompt (e.g., sound or light signal or other means), and the flight control system is locked. + +#### **8.1.1.2 Flight performance and function** + +FP-01: A CUAV is required to communicate and interact with the telecontroller or GCS to complete the its flight and state control, and can transmit the data, video or images obtained by mission payload devices back to the IaE service management system. + +FP-02: A CUAV is required to transport flight data (including flight altitude, speed, flight trajectory coordinates, flight attitude, residual energy and flight time) to the GCS and IaE service management system. + +FP-03: When a CUAV loses the remote control signal, it is required to support automatic hovering, calculate the best return route after exceeding the given time, and automatically return and land. + +FP-04: The deviation of navigation and positioning is recommended to not be greater than 1.5 m in the horizontal direction and 3 m in the vertical direction. + +FP-05: Hovering control deviation is recommended to be no more than 1.5 m in the horizontal direction with a standard deviation of no more than 0.75 m, and no more than 2 m in the vertical direction with a standard deviation of no more than 1 m. + +FP-06: When the flight speed exceeds 5 m/s, a CUAV is required to detect obstacles such as conductors and towers, within a 20 m radius. + +### **8.1.2 Requirements for the telecontroller/ground control station** + +The functions of a flight controller (including telecontroller and mobile phone) and GCS shall support the operation and control of a CUAV, including cruise flight control, mission specific flight control and landing (return), complete flight status monitoring, operation mode switching, fault diagnosis and processing. + +#### **8.1.2.1 Telecontroller** + +TE-01: A telecontroller is required to remotely control the CUAV within the visual range, and international mobile telecom system-2020 (IMT-2020) network or an industrial-scientific-medical (ISM) band can be selected to control the CUAV. + +TE-02: A telecontroller is required to have its own flight display screen to support the display of the CUAV flight trajectory on an electronic map, and be able to control the flight direction, flight speed, take-off and landing of the CUAV. + +TE-03: A telecontroller is required to operate and control the mission payload equipment to complete specific flight tasks, and can be set according to requirements to display the audio and video or images captured by the mission payload and the data obtained by the sensors on the display screen. + +TE-04: A telecontroller is required to display its remaining power and the remaining flight time of the CUAV. + +#### **8.1.2.2 Ground control station** + +GCS-01: A GCS is required to integrate and display flight parameters and mission parameters, which can be used by operators to analyse and judge the flight status of a CUAV and the status of mission load equipment. Also, a GCS can remotely operate and control the payload equipment carried by a CUAV to complete the mission tasks. + +GCS-02: A GCS is recommended to support the automatic or manual generation of a task plan. + +GCS-03: A GCS is required to display flight heading and track parameters in real time, including flight altitude, speed, flight track coordinates, flight attitude, flight time and navigation parameters of a CUAV. It is required to support set waypoint parameters and routes on the map (the number of editable waypoints to be not less than 50). It is required to support the function of automatic map roaming, scheduled flight track and real-time flight track display of a CUAV. + +GCS-04: A GCS is required to support multiple flight control modes for different environments. When the control signal is disturbed, "one key" return command shall be supported. + +### **8.1.3 Requirements for the generation of panoramic views** + +There are two ways of generating panoramic views. The first deploys a group of cameras at certain positions, and the field of view can be 180° or 360°. The original images collected by these wide-angle cameras are processed by software to eliminate some distortions, and then a panoramic video is synthesized according to the visual habits of the human eye. The second is the panoramic camera. Users can choose either according to their requirements and the application scenario. + +#### **8.1.3.1 Multi cameras acquisition and synchronization** + +AS-01: It is recommended that the number of cameras be between four and eight, and each channel correspond to one video stream, and select one channel with its video stream and audio stream (on-site pickup) synchronization. Control the cameras to power on and start synchronously to keep the interframe error of the multi-channel stream as small as possible (< 100 ms). If the synchronization error of multi-channel video continues to expand after long time streaming, it needs to be calibrated regularly. + +#### **8.1.3.2 Panoramic camera** + +PC-01: It is recommended that a panoramic camera adopt internal system on chip splicing, and support three-dimensional noise reduction as well as ITU-T H.264 coding baseline/main profile/high profile, ITU-T H.265 coding main profile. + +PC-02: It is recommended that mainstream frame rate and resolution support 50 to 60 frames/s or 20 to 30 frames/s and 3840\*2160, 1920\*1080 and 1280\*720, respectively. If a substream is enabled, then 20 to 30 frames/ s and 1280\*720, 640\*480 and 320\*240, respectively. + +### **8.1.4 Requirements for other mission payload and auxiliary equipment** + +The IaE equipment carried by a CUAV includes the following types, which can be selected according to actual requirements. + +MP-01: An HD camera is recommended to meet the application scene requirements of resolution, field angle, etc. Video and audio encoding and decoding technologies such as those specified in [ITU-T H.264] and [ITU-T H.265] are recommended, supporting frame rate and resolution of 50 to 60 frames/s or 20 to 30 frames/s and 3840\*2160, 1920\*1080 and 1280\*720, etc., respectively. + +MP-02: An infrared camera is recommended to meet the application scene requirements of working wavelength, resolution and automatic focusing, etc. It is recommended that the effective count shall not be less than 300 kpixel, and the image taken at a distance of not less than 10 m can clearly identify the fault hot spot. + +MP-03: It is recommended that payload equipment be remotely controlled for switching, recording, amplification, focusing, aperture, etc., and supporting the adjustment of pan/tilt/zoom, horizontal rotation speed and reverse direction. + +## **8.2 Requirements for network communication** + +A CUAV-based IaE service is required to use a wireless communication network to control CUAV flight and remotely operate the mission payload as well as transport audio and video. Wireless networks are required to have high bandwidth and low delay, and provide relevant control and resource reservation of network connection, access and transmission resource control, mobility management or authentication, authorization and accounting. Multiple wireless communication modes are required between a CUAV and telecontroller/GCS to maintain uninterrupted, real-time and reliable communication and transport. Means such as IMT-2020, ISM band and satellite can be used to control CUAV flight and transport data and audio or video information obtained by the mission payload. + +### **8.2.1 Requirements for networking and transport** + +NT-01: Network and transmission functions are required to support real-time and reliable flight control and flight data transport. Before transport service starts, it is necessary to establish secure transport channels for data and audio and video images, and conduct two-way authentication. + +NT-02: When IMT-2020 and ISM bands are used, the image transport bandwidth is recommended to be $> 2$ Mbit/s and the image transport delay $< 1$ s. The data transport delay is $< 500$ ms, and the bit error rate is $\leq 1 \times 10^{-6}$ . + +### **8.2.2 Requirements for non-cellular network access** + +NC-01: A non-cellular network can optionally support using satellite to access and transport application data and video or images. + +NC-02: It can optionally support the use of an ISM frequency band for flight control communication of CUAV and audio and video data transport. + +### **8.2.3 Requirements for cellular network access** + +CN-01: Establishment of a direct command and control link is required to support the CUAV controller and CUAV to communicate with each other and both are registered on the IMT-2020 network using a radio resource configured and scheduled provided by that network for their communication. + +CN-02: Support for the CUAV controller and CUAV register is required, as well as establishment of the corresponding unicast command and control communication links to the IMT-2020 network. Also, both the CUAV controller and CUAV may be registered on the IMT-2020 network via different radio access network nodes. The IMT-2020 network needs to support a mechanism to handle the reliable routing of communication. + +CN-03: It is recommended that CUAV controllers based on IMT-2020 network can support control of one or more CUAVs. + +## **8.3 Requirements for service and application support** + +The audio or video images and data obtained by fore-end devices are processed, and the corresponding objects or facilities and surroundings directly presented on the given interface (such as an HD television (TV) screen and computer screen). The inspector judges the state of these objects, or the inspector can see the surroundings all-round through VR glasses to experience the environment and objects. + +### **8.3.1 Audio and video processing** + +Videos and images captured by HD and infrared cameras can be viewed in real time or replayed at the back end. + +#### **8.3.1.1 High-definition video** + +VI-01: High-definition video is required to support audio and video storage in mainstream formats. + +VI-02: High-definition video is recommended to adopt video and audio coding technologies such as those specified in [ITU-T H.264] and [ITU-T H.265] for encoding and decoding formats, as well as support for frame rate and resolutions such as 50 to 60 frames/s or 20 to 30 frames/s and 3840\*2160, 1920\*1080 and 1280\*720, respectively. + +VI-03: For the same input, high-definition video can optionally support output of multi-rate code streams (4K, 1080P and 720P) at the same time. + +#### **8.3.1.2 Synthesis of multi view video or audio as a panoramic video** + +SY-01: It is recommended first to receive the video and audio data collected by the fore-end devices and to save them. Second the streaming server is required to complete the splicing, projection, rotation, etc. of multiple sub images to become a complete video or image, and the audio data is encoded directly. According to the configuration of the terminal, multi view video rendering can also be achieved on the streaming server to reduce the requirements on the VR terminal. Finally, the encoded video and audio data are encapsulated in the streaming server. The packaging format and storage format of multi view video is recommended to support MP4 and dynamic adaptive streaming over hypertext transfer protocol (HTTP) (DASH), etc. + +### **8.3.2 Inspection and examination service management** + +CUAV-based IaE can generally be divided into daily IaE, dedicated IaE or construction status IaE. + +Daily IaE: It is recommended that daily IaE can conduct a large-scale, non-stop inspection based on pre-set route flight, and save the obtained data and video on the IaE server. The inspector can check and compare in real time or review, summarize and mark these data or images for further inspection and verification. + +Dedicated IaE: It is recommended that for key locations and parts, suspicious points found in daily IaE, local special IaE can be carried out for rapid inspection, disposal and decision-making, such as obtaining evidence for certain behaviours and events (e.g., oil theft) or early warning (e.g., about pipeline or tank temperature, accurate location of leakage point or rapid fire detection). + +Construction status IaE: It is recommended that construction status IaE can finish construction information collection, construction progress management, construction supervision of power lines or oil pipelines and bridges, etc. + +SM-01: It is recommended each time to record the flight process, including flight speed, flight altitude and route, through continuous optimization and adjustment, so as to set the best route, flight altitude and speed for the next IaE service. + +SM-02: It is recommended that the IaE service management system shall be able to convert inspection results (including normal status, aging, damage degree and water quality level) into file data and generate an IaE report. + +SM-03: It is recommended that the IaE service management system shall be able to give action suggestions for aging, serious damage or danger, etc. such as immediate maintenance or dedicated IaE for confirmation according to IaE report results. + +SM-04: It is recommended that the IaE service management system shall be able to make statistical analysis on the historical data of the inspection results and predict the change trend (e.g., of water quality and cracks), so as to provide reference for later maintenance and treatment. + +### **8.3.3 Civilian unmanned aerial vehicle management** + +A CUAV shall be certified and authorized by the management department (e.g., of the national civil aviation authority or international civil aviation organization) before its flight, and meet its local flight supervision requirements. A CUAV can receive the latest flight notice and electronic map data of flight-prohibition zones, densely populated areas and airports from the management department, save the current flight data to the database server, and send the updated notice and data to the CUAV flight control system. + +CM-01: It is recommended that CUAV management shall provide function to register CUAV information (including its registration number, communication module identifier, model, and flight control equipment number and category) before the first flight. The national registration number is unique and traceable. + +CM-02: It is recommended that a flight control system shall support real-time flight warning in a flight-prohibition zone and densely populated area. If a CUAV flies close to or enters electronically fenced areas, then immediate warning shall be sent to users (e.g., notifying them of illegal flights with time and locations). + +## **8.4 Requirements for service presentation and playback** + +According to different application scenarios and requirements, the video or images obtained by HD cameras (including single channel and multi-channel) and infrared cameras can be presented through HD display equipment for inspectors to view. The view can be divided into real-time view and review. The images obtained by a multi-channel or panoramic camera can also be synthesized and viewed through a VR terminal. + +### **8.4.1 Requirements for a virtual reality terminal** + +A VR terminal mainly includes several modules such as access engine, unpacker, decoder, renderer and sensor (such as inertial sensor and geomagnetic sensor), which complete the acquisition of user motion signal and window metadata, as well as the unpacking, decoding and rendering of VR media content. + +VR-01: It is required that a VR terminal request VR media service from a streaming server and downloads an index file and omnidirectional media fragment file. The compressed video data and audio data are obtained by unpacking, and then the compressed data are decoded by video decoder and audio decoder, respectively. Finally, combined with the window metadata, the decoded video data is projected, rotated and rendered, and the decoded audio data is directly rendered, and all presented to the viewer. + +VR-02: It is recommended that a VR terminal support a universal serial bus interface, wireless fidelity or Bluetooth to connect to a streaming server. + +### **8.4.2 Requirements for the high-definition display device** + +HD display equipment includes an HD TV and computer, which supports HD and 4K display, frame rates of 50 to 60 frames/s or 20 to 30 frames/s and 3840\*2160, 1920\*1080 and 1280\*720 resolutions. It is recommended to support real time streaming protocol, MP4, DASH formats, 4:2:2 and 4:2:0 live broadcast and review, with pause, fast forward or rewind operations. For multi-channel cameras, it is recommended to support switching to any channel for viewing. + +HD-01: It is required to clearly display the status of power lines, insulators, iron towers and other components. It is required to support the positioning and zoom view of whether the power line is aging or damaged, and whether screws are loose, rusted or dirty. + +HD-02: It is required to clearly display the status of a bridge deck, pier, tower column, cable, tower top lightning rod, etc. It is required to support the zoom view of whether the main components of the bridge have cracks or damage, exposed steel bar corrosion, bolts falling off, etc. + +HD-03: It is required to clearly display an oil pipeline and its surrounding status, and support the identification and inspection of whether the oil pipeline is damaged, corroded or leaking. + +HD-04: It is required to clearly display the water surface and water facilities of rivers, lakes and wetlands. + +HD-05: It is required to clearly display the on-site status of natural disasters and support zoom view of drought, forest fire, flood, landslide, earthquake, etc., and to find whether there are survivors at an emergency or disaster site. + +HD-06: It is required to clearly display the status of an industrial disaster site and support the zoom view of chemical plant and nuclear power plant facilities and surroundings. + +# Appendix I + +## Examples of high-definition and virtual reality inspection and examination services using civilian unmanned aerial vehicles + +(This appendix does not form an integral part of this Recommendation.) + +Figure I.1 shows the general process of HD and VR IaE service using CUAV, including fore-end devices capturing HD images and videos of objects and surroundings, uploading images or data via networks, subsequent data processing, finally, HD images and video presented and displayed for inspection. + +![Diagram illustrating the general process of HD and VR IaE service using CUAV, showing four stages: Capturing objects and surroundings, Uploading data, Data processing, and Service presentation.](16152cf1d84aea10848758f51a91ff6a_img.jpg) + +The diagram illustrates the general process of HD and VR IaE service using CUAV, organized into four sequential stages, each enclosed in an oval: + +- Capturing objects and surroundings:** This stage shows various images captured by a CUAV, including a close-up of a camera lens, a CUAV in flight, and several aerial views of infrastructure like power lines, towers, and buildings. +- Uploading data:** This stage shows a satellite in space and a CUAV with a yellow label "IMT-2020", indicating data transmission via networks. +- Data processing:** This stage shows a cloud icon labeled "Networks" and "Access and transport", followed by a CUAV with a yellow label "IMT-2020", and finally a computer monitor displaying a 3D visualization of a power line and tower. +- Service presentation:** This stage shows a VR headset, a computer monitor displaying a 3D visualization of a power line and tower, and a desktop screen with a green background. + +Diagram illustrating the general process of HD and VR IaE service using CUAV, showing four stages: Capturing objects and surroundings, Uploading data, Data processing, and Service presentation. + +F.749.15(22) + +**Figure I.1 – The general process of high-definition and virtual reality inspection and examination services using civilian unmanned aerial vehicles** + +Some examples and scenarios of IaE services follow. See Figures I.2 to I.5. + +For hundreds of kilometres power lines, clear vision of their status, as well as insulators, iron towers and other components, is needed to determine whether the power line is aging or damaged, and whether screws are loose, rusted or dirty. + +![Two side-by-side images related to power line inspection. The left image shows a drone flying over power lines and towers. The right image shows a close-up of a power line with a yellow box overlay reading 'Displayed or presented on HD TV or computer' and a white box overlay reading 'Defect identification'.](ac852a162572ca8a8c8478c49b571af5_img.jpg) + +F.749.15(22) + +Two side-by-side images related to power line inspection. The left image shows a drone flying over power lines and towers. The right image shows a close-up of a power line with a yellow box overlay reading 'Displayed or presented on HD TV or computer' and a white box overlay reading 'Defect identification'. + +**Figure I.2 – Example of power lines and its supporting facilities inspection and examination service** + +For hundreds of kilometres oil pipelines as well as their supporting facilities, inspectors can clearly see whether installations are aging or damaged, corroded or leaking, etc. + +![Two side-by-side images related to oil pipeline inspection. The left image shows an industrial facility with complex piping. The right image shows a long, straight oil pipeline running through a landscape, supported by metal posts.](6cc4a2d5ea0462e4825d57bd689bd2b3_img.jpg) + +F.749.15(22) + +Two side-by-side images related to oil pipeline inspection. The left image shows an industrial facility with complex piping. The right image shows a long, straight oil pipeline running through a landscape, supported by metal posts. + +**Figure I.3 – Example of oil pipelines and its supporting facilities inspection and examination service** + +For bridges and viaducts over water and land, an inspector can clearly look over the main components to establish whether the bridge has cracks or damage, exposed steel bar corrosion, bolts falling off, etc.; an inspector can also regularly inspect and evaluate their appearance and structure. + +![Two side-by-side images related to bridge inspection. The left image shows a long bridge over water with a drone flying underneath. The right image shows a close-up of a bridge's underside structure with a drone flying nearby.](c3ee41c6d46565ab4198d0a9c69108c5_img.jpg) + +F.749.15(22) + +Two side-by-side images related to bridge inspection. The left image shows a long bridge over water with a drone flying underneath. The right image shows a close-up of a bridge's underside structure with a drone flying nearby. + +**Figure I.4 – Example of bridges and its supporting facilities inspection and examination service** + +For wide rivers and lakes over long distances, the inspectors can electronically survey the scene via a VR panoramic camera to observe the water surface and water facilities of rivers, lakes and wetlands. + +![Aerial view from a drone showing a river, a city skyline, and surrounding infrastructure.](fbfaa723129f8b8e3be651c6fe7cfab7_img.jpg) + +An aerial photograph taken from a drone, showing a wide river on the left, a multi-lane road and railway track running parallel to it, and a dense urban area with numerous high-rise buildings in the background under a clear sky. + +Aerial view from a drone showing a river, a city skyline, and surrounding infrastructure. + +![A person wearing a VR headset and holding a controller, standing in a room with a camera on a tripod.](2150ae7f14a2e4ad1866ac0c8ec685ad_img.jpg) + +A person wearing a black and white striped shirt is shown from the side, wearing a virtual reality (VR) headset and holding a small black controller. They are standing in a room with light-colored tiled walls. In the background, a camera is mounted on a black tripod, and a laptop is visible on a desk. + +A person wearing a VR headset and holding a controller, standing in a room with a camera on a tripod. + +F.749.15(22) + +**Figure I.5 – Example of rivers and its surroundings IaE service** + +## Bibliography + +- [b-ITU-T F.749.10] Recommendation ITU-T F.749.10 (2019), *Requirements for communication services of civilian unmanned aerial vehicles*. + + + + + +# SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.749.16-202307-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/F/T-REC-F.749.16-202307-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d4d8969f805414689ab79cfd08c333d454d482b3 --- /dev/null +++ b/marked/F/T-REC-F.749.16-202307-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:636c3839edff955a5b6208530b05fa1c983ab153d90f00216517a7a99478b988 +size 7074 diff --git a/marked/F/T-REC-F.749.16-202307-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg b/marked/F/T-REC-F.749.16-202307-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..035eddbd29caa0765c43a331f6cbc80430aca26a --- /dev/null +++ b/marked/F/T-REC-F.749.16-202307-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4233ff87fa6f8a352345788875bed93d63645f0aa72c0c2d0e4594f0b33a04d9 +size 52131 diff --git a/marked/F/T-REC-F.749.16-202307-I_PDF-E/6b7b3f3d6f9341906163682cf12d1ea1_img.jpg b/marked/F/T-REC-F.749.16-202307-I_PDF-E/6b7b3f3d6f9341906163682cf12d1ea1_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c4a76c61433c8739ef77690a1caea5b6fff165c0 --- /dev/null +++ b/marked/F/T-REC-F.749.16-202307-I_PDF-E/6b7b3f3d6f9341906163682cf12d1ea1_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ea615251e5facb516d895001138f6a41bbb955e7bceeef21f88bef2efaafc757 +size 38729 diff --git a/marked/F/T-REC-F.749.16-202307-I_PDF-E/7affafe7362a2d2d072e9d4bf515f0bb_img.jpg b/marked/F/T-REC-F.749.16-202307-I_PDF-E/7affafe7362a2d2d072e9d4bf515f0bb_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..68c04fee769b5e690bde7140377c30166c5664a1 --- /dev/null +++ b/marked/F/T-REC-F.749.16-202307-I_PDF-E/7affafe7362a2d2d072e9d4bf515f0bb_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4112bf086283479a5a200f52743eac7bf260801d5ba29ec9cd8eda634899c75f +size 63190 diff --git a/marked/F/T-REC-F.749.16-202307-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg b/marked/F/T-REC-F.749.16-202307-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..5ee7e58f53286685d735e67fdb2cf45ba2429e56 --- /dev/null +++ b/marked/F/T-REC-F.749.16-202307-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aa0bb7d68993aef6bd7a19cf5258146ac2d24e04ba15882bf5bcb663df32dc5d +size 69781 diff --git a/marked/F/T-REC-F.749.16-202307-I_PDF-E/b8205e5e617a8946ddc956c816156fec_img.jpg b/marked/F/T-REC-F.749.16-202307-I_PDF-E/b8205e5e617a8946ddc956c816156fec_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ffe4161c2d3ac92396107aee2a2161ec0248e63d --- /dev/null +++ b/marked/F/T-REC-F.749.16-202307-I_PDF-E/b8205e5e617a8946ddc956c816156fec_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5bf2740d17407f95ecf35ea7075a85d450724bfbaf246ec0ee338dc60c7231b2 +size 30812 diff --git a/marked/F/T-REC-F.749.16-202307-I_PDF-E/raw.md b/marked/F/T-REC-F.749.16-202307-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..36e280f39d67f1c8d873221e36830d8e7223e274 --- /dev/null +++ b/marked/F/T-REC-F.749.16-202307-I_PDF-E/raw.md @@ -0,0 +1,389 @@ + + +# Recommendation **ITU-T F.749.16 (07/2023)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +## **Requirements for logistics express delivery based on civilian unmanned aerial vehicles** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.749.16 + +# Requirements for logistics express delivery based on civilian unmanned aerial vehicles + +## Summary + +At present, logistics express delivery based on civilian unmanned aerial vehicles (CUAVs) is developing rapidly all over the world. Compared with general water transportation and land transportation, CUAV transportation has the advantages of low cost and flexible scheduling, and can make up for the shortcomings of traditional air transportation. It will change people's consumption mode. Recommendation ITU-T F.749.16 provides the requirements for the service system and management of CUAV logistics express delivery. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|----------------------------------------------------------------------------| +| 1.0 | ITU-T F.749.16 | 2023-07-10 | 16 | 11.1002/1000/15173 | + +## Keywords + +Civilian unmanned aerial vehicle, logistics express delivery, requirements. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2023 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|---------------------------------------------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Introduction..... | 2 | +| 7 Requirements for logistics express delivery based on civilian unmanned aerial vehicles ..... | 3 | +| 7.1 Requirements for service control and management system ..... | 4 | +| 7.2 Requirements for network communication ..... | 5 | +| 7.3 Requirements for the devices and facilities of goods delivery stations and receiving stations ..... | 5 | +| 7.4 Requirements for the civilian unmanned aerial vehicle ..... | 6 | +| Appendix I – Use case of logistics express delivery based on civilian unmanned aerial vehicle..... | 7 | +| Bibliography..... | 9 | + + + +# Recommendation ITU-T F.749.16 + +## Requirements for logistics express delivery based on civilian unmanned aerial vehicles + +## 1 Scope + +This Recommendation specifies the requirements for logistics express delivery based on civilian unmanned aerial vehicles (CUAVs), including the requirements for devices and facilities of goods provided and received, network communication and the service control and management system of CUAVs. + +NOTE 1 – The regulations and supervision of civilian unmanned aerial vehicle flights, such as those stated by the International Civil Aviation Organization (ICAO) in [b-ICAO], particularly in chapter 2 on the ICAO Regulatory Framework, chapter 3 on the Overview of Unmanned aircraft systems (UAS), chapter 4 on Legal Matters, and chapter 6 on Aircraft and Systems, are out of the scope of this Recommendation. + +NOTE 2 – Any implementation based on this Recommendation should be limited by being aligned with regional and national regulations, as in [b-ICAO] in chapter 5 on Operations and chapter 7 on Personnel. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 civilian unmanned aerial vehicle** [b-ITU-T F.749.10]: An unmanned flying device controlled by a ground control station or telecontroller via various wireless communication means. It usually consists of an aeroplane body, a power device, aviation electrical and electronic equipment and mission payload equipment, etc. and is used in non-military application areas such as industrial and consumer areas to complete the specific operation and transportation of data including audio, video and image. + +**3.1.2 ground control station** [b-ITU-T F.749.10]: A ground control station is a device which is used to realize the functions of mission planning, flight control, payload control, flight path display, parameter display, image and video display and mission information displaying, recording and distributing. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following term: + +**3.2.1 civilian unmanned aerial vehicle logistics express delivery**: A process of planning, implementing and controlling the flow of physical goods from provider to recipient, mainly using civilian unmanned aerial vehicles (CUAVs) as a transport tool to provide an express delivery service. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +CUAV      Civilian Unmanned Aerial Vehicle + +GCS        Ground Control Station + +IMT-2020 International Mobile Telecommunications 2020 + +## 5 Conventions + +In this Recommendation: + +- The keyword "shall" indicates a requirement that must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The keywords "is required to" indicate a requirement that must be strictly followed and from which no deviation is permitted if conformance with this Recommendation is to be claimed. +- The keywords "is recommended" indicate a requirement that is recommended but which is not absolutely required. Thus, this requirement needs not be present to claim conformance. +- The keywords "can optionally" indicate an optional requirement that is permissible, without implying any sense of being recommended. These terms are not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the Recommendation. + +## 6 Introduction + +A complete air logistics express delivery service includes the three stages of large manned transport aircraft; medium-sized branch civilian unmanned aerial vehicle (CUAV) transport; and small terminal CUAV transport. These stages are connected by cargo transfer stations and a CUAV logistics centre. Figure 1 shows the whole process model of an air logistics express delivery service. + +![Figure 1 – The complete air logistics express delivery chain model. The diagram illustrates a five-stage process: 1. Large manned transport aircraft (represented by a cargo plane), 2. Cargo transfer stations (represented by buildings), 3. Medium-sized branch CUAV transport (represented by a fixed-wing drone), 4. CUAV logistics centre (represented by a warehouse with conveyor belts), and 5. Small terminal CUAV transport (represented by a multi-rotor drone). The stages are connected by overlapping circles, indicating a continuous flow.](7affafe7362a2d2d072e9d4bf515f0bb_img.jpg) + +The diagram illustrates a five-stage process for air logistics express delivery, connected by overlapping circles. The stages are: 1. Large manned transport aircraft (represented by a cargo plane), 2. Cargo transfer stations (represented by buildings), 3. Medium-sized branch CUAV transport (represented by a fixed-wing drone), 4. CUAV logistics centre (represented by a warehouse with conveyor belts), and 5. Small terminal CUAV transport (represented by a multi-rotor drone). The stages are connected by overlapping circles, indicating a continuous flow. + +Figure 1 – The complete air logistics express delivery chain model. The diagram illustrates a five-stage process: 1. Large manned transport aircraft (represented by a cargo plane), 2. Cargo transfer stations (represented by buildings), 3. Medium-sized branch CUAV transport (represented by a fixed-wing drone), 4. CUAV logistics centre (represented by a warehouse with conveyor belts), and 5. Small terminal CUAV transport (represented by a multi-rotor drone). The stages are connected by overlapping circles, indicating a continuous flow. + +F.749.16(23) + +**Figure 1 – The complete air logistics express delivery chain model** + +The scope of this Recommendation is confined to logistics express delivery based on CUAV. This type of delivery is a part of air logistics, which comprises: medium-sized branch CUAV transport, a CUAV logistics centre and small terminal CUAV logistics express delivery. + +### 1) Medium-sized branch CUAV transport: + +The transportation distance is usually less than 1000 km. Use cases include transregional freight transport (fixed routes, fixed shifts, etc.), transportation of temporary materials to + +border posts, islands and so on, and freight allocation amongst multilogistics centres to supplement medium-distance transportation requirements. + +### 2) The CUAV logistics centre: + +A CUAV logistics centre should be a distributed storage and management system that can realize classified scheduling of cargoes/goods from different sources and optimize the storage location of goods according to their address and delivery time requirements, so as to create conditions for CUAV delivery. The source of cargoes/goods and the ways of sending/receiving goods include the following situations: The first case is the transport of goods from large cargo transfer stations, which are transported by medium-sized branch CUAV. After the goods (e.g., letters, small packages) have been automatically sorted by the logistics centre, they can be transported to a specific express loading platform by transmission tools (such as conveyor belts, robots) so that they can be fixed to the CUAV. The second case is delivery in the city and nearby suburbs. For example, goods suppliers in the city deliver food and daily necessities to nearby buildings, communities and office buildings, or deliver them to the suburbs close to it. The staff of the CUAV express service organization needs to load and fix the items on the CUAV; to achieve this, the express waiter is required to send the goods to the CUAV express staff. A relatively open site in a field can be selected and landing point signs can be set to land the CUAV, or drop the goods by rope and then take back the rope. In short, the receiving stations have different settings according to the location and scenario of the customer or other recipient. A multilogistics centre or warehouse centre can be set up to reduce the number of medium-sized branch CUAVs. + +### 3) Terminal CUAV transport: + +Terminal CUAV transport usually flies for a distance of less than 20 km with a load of less than 20 kg, and with a one-way flight time of 10 to 20 minutes. + +## 7 Requirements for logistics express delivery based on civilian unmanned aerial vehicles + +CUAV logistics express delivery includes medium-sized branch CUAV transport and terminal CUAV transport. Before the delivery service is provided, the sender and recipient of goods need to communicate and clarify the responsibilities of both parties. The CUAV express service organization shall comply with the corresponding national laws and regulations, and be responsible for the information security of the express delivery goods and the customer. Before the delivery service, the CUAV express service organization shall complete a flight route survey and apply and report to the local and national aviation management departments in accordance with the regulations. It is not permitted to carry out a CUAV express delivery service in flight-prohibition zones, restricted flight zones and other unapproved flight areas. Terminal CUAV transport is the key process of a CUAV logistics express. This Recommendation will focus on this service procedure. It is often not recommended that recipients have direct contact with the CUAV. + +Figure 2 presents a diagram of the service procedure of CUAV logistics express delivery. + +![Diagram of the service procedure of CUAV logistics express delivery. The diagram shows a flow from the Service control and management system (top left) and Client app (top right) connected via Network communication (center). The Network communication is connected to a CUAV delivery (center). The CUAV delivery is connected to Goods delivery stations (bottom left) and Goods receiving stations (bottom right).](a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg) + +The diagram illustrates the service procedure of CUAV logistics express delivery. At the top, the 'Service control and management system' (represented by a computer monitor and two server icons) and a 'Client app' (represented by a smartphone) are connected via 'Network communication' (represented by a radio tower and signal waves). The 'Network communication' is connected to a 'CUAV delivery' (represented by a drone carrying a package). The 'CUAV delivery' is connected to 'Goods delivery stations' (represented by a dashed box containing a package and a person) and 'Goods receiving stations' (represented by a dashed box containing a yellow vehicle, a person, and a drone). The label 'F.749.16(23)' is visible in the bottom right corner of the diagram. + +Diagram of the service procedure of CUAV logistics express delivery. The diagram shows a flow from the Service control and management system (top left) and Client app (top right) connected via Network communication (center). The Network communication is connected to a CUAV delivery (center). The CUAV delivery is connected to Goods delivery stations (bottom left) and Goods receiving stations (bottom right). + +**Figure 2 – Diagram of the service procedure of CUAV logistics express delivery** + +The specific functional requirements of each entity are described in clauses 7.1.1 to 7.1.3. + +### 7.1 Requirements for service control and management system + +The CUAV logistics express delivery service control and management system completes the real-time dynamic information management and real-time dynamic scheduling control, as well as the whole process tracking of goods. The client app completes the ordering of goods, the arrival location and storage information of received goods, and so on. + +#### 7.1.1 Requirements for service + +The CUAV express service organization shall reach an agreement with the recipient on the specific delivery method. When a CUAV cannot fly due to the weather, electromagnetic environment or airspace problems, the recipient shall be informed of the special approach in these circumstances. + +SR-01: It is required that the CUAV express service organization communicate with the recipient in advance. The interactive information includes but is not limited to express information, departure place and time, landing place and time, and name and address of the recipient. + +SR-02: It is required that the delivered goods strictly comply with the safety inspection mechanism of the logistics industry. The CUAV express service organization shall not accept dangerous goods (such as chemicals and toxic substances) for delivery, and overloading shall not be allowed. + +SR-03: It is required that a period and conditions for goods return or exchange be specified to the customer in advance of the delivery, for example if the goods delivered are found to be damaged or inconsistent with the order. + +#### 7.1.2 Requirements for route planning + +Before launching the CUAV express delivery service, the CUAV express service organization should complete the air route survey and avoid the flight-prohibition zone designated by the aviation management department. + +RP-01: It is recommended that the route be a fixed route. A field survey shall be conducted on the take-off and landing points and route areas to confirm the route grid data. The route planning between two points shall include an alternative route. + +RP-02: It is required that multiple flight tests be carried out to verify the position data of the take-off and landing points, and to check the distance from known obstacles and the performance of communication, navigation and monitoring. + +RP-03: It is required that the route be inspected regularly. + +RP-04: It is required that the route be re-planned when risks are found. + +#### **7.1.3 Requirements for flight control** + +CUAV flight control is an important part of service control and is controlled by the ground control station (GCS). + +GCS-01: It is required that the GCS completes the CUAV flight process controlling and monitoring based on the service requirements. + +GCS-02: The GCS is required to display flight direction and track parameters of each CUAV in real time, including flight altitude, speed, flight track coordinates, accumulated flight mileage and flight time, and number of flights. It is required to support the functions of automatic map roaming, scheduled flight tracking and real-time flight tracking display of CUAVs. + +GCS-03: The GCS is required to support multiple flight control modes for different environments. + +GCS-04: The GCS is required to support data playback mode. After receiving the flight data and saving the data, the whole process of the flight path can be reproduced through the data playback mode, including the position, attitude and other important data of the CUAV, which is convenient for the analysis of the flight status and mission execution status of the CUAV. + +### **7.2 Requirements for network communication** + +The communication network completes the communication between the CUAV and the service control and management system, as well as the interaction between the CUAV and the ground infrastructure (goods sending/receiving facilities) and mobile terminals (recipient). + +NC-01: Network and transmission functions are required to support real-time and reliable flight control and flight data transport. + +NC-02: Before the transport service starts, it is required to establish a secure data transport channel and audio and video image transport channel, and conduct two-way authentication. + +NC-03: It is recommended that CUAV GCS based on IMT-2020 network support the control of one or more CUAVs. + +### **7.3 Requirements for the devices and facilities of goods delivery stations and receiving stations** + +The devices and facilities of goods delivery stations and receiving stations shall have good environment and site conditions to meet the operational needs of the sender and recipient, as well as meeting the requirements for safe take-off and landing of CUAVs. + +#### **7.3.1 Requirements for the devices and facilities of goods delivery stations** + +The goods delivery station is responsible for completing the loading of goods and checking the goods to ensure the express information is accurate and confirm whether the outer package is undamaged. + +DS-01: The staff of the CUAV express service organization is required to judge whether the size of the goods meets the requirements of the CUAV. If satisfied, the goods can be directly loaded on the + +CUAV. If the requirements are not satisfied, the package can be disassembled and repacked to meet the loading regulations of the CUAV. + +DS-02: Goods from large cargo transfer stations (e.g., letters and small parcels which meet the requirements of CUAV transport), can optionally be transported to a specific express loading platform by transport tools (such as conveyor belt or robot) and loaded onto the CUAV (the process is monitored by the service control and management system). + +DS-03: It is required that the staff of the CUAV express service organization load and fix the goods on the CUAV. To achieve this, the express waiter is required to send the goods to the CUAV express staff. + +DS-04: It is required that special goods be enclosed in shock-absorption and fall-proof packaging over the original package to ensure the safety of the express goods during the whole flight. + +#### **7.3.2 Requirements for the devices and facilities of goods receiving stations** + +The goods receiving station is responsible for completing the unloading and receiving of goods. The receiving station can be set up in different ways according to the location and scenario of the recipient. + +RS-01: If the receiving station is a smart cabinet, then the CUAV sends a landing request to the smart cabinet. After receiving the landing request, the smart cabinet is required to send a signal to guide the CUAV to land on the platform at the top of the cabinet and then to unload the goods. + +RS-02: If the receiving terminal is open ground or a platform set in a hospital or outdoor domestic area, the CUAV is recommended to land directly on the ground or platform, or drop the goods by rope and then take back the rope. + +RS-03: For goods delivered to an edge transfer station by a medium-sized CUAV, the staff of the receiving station shall first receive them. According to the distance and the receiving agreement, the goods are required to be sent to the recipient by terminal CUAV, collected by the recipient or sent by the express staff. + +### **7.4 Requirements for the civilian unmanned aerial vehicle** + +The CUAV transports the goods from the delivery stations to the receiving stations according to the predetermined path, and receives the instructions from the service control and management system. The CUAV shall be certified and authorized by the management department (e.g., the national civil aviation management department or ICAO) before its flight, and shall meet the flight supervision requirements of the local management department. + +CUAV-01: The CUAV is required to transmit flight data (including but not limited to weather conditions, flight altitude, speed, geographical coordinates and status information, residual energy and flight time) to the service control and management system in real time. + +CUAV-02: It is required to check the appearance and payload safety lock device of the CUAV before and after the flight to avoid goods falling off during a flight. + +CUAV-03: The CUAV is required to fly stably or hover normally under light rain conditions or with an instantaneous wind speed not greater than 10 m/s (Level 5 wind). + +CUAV-04: On the onset of abnormal conditions during flight, the CUAV is required to support automatic hovering, and based on the predefined process and alternate procedures, select other sites for landing. + +CUAV-05: It is required that the CUAV fly to a preset place for charging after completing the mission, or it can optionally be briefly charged at the top of the smart cabinet (if certain conditions are met). + +## Appendix I + +### Use case of logistics express delivery based on civilian unmanned aerial vehicle + +(This appendix does not form an integral part of this Recommendation.) + +This appendix contains a use case of urban express service and its service procedure. + +Before an express service starts, customers download an app from the website of the CUAV express service organization to their mobile terminal and browse potential goods suppliers of their desired products (the express service organization needs to have already selected some goods suppliers suitable for working with based on the price, cost, distance and other factors determining the service mode). + +After the customer selects the goods (such as food and beverage items), the express waiter sends the goods to the nearby CUAV delivery station, and then the staff of the express service organization fixes the goods onto the CUAV (see Figure I.1). + +![Figure I.1: Express item fixed on the CUAV. The image consists of two side-by-side photographs. The left photograph shows a person's hands securing a cardboard box to the underside of a yellow quadcopter drone using white zip ties. The right photograph shows a top-down view of the drone with the box attached, resting on a white surface. A small label 'F.749.16(23)' is visible in the bottom right corner of the right image.](b8205e5e617a8946ddc956c816156fec_img.jpg) + +Figure I.1: Express item fixed on the CUAV. The image consists of two side-by-side photographs. The left photograph shows a person's hands securing a cardboard box to the underside of a yellow quadcopter drone using white zip ties. The right photograph shows a top-down view of the drone with the box attached, resting on a white surface. A small label 'F.749.16(23)' is visible in the bottom right corner of the right image. + +**Figure I.1 – Express item fixed on the CUAV** + +The express service control centre determines a suitable receiving station according to the geographical location information of the customer, sends the pre-planned route data to the CUAV, and then sends a take-off command to the CUAV. At the same time, the service control centre sends a message to tell the customer when to collect the express item from the designated smart cabinet (see Figure I.2). + +![Figure I.2: CUAV unloading item and customer collecting express item(s). The image consists of two side-by-side photographs. The left photograph shows a yellow drone with its cargo bay open, unloading a small white box into a yellow smart cabinet. A label 'Goods receiving stations' is overlaid on the bottom left. The right photograph shows a person standing in front of a row of yellow smart cabinets, interacting with one of them. A label 'Take express items' is overlaid on the bottom right. A small label 'F.749.16(23)' is visible in the bottom right corner of the right image.](6b7b3f3d6f9341906163682cf12d1ea1_img.jpg) + +Figure I.2: CUAV unloading item and customer collecting express item(s). The image consists of two side-by-side photographs. The left photograph shows a yellow drone with its cargo bay open, unloading a small white box into a yellow smart cabinet. A label 'Goods receiving stations' is overlaid on the bottom left. The right photograph shows a person standing in front of a row of yellow smart cabinets, interacting with one of them. A label 'Take express items' is overlaid on the bottom right. A small label 'F.749.16(23)' is visible in the bottom right corner of the right image. + +**Figure I.2 – CUAV unloading item and customer collecting express item(s)** + +![Procedure diagram of urban express service showing the flow from order placement to delivery and return.](4801720824e4b5e2361a5564f91cfb70_img.jpg) + +The diagram illustrates the procedure of an urban express service. It begins with an 'Order' (represented by a smartphone icon) being placed by a customer, which is received by a 'Goods supplier' (represented by a factory icon). The supplier then sends the goods to a 'Delivery station' (represented by a building icon) via step (2). At the delivery station, a 'CUAV delivery' (represented by a drone icon) is dispatched via step (3). The CUAV then flies to one of three possible receiving stations: a 'Community receiving station' (step 4), an 'Office building receiving station' (step 5), or an 'Apartment receiving station' (step 6). Each receiving station is accompanied by an icon of a smart cabinet. After delivering the goods, the CUAV returns to a 'Charging pile' (represented by a battery icon) via step (7). Finally, the CUAV flies to the designated sending station to continue its mission via step (8). The diagram is labeled F.749.16(23). + +Procedure diagram of urban express service showing the flow from order placement to delivery and return. + +**Figure I.3 – Procedure diagram of urban express service** + +Figure I.3 presents the process of a urban express service, which has the following steps for each order: + +- (1) A customer orders goods. +- (2) The express service staff sends the goods to the delivery station. +- (3) The service control system determines a suitable receiving station and the CUAV flies to the destination. +- (4) / (5) / (6) The CUAV arrives the express smart cabinet set up in the communities / office buildings / apartments and deposits the goods. +- (7) The CUAV returns to the place specified by the logistics centre for charging according to the preset path. +- (8) The CUAV flies to the designated sending station to continue the next express mission. + +## Bibliography + +- [b-ITU-T F.749.10] Recommendation ITU-T F.749.10 (2019), *Requirements for communication services of civilian unmanned aerial vehicles*. +- [b-ICAO] International Civil Aviation Organization Cir 328 (2011), *Unmanned Aircraft Systems (UAS)*. [https://www.icao.int/meetings/uas/documents/circular%20328\\_en.pdf](https://www.icao.int/meetings/uas/documents/circular%20328_en.pdf) + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.749.18-202406-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/F/T-REC-F.749.18-202406-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..38bdd7bc1d3581fbdc3cf6b035f618693dbe3846 --- /dev/null +++ b/marked/F/T-REC-F.749.18-202406-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:75efa8e3513a799767c04a0bb7bcd87a550f2378f5880629068581808c1c9f7f +size 7353 diff --git a/marked/F/T-REC-F.749.18-202406-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg b/marked/F/T-REC-F.749.18-202406-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1469d31f532896666097837b45c6ede56170f0a0 --- /dev/null +++ b/marked/F/T-REC-F.749.18-202406-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:616067e90d0443e675bb6720cf0b4c407ed7a299f53bd5c190dd82a8f427509e +size 155066 diff --git a/marked/F/T-REC-F.749.18-202406-I_PDF-E/cb4cfa42ce34febde7bdb882f3fc3094_img.jpg b/marked/F/T-REC-F.749.18-202406-I_PDF-E/cb4cfa42ce34febde7bdb882f3fc3094_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..fad61081010a5606ea17c8e1dedd1cbd0c80fc43 --- /dev/null +++ b/marked/F/T-REC-F.749.18-202406-I_PDF-E/cb4cfa42ce34febde7bdb882f3fc3094_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bc3e3df6328cb8abc8264a3296334e61a28cc02dce32271405cb24f7b1943cd1 +size 43771 diff --git a/marked/F/T-REC-F.749.2-201703-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.749.2-201703-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..22dc7ff0fb30f976cc1e576c45d2b7e65fa5ddc7 --- /dev/null +++ b/marked/F/T-REC-F.749.2-201703-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aa14933e776f7d0be8b2d60b81cbb6c1ee49deb5080e0281da05296f48189055 +size 6859 diff --git a/marked/F/T-REC-F.749.2-201703-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg b/marked/F/T-REC-F.749.2-201703-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ac55c6928e57e3de8d2cee958b7b06a66ab0cd1a --- /dev/null +++ b/marked/F/T-REC-F.749.2-201703-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:da641d38fed312d3b910232a2baaff907136a0fb69bae594157b43a7a9712c43 +size 31434 diff --git a/marked/F/T-REC-F.749.2-201703-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg b/marked/F/T-REC-F.749.2-201703-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f5cb3c68b9193f3f328da6a576d7e7e7bd815a92 --- /dev/null +++ b/marked/F/T-REC-F.749.2-201703-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8f7d91dbb275688e808ce014aa1ad963786ee5854cab0d94a3e42005bffcf558 +size 78524 diff --git a/marked/F/T-REC-F.749.2-201703-I_PDF-E/a734898ce18e972938949637c32a34f4_img.jpg b/marked/F/T-REC-F.749.2-201703-I_PDF-E/a734898ce18e972938949637c32a34f4_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..32abdd811101529cb8ff0a25fe3756468e3e26f9 --- /dev/null +++ b/marked/F/T-REC-F.749.2-201703-I_PDF-E/a734898ce18e972938949637c32a34f4_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cf490df312ca17aa80a603abde2fceb604200058558c53cf8ef313267fda0366 +size 25963 diff --git a/marked/F/T-REC-F.749.2-201703-I_PDF-E/b90144cfbb81a2d610d920240fda689d_img.jpg b/marked/F/T-REC-F.749.2-201703-I_PDF-E/b90144cfbb81a2d610d920240fda689d_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..4456d2f63c3dcf806dad7c118c7cc32939e22813 --- /dev/null +++ b/marked/F/T-REC-F.749.2-201703-I_PDF-E/b90144cfbb81a2d610d920240fda689d_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e080eff2748e1bdfb4a85ecf3ff465a7e05aadcd332e14dac59fc9ceaad723f2 +size 30224 diff --git a/marked/F/T-REC-F.749.2-201703-I_PDF-E/e636d7ccca0ad14c6b95201404324823_img.jpg b/marked/F/T-REC-F.749.2-201703-I_PDF-E/e636d7ccca0ad14c6b95201404324823_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..2ccf9e6a43faf0f2824ee5c6f0855ec5b4a8c13d --- /dev/null +++ b/marked/F/T-REC-F.749.2-201703-I_PDF-E/e636d7ccca0ad14c6b95201404324823_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:77b26a1f984f6e2bb6188b2d7e715a2b8b497492f2504a99baf07411ec82f707 +size 43020 diff --git a/marked/F/T-REC-F.749.2-201703-I_PDF-E/raw.md b/marked/F/T-REC-F.749.2-201703-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..899272b2e45082ca73b06eae22febdb3edfb0504 --- /dev/null +++ b/marked/F/T-REC-F.749.2-201703-I_PDF-E/raw.md @@ -0,0 +1,643 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.749.2** + +(03/2017) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +# --- **Service requirements for vehicle gateway platforms** + +Recommendation ITU-T F.749.2 + +**ITU-T** + +![ITU logo: A globe with a red lightning bolt, next to the text 'International Telecommunication Union'.](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with a red lightning bolt striking across it. To the right of the globe, the text "International Telecommunication Union" is written in a blue, sans-serif font, with "ITU" in a larger, bold font above the full name. + +ITU logo: A globe with a red lightning bolt, next to the text 'International Telecommunication Union'. + +## ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.749.2 + +# Service requirements for vehicle gateway platforms + +## Summary + +Recommendation ITU-T F.749.2 describes the service requirements and functional requirements for the vehicle gateway platform (VGP). The use cases and scenarios are described in Appendix I. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.749.2 | 2017-03-01 | 16 | 11.1002/1000/13183 | + +## Keywords + +Intelligent transport system, ITS, service requirements, telematics, use cases, vehicle gateway platform, VGP. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2017 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 3 | +| 6 Descriptions and characteristics of VGP ..... | 3 | +| 6.1 Descriptions..... | 3 | +| 6.2 General characteristics..... | 4 | +| 7 Communication requirements for the VGP ..... | 5 | +| 7.1 External network connection requirements ..... | 5 | +| 7.2 In-vehicle connection requirements ..... | 5 | +| 7.3 Communication management requirements ..... | 6 | +| 7.4 Communication security requirements..... | 6 | +| 7.5 Networking requirements ..... | 6 | +| 8 Service requirements for VGP..... | 7 | +| 8.1 Driver distraction management requirements ..... | 7 | +| 8.2 Session management requirements..... | 7 | +| 8.3 In-vehicle data resource access management requirements ..... | 7 | +| 8.4 Driver-vehicle access management requirements ..... | 8 | +| 8.5 High-layer security requirements ..... | 8 | +| 8.6 Software management requirements ..... | 8 | +| 8.7 Application data management requirements ..... | 8 | +| Appendix I – VGP use cases..... | 10 | +| I.1 Background..... | 10 | +| I.2 Vehicle-to-vehicle interaction use cases ..... | 10 | +| I.3 Vehicle-to-infrastructure interaction use cases ..... | 11 | +| I.4 Vehicle-to-nomadic device interaction use cases..... | 13 | +| I.5 Vehicle-to-cloud interaction use cases ..... | 14 | +| I.6 Vehicle-to-pedestrian and bicycle interaction use cases ..... | 16 | +| Bibliography..... | 17 | + + + +# Recommendation ITU-T F.749.2 + +# Service requirements for vehicle gateway platforms + +# 1 Scope + +This Recommendation specifies functional requirements for the vehicle gateway platform (VGP), including description, communication requirements and service requirements. Some use cases are included in Appendix I. + +# 2 References + +None. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 functional entity** [b-ITU-T Y.2012]: An entity that comprises an indivisible set of specific functions. Functional entities are logical concepts, while groupings of functional entities are used to describe practical, physical implementations. + +**3.1.2 intelligent transport systems (ITS)** [b-ITU-R M.1797]: ITS can be defined as systems utilizing the combination of computers, communications, positioning and automation technologies to improve the safety, management and efficiency of terrestrial transport systems. + +**3.1.3 nomadic devices** [b-ITU-T F.749.1]: Nomadic devices include all types of information and communication as well as entertainment devices that can be brought into the vehicle by the driver and/or passengers to be used while driving. Examples include mobile phones, portable computers, tablets, mobile navigation devices, portable media players and multi-functional smart phones. + +**3.1.4 telematics** [b-ISO 15638-1]: Telematics is the use of wireless media to obtain and transmit (data) from a distant source. + +**3.1.5 vehicle gateway (VG)** [b-ITU-T F.749.1]: A VG is a device in a vehicle that enables communications between a device in the vehicle and another device which may be physically located either inside the vehicle or outside the vehicle (e.g., roadside station, cloud-based server, etc.). A VG provides standardized interfaces and protocols, communications across heterogeneous networks, optimized network selection based on application needs and network QoS, arbitration and integration of network communications, security and switching network connections to maintain service continuity. + +**3.1.6 vehicle gateway platform (VGP)** [b-ITU-T F.749.1]: A VGP is the collection of ICT hardware and software in a vehicle operating as an open platform to provide an integrated runtime environment for delivering the communications services of a VG. A VGP may also provide higher layer communications services such as interaction with the driver through the driver-vehicle access services and so on. Subsystems dedicated solely to vehicle operation are not considered part of the VGP. Supported applications/services include ITS and infotainment. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 driver-vehicle interface (DVI)**: The integrated user interface for the vehicle. It includes visual displays, loudspeakers, microphones, manual input controls, etc. + +**3.2.2 remote user interface:** An approach to realize the interaction between applications in nomadic devices and the driver-vehicle interface (DVI). The UI of applications in nomadic devices can be displayed in the vehicle's touch screen, and drivers are able to control the applications through the DVIs (touch screen, buttons, etc.). A remote UI can help drivers avoid distraction from applications and thus reduce the probability of accidents. + +# **4 Abbreviations and acronyms** + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|--------|--------------------------------------------------| +| 2G | Second generation of cellular phone technologies | +| 3G | Third generation of cellular phone technologies | +| 4G | Fourth generation of cellular phone technologies | +| API | Application Programming Interface | +| ASR | Automatic Speech Recognition | +| DSRC | Dedicated Short Range Communication | +| DVI | Driver-Vehicle Interface | +| ECU | Electronic Control Unit | +| ETC | Electronic Toll Collection | +| FM-RDS | Frequency Modulation-Radio Data System | +| GNSS | Global Navigation Satellite System | +| INS | Inertial Navigation System | +| ITS | Intelligent Transport System | +| MAN | Metropolitan Area Network | +| NAT | Network Address Translation | +| NFC | Near-Field Communication | +| OSI | Open Systems Interconnection | +| POI | Point of Interest | +| PSAP | Public Service Answering Point | +| QoS | Quality of Service | +| RSU | Roadside Unit | +| TSP | Telematics Service Provider | +| TTS | Text to Speech | +| UI | User Interface | +| USB | Universal Serial Bus | +| VG | Vehicle Gateway | +| VGP | Vehicle Gateway Platform | +| VIN | Vehicle Identification Number | +| WLAN | Wireless Local Area Network | +| WWAN | Wireless Wide Area Network | + +# 5 Conventions + +In this Recommendation: + +- The keyword "shall" indicates a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "should" and "optional" indicate an optional requirement which is permissible. This term is not intended to imply that the vendor's implementation must provide the option, and the feature can be optionally enabled by the vendor. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +# 6 Descriptions and characteristics of VGP + +## 6.1 Descriptions + +Figure 1 shows the VGP positioning in the intelligent transport system (ITS) reference model; six major sets of scenarios are defined: + +### 1) Vehicle-to-vehicle + +The vehicle-to-vehicle scenario set mainly includes safety and auto-driving use cases in which some vehicles communicate with one another. + +### 2) Vehicle-to-infrastructure + +The vehicle-to-infrastructure scenario mainly includes safety, electronic toll collection (ETC) and traffic information exchange use cases in which vehicles communicate with roadside infrastructures. + +### 3) Vehicle-to-cloud-based server + +The vehicle-to-cloud-based server scenario set mainly includes emergency call and some telematics scenarios in which vehicles communicate with cloud-based servers. + +### 4) Vehicle-to-nomadic device + +The vehicle-to-nomadic device scenario set mainly includes telecommunication and remote user interface (UI) use cases in which vehicles connect to nomadic devices. + +### 5) Vehicle-to-pedestrian and bicycle + +The vehicle-to-pedestrian and bicycle scenario set mainly includes safety warning use cases in which vehicles communicate with devices carried by pedestrians and bicycles. + +### 6) Interaction with in-vehicle network + +The interaction with in-vehicle network scenario set mainly includes vehicle diagnostics, remote data collection and vehicle remote control use cases in which the VGP communicates with the in-vehicle networks. + +![Figure 1 – VGP positioning in the ITS reference model. The diagram shows a central VGP (Vehicle Gateway) unit connected to various external and internal systems. At the top, 'Cloud-based servers' are connected via a 'WWAN' (Wireless Wide Area Network) through a satellite icon. To the left, 'Infrastructure' (represented by traffic lights and a road scene) is connected via 'Short-to-medium range wireless' communication. To the right, a 'Vehicle' and 'Pedestrian/bicycle' (represented by icons) are also connected via 'Short-to-medium range wireless' communication. Inside a dashed box labeled 'Nomadic device' (containing a smartphone and tablet icon), the VGP is connected to a 'Proprietary in-vehicle network' (represented by a thick horizontal line), which is further connected to a car icon at the bottom.](a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg) + +Figure 1 – VGP positioning in the ITS reference model. The diagram shows a central VGP (Vehicle Gateway) unit connected to various external and internal systems. At the top, 'Cloud-based servers' are connected via a 'WWAN' (Wireless Wide Area Network) through a satellite icon. To the left, 'Infrastructure' (represented by traffic lights and a road scene) is connected via 'Short-to-medium range wireless' communication. To the right, a 'Vehicle' and 'Pedestrian/bicycle' (represented by icons) are also connected via 'Short-to-medium range wireless' communication. Inside a dashed box labeled 'Nomadic device' (containing a smartphone and tablet icon), the VGP is connected to a 'Proprietary in-vehicle network' (represented by a thick horizontal line), which is further connected to a car icon at the bottom. + +F.749.2(17)\_F01 + +**Figure 1 – VGP positioning in the ITS reference model** + +## 6.2 General characteristics + +#### 6.2.1 Access to external communication networks + +The VGP has the ability to access communication networks including short-to-medium range wireless communication, such as the legacy dedicated short range communication (DSRC) or IEEE 802.11 (part of wireless access in vehicular environments) and wireless wide area network (WWAN). + +### 6.2.2 Access to in-vehicle networks + +The VGP shall support connectivity to in-vehicle sensors and controller networks in order to get information from, and/or control the vehicle. + +#### 6.2.3 Data process and transmission + +The VGP shall support data transmission among in-vehicle networks, nomadic devices, cloud-based applications, roadside units (RSUs) and vehicle applications. + +The VGP shall also support data formation, quality of service (QoS) management, priority management, etc. + +#### 6.2.4 Service or application interaction + +The VGP shall encapsulate the communication and service capabilities as application programming interfaces (APIs) that can be invoked by external applications or services. + +#### 6.2.5 Secure data transmission + +The VGP shall guarantee that data transmits securely, especially data exchanges between the external network and the in-vehicle data buses (vehicle diagnostics, vehicle control, etc.), and ensure that data cannot be tampered with or stolen. + +The VGP shall support registration and authentication functions for nomadic devices and application access. + +#### **6.2.6 Management of the VGP** + +The VGP shall implement remote or local firmware and software updates, maintenance, configuration, priority strategies set and import, fault management and other functions. + +The VGP shall provide fault prediction reports, diagnostics, recovery and log management. + +# **7 Communication requirements for the VGP** + +## **7.1 External network connection requirements** + +The VGP should support ubiquitous connectivity to external communication networks. The network communication requirements for the VGP are as follows: + +- The VGP should support two-way communications with local area wireless communication networks (e.g., DSRC ranges), WWANs (wireless metropolitan area network (MAN), 3G, 4G, etc.) and very short-range or near-field communication (NFC). +- The VGP should support wired (e.g., universal serial bus (USB)) and/or wireless connectivity with nomadic devices (e.g., smart phones, tablets, audio players) and other portable electronic equipment (e.g., cameras). +- The VGP should support communication with data broadcast networks such as frequency modulation-radio data system (FM-RDS). +- The VGP should be able to communicate with a fixed roadside infrastructure system via WWANs (e.g., wireless MAN, 3G, 4G) or wireless local area networks (WLANs) (e.g., DSRC). +- The VGP should be able to communicate with ad hoc roadside infrastructure (e.g., temporary roadwork beacons). +- The VGP should be able to communicate with other vehicles via WWANs (e.g., wireless MAN, 3G, 4G) or WLANs (e.g., DSRC). +- The VGP should be able to communicate with a road safety device carried by pedestrians and bicycles via WWANs (e.g., wireless MAN, 3G, 4G) or WLANs (e.g., DSRC). + +## **7.2 In-vehicle connection requirements** + +In order to achieve vehicle diagnostics, control or vehicle data acquisition, the VGP should connect to the in-vehicle network. The in-vehicle connection requirements for the VGP are as follows: + +- The VGP shall support the wired and wireless connectivity with the in-vehicle network and allow bidirectional communication with the connected electronic control units (ECUs) and sensors in the vehicle. +- The VGP shall support the control and management functions for communicating between external applications and the ECUs. +- When applications running on nomadic devices require access to in-vehicle networks, the VGP shall support authorisation management. +- When applications running on remote servers require access to in-vehicle networks, the VGP shall support authorisation management. +- The VGP shall support authorised access for services running within the VGP to access in-vehicle networks. When services running over the VGP require access to in-vehicle networks, the VGP shall support authorisation management. + +## **7.3 Communication management requirements** + +The communication management requirements for the VGP are as follows: + +- The VGP should allow external applications accessing the VGP to select the communication interface to use for transmission and reception of data. +- The VGP should allow services running over the VGP to select the communication interface to use for transmission and reception of data. +- The VGP should allow configuration of the transmission power level used for wireless communication with a device attached to the VGP in order to support a variable distance of transmission. +- The VGP should provide communication interfaces with different levels of QoS to cater to different communication scenarios. +- The VGP should provide communication interfaces with different levels of transmission priority settings to cater to different communication scenarios. +- The VGP should support selection of the access network according to certain criteria. For example, according to the type of service or network quality. +- The VGP should have a link layer address that identifies a radio transceiver attached to the VGP. +- The VGP should keep track of the connection status and quality of the connection for both the external networks and in-vehicle networks. + +## **7.4 Communication security requirements** + +The communication security requirements for the VGP are as follows: + +- The VGP shall support access security policies for external equipment and platforms. +- The VGP shall protect against cyberattacks. +- The VGP shall support data exchange security policies on open systems interconnection (OSI) 1-4 layer. +- The VGP shall support network end-to-end security functions to secure data transmission from the VGP to external devices (e.g., cloud servers, other VGP, nomadic devices). +- The VGP shall implement medium access security for communication between the VGP and the in-vehicle network as well as between the VGP and external networks. +- The VGP shall guarantee secure data exchange between the external network and the in-vehicle data bus. +- When the relaying function (see clause I.1) is available, the VGP shall support some security functions to authenticate the request for relaying. +- When the relaying function (see clause I.1) is available, the VGP data relaying service shall be secured at every layer of the data flow. + +## **7.5 Networking requirements** + +- The VGP shall support the routing of packets between an external network and an in-vehicle network. +- The VGP shall support the routing of packets within a single external network or among multiple external networks. +- The VGP shall support IP-based and non-IP-based connections. +- The VGP shall support network address translation (NAT) functions for in-vehicle devices connected to in-vehicle networks and nomadic devices connected via an external network. +- The VGP shall support IPv4 and has optional support for IPv6 protocols. + +- The VGP should support IP session handover across different IP sub-networks using the same wireless interface or different wireless interfaces. +- The communication devices attached to the VGP shall perform unicasting and should be capable of broadcasting and multi-casting messages to other VGPs or external devices. +- The VGP should be able to obtain its location information and transmit data within certain geographical boundaries (geo-casting). +- The VGP should provide data relaying services for a VGP using the same or different wireless communication interfaces. +- Relaying services should include configurable relaying service levels. The delay and bandwidth profile of the end-to-end relaying channel can be configurable based on user profiles and QoS settings. +- The VGP should be configurable to support data relaying. +- The VGP should implement one or several routing schemes to handle the relaying of the data. + +# **8 Service requirements for VGP** + +## **8.1 Driver distraction management requirements** + +To mitigate driver distraction and workload, driver interaction with applications should be managed so that the driver is able to maintain situational awareness. Driver distraction management requirements for the VGP are as follows: + +- The VGP shall acquire situational driving information from external entities. +- The VGP, to mitigate driver distraction, shall control when and how applications are used by the driver while driving the vehicle. + +## **8.2 Session management requirements** + +The VGP is the hub of data/message exchange. Session management requirements for the VGP are as follows: + +- The VGP shall support the routing and dispatching management of data/messages. +- The VGP shall support a data format process to guarantee high efficiency of data transportation. +- The VGP should support standard APIs to communicate with the local/cloud-based applications. + +## **8.3 In-vehicle data resource access management requirements** + +In-vehicle data resource access management is an independent and isolated function to allow data exchange among external applications and the in-vehicle network. In-vehicle data resource access management requirements for the VGP are as follows: + +- The VGP shall support selecting, organizing and converting in-vehicle data required for each managed function type (e.g., air conditioning, infotainment system, seat control, power train) in order to present data in a uniform and generic way. +- The VGP shall support access control functions so that an application or a service has access to only in-vehicle resources for which it has been authorized. +- The VGP shall support external application access to in-vehicle resources in a uniform way for the particular type of in-vehicle buses. + +## **8.4 Driver-vehicle access management requirements** + +A driver-vehicle interface (DVI) is not included in the VGP. However, the VGP can control the interaction between applications and the DVI. The driver-vehicle access management requirements for the VGP are as follows: + +- The VGP shall manage requests from applications to the DVI according to pre-configured policies or driving situations. +- The VGP shall manage instructions from the DVIs to applications according to pre-configured policies or driving situations. +- The VGP should manage the remote UI to approach the interface between applications and the DVI. + +## **8.5 High-layer security requirements** + +If the higher-layer security policies are implemented, the high-layer security requirements for the VGP are as follows: + +- The VGP shall protect against software attacks which target software vulnerabilities. +- The VGP shall support access control policies of applications. +- The VGP shall support user data privacy protection. +- The VGP shall support cipher, key, digital signature and certificate management. + +## **8.6 Software management requirements** + +To maintain the VGP, improvements of current services and the potential deployment of new services inside and outside the VGP, it is required that the VGP provides the following requirements: + +- The VGP shall support software management from external devices or servers in a safe, secure and flexible way. In particular, it shall be able to manage monitoring, adding, removing and updating of software packages. +- The VGP shall support software management of services running inside the VGP in a safe way. +- The VGP shall support software management of services running over in-vehicle devices connected to the VGP through in-vehicle buses in a safe way. +- The VGP shall support software version checking to ensure software version coherency between the various devices and services. +- The VGP shall maintain and provide access to a journal containing all management operations performed by the VGP. This journal must be protected with signature and encryption schemes in order to provide reliability and verifiability of the logs. + +## **8.7 Application data management requirements** + +To maintain the VGP and improvements of current services, it is required that the VGP provides the following requirements: + +- The VGP shall support application data management from external devices or servers in a safe, secure and flexible way. In particular, it shall be able to manage reading, adding, removing and modifying application data. +- The VGP shall support, in a safe way, application data management of services running inside the VGP. +- The VGP shall support, in a safe way, application data management of services running over in-vehicle devices connected to the VGP through in-vehicle buses. + +- The VGP shall support application data version checking to ensure coherency with the software version running over the VGP or in-vehicle devices. +- The VGP shall support application data sharing among different services and access management (ownership, permissions). +- The VGP shall provide the capability to associate some metadata to data entries, such as size, data type. +- The VGP shall maintain and provide access to a journal containing all accesses and operations done to any data entries and their associated metadata. It must be protected with signature and encryption schemes in order to provide reliability and verifiability of the logs. + +# Appendix I + +## VGP use cases + +(This appendix does not form an integral part of this Recommendation.) + +## I.1 Background + +The following entities consume or initiate the interactions with the VGP. They are used to describe the various use cases: + +- Vehicle gateway platform +- Vehicle sensors and input devices (e.g., global navigation satellite system (GNSS), cameras) +- Vehicle controller system (e.g., ECUs) +- Drivers +- Passengers +- Infrastructure systems (e.g., RSU) +- External mobile devices +- In-vehicle nomadic or mobile devices (e.g., navigation devices, tablet PCs) +- Servers (e.g., cloud servers) +- Pedestrians +- Call centres. + +## I.2 Vehicle-to-vehicle interaction use cases + +Vehicle-to-vehicle interaction may occur under several usage scenarios and the interactions may be triggered by various entities, see Figure I.1. + +![Diagram illustrating a vehicle-to-vehicle interaction scenario. Two vehicles are shown, each enclosed in a dashed box labeled 'Vehicle'. Inside each box is a 'VGP' (Vehicle Gateway Platform) block and a car icon. A horizontal orange double-headed arrow connects the VGP of the left vehicle to the VGP of the right vehicle. Within the left vehicle's box, a vertical orange double-headed arrow connects the VGP block to a horizontal line labeled 'Proprietary in-vehicle network', which in turn has a vertical double-headed arrow connecting it to the car icon. The diagram is labeled F.749.2(17)_FI.1.](b90144cfbb81a2d610d920240fda689d_img.jpg) + +The diagram shows two vehicles, each represented by a dashed rectangular box labeled "Vehicle". Inside each box, there is a grey rectangular block labeled "VGP" and a grey car icon. In the left vehicle, a vertical orange double-headed arrow connects the "VGP" block to a horizontal line labeled "Proprietary in-vehicle network", which then connects via another vertical double-headed arrow to the car icon. A horizontal orange double-headed arrow connects the "VGP" block of the left vehicle to the "VGP" block of the right vehicle. The right vehicle's box also contains a "VGP" block and a car icon, but no internal connections are shown. The diagram is labeled F.749.2(17)\_FI.1. + +Diagram illustrating a vehicle-to-vehicle interaction scenario. Two vehicles are shown, each enclosed in a dashed box labeled 'Vehicle'. Inside each box is a 'VGP' (Vehicle Gateway Platform) block and a car icon. A horizontal orange double-headed arrow connects the VGP of the left vehicle to the VGP of the right vehicle. Within the left vehicle's box, a vertical orange double-headed arrow connects the VGP block to a horizontal line labeled 'Proprietary in-vehicle network', which in turn has a vertical double-headed arrow connecting it to the car icon. The diagram is labeled F.749.2(17)\_FI.1. + +Figure I.1 – Vehicle-to-vehicle scenario + +Examples of vehicle-to-vehicle use cases are described in the following clauses. + +### I.2.1 Vehicle-to-vehicle safety scenario + +In the vehicle-to-vehicle safety scenario, real-time information of a vehicle such as its speed, direction, location, etc., is collected from in-vehicle sensors (within the VGP or possibly connected to vehicle controllers) by the VGP and is broadcast to the surrounding vehicles. The surrounding vehicles will then collect this information and use it to derive the safety context of the environment. Once the safety context is fully understood, the vehicles can then derive other safety warnings such as lane change warnings, forward collision warnings, etc. Besides sending raw sensor information, the VGP may also send processed safety context information in the form of alarms or warning + +messages to other vehicles. For example, if a vehicle senses an obstacle ahead, the vehicle could send this information to other nearby vehicles. This information can be sent either in a broadcast, unicast, multicast or geo-cast manner. The processed safety information or warnings that are received by a vehicle may also be displayed by a UI attached to the VGP or sent to the UI of a nomadic mobile device associated with the VGP. + +The vehicle-to-vehicle safety messages may be set as high-priority messages and transmitted over a dedicated wireless channel (frequency or timeslot) or a dedicated radio device. For this vehicle-to-vehicle safety scenario, security functions are implemented at every stage of information gathering, transmission and processing, to ensure the integrity of the data and processed information. Security functions are implemented to ensure that the integrity of the sensory data is not compromised by cyber security attacks. Transmission of data from one hardware/software block to another or from one vehicle to another is authenticated and secured along the entire transmission medium. + +#### **I.2.2 Vehicle-to-vehicle data relaying** + +In the vehicle-to-vehicle relaying scenario, the VGP of a vehicle is used to relay data such as software updates, telematics data or application data to an end destination, which could be a vehicle or an infrastructure system. Such a use case could happen if an infrastructure system, driver, passenger (via mobile device) or vehicle chooses to use a relaying approach to reach the end destination because of lower-cost communication or perhaps due to the lack of other viable communication methods. + +The delay and bandwidth profile of the end-to-end relaying channel can be configurable based on user profiles and QoS settings. The transmission of the vehicle-to-vehicle relayed data could be based on either unicast, broadcast, multicast or geo-cast. Security functions are implemented to ensure end-to-end data integrity and security. The VGPs participating in the relaying of data should ensure that no cyber security attacks to the relayed information have occurred. + +#### **I.2.3 Platooning** + +The platooning scenario can be seen as a special case of a vehicle-to-vehicle safety use case. Real-time information of a vehicle such as its speed, direction or location is collected from the in-vehicle sensors (or possibly vehicle controllers) by the VGP and is broadcasted, unicasted or multicast to the vehicles in the platoon. The lead vehicle that controls the platoon uses this information to derive the movement control information and surrounding safety context information of the environment. + +The processing of the movement control and surrounding safety context information, which is required to implement the platooning functions, can be carried out by a separate vehicle controller or by the VGP. In either case, the movement control information and safety context information is sent to the vehicles in the platoon to control their movement via the VGP. For the platooning scenario, security functions are implemented at every stage of information gathering, transmission and processing to ensure the integrity of the data and processed information. Security functions are implemented to ensure that the integrity of the sensory data is not compromised by cyber security attacks. Transmission of data from one hardware/software block to another or from one vehicle to another is authenticated and secured along the entire transmission medium. + +## **I.3 Vehicle-to-infrastructure interaction use cases** + +Vehicle-to-infrastructure interaction may occur under several usage scenarios and the interactions may be triggered by various entities. Vehicle-to-infrastructure interaction may happen either through WWANs (e.g., cellular networks) or WLANs (e.g., DSRCs) and typically cover the interaction between vehicles and roadside infrastructure (which are normally transportation-related infrastructure physically located within the road networks); see Figure I.2. + +![Diagram of a vehicle-to-infrastructure scenario. On the left, a dashed box labeled 'Infrastructure' contains icons for traffic lights and a digital display showing 'P1 language data' and 'LIMITE 50'. On the right, a dashed box labeled 'Vehicle' contains a 'VGP' (Vehicle Gateway) unit. An orange double-headed arrow connects the VGP to the infrastructure. Below the VGP is a horizontal line representing a 'Proprietary in-vehicle network', which is connected via a white double-headed arrow to a car icon at the bottom. The text 'F.749.2(17)_FI.2' is at the bottom right.](76b0cd79baaedd942af4dc42f2e764b8_img.jpg) + +Diagram of a vehicle-to-infrastructure scenario. On the left, a dashed box labeled 'Infrastructure' contains icons for traffic lights and a digital display showing 'P1 language data' and 'LIMITE 50'. On the right, a dashed box labeled 'Vehicle' contains a 'VGP' (Vehicle Gateway) unit. An orange double-headed arrow connects the VGP to the infrastructure. Below the VGP is a horizontal line representing a 'Proprietary in-vehicle network', which is connected via a white double-headed arrow to a car icon at the bottom. The text 'F.749.2(17)\_FI.2' is at the bottom right. + +**Figure I.2 – Vehicle-to-infrastructure scenario** + +Examples of vehicle-to-infrastructure use cases are described in the following clauses. + +### I.3.1 Vehicle-to-infrastructure safety + +Vehicle-to-infrastructure safety scenarios involve the interaction between a vehicle and roadside infrastructure to provide navigation safety to drivers. One example is an intersection safety system which consists of roadside sensors that collect contextual information of positions and movement of vehicles and other road users. The system will then process the trajectory of the detected objects and provide warnings to vehicles or drivers navigating an intersection. + +Another vehicle-to-infrastructure safety service is the traffic controller signal phase and timing information, which can be used by the vehicle or driver to plan their movement through an intersection. In these vehicle-to-infrastructure examples, processed safety information can be broadcast to the vehicles from the roadside infrastructure. Vehicles can also assist the safety system by sending their real-time information such as speed, direction, location, etc., which is collected from the in-vehicle sensors (within the VGP or possibly connected to vehicle controllers) by the VGP to the roadside infrastructure. For example, if a vehicle senses an obstacle ahead, the vehicle could send this information to nearby vehicles. The information from the roadside infrastructure can be sent either in a broadcast, unicast or geo-cast manner. The processed safety information or warnings that are received by a vehicle can be displayed by a UI attached to the VGP or the UI of a nomadic device associated to the VGP. Some vehicle-to-infrastructure safety services such as vehicle priority at intersections may require a longer distance communication with the roadside infrastructure. In such cases, the VGP in the vehicle requesting the service may select a different radio communication unit or change its power setting to ensure a longer distance communication. While the examples given above describe a permanent roadside infrastructure setup, there can also be ad hoc deployment of roadside infrastructure such as temporary road works. The interaction of such a system is expected to be the same as the permanent setup. The vehicle-to-infrastructure safety messages may be set as high priority messages and transmitted over a dedicated wireless channel. The messages sent from the roadside infrastructure and vice versa is secured at every stage of the message flow. Security functions are implemented to ensure that the integrity of the sensory data from the vehicle and the messages from the infrastructure are not compromised by cyber security attacks. + +### I.3.2 Vehicle-to-infrastructure traffic management + +The vehicle-to-infrastructure traffic management use case involves the use of roadside infrastructure to provide traffic management functions and services such as ETC, probe data collection from vehicles, traffic information upload and download. The interaction between the VGP and the roadside infrastructure is similar for these services. In the road toll collection system, the VGP may be connected to a payment system (pre-paid or post-paid). The payment transaction for the ETC system involving the VGP and the roadside infrastructure must be secured to ensure no fraudulent payment. For the probe data collection, the VGP may upload a pre-recorded probe data collected by the VGP + +or real-time data to the infrastructure device. Similar to the vehicle-to-infrastructure safety use case, security functions are implemented to ensure that the integrity of the data from the vehicle to the infrastructure and vice versa is not compromised by cyber security attacks. + +#### I.3.3 Vehicle-to-infrastructure commercial and access control + +Vehicle-to-infrastructure commercial and access control services could be services provided by commercial companies, transport authorities or other governmental agencies. For example, commercial services could include parking management, rental car transaction, vehicle diagnostic and fuelling station. In most cases, the interaction between the infrastructure and vehicle is similar to the vehicle-to-infrastructure traffic management use case, whereby there is an infrastructure device at the roadside. In some cases, it may involve the interaction between a third-party mobile device (external but nearby the vehicle) and the VGP. One such case is the use of mobile diagnostic equipment carried by vehicle service personnel to query information about a vehicle's status. In this case, the VGP will allow authorised access to the vehicle's ECUs. In another case, a parking attendant may use a mobile device to communicate with the VGP to request parking status from the vehicle. Some of these services may even use an NFC to communicate between the external mobile device and the VGP. Access control services include border crossing applications and drive-through applications. In these applications, the use of secure credentials and unique identification to identify the vehicle or its owner is important. In some cases, for example, when a vehicle is in a service centre, the authentication approach may even involve two-level authentication, which requires the user to key in a special code via the VGP's UI. + +## I.4 Vehicle-to-nomadic device interaction use cases + +Vehicle-to-nomadic device interaction use cases include the interoperation among the VGP and nomadic electronic products such as mobile devices, personal productivity devices (e.g., notebooks, tablets) and navigation devices. See Figure I.3. + +As nomadic devices can cause driving distractions, the interaction and interfaces to the applications of vehicle-to-nomadic devices need to be carefully designed. Nomadic devices may be used for starting voice calls or accessing telematics applications. Nomadic devices may also be used by drivers or passengers within a vehicle to access Internet applications. Some examples of vehicle-to-nomadic device usage categories are as follows: + +![Diagram illustrating the scenario of vehicle-to-nomadic device interaction. A dashed box labeled 'Vehicle' contains a 'Nomadic device' (represented by a smartphone and a tablet) connected via a double-headed orange arrow to a 'VGP' (Vehicle Gateway) box. The 'VGP' is connected via a double-headed orange arrow to a horizontal line representing the 'Proprietary in-vehicle network'. Below the network line, a car icon is shown, connected to the network by a double-headed black arrow. The text 'F.749.2(17)_FI.3' is visible in the bottom right corner of the diagram area.](a734898ce18e972938949637c32a34f4_img.jpg) + +The diagram shows a 'Vehicle' (enclosed in a dashed box) containing a 'Nomadic device' (smartphone and tablet), a 'VGP' (Vehicle Gateway), and a 'Proprietary in-vehicle network' (represented by a horizontal line). The 'Nomadic device' is connected to the 'VGP' via a double-headed orange arrow. The 'VGP' is connected to the 'Proprietary in-vehicle network' via a double-headed orange arrow. Below the network line, a car icon is shown, connected to the network by a double-headed black arrow. The text 'F.749.2(17)\_FI.3' is visible in the bottom right corner of the diagram area. + +Diagram illustrating the scenario of vehicle-to-nomadic device interaction. A dashed box labeled 'Vehicle' contains a 'Nomadic device' (represented by a smartphone and a tablet) connected via a double-headed orange arrow to a 'VGP' (Vehicle Gateway) box. The 'VGP' is connected via a double-headed orange arrow to a horizontal line representing the 'Proprietary in-vehicle network'. Below the network line, a car icon is shown, connected to the network by a double-headed black arrow. The text 'F.749.2(17)\_FI.3' is visible in the bottom right corner of the diagram area. + +**Figure I.3 – Scenario of vehicle-to-nomadic device interaction** + +**Example 1:** A driver can start/hang up a voice call through the UI (e.g., touch screen, button, microphone) attached to the VGP and through the communication link in a nomadic device. The VGP may synchronize the application data (e.g., address book) with the nomadic device. + +**Example 2:** The VGP can be used as a personal hotspot for nomadic devices carried by a driver or passengers. Pre-configured pairing or per-session pairing (with perhaps manual password entry) to access to the in-vehicle hotspot is possible. In this scenario, the VGP must ensure that the UI operation causes minimal driver distraction. + +**Example 3:** Nomadic devices may also interoperate with the VGP to access UI support functions such as automatic speech recognition (ASR)/text to speech (TTS)/remote UI technologies that can reduce driver distraction. For example, a personal navigation device can be used for navigation purposes. In this scenario, the device may have its own sensors such as GNSS/inertial navigation system (INS) or may access the VGP to obtain geographic information. The navigation device can use remote UI technology to project the applications on the vehicle's screen via the VGP. The navigation device may also access the audio devices through the VGP to output the audio feedback of route guidance or use the TTS/ASR applications to provide an audio input from the driver to control the navigation device. The VGP may connect to a sensor such as a video camera that can track a driver's distraction level to decide on what kind of UI can be used by the driver when using nomadic devices. + +## I.5 Vehicle-to-cloud interaction use cases + +The vehicle-to-cloud use cases include the interoperation between the vehicle and the cloud-based servers; see Figure I.4. + +Within a vehicle, the originator of information could be a passenger, the driver, in-vehicle mobile devices, VGPs or the vehicle owners. In the vehicle-to-cloud use cases, the interaction occurs through the WWAN (e.g., cellular networks) or short/medium range (e.g., DSRC, IEEE 802.11) communication infrastructure. + +Some examples of vehicle-to-cloud use cases are described in the following clauses. + +![Diagram illustrating the scenario of vehicle-to-cloud-based server interaction.](e636d7ccca0ad14c6b95201404324823_img.jpg) + +The diagram illustrates the interaction between a vehicle and cloud-based servers. At the top, a dashed box labeled 'Cloud-based servers' contains three icons: 'Emergency call system', 'TSP', and 'UBI'. Below this box is a cellular tower icon with signal waves, labeled 'WWAN'. A double-headed orange arrow connects the cloud servers to the tower. Below the tower is another dashed box labeled 'Vehicle'. Inside this box, a car icon is at the bottom, connected by a double-headed white arrow to a horizontal line labeled 'Proprietary in-vehicle network'. Above this line is a grey box labeled 'VGP', which is connected to the network line by a double-headed orange arrow. The VGP is also connected to the WWAN tower by a double-headed orange arrow. The text 'F.749.2(17)\_FI.4' is located at the bottom right of the diagram. + +Diagram illustrating the scenario of vehicle-to-cloud-based server interaction. + +**Figure I.4 – Scenario of vehicle-to-cloud-based server** + +### **I.5.1 Vehicle-to-cloud emergency scenario** + +When a crash or accident occurs, an emergency call to the call centre of the telematics service provider (TSP) or public service answering point (PSAP) is initiated automatically or by manually pressing the emergency call button in the vehicle. At the same time, the geolocation data, accident information and vehicle identification number (VIN) of the vehicle are transmitted to the TSP or PSAP. The TSP or PSAP operator communicates to the public rescue agency immediately to take action and provide medical and/or roadside assistance. A good example of this use case is the European Union's eCall plan for all of Europe, which strives to shorten rescue time in accidents and save many lives each year. + +### **I.5.2 Vehicle-to-cloud telematics scenario** + +Some examples of a TSP's services include: emergency call, B-call, anti-theft, remote diagnostic and control, navigation assistance, eco-driving, infotainment and remote software update. + +The following is a non-exhaustive list of vehicle-to-cloud telematics scenarios: + +- B-call scenario: When a vehicle breaks down, the driver presses a button in the vehicle to start a call to the call centre and transmits the GNSS information of the vehicle. The operator at the call centre contacts the rescue agency to provide quick roadside assistance. +- Anti-theft scenario: When a vehicle is stolen, the driver can start a call to the call centre. The operator can communicate with the vehicle and remotely track the vehicle. The operator can take proper actions, such as alerting the police, immobilizing the vehicle, etc. +- Remote control scenario: Drivers can remotely control heating and ventilation in the vehicle to start during winter/summer by using a mobile phone or calling the call centre. Also, in this scenario, when a driver forgets their key, they can unlock the vehicle by using an authorised mobile phone or calling the call centre. +- Eco-driving scenario: Eco-driving can promote good driving behaviour, increase fuel-efficiency and reduce carbon emissions. The VGP collects real-time vehicle operation information and transmits this information to a remote telematics platform. The telematics platform analyses the data and sends driving suggestion reports to the driver. +- Navigation assistance scenario: Navigation assistance provides the functions of path planning and navigation. Drivers can press a button to start a call to the call centre and speak with an operator. The driver can provide their destination and require assistance. The operator searches for the best routes and sends this information to the VGP. The VGP can either transmit this information to the vehicle's built-in navigation system or to a nomadic device connected to the VGP. + +Other examples of telematics applications are infotainment applications. These applications include: weather conditions, point of interest (POI) searching (e.g., gas stations, parks and restaurants), booking services (e.g., plane tickets, restaurants, hotels), online news and online music. + +Most of the telematics applications mentioned above require a remote software update feature. Remote service or application update may occur in the VGP or nomadic devices attached to the VGP. The VGP may periodically connect to various service providers' cloud servers to retrieve software binary updates and to apply bug fixes or version updates to provide new features. Once the files are downloaded, the VGP can help coordinate the installation of the files to the corresponding modules. The entire process of software updates to the nomadic devices or the VGP must be secure to protect against cyber security attacks. For applications downloaded to a nomadic device, proper authentication with the VGP must be present to ensure secure communication. + +#### **I.5.3 Vehicle-to-cloud usage-based insurance** + +Usage-based insurance (UBI) service involves the collection, by insurance companies, of actual driver behaviour data and is used for linking premium pricing schemes in accordance with actual mileage and driving behaviour. + +Data can be collected with the aid of the VGP, a digital map with current speed limits for various road links, etc., and matching how, where, when, and at what speeds the vehicle is driven. These data are transmitted to the insurance company and reduce ambiguous information about driving patterns making it possible to charge an individual a customised premium. + +## **I.6 Vehicle-to-pedestrian and bicycle interaction use cases** + +The vehicle-to-pedestrian and bicycle use cases cover mainly the safety aspects of pedestrians or vulnerable users. In the vehicle-to-pedestrian and bicycle use cases, a mobile device carried by road users would send out its location via a dedicated short-range communication medium (e.g., DSRC) to warn drivers. For example, when a pedestrian appears from behind a parked car or other obstruction, the driver in the vehicle would receive an alert. The pedestrian may also get an alert on their mobile device. The authentication of the mobile device in this use case is important to protect against cyber security attacks or unwanted disruption to the operation of the vehicle. + +# Bibliography + +- [b-ITU-T F.749.1] Recommendation ITU-T F.749.1 (2015), *Functional requirements for vehicle gateways*. +- [b-ITU-T Y.2012] Recommendation ITU-T Y.2012 (2010), *Functional requirements and architecture of next generation networks*. +- [b-ITU-R M.1797] Recommendation ITU-R M.1797 (2007), *Vocabulary of terms for the land mobile service*. +- [b-ETSI EN 302 665] ETSI EN 302 665 v1.1.1 (2010-09), *Intelligent Transport Systems (ITS); Communications; Architecture*. +- [b-ISO 7498-1] ISO 7498-1 (1994), *Information technology – Open Systems Interconnection – Basic Reference Model: The Basic Model*. +- [b-ISO 15638-1] ISO 15638-1 (2012), *Intelligent transport systems – Framework for collaborative Telematics Applications for Regulated commercial freight Vehicles (TARV) – Part 1: Framework and architecture*. +[\[https://www.iso.org/standard/59184.html\]](https://www.iso.org/standard/59184.html) +- [b-ISO 21210] ISO 21210 (2012), *Intelligent transport systems – Communications access for land mobiles (CALM) – IPv6 Networking*. +- [b-ISO 21212] ISO 21212 (2008), *Intelligent transport systems – Communications access for land mobiles (CALM) – 2G Cellular systems*. +- [b-ISO 21213] ISO 21213 (2008), *Intelligent transport systems – Communications access for land mobiles (CALM) – 3G Cellular systems*. +- [b-ISO 21214] ISO 21214 (2015), *Intelligent transport systems – Communications access for and mobiles (CALM) – Infra-red systems*. +- [b-ISO 21216] ISO 21216 (2012), *Intelligent transport systems – Communications access for land mobiles (CALM) – Millimetre wave air interface*. +- [b-ISO 24102-1] ISO 24102 (2013), *Intelligent transport systems – Communications access for land mobiles (CALM) – ITS station management – Part 1: Local management*. +[\[https://www.iso.org/standard/61561.html\]](https://www.iso.org/standard/61561.html) +- [b-ISO 24102-2] ISO 24102 (2015), *Intelligent transport systems – Communications access for land mobiles (CALM) – ITS station management – Part 2: Remote management of ITS-SCUs*. +[\[https://www.iso.org/standard/61563.html\]](https://www.iso.org/standard/61563.html) +- [b-ISO 24102-3] ISO 24102 (2013), *Intelligent transport systems – Communications access for land mobiles (CALM) – ITS station management – Part 3: Service access points*. +[\[https://www.iso.org/standard/61564.html\]](https://www.iso.org/standard/61564.html) +- [b-ISO 24102-4] ISO 24102 (2013), *Intelligent transport systems – Communications access for land mobiles (CALM) – ITS station management – Part 4: Station-internal management communications*. +[\[https://www.iso.org/standard/61565.html\]](https://www.iso.org/standard/61565.html) +- [b-ISO 24102-5] ISO 24102 (2013), *Intelligent transport systems – Communications access for land mobiles (CALM) – ITS station management – Part 5: Fast service advertisement protocol (FSAP)*. +[\[https://www.iso.org/standard/61566.html\]](https://www.iso.org/standard/61566.html) + +[b-ISO 29281] + +ISO 29281 (2011), *Intelligent transport systems – Communications access for land mobiles (CALM) – Non-IP networking*. +[] + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.749.3-202008-I_PDF-E/raw.md b/marked/F/T-REC-F.749.3-202008-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..1f62670325302dcaba3be59015d8c0df328b9981 --- /dev/null +++ b/marked/F/T-REC-F.749.3-202008-I_PDF-E/raw.md @@ -0,0 +1,1618 @@ + + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.749.3** + +(08/2020) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Use cases and requirements for vehicular +multimedia networks** + +Recommendation ITU-T F.749.3 + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +## Recommendation ITU-T F.749.3 + +## Use cases and requirements for vehicular multimedia networks + +## Summary + +Recommendation ITU-T F.749.3 specifies use cases and requirements for vehicular multimedia networks (VMN), including an overview, connectivity, intelligent human machine interfaces (HMI) for the vehicle multimedia system (VMS), privacy considerations, content rights protection in a converged network environment, copyright and rights management support for content delivery, security, safety, and definitions of vehicular multimedia configurations. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.749.3 | 2020-08-13 | 16 | 11.1002/1000/14330 | + +## Keywords + +Copyright, human machine interface, network convergence transmission, privacy, safety, security, vehicular multimedia. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2020 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | | Page | +|------|-----------------------------------------------------------------------------------------------------------------------------------|------| +| 1 | Scope ..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 1 | +| 4 | Abbreviations and acronyms ..... | 2 | +| 5 | Conventions ..... | 4 | +| 6 | Overview of the vehicular multimedia networks..... | 4 | +| 6.1 | Relationship with SAE J3016 levels of driving automation ..... | 6 | +| 7 | Connectivity..... | 8 | +| 7.1 | Distinction between brought-in, built-in and hybrid connectivity ..... | 8 | +| 7.2 | Multiple network connectivity in the vehicular multimedia context..... | 10 | +| 7.3 | Network convergence transmission and vehicle-onboard processing..... | 12 | +| 7.4 | Networking connection among multiple terminals within the vehicle..... | 13 | +| 7.5 | Continuous and consistent services across various networks and environments (from/to VMS)..... | 15 | +| 8 | Intelligent human machine interfaces (HMI) for VMS ..... | 17 | +| 8.1 | The acoustic system in the vehicular multimedia..... | 19 | +| 8.2 | Gesture control in vehicular multimedia system ..... | 21 | +| 8.3 | Eye-movement control in VMS..... | 21 | +| 8.4 | Data formats and storage ..... | 23 | +| 9 | Privacy considerations ..... | 23 | +| 9.1 | Background..... | 23 | +| 9.2 | General privacy requirements..... | 24 | +| 9.3 | Use case – Private vehicle with one single user ..... | 25 | +| 9.4 | Use case – Long distance shared vehicle with multi users (shuttle all users are seated)..... | 25 | +| 9.5 | Use case – Public transportation shared vehicle with multi users (shuttle short distance trips – seated and standing users) ..... | 26 | +| 10 | Content rights protection in a converged network environment ..... | 26 | +| 10.1 | Background..... | 26 | +| 10.2 | Use cases ..... | 28 | +| 10.3 | Gap analysis..... | 29 | +| 10.4 | Requirements – Content rights protection requirements for CA..... | 30 | +| 11 | Copyright and rights management support for content delivery ..... | 30 | +| 11.1 | Use case – Content payment and copyright protection ..... | 30 | +| 11.2 | Privilege and account management ..... | 31 | +| 11.3 | Payment services in shared vehicle ..... | 33 | + +| | Page | +|-------------------------------------------------------------|------| +| 12 Security ..... | 33 | +| 12.1 Use case – User data protection..... | 33 | +| 12.2 Use case – VMN application security ..... | 34 | +| 12.3 Use Case – VMS software security ..... | 34 | +| 13 Safety ..... | 34 | +| 13.1 Use case – Driving safety: Speed control..... | 34 | +| 14 Definition of vehicular multimedia configurations ..... | 35 | +| 14.1 Use case – Vehicular multimedia configurations..... | 35 | +| 15 Matching of described use cases with driving levels..... | 36 | +| Appendix I – Examples of VMS reference configurations ..... | 37 | +| I.1 M0 configuration ..... | 37 | +| I.2 M1 configuration ..... | 37 | +| I.3 M2 configuration ..... | 38 | +| I.4 M3 configuration ..... | 39 | +| I.5 M4 configuration ..... | 40 | +| I.6 M5 configuration ..... | 41 | +| Bibliography..... | 44 | + +# Recommendation ITU-T F.749.3 + +## Use cases and requirements for vehicular multimedia networks + +## 1 Scope + +This Recommendation describes the use-cases and requirements for vehicular multimedia networks (VMN), taking into account the autonomous levels defined by [SAE J3016] and used in conjunction with advanced driver assistance system (ADAS), automatic driving technologies, maps, voice commands or natural language processing. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-R BT.2055-1] Recommendation ITU-R BT.2055-1 (2018), *Content elements in multimedia broadcasting systems for mobile reception*. + +[SAE J3016] SAE, Standard J3016 (2018), *Taxonomy and Definitions for Terms Related to Driving Automation Systems for On-Road Motor Vehicles*. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 vehicle gateway (VG)** [b-ITU-T F.749.1]: A VG is a device in a vehicle that enables communications between a device in the vehicle and another device which may be physically located either inside the vehicle or outside the vehicle (e.g., roadside station, cloud-based server, etc.). A VG provides standardized interfaces and protocols, communications across heterogeneous networks, optimized network selection based on application needs and network QoS, arbitration and integration of network communications, security and switching network connections to maintain service continuity. + +**3.1.2 vehicle gateway platform (VGP)** [b-ITU-T F.749.1]: A VGP is the collection of ICT hardware and software in a vehicle operating as an open platform to provide an integrated runtime environment for delivering the communications services of a VG. A VGP may also provide higher layer communications services such as interaction with the driver through the driver-vehicle access services and so on. Subsystems dedicated solely to vehicle operation are not considered part of the VGP. Supported applications/services include ITS and infotainment. + +**3.1.3 telematics** [b-ISO 15638-1]: Telematics is the use of wireless media to obtain and transmit (data) from a distant source. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +- 3.2.1 vehicular multimedia networks (VMN):** The VMN consist of the vehicular multimedia service platform (VMSP), broadcast and communication networks, and the vehicle multimedia system (VMS) in the vehicle. +- 3.2.2 vehicular multimedia service platform (VMSP):** The VMSP is a platform in the cloud that enables the multimedia service for end-user(s) in the vehicle. +- 3.2.3 vehicle multimedia system (VMS):** The VMS consists of vehicle multimedia system inputs (VM I/P), vehicle multimedia unit (VMU) and vehicle multimedia system outputs (VM O/P). +- 3.2.4 VMN services (VMNS):** The VMNS are the vehicular multimedia services provided by the VMSP to the end-user(s) in the vehicle. +- 3.2.5 VMN application:** The VMN application uses the underlying vehicle multimedia system (VMS) capabilities to consume and present VMN services (VMNS) to end-user(s) in the vehicle. +- 3.2.6 in vehicle infotainment (IVI) service:** The IVI service is a service that provides both information and/or entertainment for the end-user(s) in the vehicle. +- 3.2.7 network convergence transmission:** Network convergence transmission is an intelligent transmission technique used by the vehicular multimedia service platform (VMSP) to efficiently and timely deliver the VMN services (VMNS) data to the vehicle multimedia system (VMS) via broadcasting and communication networks. +- 3.2.8 vehicle-onboard processing:** Vehicle-onboard processing is the signal processing of the vehicle multimedia system (VMS), which includes the reception, consumption, and redistribution of VMN services (VMNS) data in the vehicle. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms. + +| | | +|------|------------------------------------------| +| ADAS | Advanced Driver Assistance System | +| AEC | Acoustic Echo Cancelation | +| AM | Amplitude Modulation | +| APP | Application | +| AR | Augmented Reality | +| BCM | Body Control Module | +| CA | Conditional Access | +| CDR | China Digital Radio | +| CW | Control Word | +| DAB | Digital Audio Broadcast | +| DRM | Digital Rights Management | +| DTMB | Digital Terrestrial Multimedia Broadcast | +| DVB | Digital Video Broadcasting | +| DVR | Digital Video Recorder | +| ECM | Entitlement Control Message | +| ECU | Electronic Control Unit | +| EMM | Entitlement Management Message | +| FG | Focus Group | + +| | | +|--------|----------------------------------------| +| FM | Frequency Modulation | +| GDPR | General Data Privacy Regulation | +| GNSS | Global Navigation Satellite System | +| HD | High Definition | +| HMI | Human Machine Interface | +| HUD | Heads Up Display | +| HUT | Infotainment Head Unit | +| ID | Identification | +| ITS | Intelligent Transport System | +| IVI | In Vehicle Infotainment | +| KWS | Keyword Spotter | +| LCD | Liquid Crystal Display | +| LEO | Low Earth Orbit (Satellite) | +| LIDAR | Light Detection and Ranging | +| MMS | Multimedia Messaging Service | +| OEM | Original Equipment Manufacturer | +| OLED | Organic Light Emitting Diode | +| PDK | Personal Distribution Key | +| PII | Personally Identifiable Information | +| PK | Personalisation Key | +| PK | Product Key | +| QoS | Quality of Service | +| RDS | Radio Data System | +| RF | Radio Frequency | +| SMS | Short Message Service | +| T-Box | Telematics System | +| TM-Box | Telematics Multimedia System | +| TS | Transport Stream | +| USB | Universal Serial Bus | +| VG | Vehicle Gateway | +| VGP | Vehicle Gateway Platform | +| VM I/P | Vehicle Multimedia System Inputs | +| VM O/P | Vehicle Multimedia System Outputs | +| VMN | Vehicular Multimedia Networks | +| VMNS | Vehicular Multimedia Networks Services | +| VMSP | Vehicular Multimedia Service Platform | +| VMS | Vehicle Multimedia System | + +| | | +|------|-------------------------------| +| VMU | Vehicle Multimedia Unit | +| VR | Virtual Reality | +| WLAN | Wireless Local Area Network | +| WG | Working Group | +| ZIC | Zone Interference Cancelation | + +## 5 Conventions + +In this Recommendation: + +- The keywords "shall" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "should" or "may" indicate an optional requirement which is permissible. This term is not intended to imply that the vendor's implementation must provide the option, and the feature can be optionally enabled by the vendor. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +## 6 Overview of the vehicular multimedia networks + +As shown in Figure 1, vehicle multimedia networks (VMN) consist of the vehicular multimedia service platform (VMSP) in the cloud, broadcast and communication networks, and the vehicle multimedia system (VMS) in the vehicle. + +![Figure 1 – Vehicular multimedia networks. A diagram showing the VMSP (vehicular multimedia service platform) connected to a Cellular Network and a Broadcast Network. The Cellular Network is connected to a Short range network, which is connected to the VMS vehicle multimedia system in the vehicle. The Broadcast Network is also connected to the VMS vehicle multimedia system in the vehicle. The VMS vehicle multimedia system is connected to Other domains/subsystems (i.e., ADAS) and Nomadic and vehicle devices. The VMSP provides Multimedia services, Infotainment application, Intelligent voice interaction, High precision navigation (maps), Security update, and (software, certificates).](053f1077d592e6622cd21dc4bb4cb366_img.jpg) + +The diagram illustrates the architecture of vehicular multimedia networks (VMN). At the top left, the VMSP (vehicular multimedia service platform) is shown as a blue oval containing a list of services: Multimedia services, Infotainment application, Intelligent voice interaction, High precision navigation (maps), Security update, and (software, certificates). It has a bidirectional connection to a 'Cellular Network' cloud. The 'Cellular Network' cloud has a bidirectional connection to a 'Short range network' cloud (represented by a Wi-Fi icon) and a unidirectional connection to a 'Broadcast Network' cloud (represented by a satellite and antenna icon). The 'Short range network' cloud has a bidirectional connection to the 'VMS vehicle multimedia system' inside a car. The 'VMS vehicle multimedia system' is connected to 'Other domains/subsystems (i.e., ADAS)' and 'Nomadic and vehicle devices' (represented by icons of a smartphone, tablet, and microphone). The 'Broadcast Network' cloud also has a unidirectional connection to the 'VMS vehicle multimedia system' inside the car. The car itself contains the 'VMS vehicle multimedia system', 'Other domains/subsystems (i.e., ADAS)', and 'Nomadic and vehicle devices'. + +Figure 1 – Vehicular multimedia networks. A diagram showing the VMSP (vehicular multimedia service platform) connected to a Cellular Network and a Broadcast Network. The Cellular Network is connected to a Short range network, which is connected to the VMS vehicle multimedia system in the vehicle. The Broadcast Network is also connected to the VMS vehicle multimedia system in the vehicle. The VMS vehicle multimedia system is connected to Other domains/subsystems (i.e., ADAS) and Nomadic and vehicle devices. The VMSP provides Multimedia services, Infotainment application, Intelligent voice interaction, High precision navigation (maps), Security update, and (software, certificates). + +Figure 1 – Vehicular multimedia networks + +A reference model of the VMS is shown in Figure 2, where the overall boundary of the VMS is given. The VMS consists of vehicle multimedia system inputs (VM I/P), vehicle multimedia unit (VMU) and vehicle multimedia system outputs (VM O/P). + +![Figure 2: A reference model for data input/output of the vehicle multimedia system. The diagram shows a central VMU (Vehicle Multimedia Unit) connected to various input and output devices. Inputs include ADAS, BCM, etc., Camera, Microphone, Hard keys, A/V devices, and various antennas (Broadcast satellite Rx, Cellular Com. Rx, LEO satellite Rx, Short-range Com. Rx). Outputs include Display, Speaker, and various antennas (Cellular Com. Tx, LEO satellite Tx, Short-range Com. Tx). Nomadic devices are also connected to the VMU.](cfef993dcc8fb513de79eb1f93cf26ae_img.jpg) + +``` + +graph TD + subgraph VMS + subgraph VM_IP [VM I/P] + Camera + Microphone + HardKeys[Hard keys] + AVDevices[A/V devices] + BSRx[Broadcast satellite Rx Ant.] + CCRx[Cellular Com. Rx Ant.] + LSORx[LEO satellite Rx Ant.] + SRCRx[Short-range Com. Rx Ant.] + end + + VMU + DP[Display-processing] + + subgraph VM_OP [VM O/P] + Display + Speaker + CCTx[Cellular Com. Tx Ant.] + LSOTx[LEO satellite Tx Ant.] + SRCTx[Short-range Com. Tx Ant.] + end + end + + ADAS -- ADAS input --> VMU + BCM[BCM, etc.] -- Data --> VMU + + Camera -- Video --> VMU + Microphone -- Audio --> VMU + HardKeys -- Control --> VMU + AVDevices -- A/V --> VMU + BSRx -- RF --> VMU + CCRx -- RF --> VMU + LSORx -- RF --> VMU + SRCRx -- RF --> VMU + + VMU -- Display input --> DP + DP -- Data --> Display + DP -- Video --> Display + VMU -- Audio --> Speaker + VMU -- RF --> CCTx + VMU -- RF --> LSOTx + VMU -- RF --> SRCTx + + NomadicDevices[Nomadic devices] <--> VMU + +``` + +The diagram illustrates the reference model for data input/output of the vehicle multimedia system. At the top, two external components, ADAS and BCM, etc., provide input to the system. ADAS provides 'ADAS input' and BCM, etc. provide 'Data'. These inputs enter a dashed box labeled 'VMS' (Vehicle Multimedia System). Inside the VMS box, on the left, is the 'VM I/P' (Vehicle Multimedia Input) section, which contains several input devices: Camera (sending 'Video'), Microphone (sending 'Audio'), Hard keys (sending 'Control'), A/V devices (sending 'A/V'), Broadcast satellite Rx Ant. (sending 'RF'), Cellular Com. Rx Ant. (sending 'RF'), LEO satellite Rx Ant. (sending 'RF'), and Short-range Com. Rx Ant. (sending 'RF'). All these signals are directed to the central 'VMU' (Vehicle Multimedia Unit). The VMU is connected to a 'Display-processing' unit. The Display-processing unit sends 'Display input', 'Data', and 'Video' to the 'VM O/P' (Vehicle Multimedia Output) section on the right. The VM O/P section contains output devices: Display, Speaker (receiving 'Audio'), Cellular Com. Tx Ant. (receiving 'RF'), LEO satellite Tx Ant. (receiving 'RF'), and Short-range Com. Tx Ant. (receiving 'RF'). Below the VMS box, 'Nomadic devices' are connected to the VMU via a bidirectional arrow. The diagram is labeled 'F.749.3(20)\_F02' in the bottom right corner. + +Figure 2: A reference model for data input/output of the vehicle multimedia system. The diagram shows a central VMU (Vehicle Multimedia Unit) connected to various input and output devices. Inputs include ADAS, BCM, etc., Camera, Microphone, Hard keys, A/V devices, and various antennas (Broadcast satellite Rx, Cellular Com. Rx, LEO satellite Rx, Short-range Com. Rx). Outputs include Display, Speaker, and various antennas (Cellular Com. Tx, LEO satellite Tx, Short-range Com. Tx). Nomadic devices are also connected to the VMU. + +**Figure 2 – A reference model for data input/output of the vehicle multimedia system** + +System inputs may include audio and video signals from camera, microphone and audio/video devices such as universal serial bus (USB) devices, SD-cards, CD/DVD/Blu-ray and Aux-In, control signals from hard keys, and radio frequency (RF) signals from reception antennas. Connectivity functions and the related recommendations for intelligent transport systems have been discussed in [b-ITU-T Q27/16] and may be reused for the connectivity in vehicle multimedia systems e.g., [b-ITU-T F.749.1], [b-ITU-T F.749.2], [b-ITU-T H.550] and [b-ITU-T H.560]. + +VMU is a central processing unit that processes the received input and output audio via the speakers, video via displays, etc. + +Antenna systems for cellular communications, low earth orbit (LEO) satellites and short-range communications are used for reception and transmission. + +VMU is not a single physical unit. It may consist of multiple electronic control units (ECUs). Options include either (a) infotainment head unit (HUT) and telematics system (T-Box), (b) telematics multimedia system (TM-Box), or (c) any other integrated solutions customized by automotive manufacturers such as a HUT. + +Note that a T-box alone that is compatible with [b-ITU-T F.749.2] does not satisfy the requirements of a vehicle multimedia system, which include but is not limited to the connectivity requirements and the multimedia processing requirements. Nevertheless, the connectivity capability of the T-box + +could be used to partially satisfy the connectivity requirements of the vehicle multimedia system. In such a case, option (a) as described above is a sound solution for the vehicle multimedia system, where HUT together with T-Box fulfils all the requirements of a vehicle multimedia system. + +A VMN application is either a pre-installed or a downloaded application on the VMS available in the vehicle. The VMS application makes use of the VMS capabilities to present the VMNS to end-user(s) in the vehicle. The VMN application is configured to retrieve the VMS capabilities and connect to the VMNS. There may be multiple VMN applications in a vehicle and at different seat/screen positions. + +Other system inputs such as those from the advanced driver assistance system (ADAS), body control module (BCM), etc., are processed by the VMU mainly for display. Nevertheless, for autonomous driving capable vehicles, the interaction between autonomous driving domains/function blocks or ADAS and the VMS need to be addressed in detail in the future. For example, the system could limit some VMS functions (e.g., no broadcasting or no conference calls) based on ADAS domain input that the vehicle is operating at level 3 or below, while it would be allowed for level 4/5 particularly for the driver's position. + +Nomadic devices such as mobile phones and handheld wireless devices are another group of devices whose data are processed by VMU for audio, video and control, but they are not part of the VMS. + +With the development of ADAS and autonomous driving, the multimedia trend for a vehicular multimedia system includes: + +- Multi-screens; +- Multi-instruments: e.g., glass to replace the screen of the in-vehicle infotainment; +- Multi-screen interaction; +- Multi-purpose: used for entertainment, official, social, gaming, etc; +- Holographic projection will be popular in the future. + +### **6.1 Relationship with SAE J3016 levels of driving automation** + +For the purpose of a common understanding the [SAE J3016] "Levels of Driving Automation" responsibilities of the driver per driving level are explained in this clause. See also Figure 3. + +| | SAE LEVEL 0 | SAE LEVEL 1 | SAE LEVEL 2 | SAE LEVEL 3 | SAE LEVEL 4 | SAE LEVEL 5 | +|------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------------------------------------| +| |

Image: Eyes on icon

Eyes on

Image: Hands on icon

Hands on |

Image: Eyes on icon

Eyes on

Image: Hands on icon

Hands on |

Image: Eyes on icon

Eyes on

Image: Hands on icon

Hands on |

Image: Eyes on icon

Eyes on

Image: Hands off icon

Hands off |

Image: Eyes off icon

Eyes off

Image: Hands off icon

Hands off |

Image: Eyes off icon

Eyes off

Image: Hands off icon

Hands off | +| What does the human in the driver's seat have to do? | You are driving whenever these driver support features are engaged – even if your feet are off the pedals and you are not steering
You must constantly supervise these support features; you must steer, brake or accelerate as needed to maintain safety | | | You are not driving when these automated driving features are engaged – even if you are seated in "the driver's seat"
When the feature requests
You must drive | | | +| | These are driver support features | | | These are automated driving features | | | +| What do these features do? | These features are limited to providing warnings and momentary assistance | These features provide steering OR brake/acceleration support to the driver | These features provide steering AND brake/acceleration support to the driver | These features can drive the vehicle under limited conditions and will not operate unless all required conditions are met | | These features can drive the vehicle under all conditions | +| Example features |
  • Automatic emergency braking
  • Blind spot warning
  • Lane departure warning
|
  • Lane centering OR
  • Adaptive cruise control
|
  • Lane centering AND
  • Adaptive cruise control at the same time
|
  • Traffic jam chauffeur
|
  • Local driverless taxi
  • Pedals/steering wheel may or may not be installed
|
  • Same as level 4, but feature can drive everywhere in all conditions
| + +F.749.3(20)\_F03 + +**Figure 3 – SAE J3016 definition of driving levels** + +(Sources: SAE International Release update (11.12.2018), illustration Mike Lemanski (TRW, ZF) (16.04.2017)) + +In levels 0, 1 and 2 the driver has to keep the eyes on the road and the hands on the steering wheel. The driver is fully in charge of the vehicle. + +In level 2 driving, safety features will support the driver in his driving tasks and will make driving safer by intervening with corrective actions if necessary, e.g., emergency braking or lane keeping. However, at no time will the vehicle drive autonomously. + +Level 3 driving is a first level of partly automated driving. Due to available driving features and legal constraints those automated driving features are allowed for use in limited or geofenced areas, such as on highways, or under a certain speed, e.g., in traffic jams with a speed up to 40 km/h (when speed gets faster the driver is requested to take over control again). However, due to legislation the driver is still fully in charge for the vehicle even when driving in autonomous mode. This means that even if the driver takes the hands off the steering wheel while driving autonomously, the driver still needs to keep the eyes on the road and be able to take over control from the vehicle at any time and within a very short timeframe in case of unexpected situations. + +In driving levels 4 and 5 vehicles no longer need steering wheels or pedals. Vehicles drive autonomously. All persons in the vehicle become passengers. The vehicle has no driver. + +Level 5 is the highest level of automation allowing vehicles to drive autonomously in any condition and every geographic area. The vehicle can drive fully autonomously without having any persons on board. + +Driving levels mainly apply to the driver and make a difference for him being in charge of the vehicle or not. However, all other persons on board a vehicle no matter if on the front passenger seat or on rear seats are passengers no matter which driving level applies. + +#### 6.1.1 Matching of generic multimedia use cases with driving levels + +Figure 4 matches generic use cases and services with the driving levels. All activities that are mentioned in the table should be considered for a driving vehicle. When the vehicle parks or stands still, all services are available for the driver as well as for the passengers. Excluded in this case is a standing vehicle at traffic lights or other traffic signs as in this case the vehicle and the driver are still considered as active traffic participants similar to a driving car. + +Within the driving levels a distinction on what drivers (D) and passengers (P) are allowed to do in the framework of these use cases and services is taken into consideration. + +| Use cases | SAE level 0 | | SAE level 1 | | SAE level 2 | | SAE level 3 | | SAE level 4 | | SAE level 5 | | +|-----------------------------------------------|-------------|---|-------------|---|-------------|---|-------------|---|-------------|---|-------------|---| +| | D | P | D | P | D | P | D | P | D | P | D | P | +| Listen to audio services | X | X | X | X | X | X | X | X | X | X | X | X | +| Use video, TV and gaming services | | X | | X | | X | | X | X | X | X | X | +| Use mobile devices | | X | | X | | X | (X) | X | X | X | X | X | +| Use mobile devices through VMS | X | X | X | X | X | X | X | X | X | X | X | X | +| Phone conferencing (handsfree) | X | X | X | X | X | X | X | X | X | X | X | X | +| Videoconferencing | | X | | X | | X | | X | X | X | X | X | +| Reading in vehicle (books, mails, news, etc.) | | X | | X | | X | | X | X | X | X | X | +| Getting read content (text-to-speech) | X | X | X | X | X | X | X | X | X | X | X | X | +| Working in vehicle (office applications) | | X | | X | | X | | X | X | X | X | X | + +X = Allowed + +(X) = Partly allowed (in autonomous driving mode) + +**Figure 4 – Matching of generic multimedia use cases with SAE driving levels** + +The table shown in Figure 4 follows the guidelines of current legal frameworks in place in some jurisdictions, (e.g., US and EU. Note that some other jurisdictions may use similar or different classifications than those in [SAE J3016]). However, from an ergonomic point of view phoning or getting read content while driving through text-to-speech can lead to a cognitive overload for the driver and can lead to critical driving situations. Therefore, for driving levels 0 to 3, it is suggested to integrate driver monitoring into the vehicle in order to measure the cognitive load of the driver and to adapt access and content to driving situations and distraction level. + +In addition, driver monitoring in driving level 3 facilitates tracking if the driver has taken back control when requested or if alternative actions need to be taken in order to bring the vehicle to a safe stop. + +The vehicle can be fully autonomous, even limiting any direct user interaction (apart from emergency stop) regarding driving, choice of destination and itinerary to follow. In this case the driver and passenger roles are mostly identical. + +## 7 Connectivity + +### 7.1 Distinction between brought-in, built-in and hybrid connectivity + +In-vehicle connectivity can be distinguished in three connectivity cases, that is, brought-in, built-in and hybrid connectivity. These are illustrated in Figure 5. + +#### 7.1.1 Brought-in connectivity + +The multi-media system of the vehicle is neither connected to the internet nor to other connected services and applications (maybe with the exception of radio FM). Due to legal requirements in certain jurisdictions the only VMS connectivity that is built-in might be for emergency calls. However, the limited bandwidth for emergency calls might not allow access or stream to any other connected services and applications. + +The user can connect the VMS in this case through his mobile device that he brings into the vehicle. The mobile device will connect with the car through USB or other connections. + +Content of the mobile device such as applications and mobile services will be displayed on the VMS through some tools. + +All connected applications and services are accessible through the VMS. + +#### 7.1.2 Built-in connectivity + +The vehicle is fully connected to the Internet, applications and services through its built-in connectivity. + +All applications and services can be accessed through the VMS. + +#### 7.1.3 Hybrid connectivity + +Hybrid connectivity is a mixture of brought-in and built-in connectivity. + +The vehicle has its own built-in connectivity. However, the user can bring-in additional connected services and applications through its mobile devices. As for the brought-in connectivity, the mobile device connects to the vehicle through USB or other types of connection and its services and applications can be displayed on the VMS through some tools. The user accesses all vehicle connected services and applications through the VMS. + +![Icon representing a vehicle's infotainment system (VMS) connected to a mobile device via a short-range connection.](ded1520e45c0c3e4c46c58984602bf0e_img.jpg) + +The image shows a stylized icon of a car's dashboard with a screen displaying a grid of icons. To the right of the dashboard icon, there is a small icon of a smartphone with a signal wave between them, indicating a connection. + +Icon representing a vehicle's infotainment system (VMS) connected to a mobile device via a short-range connection. + +| Brought-in | Build-in | Hybrid connectivity | +|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +|
  • – Vehicle has no own connectivity
  • – User will connect through his brought-in mobile device
  • – Mobile device will connect through short-range connections to VMS
  • – Content will be made visible on VMS
  • – All APPs and services are accessible through VMS
|
  • – Vehicle is fully equipped to connect to Internet and other connected services to the vehicle
  • – All APPs and services are accessible through VMS.
|
  • – Mixed of brought-in and build-in connectivity.
  • – Vehicle has own connectivity.
  • – Some APPs and services are only accessible through mobile device brought into the vehicle
  • – Mobile device connects to vehicle VMS through short-range connections.
  • – User accesses all vehicle connected APPs and services through VMS.
| + +Figure 5 – Vehicle connectivity + +Ideally, in all the above cases, access to services, applications and content displayed are adapted to driving situations and to the location of users within the vehicle in order to minimize distraction for the driver. + +### **7.2 Multiple network connectivity in the vehicular multimedia context** + +#### **7.2.1 Background** + +Most in vehicle infotainment (IVI) systems have at least FM/AM radio functions, while digital radio and sound broadcasting receivers are becoming more common [b-ITU-R BS.2384], due particularly to the European Union mandating DAB+ receiver in all new cars. With the improvements of networking technologies, online radio and audio/video entertainment services in intelligent connected vehicles are becoming more and more popular. + +On one hand, traditional one-way broadcasting systems, i.e., FM/AM, terrestrial digital broadcasting, and satellite broadcasting, only support one-way transmission mode. Hence, interactive and personalized services can hardly be achieved by such broadcasting networks. + +On the other hand, mobile networks support two-way transmission modes. By using mobile network, online multimedia entertainment services can be provided to users to satisfy interactive and personalized demands. However, there are three drawbacks: + +- 1) In certain situations, the cost of mobile Internet is an important factor impeding the wide deployment of VMNS in intelligent connected vehicles. +- 2) The wireless coverage of mobile networks is not fully guaranteed. For example, the signal strength can be limited or non-existent in certain locations, and, therefore, users may experience a poor connection and experience when at those locations. +- 3) The mobile network can be congested, which impacts the smoothness of online audio/video playback and degrades the user experience. + +##### **7.2.1.1 Converged network empowers migration to more advanced services** + +Network convergence transmission takes advantage of the complementary strengths of broadcasting and mobile Internet infrastructures to overcome their limitations. Based on the converged network transmission channels, the rich media contents and information can be delivered to all kinds of terminals in an efficient and timely manner, i.e., intelligent connected vehicles, personal mobile terminals, etc. + +The VMS retrieves rich media contents from a VMSP through various networks and acts as a local infotainment service provider. The VMS can not only play the contents locally on the built-in vehicle screens and speakers but can also share the contents with passengers' personal devices (for example, smart phones and tablet computers using in-vehicle local area wireless connections). + +The VMS seamlessly integrates the services received from various broadcasting networks and communication networks. In particular, when the VMS is in-coverage of multiple broadcasting networks, it could intelligently select one of the broadcasting networks to receive the service. Such choice can be made based on various factors, such as network signal strength, terminal processing power, cost, etc. + +#### **7.2.2 Use-cases** + +##### **7.2.2.1 Use case A – In-city commuting/native life service as driving purpose** + +1. Satisfy the needs of personalized and two-way interactions. +2. Provide undegraded quality of service (QoS) for audio/video streaming services. + +The coverage of cellular mobile networks in densely populated city areas is fairly good. However, as more and more users are trying to access the cellular mobile network, the connection speed for + +each user becomes lower and lower. In such a case, mobile Internet by itself could not satisfy the user experience. + +##### **7.2.2.2 Use case B – Out-city driving** + +Traveling for long distances away from cities using family cars or SUVs during weekends or holidays is becoming a common trend. In some regions, large populations live outside of the cities and using a vehicle is a necessity. In environments such as mountain valleys, forests or deserts, the mobile network coverage could be much poorer than in cities, which in turn cause the degradation of multimedia services. + +#### **7.2.3 Characteristics of the VMN connectivity** + +The VMN could have the following characteristics: + +- 1) Be available on a connected vehicle (or/and autonomous vehicle). +- 2) Includes a connectivity layer allowing services to seamlessly use multiple bearers and protocols (satellite, cellular, short range / broadcast, streaming, download). That connectivity layer is usable by various sub-systems and services such as infotainment, maps or ADAS. +- 3) The VMS may connect to one or more communication networks, in particular: + - 3G/4G/5G cellular networks. + - Satellites (e.g., L Band Broadcast) with both broadcast and bi-directional communications. + - Low earth orbit (LEO) bi-directional communication networks. + - Short range networks and/or local area wireless networks. + - Terrestrial broadcast networks. + +Further, the following scenarios should be envisaged for VMN: + +- The vehicle drives away from the coverage area of the wireless local area network (WLAN), but access to services continue to be available via cellular network or satellite network. +- The vehicle drives away from the coverage area of the cellular network, but access to services continue to be available via satellite communication. +- The vehicle drives away from the coverage area of the satellite network, but access to services continue to be available via cellular communication (e.g., entering an underground parking). +- The vehicle drives away from the coverage area of the mobile network in a region where satellite broadcasting systems are not deployed, but access to service continue to be available via terrestrial broadcast communication. +- The VMS may prefer that large downloads such as map updates and security updates be performed only on higher bandwidth network such as WLAN. +- Network connectivity may be billed to the driver, or it may be billed to the vehicle manufacturer. In either case, the driver/manufacturer may want to limit types of data transmitted over higher-cost networks. For example, diagnostic data can be uploaded at a later time when WLAN is available, whereas safety related information may be downloaded via any means available. +- The VMS may receive indicators over a broadcast network and connect to a server over a cellular or short-range data connection to download metadata about the program. + +- The VMS may transmit/receive multimedia information such as image/video, voice and broadcast messages (such as emergency messages received and retransmitted to nearby vehicles) through wireless short-range communication networks. + +#### 7.2.4 Requirements – VMS connectivity + +Con-R1: The VMS shall be inclusive of both bi-directional communication and one-directional broadcast networks. + +Con-R2: The VMS should be designed to allow access via multiple communication networks. + +Con-R4: The VMS may connect via multiple communication networks simultaneously. + +Con-R5: The VMS may provide seamless handoffs between communication networks. + +Con-R5: The VMS shall allow the user to configure or limit types of data transmission over various communication networks when multiple networks are accessible. + +### 7.3 Network convergence transmission and vehicle-onboard processing + +The network architecture of convergence transmission, as shown in Figure 6, illustrates how the network convergence greatly improves system efficiency and user experience. + +![Figure 6: The network architecture of convergence transmission. The diagram shows a Service platform connected to a Convergence gateway, which is connected to a CDN and the Internet. The Convergence gateway is also connected to a Broadcasting network (via a satellite) and a Mobile communication network (via a base station). The Broadcasting network is connected to a Vehicle (via a Broadcasting link) and Mobile terminals (via a Mobile communication link). The Mobile communication network is also connected to the Vehicle and Mobile terminals. The Vehicle and Mobile terminals are shown within a dashed green box labeled 'Vehicle' and 'Mobile terminals'.](3fa8bfee86764e3c3a1a6fbbe61bbd52_img.jpg) + +The diagram illustrates the network architecture for convergence transmission. On the left, a 'Service platform' (represented by icons of a database, server, and laptop) is connected to a 'Convergence gateway' (a server icon). The gateway is connected to a 'CDN' (Content Delivery Network, represented by three server icons) and the 'Internet'. The Convergence gateway is also connected to two network clouds: a 'Broadcasting network' (containing a satellite and a ground station) and a 'Mobile communication network' (containing a base station). The Broadcasting network connects to a 'Vehicle' (a red car) via a 'Broadcasting link' (indicated by a lightning bolt). The Mobile communication network connects to the 'Vehicle' and 'Mobile terminals' (represented by a smartphone and a tablet) via a 'Mobile communication link' (also indicated by a lightning bolt). The Vehicle and Mobile terminals are grouped within a dashed green box. A label 'F.749.3(20)\_F06' is present at the bottom right of the diagram. + +Figure 6: The network architecture of convergence transmission. The diagram shows a Service platform connected to a Convergence gateway, which is connected to a CDN and the Internet. The Convergence gateway is also connected to a Broadcasting network (via a satellite) and a Mobile communication network (via a base station). The Broadcasting network is connected to a Vehicle (via a Broadcasting link) and Mobile terminals (via a Mobile communication link). The Mobile communication network is also connected to the Vehicle and Mobile terminals. The Vehicle and Mobile terminals are shown within a dashed green box labeled 'Vehicle' and 'Mobile terminals'. + +Figure 6 – The network architecture of convergence transmission + +#### 7.3.1 Use cases + +Scenario 1: Vehicle A commutes in metropolitan areas with satellite broadcasting signal being intermittently blocked by high-rise buildings; vehicle B takes a weekend excursion to remote areas where 4G/5G cellular coverage is limited; vehicle C is on a long road trip across country during which favourite channels from home city are played throughout the whole trip. In the above cases, as long as one bearer of the converged networks has signal coverage, the services available are uninterrupted while the overall data charges are kept at minimum. + +Scenario 2: Users enjoy rich media content and various services, such as live/on-demand audios and videos, pushed data, on-the-go Internet, emergency alert, global navigation satellite system (GNSS) etc., on one unified in-vehicle infotainment platform over converged networks, without having to care about the actual bearer that carries the service. + +Scenario 3: The driver in a vehicle equipped with vehicular multimedia system (VMS) can enjoy high-quality audio, videos and other value-added services through in-vehicle screen and speakers. + +Scenario 4: Each passenger inside a vehicle equipped with VM system can enjoy their own choice of high-quality audio, video, and other value-added services on their personal smartphones or tablet computers through local area wireless access. + +Scenario 5: Users who have downloaded the application on their mobile phones/pads can continue enjoying the same program at the same quality of service using their own Internet connections, when out of the vehicle local area wireless coverage. + +#### **7.3.2 Gap analysis** + +Various media services, i.e., live/on-demand audio/video service, pushed data service, on-the-go Internet service, should be supported by the VMSP and the VMS over a variety of underlying broadcast and communication networks. In order to guarantee the QoS of these services and improve the network efficiency, carefully designed transmission protocols and reception methods should be pursued. Firstly, the transmission protocols and reception methods should support various underlying physical-layer transmission standards, i.e., various wireless broadcast standards and wireless access standards, so that the QoS of these media services can be guaranteed while the vehicle is roaming across different networks. Secondly, service providers may use various application layer protocols to support similar media services. Therefore, the transmission protocols and reception methods should be transparent to the upper layer protocols so that the vehicular multimedia network can be used by various service providers to run their own services. Thirdly, the transmission protocols and reception method should be able to take full advantage of the characteristics of various wireless networks to improve the QoS of these media services, or to optimize the overall network efficiency. + +#### **7.3.3 Requirements – VMS converged network** + +The requirements on network convergence transmission and vehicle-onboard processing are listed as follows: + +Con-T-R1: The VMS shall require none or minimum modifications to the existing broadcasting and mobile communication infrastructures. + +Con-T-R2: Network convergence transmission and reception shall be agnostic to the underlying physical layer transmission standards and be transparent to the upper layer applications. + +Con-T-R3: The VMS should take full advantage of different network characteristics to ensure uninterrupted and seamless services (e.g., to keep the overall network costs minimum). + +Con-T-R4: The increase in the end-to-end transmission delays due to the network convergence transmission and vehicle-onboard processing should be minimized. + +Con-T-R5: The increase in the overall system start-up and channel switching time due to the network convergence transmission and vehicle-onboard processing shall be minimized. + +### **7.4 Networking connection among multiple terminals within the vehicle** + +#### **7.4.1 Use case A – Nomadic device connectivity** + +The VMS provides in-vehicle media content distribution over a local wireless network. Various types of terminal (such as various types of in-vehicle screens, mobile phones, tablet computers, etc.) can connect to the VMS and receive content adapted to the characteristics of the different terminals. + +#### **7.4.2 Use case B – Screen casting** + +Support the interactive connection between the VMS and a user's mobile device. Mobile devices use a variety of existing screen casting systems. + +#### **7.4.3 Requirements – Display Screens and terminals connectivity** + +Net-C-R1: The VMS should be able to connect to various terminals (such as different types of in-vehicle screens, mobile phones, tablet computers, etc.) based on their capabilities. + +Net-C-R2: The VMS should support the interactive connection with mobile devices. + +Net-C-R3: The VMS should be able to selectively limit bandwidth allocated to the connected mobile devices, based on total VMS vehicle available bandwidth and user profile. + +Net-C-R4: Upon detection of change of the VMS user, the VMS should wipe personal data transferred through the mobile device connected to the VMS in order to respect privacy of personal data. + +#### **7.4.4 Use case C – Comfort and well-being on board** + +The emotional and physical conditions of driver and passengers are measured either through smart watches connected with the VMS, through scanning of facial expressions and postures with cameras or other sensors on board (e.g., sensors integrated in steering wheel or seats) that are connected to the VMS. In order to stimulate the comfort and well-being of all passengers on board and to ensure the non-drowsiness and attentiveness of the driver, the senses of persons on board are stimulated through display content, ambient light, sound, scents, as well as air distribution and flow. + +In case of critical situations, such as drowsiness or distraction of the driver, VMS displays alert messages and gives audible alerts, or shuts down its displays in case the driver is looking away from the road for a considerably long period of time. + +The information collected for each passenger is integrated in each personal profile, possibly along with other environmental information (localization, number of persons aboard, air condition, temperature, vehicle telemetry, etc.). + +Over time, an individual evolution measured per passenger as well as an individual profile for well-being will be created with the help of artificial intelligence running in the background of the VMS and be deposited in the VMSP. + +The information collected by the VMS regarding the emotional and physical status of the users, can be made available to other well-being and health related value-added services to which the user has subscribed. These services can provide information on user well-being and health needs to the VMS. + +##### **7.4.4.1 Requirements** + +Net-WB-R1: VMS should have or should be connected to sensors sensing the cognitive load of the driver, and the emotional and physical conditions of all vehicle users. + +Net-WB-R2: VMS should have or be connected to all features that can change the ambience in the vehicle such as a library of music and messages, lighting, perfume deodorizer, etc. + +Net-WB-R3: VMS should have integrated algorithms that react to and initiate actions instantaneously including outside of network coverage. + +Net-WB-R4: VMS should be able to connect to external well-being and health services and transmit information collected regarding each user. + +Net-WB-R5: VMS shall have storage space to save each user profile data and to send them to the VMSP when network coverage becomes available. + +Net-WB-R6: Upon detection of change of the VMS user, the VMS should wipe personal data in order to respect privacy of the user. + +Net-WB-R7: Each individual user should have the possibility to shut down or unsubscribe the function at any time temporarily or irrevocably. + +#### **7.4.5 Use case D – On-demand functions for privately owned vehicle** + +Through vehicle applications available on smart devices and on the VMS, the vehicle user can demand vehicle functions that he can subscribe to at a fee for one-time use, a certain duration, or over the vehicle's lifetime. Examples for such functions could be premium on-demand autonomous driving functions, longer battery range, on-demand navigation, 3D video, AR/VR games, digital radio, etc. The requirement is that safety functions should not be part of such on-demand functions. + +Based on user interests, services can be proposed to the customer actively through the application. Some of these on-demand functions will be directly displayed or linked to the VMS. + +In the vehicle the VMS displays a choice of on-demand functions. The passenger can choose a function and the duration of the subscription through interaction with the VMS. Based on the VMS payment profile for that user deposited in the cloud, a passenger can pay after secure identification through the VMS. Alternatively, payment may not require a user profile and/or an in-application payment could be used. VMS confirms the purchase and displays the chosen function that is ready for use. + +##### **7.4.5.1 Requirements** + +**Net-PV-R1:** In a multi-passenger vehicle, the VMS should enable a centralized or individual access to the VMNS via applications for on-demand functions and payment functions. + +**Net-PV-R2:** In a multi-passenger vehicle, the VMS should have access to the VMS user profiles at a centralized or individual position for on-demand functions and payment functions. + +**Net-PV-R3:** VMS should be equipped with sensors allowing the identification passenger(s) at different positions for usage and payment functions or with an input function allowing the user to identify himself through a password. + +**Net-PV-R4:** VMS should support on-demand functions that are meant to run locally on VMS such as on-demand navigation, 3D video, AR/VR games, digital radio, etc. + +- VMS should have enough storage space. +- VMS should have enough computing power. +- For multi-passenger vehicles individual VMS at each passenger seat should respond to these prerequisites. + +### **7.5 Continuous and consistent services across various networks and environments (from/to VMS)** + +#### **7.5.1 Use case A – Continuity of service** + +During breakfast, a person starts to watch a live streaming of a basketball game using an application on a tablet computer through the local area wireless network at home. After breakfast, the video of the basketball game is paused. Meanwhile, the viewing records (i.e., the program that is being watched, the time when the video is paused, the preferred audio commentary track, etc.) are uploaded to a cloud database automatically. The individual steps into a waiting vehicle and subsequently decides to continue to watch the live broadcasting of the basketball game en route to the office with the in-vehicle display or windshield display through the converged network. In the vehicle, the individual may log into an application in the VMS with the same user account to continue viewing. Meanwhile, the viewing records are downloaded from the cloud database to the VMS. In such a case, the account information and viewing records can be used to support continuous and consistent services across various networks and on multiple screens providing an improved viewing experience. For example, + +- 1) Suppose the live streaming of the basketball game is available on both satellite broadcast networks and cellular communication networks. If the person has a premium subscription, then the high-definition (HD) live broadcasting of the basketball game with HD surround + +sound available on the satellite broadcast network may be enjoyed. If the person has a non-premium subscription, then the service could be a best-effort live streaming of the basketball game available on the cellular communication network. + +- 2) If the person decides to resume the play of the basketball game from the time that the video was paused previously, then the buffered recording of the basketball game on the cellular communication network should be supported. Meanwhile, the live stream of the basketball game on the satellite network may be available in a picture-in-picture mode. +- 3) When the person resumes the video of the basketball game on the application in the VMS, it should automatically choose the preferred audio commentary track based on the viewing records. + +##### **7.5.1.1 Requirements** + +To support continuous and consistent services across various networks and on multiple screens, the following requirements are proposed: + +HET-SC-R1: The VMSP and the VMS should allow a user to log in with the same user account by using various devices, i.e., VMS itself, mobile phones, tablet computers, etc. + +HET-SC-R2: The VMSP and the VMS should support the uploading of the user's viewing records to the VMSP. + +HET-SC-R2.1: The VMSP and the VMS should support the configuration, by the user, of the scope of his/her viewing records that are to be uploaded. + +HET-SC-R2.2: The VMSP and the VMS should support the configuration, by the user, of the frequency of his/her viewing records that are to be uploaded. + +HET-SC-R3: The VMSP and the VMS should support the download of the user's viewing records to various devices. + +HET-SC-R3.1: The VMSP and the VMS should support the configuration, by the user, of the frequency of his/her viewing records that are to be downloaded. + +HET-SC-R4: The VMS should support the reception, processing, and display of multiple media channels from various networks and on different screens simultaneously. + +#### **7.5.2 Use case B – Cloud based services** + +Seamless connectivity between home, office and in-vehicle services will allow users to always find the same software applications, user interfaces, language, etc., regardless of their location (home, in the vehicle or at the office). Multimedia content, office applications and services will be the same. However, for security and confidentiality reasons corporate data will not be made available or be transferred to a VMS centrally managed user profile. + +The VMS displays this content once the user is identified and allows access to individual data from the cloud or from the user's mobile phone. + +##### **7.5.2.1 Requirements** + +HET-CL-R1: A central user profile with content, applications and services excluding any corporate data should be deposited in the cloud or locally in the user's mobile phone. Each user may have different profile versions, depending on the roles he can assume (on business, on leisure, family, etc.). + +HET-CL-R2: VMS should identify driver/passenger(s) through its mobile phone (NFC, and other) or cameras or sensors (scan of fingerprint, iris, face, other). + +HET-CL-R3: VMS should identify which seat in the vehicle passenger has occupied in order to display the right content at the right place. + +HET-CL-R4: VMS should access cloud content of passenger or the user provided profile of the mobile phone, display it at the place user is occupying in the vehicle and adapt the passenger comfort of the specific place (seat and screen positioning, audio levels, temperature etc.). In case of multiple user profiles, the user will be asked to choose the profile to apply or the VMS will choose the one that corresponds to the current situation. + +- Changes made to the profile on the VMS within the vehicle, e.g., new applications, user surfaces, comfort choices etc. should be sent back from the VMS to the cloud, or stored to the user's mobile phone, where they will be saved and are accessible later from any other vehicle, at home or at work. +- Access of VMS to passenger specific cloud data from anywhere should be secured against intrusion and cyber-attacks. +- VMS should have storage space to save profile data and to send them to the cloud when network coverage becomes available. + +HET-CL-R5: Upon detection of change of the VMS user, the VMS should wipe personal data in order to respect privacy of the user. + +## 8 Intelligent human machine interfaces (HMI) for VMS + +In keeping with current development trends of intelligent vehicles, the integration of various interactive control schemes for safe driving such as voice control, gesture control, eye movement control, etc., will become mainstream in the future. This is illustrated in Figure 7. + +![Diagram of a vehicle cockpit showing input and output devices for the VMS. Input devices are listed in green and output devices in blue. Red boxes and circles highlight specific components like the steering wheel, dashboard displays, and center console.](d9c0a780cd22626253dab4aa41699e2f_img.jpg) + +Input +Output + +(Natural) speech recognition + Gesture recognition + Facial recognition + Gaze detection/eye tracking + Driver/passenger monitoring + +Head-up-displays (HUD) + Instrument cluster + Acoustic warning signals +Steering wheel buttons +Touch screen +Fingerprint sensor +Iris scanner + +Passenger display(s) +Touch screen +Intelligent surfaces (fabric, windows) +Push buttons + Text-to-speech + Sound + +Centre stack display(s) + Intelligent surfaces + Ambient lighting +Touch screen +Touchpad +Push buttons/turn-press knobs +Intelligent surfaces + +Diagram of a vehicle cockpit showing input and output devices for the VMS. Input devices are listed in green and output devices in blue. Red boxes and circles highlight specific components like the steering wheel, dashboard displays, and center console. + +**Figure 7 – VMS input and output devices, cockpit view** + +When interacting with the VMS there should be a distinction between input and output devices. All input and output devices should be intuitive, easy to use and be placed preferably in ergonomically areas so that they are easy to reach and allow simple interaction. Especially in driving levels 0 to 3, driver distraction and cognitive overload for the driver should be avoided so that the driver can + +safely operate the vehicle in all driving situations (hands on the steering wheel and eyes on the road) [SAE J3016]. + +**Input devices** allow the driver and passengers to input data and information into the VMS by directly interacting with the vehicle. The following input devices can be distinguished: + +- Centre stack and tunnel area + - Touch screen(s) + - Touchpad + - Push buttons + - Turn-press-knob +- Interior camera(s) for + - Person identification (e.g., eye scanner, facial recognition) + - Gesture control + - Gaze detection/eye tracking + - Driver and passenger monitoring (sensing, e.g., of drowsiness, distraction, emotional state, etc.) +- (Natural) Speech recognition +- Driver area + - Steering wheel buttons + - Touch screen on steering wheel + - Fingerprint sensor +- Passenger areas + - Touch screens + - Push buttons +- Intelligent surfaces (e.g., fabrics, cockpit surface, windows) + +**Output devices** can give information specific to the driver as well as for all other passengers on board the vehicle: + +- Driver area + - Head-up display (HUD) (limited to driver, or augmented over a big part or the whole windshield) + - Instrument cluster (numeric, or digital display) + - Touch screen in the steering wheel +- Loudspeakers either in doors, or passenger specific for individual sound bubbles in headrests + - Acoustic warning signals + - Text-to-speech (e.g., reading of messages, mails, other information) + - Sound (e.g., multimedia content, phone/video conferencing) +- Centre stack display(s) +- Passenger displays +- Intelligent surfaces (e.g., fabrics, cockpit surface, windows) +- Ambient lighting (light integrated, e.g., around steering wheel, in cockpit, doors, etc.) + +### 8.1 The acoustic system in the vehicular multimedia + +#### 8.1.1 Background + +As stated in the terms of reference of the FG-VM, a vehicle may become a user's third entertainment space, and could be highly personalized. A VMS may support multiple screens and may be used not only to listen to radio but also possibly to watch television, movies, play games, either collectively or individually. At the same time, a vehicle may comprise a voice recognition or natural language processing system. It is expected that one or more multimedia sources and users interacting with one another or with a multimedia system will interfere with the voice commands. + +It becomes apparent that the determination of the location (zone) of the various emitting sources (talkers) in the cabin and the acoustic treatment of each transmitted signal (voice command) from each location (zone) will facilitate the correct processing of voice commands by the voice recognition system. + +#### 8.1.2 Use-cases + +##### 8.1.2.1 Use case A – Initiating a voice recognition session + +A person in a vehicle containing multiple occupants wishes to initiate a voice recognition session by uttering a keyword. Each occupant is in a separate zone of the cabin. The cabin may contain one or more microphones which may or may not be dedicated for each zone. Each microphone picks up the voice of the occupant, but also the voices of other occupants, or "interference speech". One or more multiple microphone signals (or audio channels) may be available to a keyword spotter (KWS), which must decide not only whether/when the keyword was spoken, but also from which zone the keyword was spoken. + +The following problem scenarios may result in inadequate behaviour of the KWS: + +- A-1 If there is no dedicated microphone for each zone, or no means to identify the zone of the target talker, the command may not be detected, may be rejected or wrongly executed. +- A-2 Otherwise: + - A-2A Interfering speech may cause a KWS to fail to detect (false reject) the keyword spoken by the target talker in the target zone microphone. + - A-2-B Concurrent sources (e.g., music, video) played into the vehicle, resulting in echo on the microphones, may cause a KWS to fail to detect (false reject) the keyword spoken by the target talker in the target zone microphone. + - A-2-C Interference of the target talker onto microphones outside of the target zone may cause the KWS to detect the keyword but from the wrong zone. + +![Figure 8: Acoustic processing (AEC and ZIC) on each zone dedicated microphone. The diagram shows three microphones (Mic 1, Mic 2, Mic N) on the left, each with a complex waveform representing target speech, interfering speech, and echo. Arrows lead from each microphone to a blue rectangular block labeled 'AEC'. From the 'AEC' block, arrows lead to another blue rectangular block labeled 'ZIC'. Finally, arrows lead from the 'ZIC' block to a blue rectangular block labeled 'KWS'. The waveforms are shown at each stage, illustrating the processing flow. A small label 'F.749.3(20)_F08' is in the bottom right corner of the diagram area.](ea4fd10a9a501c602f2bea0f7f711877_img.jpg) + +``` +graph LR + Mic1[Mic 1] -- Waveforms --> AEC[AEC] + Mic2[Mic 2] -- Waveforms --> AEC + MicN[Mic N] -- Waveforms --> AEC + AEC -- Processed Waveforms --> ZIC[ZIC] + ZIC -- Cleaned Waveforms --> KWS[KWS] +``` + +Figure 8: Acoustic processing (AEC and ZIC) on each zone dedicated microphone. The diagram shows three microphones (Mic 1, Mic 2, Mic N) on the left, each with a complex waveform representing target speech, interfering speech, and echo. Arrows lead from each microphone to a blue rectangular block labeled 'AEC'. From the 'AEC' block, arrows lead to another blue rectangular block labeled 'ZIC'. Finally, arrows lead from the 'ZIC' block to a blue rectangular block labeled 'KWS'. The waveforms are shown at each stage, illustrating the processing flow. A small label 'F.749.3(20)\_F08' is in the bottom right corner of the diagram area. + +**Figure 8 – Acoustic processing (AEC and ZIC) on each zone dedicated microphone** + +Figure 8 is an illustration of use case A-2 involving KWS with N microphones/zones in a vehicle, depicting the waveforms. Each microphone contains target speech, interfering speech and echo. + +(black). Talker in zone 1 is yellow, talker in zone 2 is red and talker in zone 3 is blue. Acoustic echo cancellation (AEC) is used to subtract the echo from each microphone, and zone interference cancellation (ZIC) is used to isolate the target speech from interfering speech in each microphone. + +##### 8.1.2.2 Use case B – Interference during a voice recognition session + +Once a voice recognition session has been initiated and the target zone has been identified (e.g., using KWS or push-to-talk), an occupant in the target zone will use voice commands to interact with the voice recognition system. The target speech in the target zone will potentially be mixed with interfering speech from other zones, and echo from media sources playing into the cabin. These may result in incorrectly detected speech and difficulties in accomplishing the voice command. + +![Figure 9 – Illustration of use case B, where zone 2 has been identified as the target zone](90ddb84c323b956e2d50a54d3f870566_img.jpg) + +The diagram shows a signal processing pipeline for three microphone inputs: Mic 1 (yellow label), Mic 2 (red label), and Mic N (blue label). Each microphone input consists of a mixture of three colored waveforms (yellow, red, and blue). These inputs pass through three sequential processing blocks represented by blue rectangles: 'AEC' (Acoustic Echo Cancellation), 'ZIC' (Zone Interference Cancellation), and 'VR' (Voice Recognition). After the AEC block, the signals remain mixed. After the ZIC block, only the red waveform (representing the target zone 2) remains. This single red waveform is then fed into the VR block. The diagram is labeled F.749.3(20)\_F09 at the bottom right. + +Figure 9 – Illustration of use case B, where zone 2 has been identified as the target zone + +**Figure 9 – Illustration of use case B, where zone 2 has been identified as the target zone** + +#### 8.1.3 Gap analysis + +ITU-T SG12 is currently working on a set of recommendations relevant for VMS: + +- [b-ITU-T P.1150] In-car communication utilizes the integrated microphones and speakers in the motor vehicle cabin to amplify conversation to provide an improved communication between all occupants in a motor vehicle. Furthermore, the quality of voice is ensured such that the motor vehicle driver does not feel it necessary to turn their head to amplify their voice when talking to other passengers. However, it is not immediately apparent that [b-ITU-T P.1150] addresses the requirements associated with the use of voice recognition in a vehicular multimedia context. +- [b-ITU-T P.11.x] Hands-free communications in vehicles for narrowband, wideband, super-wideband and full-band and associated subsystems provides useful conformance points to improve signal processing for hands-free communication within a vehicle. However, it is not immediately apparent that [b-ITU-T P.11.x] can address the use cases and requirements for multiple talkers, KWS, voice commands and voice recognition sessions. + +#### 8.1.4 Requirements – Acoustic + +In order for a KWS, a voice recognition or natural language processing solution to operate effectively in a multiple-talker scenario with background media in a vehicle, the following requirements are proposed: + +AUD-R1: It shall be possible to initiate a voice recognition session from any zone in the vehicle. + +AUD-R1.1: It shall be possible to initiate a voice recognition session from any zone in presence of speech interference and noise interference from the same or a different zone. + +AUD-R2: It shall be possible to process each channel/zone independently. + +AUD-R3: It shall be possible to identify a target zone/target speech (e.g., scenario A-1). + +AUD-R4: ZIC shall be performed to suppress interfering speech from a zone whilst preserving the target speech in that zone. (scenarios A-2-A, A-2-C and B). + +AUD-R5: AEC shall be performed on each channel to suppress the echo of media sources that are picked up in each zone (scenario A-2-B). + +AUD-R6: The VMS or vehicle should provide at least one voice/speech/audio channel per zone (e.g., one microphone per zone). + +AUD-R7: When interfacing with cloud recognition services that are not trained specifically with vehicle noise, it should be possible to perform noise reduction in the target zone to suppress road noise or wind noise. + +### 8.2 Gesture control in vehicular multimedia system + +Applying gesture recognition as part of intelligent human machine interface (HMI) and of ADAS can improve driving safety. Drivers can use gesture to control various functions, including in-vehicular multi-media functions, or change various settings, such as the temperature of air-conditioners or sound volume in the vehicle, thus paying more attention to the road and reducing the occurrence of traffic accidents. + +Through a simple and easy to use gesture control system, drivers can interact with multimedia system functions while looking directly at the road ahead. + +### 8.3 Eye-movement control in VMS + +#### 8.3.1 Real time eye gaze tracking scenario + +Eye gaze tracking is the process of measuring either the point of gaze (where one is looking) or the motion of an eye relative to the head. An eye tracker is a device for measuring eye positions and eye movement, such as the one illustrated in Figure 10. + +![Figure 10: Example of interactive eye-gaze-controlled HUD. The image shows a driver's perspective from inside a car, looking through the windshield at a busy street with motorcycles and cars. A transparent Heads-Up Display (HUD) is overlaid on the view, showing various information like speed, navigation, and vehicle status. A small inset image in the top right corner shows a close-up of a person's eye with a tracking device attached to the forehead, demonstrating the eye gaze tracking technology.](a3f81dedbdc5b702f397d07ef476d53e_img.jpg) + +Figure 10: Example of interactive eye-gaze-controlled HUD. The image shows a driver's perspective from inside a car, looking through the windshield at a busy street with motorcycles and cars. A transparent Heads-Up Display (HUD) is overlaid on the view, showing various information like speed, navigation, and vehicle status. A small inset image in the top right corner shows a close-up of a person's eye with a tracking device attached to the forehead, demonstrating the eye gaze tracking technology. + +**Figure 10 – Example of interactive eye-gaze-controlled HUD** + +With the progress in processor speed and image processing algorithms, it is now also possible to use the gaze tracking technology in real time to control a screen pointer in a direct manipulation interface. Eye gaze tracking technology is well explored for consumer electronics and gradually making way to the automotive and aviation sector. It has already been explored for the attention analysis of drivers and pilots and designing cockpits to minimize eye gaze movement, highlighting critical elements and so on. The following use cases emphasize real time use of eye tracking in a + +moving vehicle as a direct controller of electronic displays inside a car and to analyse drivers' cognitive load while driving. + +- Driver wants to change radio channel or accept a call. He does not need to take his eyes off the road nor look down at the dashboard, and neither to take his hands off the steering wheel. The whole dashboard is reflected on a heads-up display and he operates the display just like any graphical user interface through fixating attention on the required icons. The fixation duration is less than 2 seconds and even when he is operating the display, he can use his peripheral vision to observe the road. +- Driver is distracted, feeling stressed or bored, eye tracker detects driver's state from ocular parameters like pupil dilation or fixation rate. Based on the gaze tracker inputs the ADAS system that will trigger the VMS to reduce secondary task load (e.g., VMS media playback) and alerts driver and passengers. +- Eye tracker tracks eyes of passengers of an autonomous vehicle. The VMS uses it to operate entertainment systems on back seats and can be used to alert passengers if the passenger in the driver's seat is distracted or stressed. + +##### **8.3.1.1 Gap analysis** + +Required improvement for the use of eye gaze trackers in automotive environment includes: + +- Commercial eye gaze trackers which have not yet been tested for automotive compliance +- Robust algorithm for affective state detection +- Improving accuracy and latency of commercial eye gaze trackers in automotive environment +- Combining eye gaze control system with other existing modalities +- Testing the accuracy of driver distraction and affective state detection system in automotive environment. + +#### **8.3.2 Use case – Gaze tracking** + +A dynamic HMI responds to the user's gaze. Elements of the UI can be configured to change based on whether or not the user is looking at them. For example, an element on the screen can be set to be transparent when the user is not looking at it, and immediately return to the visible state once the user is looking at it. + +Controlling the vehicle multimedia by eye movement can be implemented to improve driving safety. + +##### **8.3.2.1 Requirements** + +GC-R1: Accuracy and latency requirement for gaze control system integrated for use in the HMI of a VMS shall not require fixating attention for more than 2 seconds or the maximum time defined by local regulations in the automotive environment on the HMI. + +#### **8.3.3 Use case – Controllable camera through eyes** + +Users can also control an external camera through their eyes, which is similar to the head-tracking camera. + +NOTE – This is not a digital video recorder (DVR) camera which is mandated in certain countries to record. DVR cameras are not user controllable. + +### **8.4 Data formats and storage** + +#### **8.4.1 Use case and considerations** + +There is a need to define a minimum set of data formats to be supported by a VMS to allow for interoperability between different VMS and different VMNS providers. The following aspects should be considered: + +- Codecs (and protocols) used by broadcast and interactive systems to and from a VMS. +- Contents on nomadic devices (mobile phones) when connecting to the VMS (e.g., to share content via the VMU). +- Existing codecs usually supported by current infotainment solution. +- Additional sensors and their storage/compression requirements for VM system services (e.g., maps, AR, telematics). + +The main electronic control unit (ECU) may be built on a powerful processor, while processors with lower capabilities (e.g., limited set of codecs) may be available in the vehicle (e.g., with mounted display in back seat positions). As such, it can be desirable to transcode content received at the main ECU in formats supported throughout the vehicle. + +Further, as the amount of data exchanged between devices and infrastructure is becoming a bottleneck, evaluation of codec performances (compression, lossless/lossy capabilities) should be taken in account when deciding on codec support. In the context of a VMS, this data comprises at least audio, video, images and point-cloud, such as light detection and ranging (LIDAR), which can be used both for entertainment and safety purposes. + +In addition, VMS and VMU will need to support software updates. These updates and software codecs should be securely provided. Their use and access by different VMNS may be restricted. + +##### **8.4.1.1 Requirements – Data format** + +DFS-R1: For multimedia broadcasting services provided by a VMNS provider, the media types, content navigation methods and interactivity methods recommended in [ITU-R BT.2055-1] shall be supported by the VMS and the VMNP. + +DFS-R2: Additional media types may be supported by the VMS and the VMNP to address some regional requirements. + +DFS-R3: A VMN application may support additional software video and/or audio codecs to address the need of a VMNS provider. + +DFS-R4: For interactive, and non-broadcasted VMNS, the VMNP and VMS should support the same multimedia broadcast media types. + +## **9 Privacy considerations** + +### **9.1 Background** + +A VMS could have the following characteristics: + +- 1) Be available on a connected vehicle (or/and autonomous vehicle) +- 2) Be personalisable, like a phone, a subscription or a service tailored to the user and to the user's interests + +Countries and states have started to regulate connected vehicles or release broad privacy regulations. Among them: + +- The US Department of Transport requires, among other things, that connected vehicle safety applications messages shall not contain any personally identifying information (PII). [b-DoT PII-1] + +- In Europe, the general data privacy regulation (GDPR) came into effect in May 2018 and requires, among other things, that only the minimum necessary personal information are collected, consent shall be obtained in a clear manner, collected data shall be accessible by the user and data shall be deleted when no longer relevant or when consent is withdrawn [b-DoT PII-2] +- In China, the Cybersecurity Law of the People's Republic of China came into effect in June 2017 and requires that network operators collecting and using personal information shall abide by the principles of legality, propriety, and necessity. Besides, a guideline for Internet personal information security protection is being formulated to ensure that the rights of data subjects are not to be violated. + +Further, the following scenario should be envisaged: + +- The vehicle is owned by the user of the VMS +- The VMS may be temporarily used by different individuals (driver or passenger) of the vehicle, for instance in a family, shared car, rental car, or taxi model. +- The VMS may be shared by different individuals in the vehicle (driver and passengers) with no relation between them, for instance in a public transportation shuttle. +- The personalization of a VMS may be provided by a centralized service provider. + +In network communication protocols, at various layers identifiers are used to identify software and hardware (e.g., IP address, MAC address, E.164 "phone number", etc.). Some of these identifiers constitute PII in some instances, and protocol designers in the past did not always take this into account. In the design of systems and protocols today, there needs to be intentional separation of PII identities (e.g., username, email address [b-IETF e-mail] and [b-ITU-T E.164]) and non-PII identities (e.g., IP address, MAC address, etc.) used at various layers in the communication stack. + +Further, user's privacy and PII (such as viewing history, history of interactions, profiles and preferences) needs also to be protected in interactive broadcasting system, both at the broadcast receiver (e.g., the VMS or VMS application), on the interactive link and at the service provider level as highlighted in [b-ITU-R BT.2052]. + +From the above it becomes apparent that VMNS need to allow for different use cases under different regulatory requirements. The high-level privacy requirements in clause 9.2 are proposed: + +### 9.2 General privacy requirements + +PrivGen-R1: The VMN and VMS shall be designed to allow for its use under different privacy regulatory environments. + +PrivGen-R2: The VMN and VMS shall prevent network-layer identifiers (such as MAC address) to be used as personally identifying information. + +PrivGen-R3: The VMS shall protect any permanent hardware identifiers and only allow access for authentication purposes. + +PrivGen-R4: In addition to the specified requirements for in-vehicle entertainment and multimedia systems, it is required that users' (driver, passengers) privacy be protected, and that private conversations not be directed to the voice-recognition system, be protected and that un-authorized sharing of such conversations be hindered. + +PrivGen-R5: Consideration should be given to the case where the voice recognition system is in-vehicle or is cloud-based. + +PrivGen-R6: The VMS should have the capability to wipe personal data transferred through the mobile device connected to the VMS in order to respect privacy of personal data. + +### **9.3 Use case – Private vehicle with one single user** + +The end user is identified when entering his privately-owned vehicle. VMS cloud platform has access to the agenda of vehicle owner. It therefore knows if the user is driving to a meeting, a birthday party, a restaurant, etc., and the time of arrival at the destination. Based on the time of day and geo-localization of the vehicle, the platform sends proposals to the VMS that are displayed on the VMS or read to the end user through the VMS (text-to-speech) e.g., restaurant reservation before a meeting, buying flowers on the way to a dinner or a gift on the way to a birthday party, etc. The vehicle user can through direct interaction (voice recognition, touch screen, other) with the VMS make use of these proposals or deny proposals. Furthermore, having deposited payment preferences in the cloud, the vehicle owner can directly pay through the VMS. Based on the inputs into the VMS, the vehicle might be rerouted to another itinerary in order to pick up the goods or to go to the restaurant chosen during the ride. + +#### **9.3.1 Requirements** + +**PrivPV-R1:** VMSP should have access to a commercial platform linking location-based data with the information of shops, restaurants and current commercial offers. + +**PrivPV-R2:** VMS should have access to personal payment data that are deposited in the cloud. + +**PrivPV-R3:** VMS should have a feature for secure identification of the user before payment process (e.g., fingerprint sensor, iris scanner). + +### **9.4 Use case – Long distance shared vehicle with multi users (shuttle all users are seated)** + +End users are identified when entering the shuttle. Each user of the shuttle has his own seat in the vehicle. VMS identifies where they are seated within the shuttle in order to adapt content, based on the capabilities of each seated position (some might have screen, some just audio, the possibility of seat comfort adaptation etc.). VMS has access to agenda or agenda parts of each shuttle user (either from the cloud, following authorization from the user, or from the user's mobile phone). The itinerary has been optimized in order to bring each vehicle user on an optimized route within the shortest time to its destination. In addition, due to the link of the cloud platform with the agenda of the multiple users, the VMS knows for each single user if he drives to a meeting, birthday party, etc. Based on optimization algorithms the VMS proposes services (restaurant reservations, buying flowers or a gift) to each individual shuttle user based on his respective agenda. Each vehicle user can pay through direct interaction with its VMS within the shuttle. + +#### **9.4.1 Requirements** + +**Priv-SVS-R1:** VMS should have the ability to identify the different users and retrieve their personal and payment data. + +**Priv-SVS-R2:** VMS should support receiving agenda and personal information from user mobile devices. + +**Priv-SVS-R3:** VMS should allow the user to enter destination without revealing his agenda and other personal information (via dedicated application, or in-vehicle screen with coordination and synchronization with user mobile phone). + +**Priv-SVS-R4:** The VMSP with data concerning destination and agenda of all users of the shuttle and the commercial decisions all users take, should provide these data to the cloud navigation application and receive from it, an optimized proposal for the different users and for the routing of the shuttle. + +**Priv-SVS-R5:** User should be able to change itinerary, overwriting agenda information and previous instructions in the VMS at any time. + +Priv-SVS-R6: Upon detection of change of the VMS user, the VMS should wipe personal data transferred through the mobile device connected to the VMS in order to respect privacy of personal data. + +### **9.5 Use case – Public transportation shared vehicle with multi users (shuttle short distance trips – seated and standing users)** + +End users are identified when entering the shuttle (via RFID, mobile phone NFC, or face recognition for registered users). VMS identifies the position of the user in the vehicle. + +VMS has access to agenda or agenda parts of each shuttle user (either from the cloud, following authorization from the user, or from the user's mobile phone). The itinerary has been optimized in order to bring each vehicle user on an optimized route within the shortest time to its destination. + +VMS provides personalized information on the public display in the bus informing users when approaching their bus stop (by preserving their privacy). VMS can send messages to mobile or wearable devices of user with more private information, if this is authorized by the user. + +VMS presents infotainment information on the public screens, based on interests of the user and their proximity to the screen. As the journey progress, information about the city, about events near to the route, about special offers of shops in the trajectory and other interesting information that are displayed on the public screen. User can also ask to send this information directly on their smart device while on board the vehicle. + +#### **9.5.1 Requirements** + +**Priv-SVM-R1:** VMSP should be able to push non-personalized messages on VMS or directly on user's mobile devices when users have opted-in to the service. + +**Priv-SVM-R2:** Connectivity from mobile devices to VMS should be allowed. + +**Priv-SVM-R3:** VMS should be able to receive agenda and personal information from user mobile devices. + +**Priv-SVM-R4:** VMS should be able to allow the user to enter destination without revealing his agenda and other personal information (via dedicated application, or in-vehicle screen with coordination and synchronization with user mobile phone). + +**Priv-SVM-R5:** VMSP with data concerning the destination and agenda of all users of the shuttle shall optimize content proposals to the different users and the routing of the shuttle. + +**Priv-SVM-R6:** User should be able to change itinerary, overwriting agenda information and previous instructions via VMS or via their mobile phone at any time. + +**Priv-SVM-R7:** User should have the possibility to opt-in or opt-out of the service at any time either through an interface of the VMS or through its mobile device. + +**Priv-SVM-R8:** Upon detection of change of the VMS user, the VMS should wipe personal data transferred through the mobile device connected to the VMS in order to respect privacy of personal data. + +## **10 Content rights protection in a converged network environment** + +### **10.1 Background** + +The convergence of different networks combines features of unidirectional networks and bi-directional networks. Such converged networks provide a variety of advanced multimedia information services to vehicle terminals, i.e., live television service, online video streaming service, multimedia content delivery service, etc. The architecture of the converged networks is illustrated in Figure 11. + +![Figure 11: Convergence of different networks for vehicular multimedia content. The diagram shows a 'Service platform' connected to a central cloud containing 'Short range network', 'LEO network', and 'Cellular' icons. Above this cloud is a 'Broadcast' satellite. A 'One-directional network link' connects the satellite to 'Vehicular multimedia' (represented by a screen). A 'Bi-directional network link' connects the central cloud to a car. A label 'F.749.3(20)_F11' is in the bottom right.](08dce7ad4c512fdf0c0cde60415fade6_img.jpg) + +Figure 11: Convergence of different networks for vehicular multimedia content. The diagram shows a 'Service platform' connected to a central cloud containing 'Short range network', 'LEO network', and 'Cellular' icons. Above this cloud is a 'Broadcast' satellite. A 'One-directional network link' connects the satellite to 'Vehicular multimedia' (represented by a screen). A 'Bi-directional network link' connects the central cloud to a car. A label 'F.749.3(20)\_F11' is in the bottom right. + +**Figure 11 – Convergence of different networks for vehicular multimedia content** + +Consider live television service as an example. Currently, live television service is delivered solely by satellite broadcast networks. To protect the television video content from being received at unauthorized terminals, the following conditional access (CA) technique is applied. The architecture of the satellite broadcast system with CA support is illustrated in Figure 12. + +![Figure 12: Diagram of the satellite broadcast system with CA support. The diagram shows the flow of data from multiple channels (A/V CH 1 to A/V CH N) through 'Scrambler' units. Each channel also has a 'Control word generator' producing a 'CW 1' for scrambling and encryption. A 'Product key generator' produces a 'PK' for encryption. A 'Personal distribution key' (PDK) is used to encrypt the PK into an 'EMM'. All scrambled data, 'ECM's, and 'EMM's are multiplexed ('MUX') into a 'Satellite broadcast channel'. At the receiver, the stream is demultiplexed ('De-MUX'). The 'ECM' is decrypted using the 'PDK' to reveal the 'PK'. The 'PK' is then used to decrypt the 'EMM' to reveal the 'CW 1'. Finally, the 'CW 1' is used to de-scramble the audio/video data to recover 'A/V CH 1'. A label 'F.749.3(20)_F12' is in the bottom right.](b5335262987c819d7f71ce40f99cb71b_img.jpg) + +Figure 12: Diagram of the satellite broadcast system with CA support. The diagram shows the flow of data from multiple channels (A/V CH 1 to A/V CH N) through 'Scrambler' units. Each channel also has a 'Control word generator' producing a 'CW 1' for scrambling and encryption. A 'Product key generator' produces a 'PK' for encryption. A 'Personal distribution key' (PDK) is used to encrypt the PK into an 'EMM'. All scrambled data, 'ECM's, and 'EMM's are multiplexed ('MUX') into a 'Satellite broadcast channel'. At the receiver, the stream is demultiplexed ('De-MUX'). The 'ECM' is decrypted using the 'PDK' to reveal the 'PK'. The 'PK' is then used to decrypt the 'EMM' to reveal the 'CW 1'. Finally, the 'CW 1' is used to de-scramble the audio/video data to recover 'A/V CH 1'. A label 'F.749.3(20)\_F12' is in the bottom right. + +**Figure 12 – Diagram of the satellite broadcast system with CA support** + +At the head end, the multimedia stream for each channel is scrambled by a control word (CW). The CW is encrypted by the product key (PK) to obtain the entitlement control message (ECM). The PK is encrypted by the personal distribution key (PDK) to obtain the entitlement management message (EMM). Scrambled data, ECMs, and EMMs are multiplexed into a transport stream (TS) before it is transmitted over satellite broadcast channels. At the terminal, the transport stream is demultiplexed to recover the scrambled data, ECMs, and EMMs. Then, the PK is decrypted from the EMM by the + +PDK stored at the CA terminal. Next, the CW 1 for channel 1 is decrypted from the ECM by the PK. Finally, the multimedia stream for channel 1 can be obtained by descrambling the data for channel 1 with CW 1. Note that PDK is unique for each terminal, and, therefore, the EMM for a terminal would be different from those for other terminals. By using the above CA technique, for each terminal the access to the live television service can be controlled (i.e., access to the content is either allowed or denied). + +### 10.2 Use cases + +The conditional access system shown in Figure 12 has several drawbacks. In the following use cases, the limits of the above conditional access system are analysed and solutions to improve the system performance with converged networks are described. + +#### 10.2.1 Use case A – VMS with flexible EMM + +Consider a live television broadcast service delivered solely by satellite networks. In such a case, the scrambled multimedia data for each channel, ECM for each channel, and EMM for each terminal should be transmitted over the satellite broadcast channel. Typically, the size of an EMM is about dozens of bytes. If tens of thousands of new users are trying to activate their devices in a short period of time, then the bandwidth required for the transmission of the EMMs for these new devices may not be available in a satellite broadcast channel. As a result, some of these users may experience service activation delay or failure. + +For a service operator with access to converged networks, such circumstance could be avoided by offloading the EMM transmission to the cellular communication network. The architecture of the multimedia system with flexible EMM transmission is illustrated in Figure 13, where the scrambled data and ECMs are transmitted over satellite broadcast channel and the EMMs are transmitted over cellular communication networks. + +![Figure 13: Diagram of the multimedia system with flexible EMM transmission. The diagram shows the flow of data from multiple channels (A/V CH 1, A/V CH 2, A/V CH N) through scramblers to a multiplexer (MUX). The MUX outputs a satellite broadcast channel. On the receiver side, a de-multiplexer (De-MUX) separates the data. One path goes to a de-scrambler to produce A/V CH 1. Another path goes to a decryption block. A third path involves a 'Cellular communication networks' block with 'Request' and 'Response' signals. The decryption block uses a Personal Distribution Key (PDK) and a Product Key (PK) to decrypt the ECM and EMM. The decrypted ECM is then used by a control word generator to produce CW 1, which is fed back to the scrambler for A/V CH 1.](a1890b9a9b85f13e67ed59bbad623659_img.jpg) + +The diagram illustrates the architecture of a multimedia system with flexible EMM transmission. On the transmitter side, multiple audio/video channels (A/V CH 1, A/V CH 2, A/V CH N) are input to individual scramblers. Each scrambler also receives a control word (CW 1) from a control word generator. The scrambled data from each channel is then multiplexed (MUX) into a single satellite broadcast channel. Below the MUX, a control word generator receives CW 1 and an ECM (Encrypted Control Word) from an encryption block. The encryption block takes a Product Key (PK) from a product key generator and a Personal Distribution Key (PDK) from a personal distribution key generator. The ECM is then sent to the MUX. The satellite broadcast channel is received by a De-MUX. One output of the De-MUX goes to a de-scrambler, which produces the final A/V CH 1 output. Another output of the De-MUX goes to a decryption block. This block also receives a PK and a PDK. It decrypts the ECM to provide CW 1 to the control word generator and the EMM (Entitlement Management Message) to the cellular communication networks. The cellular communication networks send a 'Request' and a 'Response' signal to the decryption block. The decryption block also decrypts the EMM to provide the PDK to the personal distribution key generator. + +Figure 13: Diagram of the multimedia system with flexible EMM transmission. The diagram shows the flow of data from multiple channels (A/V CH 1, A/V CH 2, A/V CH N) through scramblers to a multiplexer (MUX). The MUX outputs a satellite broadcast channel. On the receiver side, a de-multiplexer (De-MUX) separates the data. One path goes to a de-scrambler to produce A/V CH 1. Another path goes to a decryption block. A third path involves a 'Cellular communication networks' block with 'Request' and 'Response' signals. The decryption block uses a Personal Distribution Key (PDK) and a Product Key (PK) to decrypt the ECM and EMM. The decrypted ECM is then used by a control word generator to produce CW 1, which is fed back to the scrambler for A/V CH 1. + +Figure 13 – Diagram of the multimedia system with flexible EMM transmission + +#### 10.2.2 Use case B – VMS with extended EMM + +Bearing in mind that the conventional conditional access system could support simple access control at terminals, i.e., the play of multimedia streams can be either allowed or denied at a + +terminal. However, for vehicle multimedia service operators, more rights control options shall be available. The following should be noted for a live television programme: + +- 1) the reception of this live television program should be controlled (i.e., allowed or denied); +- 2) the recording and local storage of this live television program should be controlled (i.e., allowed or denied); +- 3) if the recording and local storage of this live television program is allowed, then a maximum number of plays of the recorded video can be set; +- 4) the forwarding of this live television program to other nomadic devices in vehicle should be controlled (i.e., allowed or denied). + +To support the above rights control options, rules shall be delivered from the head end to the terminals. For each terminal, specific rights rules may be applied. For example, specific to each terminal, the rights rules can be delivered together with the EMM over the cellular communication networks. A diagram of the multimedia system with extended EMM to support more rights control options is shown in Figure 14, where the scrambled data and ECMs are transmitted over satellite broadcast channel and the extended EMMs are transmitted over cellular communication networks. Note that this is a general framework and the specific scheme to achieve rights rule control at terminals is an implementation issue. + +![Diagram of the multimedia system with extended EMM to support more rights control options. The diagram shows the flow of data from multiple A/V channels through scrambling, multiplexing, and satellite transmission to a de-multiplexer and de-scrambler. It also details the generation and transmission of ECMs and Extended EMMs via cellular networks, incorporating right rules and personal distribution keys for decryption and control.](6e15fc9ea763541c5913d26f85072ae1_img.jpg) + +The diagram illustrates a multimedia system architecture. On the left, three input channels (A/V CH N, A/V CH 2, and A/V CH 1) are processed by individual 'Scrambler' blocks. Each scrambler outputs 'Scrambled data' to a central 'MUX' block. Below the MUX, a 'Control word generator' produces 'CW 1', which is used by the scramblers and also passed through an 'Encryption' block to generate an 'ECM'. The 'MUX' outputs a combined signal to a 'Satellite broadcast channel' (represented by an oval). This signal is then processed by a 'De-MUX' block. The 'De-MUX' outputs 'Scrambled data' to a 'De-scrambler' block, which produces the final output 'A/V CH 1'. The 'De-scrambler' uses 'CW 1' and an 'ECM' for decryption. Below the De-scrambler, a 'PK' (Product key) and a 'PDK' (Personal distribution key) are used by a 'Decryption' block. This block also receives an 'EMM' (Entitlement Management Message) from an 'EMM de-extension' block. The 'EMM de-extension' block receives an 'Extended EMM' from a 'Cellular communication networks' block. The 'Cellular communication networks' block sends a 'Request' and receives a 'Response' from the 'EMM extension' block. The 'EMM extension' block receives an 'EMM' from an 'Encryption' block. This 'Encryption' block uses a 'PK' from a 'Product key generator' and a 'PDK' from a 'Personal distribution key' block. It also receives 'RR' (Right rules) from a 'Right rules' block. The 'Right rules' block contains a list of control options: 1) Play control (Allowed/denied, Maximum times of play), 2) Record control (Allowed/denied), 3) Forward control (Allowed/denied), and Etc. The 'EMM extension' block outputs an 'Extended EMM' to the 'Cellular communication networks' block. The 'EMM de-extension' block outputs an 'EMM' to the 'Decryption' block and an 'RR' (Right rules) to a 'Right rules control' block. The 'Right rules control' block is connected to a 'Personal distribution key' block. The diagram is labeled 'F.749.3(20)\_F14' in the bottom right corner. + +Diagram of the multimedia system with extended EMM to support more rights control options. The diagram shows the flow of data from multiple A/V channels through scrambling, multiplexing, and satellite transmission to a de-multiplexer and de-scrambler. It also details the generation and transmission of ECMs and Extended EMMs via cellular networks, incorporating right rules and personal distribution keys for decryption and control. + +**Figure 14 – Diagram of the multimedia system with extended EMM to support more rights control options** + +### 10.3 Gap analysis + +At present, multimedia information service is mainly based on either the unidirectional network or on bidirectional networks. + +For multimedia information systems based on unidirectional networks, the following conventional CA schemes can be applied to provide simple access control at terminals. Two major drawbacks for such systems have been analysed in clause 10.2. + +- 1) ETSI Standard: Support for use of scrambling and conditional access (CA) within digital broadcasting systems ETSI ETR 289 ed.1 (1996-10) [b-ETSI ETR 289] +- 2) ATSC Standard: Conditional access system for terrestrial broadcast (A/70 Part 1:2010) [b-ATSC] +- 3) China's radio and television industry standard: Specification of conditional access system for digital television broadcasting GY/Z 175-2001 [b-CN-CAS] + +For multimedia information systems based on bi-directional networks, the following conventional digital rights management (DRM) schemes can be applied. A major drawback of such schemes is that they cannot be seamlessly integrated with conventional conditional access schemes. Nevertheless, the concept of rights rules in these DRM schemes can be used to enhance the conventional CA schemes. + +- 1) Open Mobile Alliance: OMA digital rights management (DRM) [b-OMA DRM] +- 2) ChinaDRM: Technical specification of digital rights management for internet television, GY/Z 277-2014 [b-ChinaDRM] + +### **10.4 Requirements – Content rights protection requirements for CA** + +The requirements for the content rights protection in converged networks are as follows: + +RP-R1: The integrity of rights rules for a terminal shall be guaranteed during its transmission from the head-end to the terminal. + +RP-R2: The rights control options shall be extensible for additional rules to be added in the future. + +RP-R3: Conventional conditional access schemes for unidirectional networks may be supported. + +RP-R4: Entitlement management messages in conventional CA schemes may be delivered over bidirectional networks. Further: + +- a) A CA agent on the VMS should support transmitting a request of its EMM to the head-end. +- b) The head-end should support transmitting a response with the EMM to a terminal. +- c) The head-end may push the EMM to a VMS without being requested. + +RP-R5: Rights rules for a terminal may be transmitted along with the EMM to the terminal. + +RP-R6: Rights rules for a terminal may be transmitted in an independent message to the terminal. + +## **11 Copyright and rights management support for content delivery** + +### **11.1 Use case – Content payment and copyright protection** + +For some content providers the in-vehicle screen and mobile phones are two different distribution channels with different service costs and price offerings. The price for an end-user to receive content in a vehicle can be different on a mobile phone than on the in-vehicle screen. Sometimes the end-user should pay again to enjoy the same content in a vehicle while the content was already paid for by mobile phone consumption. Alternatively, a content provider may allow its users to enjoy the same content freely in vehicle if users have a paid content subscription for mobile phone consumption. + +These different commercial policies across distribution channels should be enabled by the VMN and VMS. The VMN should enable content providers to manage content-services-fee-payment across distribution channels. + +#### 11.1.1 Requirements – Copy protection + +RM-R1: The design of the VMS should be capable of preventing multi-media contents distributed to the VMS from being copied from the VMS to mobile phones and/or be redistributed through the VMS hardware to external devices. + +NOTE – This does not include the possibility that users record screens with other recording devices. + +### 11.2 Privilege and account management + +#### 11.2.1 Use cases + +##### 11.2.1.1 Use case A – Parental control + +Parental control may be enabled for children or teenager passengers to prevent access to restricted content. + +An identification (ID) system (e.g., traditional ID such as mobile phone number or biometric technology ID such as face or voice recognition) and corresponding database may be necessary to carry out parental control. The passenger may be identified as either an adult or a child. + +##### 11.2.1.2 Use case B – Content classification + +To enable content and services rights management, identifying content classification may be necessary to replicate common practice for multi-media content on other channels (e.g., movies in theatre versus DVD). Content and services can be classified into different categories, based on age or subscription status. + +##### 11.2.1.3 Use case C – Privilege management in single use vehicle + +Depending on different business models, content and service privilege management may differ. + +An original equipment manufacturer (OEM) vehicle maker may have a business agreement with a content or service provider. The OEM vehicle maker has paid for the subscription fee for its customers. The vehicle owner, upon acquiring the vehicle can receive the content subscription for personal use, but only in this very vehicle. In this situation, privilege management is based on hardware identification, as the objective of the OEM vehicle maker by paying for the subscription is to sell more vehicles by offering such services and entertainment content privilege. + +##### 11.2.1.4 Use case D – Privilege management in shared vehicle + +In the car sharing industry, several different business models lead to differing requirements for the VMS. One vehicle will welcome many different passengers per day. There are two categories of companies in the car sharing industry: + +###### Category A + +Rental companies who provide only vehicles without drivers may pay content/service subscriptions for their customer as a strategy to enhance their market competitive strength. In that situation, the subscription may be associated with a *vehicle hardware identifier*. In addition, such rental company can also allow users to log into their own third-party content providers accounts to guarantee a similar user experience on the rental car, consistent with the one on a user's own private vehicle. + +###### Category B + +For taxi companies, who are not only providing vehicles but also drivers, the number of customers is higher than for companies in category A. It can be very difficult to satisfy all passengers by a content/service subscription tied to the vehicle hardware as it may be impossible to predict what kind of music the next passenger may enjoy or desire to hear. For these companies, a better system may be to enable access to cloud-based content/service by each customer based on a *user identifier*. + +Independently of which taxi a user is sitting in, as long as the user logs into his or her account, the user can immediately enjoy the same services and the same user-experience in different vehicles. The VMS will retrieve the user history, default recommendations, and other user-specific data from the system. + +##### 11.2.1.5 Use case E – Privilege management for local media content + +Hardware and/or software suppliers may pre-load content made by themselves into vehicles (local media content). User's privilege management for local media content could differ from third party's online media content. + +##### 11.2.1.6 Example of mapping of different ID system + +Figure 15 is an example illustrating how two ID systems (User ID and IVI system/hardware ID) can be used concurrently to provide a multimedia service. + +![Diagram illustrating the mapping of different ID systems for a multimedia service. It shows two users, User(1) and User(N), connected to an IVI (In-vehicle infotainment) system. The IVI system is connected to a Telematics service platform, which is in turn connected to a CP/SP (Content Provider/Service Provider). Login requests are shown: User(N) logs in with UserID(N) to the Telematics service platform, User(1) logs in with UserID(1) to the Telematics service platform, and the IVI logs in with (DeviceID, VIN) to the Telematics service platform. The Telematics service platform is connected to the CP/SP.](382a9c9e4816bd229191ab4591424dd8_img.jpg) + +``` + +graph LR + User1[User(1)] <--> IVI[IVI In-vehicle infotainment] + UserN[User(N)] <--> IVI + UserN -- "Login with UserID(N)" --> Telematics[Telematics service platform] + User1 -- "Login with UserID(1)" --> Telematics + IVI -- "Login with (DeviceID, VIN)" --> Telematics + Telematics <--> CPSP[CP/SP] + +``` + +F.749.3(20)\_F15 + +Diagram illustrating the mapping of different ID systems for a multimedia service. It shows two users, User(1) and User(N), connected to an IVI (In-vehicle infotainment) system. The IVI system is connected to a Telematics service platform, which is in turn connected to a CP/SP (Content Provider/Service Provider). Login requests are shown: User(N) logs in with UserID(N) to the Telematics service platform, User(1) logs in with UserID(1) to the Telematics service platform, and the IVI logs in with (DeviceID, VIN) to the Telematics service platform. The Telematics service platform is connected to the CP/SP. + +**Figure 15 – ID systems** + +For devices, "Device ID" or "Vehicle Identification" (such as a VIN code) or other identifiers may be used to log into OEM vehicle maker management platforms (e.g., telematics service platform). For end users, "User ID" may be used to access services and to obtain personalized content and recommendations. + +NOTE – Privacy and regulatory aspects will need to be taken in account when selecting specific ID mechanisms. + +#### 11.2.2 Requirements + +##### 11.2.2.1 Privilege management requirements + +PM-R1: VMN should protect and manage all content copying on the platform. + +PM-R2: User classification (e.g., based on age) and content classification should be supported to carry out multi-media service privilege management. + +PM-R3: VMN should support parental control function. + +PM-R4: VMN may support an ID system and corresponding database to enable content rights and privilege management and to provide customized services. + +Example of user-ID may comprise a mobile phone number and/or biometric (e.g., face recognition, voice recognition, or finger-print information, etc.). + +PM-R5: VMS should be able to apply privilege management to preloaded content associated with VMS hardware. When restrictions apply, the VMS should prevent copying the preloaded content to other devices. + +NOTE – As long as there is a local record of a subscription on the VMS, a user can enjoy the privilege in that vehicle. + +##### 11.2.2.2 Account management requirements + +AM-R1: The personalization of the multimedia portal should be enabled by the VMN account management system. + +AM-R2: The VMS should support multiple users to login concurrently in the same vehicle with different user IDs for personalization purposes. + +AM-R3: After logging in with different IDs, the displays of VMS for different users may be personalized, including greetings, interfaces and services. + +AM-R4: If one user logs into different vehicles with the same user ID, then the VMS may support retrieving and displaying the user personalized interfaces and recommended contents to ensure a consistent user experience across VMS enabled vehicles. + +AM-R5: The VMS should be capable of matching a vehicle hardware ID (such as key fobs or biometric based ID system) with user IDs to enable access to third party services (e.g., content provider or services provider). VMS should be able to access IDs available locally and in the cloud. + +### **11.3 Payment services in shared vehicle** + +#### **11.3.1 Use case – Time shared vehicle** + +Passengers of a time-shared vehicle in exclusive use (drivers and/or passengers, no additional passengers will be picked-up during the journey) are identified (for example by facial recognition, fingerprint, iris scan, RFID in personal ID card, etc.) when entering the vehicle and then identified at each individual seat by the VMS. These passengers have deposited their payment preferences in the cloud. The payment of services and goods like parking, fuel, charging of the EV, food at a drive-in, and usage fees for a shared vehicle will be executed after confirmation through the VMS based on person identification and payment preferences. + +For multi-passengers in a shared vehicle the person who has initiated the process, e.g., parking, charging, refuelling, drive-in will be asked by the VMS if he wants to pay individually or if he would like to share costs with the other passengers. In case this person wants to share costs the VMS dispatches messages to the individual VMS's of the other passengers asking confirmation of the other passengers for participation in the payment. These passengers confirm or deny the payment at their individual VMS within the vehicle using, e.g., touch screens, voice, mobile phone confirmation or other available means (like scanning RFID identification card, replying to keyword challenge they have previously established). + +##### **11.3.1.1 Requirements** + +PTS-R1: VMS may have access to different cloud platforms in which the different users have deposited their personal and payment data. In case user has no payment profile, VMS should support payment through credit card. + +PTS-R2: The VMS shall be able to dispatch VMS messages and processes to individual VMS terminals within the vehicle + +PTS-R3: VMS should be able to connect to user personal mobile phone for personal information, communication, and confirmation in case vehicle would not be equipped at all seats with displays and vehicle VMS interfaces. + +PTS-R4: Upon detection of change of the VMS user, the VMS should wipe personal data transferred through the mobile device connected to the VMS in order to respect privacy of personal data. + +## **12 Security** + +### **12.1 Use case – User data protection** + +As vehicles become connected and offer more interactive services, more user data and privacy-related information (such as users viewing preferences and habits) could be generated and stored in the VM system (vehicle) and in the VM service (cloud/servers). The user data and privacy-related + +information should not be transmitted between systems without user consent. Similarly, these data should not be accessed by another system without user consent. When the user data and privacy-related information are transmitted between systems, the protection of user data shall be ensured and becomes an important consideration in the performance and security evaluation of vehicle multimedia systems. + +The system should guarantee that the user data cannot be given to third parties without explicit authorization from the user. + +#### **12.1.1 Requirements** + +UsrDP-R1: The in-vehicle multimedia platform should provide end-to-end data protection to ensure confidentiality and integrity of user data, including data protection at rest in local terminals, during transmission over different channels and when processed at the cloud platform. + +### **12.2 Use case – VMN application security** + +A VMN application should be securely developed according to best practices and applicable local regulations and not bring additional risks within the vehicle and should not be tampered with or corrupted. + +#### **12.2.1 Requirements** + +AppSec-R1: In order to verify the application and to ensure its origin authorization, a VMN application should be signed with a digital signature. + +AppSec-R2: The VMS should be able to verify the signature of the VMN application in actual use prior to allowing access to APIs (e.g., communication APIs). + +AppSec-R3: It shall be possible to wipe all the data (application, content and data) on the VMS to return the VMS to its factory settings. + +AppSec-R4: It shall be possible to wipe a subset of the data on the VMS. + +AppSec-R5: The wipe shall be able to be performed by different authorised entities such as the VMNS provider, the end-user or the VMS manufacturer. + +### **12.3 Use Case – VMS software security** + +Due to increasing system complexity, software error or software fault may exist in VMS, which may cause crashes. The system should guarantee that such fault would not cause operation errors or abnormal stop that would threaten security, and would cause the entire vehicle system may not work properly. + +#### **12.3.1 Requirements** + +SwSec-R1: The vehicular multimedia systems shall be segregated from other vehicular control systems to avoid the failure and system fault of the whole vehicle systems. The vehicular multimedia systems should also provide physical or logical access controls for the isolation of sensitive applications, application data, or systems. + +## **13 Safety** + +### **13.1 Use case – Driving safety: Speed control** + +Before L4 and L5 autonomous vehicles are commercialized, consuming audio and video content in vehicles could easily be a cause for driver distraction. In accordance with the local regulation, when the vehicle speed reaches a certain threshold (it differs based on local regulations, for instance from 0 km/s to 15 km/s per hour), the VMS shall automatically block entertainment content on the + +driver-side display. Speed control is a common practice already in use in countries around the world. + +#### 13.1.1 Requirement + +Saf-R1: In accordance with local regulation, when the vehicle speed is above a given threshold, the VMS shall be able to automatically block entertainment content on the driver-side display. + +## 14 Definition of vehicular multimedia configurations + +As smart connected vehicles are more widely available, vehicular multimedia services are becoming richer and more complex than before. Vehicular multimedia configurations for smart connected vehicles could be used to manufacture and qualify vehicles accordingly. Customers can know which multimedia capabilities his/her vehicle multimedia offers. Examples of VMS reference configurations can be found in Appendix I. + +### 14.1 Use case – Vehicular multimedia configurations + +Currently, when customers choose a vehicle to acquire, the vendor of the vehicle generally provides them with a limited configuration sheet which may not have detailed information on the multimedia capability of the vehicle (e.g., FM/CD/DVD and some other storage disks such as USB disk/SD card/TF card). With the specification of VMS, and the deployment of smart connected vehicle/autonomous driving/ADAS, vehicular multimedia system (VMS) configurations should be defined and include network connectivity, media display, HMI controls, PII protection and security capabilities. + +- 1) Networks connectivity capabilities: +VMS configurations should support one or more of the following: + - 3G/4G/5G cellular networks. + - Satellites bi-directional communication networks, including low earth orbit bi-directional communication networks. + - Satellites broadcast networks. + - Terrestrial broadcast networks. + - Short range networks and/or local area wireless networks. + - V2V and V2I networks. + - Brought-in, built-in and hybrid connectivity. +- 2) Media display and media format capabilities: +VMS configurations should support one or more of: + - Text. + - Image. + - Audio: Normal audio and high-fidelity audio. + - Video: Normal video, 4K/8K video, AR and VR. + - Holographic projection and image. +- 3) HMI control capabilities +VMS configurations should support one or more of the following: + - Button + - Touch screen + - Intelligent control (speech control/gesture control/eye movement control) +- 4) Personally identifying information protection capabilities + +5) Security capabilities. + +#### 14.1.1 Requirements – Vehicular multimedia configurations + +The above five capabilities may be considered as the main aspects to define a set of VMS configurations. + +Moreover, different regions for the above aspects should be taken into consideration. The configurations should allow for geographic flexibilities to adapt to the regulatory requirements of different countries. + +NOTE – Configurations may apply differently to public and private vehicle and to different levels of vehicle autonomous capabilities [SAE J3016]. + +## 15 Matching of described use cases with driving levels + +The table in Figure 16 matches use cases and services that have been defined within the focus group with the SAE driving levels. All activities that are mentioned in the table should be considered for a driving vehicle. When the vehicle parks or stands still, all services are available for the driver as well as for the passengers. Excluded in this case is a standing vehicle at traffic lights or other traffic signs as in this case the vehicle and the driver are still considered as active traffic participants similar to a driving car. + +Within the driving levels, a distinction on what drivers (D) and passengers (P) are allowed to do in the framework of these use cases and services is taken into consideration. + +| Use cases | SAE level 0 | | SAE level 1 | | SAE level 2 | | SAE level 3 | | SAE level 4 | | SAE level 5 | | +|-------------------------------------------------|-------------|---|-------------|---|-------------|---|-------------|---|-------------|---|-------------|---| +| | D | P | D | P | D | P | D | P | D | P | D | P | +| 7.3.1 Multimedia content | | X | | X | | X | (X) | X | X | X | X | X | +| 7.4.1/2 Nomadic device connectivity | X | X | X | X | X | X | X | X | X | X | X | X | +| 7.4.4 Comfort and well-being on board | X | X | X | X | X | X | X | X | X | X | X | X | +| 7.4.5 On-demand functions private vehicle | X | X | X | X | X | X | X | X | X | X | X | X | +| 7.5.1 Continuity of service – Video streaming | | X | | X | | X | (X) | X | X | X | X | X | +| 7.5.2 Cloud based services | | X | | X | | X | (X) | X | X | X | X | X | +| 8.1.2.1/2, Voice recognition session | X | X | X | X | X | X | X | X | X | X | X | X | +| 8.3 Gaze tracking | X | X | X | X | X | X | X | X | X | X | X | X | +| 9.3 Private vehicle-single user | | X | | X | | X | (X) | X | X | X | X | X | +| 9.4 Long distance shared vehicle | | X | | X | | X | (X) | X | X | X | X | X | +| 9.5 Public transport shared vehicle multi users | | X | | X | | X | | X | | X | | X | +| 10.2.1 VMS with flexible EMM | | X | | X | | X | (X) | X | X | X | X | X | +| 10.2.2 VMS with extended EMM | | X | | X | | X | (X) | X | X | X | X | X | +| 11.2 Privilege and account management | (X) | X | (X) | X | (X) | X | (X) | X | X | X | X | X | +| 11.3 Payment services shared vehicles | (X) | X | (X) | X | (X) | X | (X) | X | X | X | X | X | + +X = Allowed + +(X) = Partly allowed (in autonomous driving mode or when it is an audio application only, or according to details that are defined in the respective use-cases). Local or global regulations apply. + +**Figure 16 – Matching of specific multimedia use cases with SAE driving levels** + +# Appendix I + +## Examples of VMS reference configurations + +(This appendix does not form an integral part of this Recommendation.) + +M0-M5 configurations are example capabilities and use cases. + +M0 and M1 are for after-market solutions while M2 to M5 are embedded solutions with increasing capabilities. + +### I.1 M0 configuration + +- Aftermarket VMS. Fitted after buying the car. +- Not connected to vehicle network. +- Brought-in connectivity. Mobile device connected to car through USB. +- Connected to vehicle only for the operating power requirements. + +Table I.1 shows M0 configuration. + +**Table I.1 – M0 configuration** + +| Entertainment capability | Social usage capability | Office use capability | Game play capability | +|------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------------------------------------------------|--------------------------| +| 1. Music playback:
Local music/CD-
AUX-SD-USB
2. Radio: FM-AM | 1. Text social: Support for SMS. Received SMS displayed in VMS display using short-range connections protocol feature, brought-in connectivity
2. Communication social: Only a short-range connection phone is supported, and only one person can be connected to at a time. Restricted by design | Hands-free-phone through the brought-in connectivity | No. Restricted by design | + +### I.2 M1 configuration + +Provides the following capabilities above M0: + +- Provides advance mobile connectivity solutions such as screen mirroring. +- Interactive applications and control sharing between VMS and brought-in connectivity. +- Connected to vehicle for power requirements and game control requirements. + +Table I.2 shows M1 configuration. + +**Table I.2 – M1 configuration** + +| Entertainment capability | Social usage capability | Office use capability | Game play capability | +|---------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------|-----------------------------| +| 1. Music playback: Local music/CD-AUX-SD-USB
2. Radio: FM-AM
3. Video playback: Local video
4. Support mobile connectivity solutions
5. Screen casting. | 1. Text and pictures social: Support for SMS and MMS. Received SMS and MMS displayed in VMS display using short-range connections protocol feature
2. Communication social: Support for multi-way short-range connections and single-way telephone access dial-out, car can be connected at the same time, no need to reconnect | 1. Hands-free-phone through the brought-in connectivity
2. Calendar events sync from mobile phone to VMS (viewable from car) | Interactive gaming | + +### **I.3 M2 configuration** + +- Connected to vehicle network. +- Hybrid connectivity. Vehicle has its own built-in connectivity to access applications and other services also possible to connect mobile phone connectivity through USB or other solutions. +- Connectivity through 2G/3G/4G/5G. +- Service continuity between different use places. +- Voice interaction controls. + +Table I.3 shows M2 configuration. + +**Table I.3 – M2 configuration** + +| Entertainment capability | Social usage capability | Office use capability | Game play capability | +|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-----------------------------| +| 1. Music playback: Local music/CD-AUX-SD-USB
2. Radio: FM-AM/Online Internet radio
3. Video playback: Local video
4. Support for mobile connectivity solutions
5. Screen casting | 1. Text, pictures and voice social: Support for SMS, MMS and social media connectivity
2. Support for multi-channel short-range connections and headphone bone conduction calls (single person support) | 1. Support multi-person car conference call
2. Calendar events sync from phone to car (viewable from car)
3. Nomadic devices connectivity.
4. Continuity in service between different use places viz home/office/car using local area wireless | Interactive gaming | + +**Table I.3 – M2 configuration** + +| Entertainment capability | Social usage capability | Office use capability | Game play capability | +|--------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|----------------------------------|----------------------| +| 6. Voice interaction:
Specific instruction
voice interaction | 3. Support for simple
team travel: location
real-time sharing
4. Continuity in service
between different use
places viz
home/office/car using
local area wireless
connections, APP
connectivity | connections, APP
connectivity | | + +### I.4 M3 configuration + +- Connected to vehicle network. +- Hybrid connectivity. Vehicle has its own built-in connectivity to access applications and other services possible to connect mobile phone connectivity through short-range connections or other solutions. +- Single display. +- Continuity in service between different use places viz home/office/car using local area wireless networks, application connectivity. +- Advance voice control. +- Shared mobility applications capabilities. +- Build-in game applications capabilities. +- Basic security capabilities. + +Table I.4 shows M3 configuration. + +**Table I.4 – M3 configuration** + +| Entertainment capability | Social usage capability | Office use capability | Game play capability | +|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| 1. Music playback:
Local music/CD-
AUX-SD-USB
2. Radio: FM-
AM/DAB/Online
Internet radio
3. Video playback:
Local video
4. Support mobile
connectivity solutions
5. Voice interaction:
Smart voice
interaction through
Internet services | 1. In-car social support
is in the following
formats: Text,
pictures, voices, and
small video social
2. Communication
social:
Support for short-
range connections
phones and seat bone
conduction calls
(headphone-free,
single-way calls) /
headphone bone
conduction calls | 1. Support for on-board
video conference
calls (support for live
screen; no support
for file picture
sharing), support for
voice memos, voice
memos to text and
memos to share to
mail
2. Two-way
synchronization of
mobile phones and
car for calendar
events (not only can | 1. Mini-games
(question-and-
answer games based
on voice interaction)
2. The on-board
version of the online
competitive game is
supported when
parking;
3. Support for action
adventure game
machine version
when parking | + +**Table I.4 – M3 configuration** + +| Entertainment capability | Social usage capability | Office use capability | Game play capability | +|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|----------------------| +| 6. AR enhanced display APP on Centre Screen
7. Simple multi-screen interaction: Display entertainment function information on the instrument screen/HUD/air conditioning control screen, which can be easily interacted by protocol
8. Anthropomorphic interactions: Emotional interaction robots (e.g. NOMI, Small-scale car robots) | (supporting single person)
3. Support complex group travel: positioning and navigation route sharing | be seen on board, but can also be edited)
3. Support event trigger navigation
4. The computer can use the computer local area wireless networks, the computer edited mail can be sent through the car network | | + +### I.5 M4 configuration + +- Connected to vehicle network. +- Hybrid connectivity. Vehicle has its own built-in connectivity to access applications and other services possible to connect mobile phone connectivity through short-range connections or other solutions. +- Higher display capability. +- Multi-screen interaction capability. +- Continuity in service between different use places viz home/office/car using local area wireless networks, application connectivity. +- Occupant health and comfort monitoring capability. +- Advance HMI capability. +- Advanced integrated control capability for game play. +- Shared mobility and connected cars applications capabilities. +- PII protection and security capability. + +Table I.5 shows M4 configuration. + +**Table I.5 – M4 configuration** + +| Entertainment functions | Social use functions | Office use functions | Game play functions | +|----------------------------------------------|--------------------------------------------------------------|------------------------------------------------------------------|-------------------------------------------------------------| +| 1. Music playback: Local music/CD-AUX-SD-USB | 1. Pre-install edgy social software for the car, and the in- | 1. Support for on-board video conference calls (support for live | 1. Co-driver or main driver parking can support small games | + +**Table I.5 – M4 configuration** + +| Entertainment functions | Social use functions | Office use functions | Game play functions | +|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +|

2. Radio: FM-AM/DAB/SDARS/Online Internet radio

3. Video playback: Local video

4. Support mobile connectivity solutions

5. Voice interaction: Smart voice interaction to Get Through the Internet Service Ecology

6. Simple multi-screen interaction: Display entertainment function information on the instrument screen/HUD/air conditioning control screen, which can be easily interacted by protocol

7. Multimedia entertainment information system platform and rich content resources with the fusion of multi-satellite broadcasting network and terrestrial cellular networks

8. AR Enhanced display APP on HUD and Centre Screen

9. Holographic projection display in specific areas of the vehicle, but only for display and not for interaction

|

car social support is in the following formats: Text, pictures, voice, small video social

2. Communication social: Support for short-range connections phones and seat bone conduction calls (headphone-free, dual-way calls supported) and headphone bone conduction calls (single-supporting)

3. Support complex team travel: positioning, navigation route sharing, support mobile bus route planning sharing

|

screen, no support for file picture sharing), support for voice memos, voice memos to text and memos shared to mail/social software

Back support for hand-written board recording MEMO

2. Calendar event mobile phone, car two-way synchronization (the car can not only be seen, but also edited)

3. The rear screen of the car system supports checking and editing mail;

4. Support handwriting/review/forward mail when parking

|

based on gesture recognition, such as cutting fruit

2. Racing games based on the steering wheel when parking

3. Mini games (question and answer games based on voice interaction)

4. Parking support: on-board version of online competitive games

| + +### I.6 M5 configuration + +- Connected to vehicle network. +- Hybrid connectivity. Vehicle has its own built-in connectivity to access applications and other services also possible to connect mobile phone connectivity through short-range connections or other solutions. + +- Satellite media reception capability. +- LCD screen support for hetero-shaped screen/OLED flexible screen/all-in-one screen. +- Advanced multi display interaction and HMI capabilities such as gesture recognition, voice capability. +- Continuity in service between different use places viz home/office/car using local area wireless networks, application connectivity. +- Advanced Audio, video formats playing capabilities. +- Advanced connected car, shared mobility features. +- PII protection and security capability (such as driver monitoring cameras). + +Table I.6 shows M5 configuration. + +**Table I.6 – M5 configuration** + +| Entertainment functions | Social use functions | Office use functions | Game play functions | +|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| 1. Music playback: Local music/CD-AUX-SD-USB
2. Radio: FM-AM/Online Internet radio
3. Video playback: Local video
4. Support mobile connectivity solutions
5. Voice interaction: Smart voice interaction to Get Through the Internet Service Ecology
6. Simple multi-screen interaction: Display entertainment function information on the instrument screen/HUD/air conditioning control screen, which can be easily interacted by protocol
7. Multimedia entertainment information system platform and rich content resources with the fusion of satellite broadcasting networks multi-satellite and terrestrial cellular networks
8. AR enhanced display can be applied on any screen, including HUD | 1. Pre-installed customized social software for the car and in-car social support is in the following formats: text, pictures, voices, small video social and Live social, interactive games social
2. Communication social: Support for short-range connections phones (headphone-free, multiple-way calls, unheard-of content between different passengers) and headphone conduction calls (single support)
3. Support complex team travel: positioning, navigation screen sharing, support mobile bus route planning sharing, route 3D visualization | 1. Support for on-board video conferencing (supports real-time screens, does not support file picture sharing), voice memos, voice memos, text and voice memos, and memos shared to mail/social software, Support for tablet recording memo
2. Calendar event phone, two-way synchronization (car cannot only see, but also can be edited)
3. The rear windshield supports the inspection and editing of mail
4. Support for handwriting/review/forwarding messages when parking
5. Front row supports on-board custom office software operations based on holographic projection/gesture recognition/eye | 1. Co-driver or main driver parking can support small games based on gesture recognition, such as cutting fruit
2. Racing games based on the steering wheel when parking
3. Mini games (question and answer games based on voice interaction)
4. Parking support: on-board version of online competitive games;
5. Games that support holographic projection and holographic interaction | + +**Table I.6 – M5 configuration** + +| Entertainment functions | Social use functions | Office use functions | Game play functions | +|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-----------------------------|-----------------------------------------------|----------------------------| +|

and Centre Screen

9. Customized holographic projection, users can display the contents of the screen in multiple areas of the car, not only to support display, but also to support interaction, similar to virtual assistant

10. Driver video recording, Driver Monitoring System, Around view Monitoring Cameras

11. Gesture Recognition capability

12. Voice Biometric identity

| |

control (in conjunction with hardware)

| | + +## Bibliography + +- [b-ITU-T E.164] Recommendation ITU-T E.164 (2010), *The international public telecommunication numbering plan*. +- [b-ITU-T F.749.1] Recommendation ITU-T F.749.1 (2015), *Functional requirements for vehicle gateways*. +- [b-ITU-T F.749.2] Recommendation ITU-T F.749.2 (2017), *Service requirements for vehicle gateway platforms*. +- [b-ITU-T H.550] Recommendation ITU-T H.550 (2017), *Architecture and functional entities of vehicle gateway platforms*. +- [b-ITU-T H.560] Recommendation ITU-T H.560 (2017), *Communications interface between external applications and a vehicle gateway platform*. +- [b-ITU-T P.1150] Recommendation ITU-T P.1150 (2020), *In-car communication audio specification*. +- [b-ITU-T P.11.x] Recommendation ITU-T P.1100.x-series, *Communications involving vehicles*. +- [b-ITU-R BS.2384] Report ITU-R BS.2384-1 (2019), *Implementation considerations for the introduction and transition to digital terrestrial sound and multimedia broadcasting*. . +- [b-ITU-R BT.2052] Recommendation ITU-R BT-2052-1 (2015), *Planning criteria for terrestrial multimedia broadcasting for mobile reception using handheld receivers in VHF/UHF bands*. . +- [b-3GPP TR38.885] 3GPP TR38.885, *Study on 5G New Radio Vehicle-to-Everything (V2X)*. +- [b-3GPP TS36.321] 3GPP TS36.321, *Evolved Universal Terrestrial Radio Access (E-UTRA); Medium Access Control (MAC) protocol specification*. +- [b-ATSC] ATSC [Online]. Available: . +- [b-ChinaDRM] ChinaDRM [Online]. Available: . +- [b-CN-CAS] CN-CAS [Online]. Available: . +- [b-DoT PII-1] DoT PII [Online]. Available: [https://www.its.dot.gov/factsheets/pdf/Privacy\\_factsheet.pdf](https://www.its.dot.gov/factsheets/pdf/Privacy_factsheet.pdf). +- [b-DoT PII-2] GDPR (2018), *Reform of EU data protection rules* [Online]. Available: [https://ec.europa.eu/commission/priorities/justice-and-fundamental-rights/data-protection/2018-reform-eu-data-protection-rules\\_en](https://ec.europa.eu/commission/priorities/justice-and-fundamental-rights/data-protection/2018-reform-eu-data-protection-rules_en). +- [b-ETSI ETR 289] ETSI ETR 289 ed.1 (1996), *Digital Video Broadcasting (DVB); Support for use of scrambling ad Conditional Access (CA) within digital broadcasting systems*. [Online]. Available: [https://www.etsi.org/deliver/etsi\\_etr/200\\_299/289/01\\_60/etr\\_289e01p.pdf](https://www.etsi.org/deliver/etsi_etr/200_299/289/01_60/etr_289e01p.pdf). +- [b-IEEE 802.11p] IEEE 802.11p(2010) *IEEE Standard for Information technology—Local and metropolitan area networks—Specific requirements—Part 11: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) Specifications Amendment 6: Wireless Access in Vehicular Environments*. + +- [b-IETF e-mail] IETF e-mail [Online]. Available: . +- [b-ISO 15638-1] ISO 15638-1 (2012), *Intelligent transport systems – Framework for collaborative Telematics Applications for Regulated commercial freight Vehicles (TARV) – Part 1: Framework and architecture*. +- [b-ISO 23239] ISO 23239-1.2, *Road vehicles – Vehicle domain service (VDS) – Part 1: General information and use case definitions* [Draft; Online]. Available: . +- [b-OMA DRM] OMA DRM [Online]. Available: [http://www.openmobilealliance.org/release/DRM/V2\\_2-20110419-C/](http://www.openmobilealliance.org/release/DRM/V2_2-20110419-C/). +- [b-ITU-T Q27/16] ITU-T SG16 Question 27/16 (2016-2020), + + + + + +# SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.749.4-202106-I_PDF-E/16152cf1d84aea10848758f51a91ff6a_img.jpg b/marked/F/T-REC-F.749.4-202106-I_PDF-E/16152cf1d84aea10848758f51a91ff6a_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ced482d2e367a297ffeca4f904014ee3be919695 --- /dev/null +++ b/marked/F/T-REC-F.749.4-202106-I_PDF-E/16152cf1d84aea10848758f51a91ff6a_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7403ea9c473da58ef78bf0fefb171900caacfc1210369e9d829327afb582e790 +size 30212 diff --git a/marked/F/T-REC-F.749.4-202106-I_PDF-E/16c1175b5f05a4b55e6d396fc51b15b3_img.jpg b/marked/F/T-REC-F.749.4-202106-I_PDF-E/16c1175b5f05a4b55e6d396fc51b15b3_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a11da4cd24968190d9a777b15140a337be690b28 --- /dev/null +++ b/marked/F/T-REC-F.749.4-202106-I_PDF-E/16c1175b5f05a4b55e6d396fc51b15b3_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e31db2fc12923d0ab285973c11e734792116c668114d38ec88e7f0e44597f71f +size 62221 diff --git a/marked/F/T-REC-F.749.4-202106-I_PDF-E/54fabc351eda5228d2fa28cd9ba07971_img.jpg b/marked/F/T-REC-F.749.4-202106-I_PDF-E/54fabc351eda5228d2fa28cd9ba07971_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..2fcdc284f51f72d5c03c008705babbe0a70e3f10 --- /dev/null +++ b/marked/F/T-REC-F.749.4-202106-I_PDF-E/54fabc351eda5228d2fa28cd9ba07971_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc6456612811ce662b37e66777e49edaf35a78fb3a38106037b104be77a464a1 +size 87875 diff --git a/marked/F/T-REC-F.749.4-202106-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.749.4-202106-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ec4bb3f982e1ff763068a22452738a8eebe99ed5 --- /dev/null +++ b/marked/F/T-REC-F.749.4-202106-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:581d36ac6e165178318a73e7c9b189ab80b8fef8b136e2f95c74567800409ce5 +size 6884 diff --git a/marked/F/T-REC-F.749.4-202106-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.749.4-202106-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..2b00bd3efcf0cec21b50e5d8fc8c6ec952a39b63 --- /dev/null +++ b/marked/F/T-REC-F.749.4-202106-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1b496d1649aeac4cec3fd7cc3ecd5e484a2d00f80616037cbbe932d29398f17e +size 103997 diff --git a/marked/F/T-REC-F.749.4-202106-I_PDF-E/raw.md b/marked/F/T-REC-F.749.4-202106-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..7fe499b552d8c204e7bffe21f87574628782c0ea --- /dev/null +++ b/marked/F/T-REC-F.749.4-202106-I_PDF-E/raw.md @@ -0,0 +1,506 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.749.4** + +(06/2021) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Use cases and requirements for multimedia +communication enabled vehicle systems using +artificial intelligence** + +Recommendation ITU-T F.749.4 + +![ITU logo](84a1d09fb489061482111515543b60dc_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with a white grid pattern, overlaid by a stylized 'ITU' text in blue. + +ITU logo + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.749.4 + +# Use cases and requirements for multimedia communication enabled vehicle systems using artificial intelligence + +## Summary + +Recommendation ITU-T F.749.4 describes use cases and scenarios, high-layer architecture, service and network requirements, functional requirements and non-functional requirements for multimedia communication enabled vehicle systems using artificial intelligence. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.749.4 | 2021-06-13 | 16 | 11.1002/1000/14683 | + +## Keywords + +ADAS, artificial intelligence, autonomous vehicles, ITS, requirements, use cases. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2021 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|----------------------------------------------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Overview ..... | 2 | +| 6.1 Environment perception scenario ..... | 3 | +| 6.2 Prediction scenario ..... | 3 | +| 6.3 Planning scenario..... | 4 | +| 6.4 Human-machine interaction scenario ..... | 4 | +| 6.5 Model training/deployment scenario ..... | 4 | +| 7 High-layer architecture ..... | 4 | +| 8 Requirements ..... | 5 | +| 8.1 Services and network requirements ..... | 5 | +| 8.2 Functional requirements ..... | 6 | +| 8.3 Non-functional requirements ..... | 7 | +| Appendix I – Use cases of multimedia communication enabled vehicle systems using artificial intelligence ..... | 8 | +| I.1 Use cases ..... | 8 | +| Bibliography..... | 12 | + + + +# Recommendation ITU-T F.749.4 + +# Use cases and requirements for multimedia communication enabled vehicle systems using artificial intelligence + +# 1 Scope + +This Recommendation specifies use cases and requirements for multimedia communication enabled vehicle systems using artificial intelligence, including overview, use cases, high-layer architecture, service and network requirements, functional requirements, and non-functional requirements. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +- [ITU-T X.805] Recommendation ITU-T X.805 (2003), *Security architecture for systems providing end-to-end communications*. +- [ITU-T X.1111] Recommendation ITU-T X.1111 (2007), *Framework of security technologies for home network*. +- [ITU-T X.1371] Recommendation ITU-T X.1371 (2020), *Security threats to connected vehicles*. +- [ITU-T X.1372] Recommendation ITU-T X.1372 (2020), *Security guidelines for vehicle-to-everything (V2X) communication*. +- [ITU-T X.1373] Recommendation ITU-T X.1373 (2017), *Secure software update capability for intelligent transportation system communication devices*. +- [ITU-T X.1811] Recommendation ITU-T X.1811 (2021), *Security guidelines for applying quantum-safe algorithms in IMT-2020 systems*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +- 3.1.1 application** [b-ITU-T Y.2091]: A structured set of capabilities, which provide value-added functionality supported by one or more services, which may be supported by an API interface. +- 3.1.2 service** [b-ITU-T Y.2091]: A set of functions and facilities offered to a user by a provider. +- 3.1.3 intelligent transport systems (ITS)** [b-ITU-R handbook]: ITS can be defined as systems utilizing the combination of computers, communications, positioning and automation technologies to improve the safety, management and efficiency of terrestrial transport systems. +- 3.1.4 artificial intelligence (AI)** [b-ISO/IEC 2382]: An interdisciplinary field, usually regarded as a branch of computer science, dealing with models and systems for the performance of functions generally associated with human intelligence, such as reasoning and learning. + +**3.1.5 deep learning** [b-ISO/IEC TR 29119-11]: Approach to creating rich hierarchical representations through the training of neural networks with one or more hidden layers. + +## **3.2 Terms defined in this Recommendation** + +This Recommendation defines the following term: + +**3.2.1 advanced driver-assistance systems (ADAS)**: ADAS are systems to help the driver in the driving process. When designed with a safe human-machine interface, they should increase car safety and more generally road safety. + +# **4 Abbreviations and acronyms** + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-------|------------------------------------| +| ADAS | Advanced Driving Assistance System | +| AI | Artificial Intelligence | +| C-V2X | Cellular-V2X | +| HUD | Head Up Display | +| Lidar | Light detection and ranging | +| RSU | Roadside Unit | +| WWAN | Wireless Wide Area Network | + +# **5 Conventions** + +In this Recommendation: + +- The keywords "shall" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "should" indicate an optional requirement which is permissible. This term is not intended to imply that the vendor's implementation must provide the option, and the feature can be optionally enabled by the vendor. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +# **6 Overview** + +Artificial intelligence technology can be applied to many scenarios of multimedia communication enabled vehicle systems, such as lane keep assistance, lane departure warning system, traffic sign recognition, driver drowsiness detection, blind zone detection, obstacle detection, route planning, etc. + +Figure 1 shows scenarios of multimedia communication enabled vehicle systems using artificial intelligence. + +![Diagram illustrating scenarios of multimedia communication enabled vehicle systems using artificial intelligence. The diagram shows three main components: Cloud server, Vehicle, and Infrastructure/RSU/vehicle, connected by WWAN and Short-range communication. The Cloud server contains Data storage and processing and Model training. The Vehicle contains Camera, Sensors, Lidar, Radar, Environment perception, Prediction, Planning, Model deployment, and Human-machine interaction. The Infrastructure/RSU/vehicle contains Assisted environment perception. External boxes show Traffic light/traffic sign and Pedestrian/non-motor vehicle/vehicle.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +The diagram illustrates the architecture of a multimedia communication enabled vehicle system. At the top, a **Cloud server** box contains two dashed sub-boxes: **Data storage and processing** and **Model training**. Below it, a **Vehicle** box contains several components: a vertical stack of **Camera**, **Sensors**, **Lidar**, and **Radar**; three dashed boxes for **Environment perception**, **Prediction**, and **Planning**; a dashed box for **Model deployment**; and a dashed box for **Human-machine interaction** next to a car icon. At the bottom, an **Infrastructure/RSU/vehicle** box contains a dashed box for **Assisted environment perception** next to a road sign icon and a car icon. Connections include a double-headed arrow labeled **WWAN** between the Cloud server and the Vehicle, and a double-headed arrow labeled **Short-range communication** between the Vehicle and the Infrastructure/RSU/vehicle. To the left, two external boxes are shown: **Traffic light/traffic sign** (with a traffic light, road intersection, and sign icons) and **Pedestrian/non-motor vehicle/vehicle** (with icons of a pedestrian, bicycle, motorcycle, and car). Arrows point from these external boxes towards the Vehicle box. + +Diagram illustrating scenarios of multimedia communication enabled vehicle systems using artificial intelligence. The diagram shows three main components: Cloud server, Vehicle, and Infrastructure/RSU/vehicle, connected by WWAN and Short-range communication. The Cloud server contains Data storage and processing and Model training. The Vehicle contains Camera, Sensors, Lidar, Radar, Environment perception, Prediction, Planning, Model deployment, and Human-machine interaction. The Infrastructure/RSU/vehicle contains Assisted environment perception. External boxes show Traffic light/traffic sign and Pedestrian/non-motor vehicle/vehicle. + +F.749.4(21)\_F01 + +**Figure 1 – Scenarios of multimedia communication enabled vehicle systems using artificial intelligence** + +There are five major scenarios. Environment perception scenario, prediction scenario, planning scenario and human-machine interaction scenario describe the application scenarios inside/outside the vehicle. Model training/deployment scenario describes the common functional scenario which can be used in all other scenarios. + +## 6.1 Environment perception scenario + +The environment perception scenarios mainly include the scenarios that use cameras, radars/light detection and ranging (Lidar) and sensors in vehicles for obtaining data from the vehicle driving environment, and then use machine learning models for data analysis. The data should include lane information, traffic sign information, traffic light information, pedestrian information, bicycle information, other vehicles information, location information, etc. Environment perception scenarios should also include the scenarios that use V2X communication technology to exchange assisted environment perception information (such as the speed/location of bicycle/other vehicles, blind zone information, etc.) and use machine learning models to analyse the data for driving environment perception. + +## 6.2 Prediction scenario + +The prediction scenarios mainly include the scenarios that use machine learning models to predict the action of moving objects to calculate the route of the objects. The prediction function shall achieve high accuracy and low latency for safety. Prediction should also support the function of continuous acquisition of new knowledge based on driving data to improve prediction accuracy using a machine learning model. + +## 6.3 Planning scenario + +The planning scenarios mainly include the scenarios that plan the proper trajectory of the surrounding vehicle in various scenes (such as turning, obstacle avoidance, overtaking, etc.) according to the results of environment perception and surrounding objects moving trajectory prediction using a machine learning model. + +## 6.4 Human-machine interaction scenario + +The human-machine interaction scenarios mainly include the scenarios that achieve intelligent cockpit (such as intelligent central screen, head up display (HUD), rear mirror and telematics unit, etc.) and analyse image, video, speech and text information for gesture recognition, voice recognition/nature language processing, face recognition, driver monitoring, etc. using a machine learning model. + +## 6.5 Model training/deployment scenario + +The artificial intelligence model training/deployment scenarios mainly include the scenarios that continuously train and optimize the models in the cloud servers using machine learning algorithms based on the data that is collected from vehicles and roadside units (RSUs). Once tested, the models should be deployed to the vehicles from cloud servers to continuously improve the advanced driving assistance system (ADAS)/autonomous driving capability. + +# 7 High-layer architecture + +Figure 2 presents the high-layer architecture for multimedia communication enabled vehicle systems using artificial intelligence according to the use cases. + +![Figure 2: High-layer architecture for multimedia communication enabled vehicle systems using artificial intelligence. The diagram shows a layered structure within a 'Vehicle system' box. The top layer is 'Services' containing Environment perception, Prediction, Planning, and Human-machine interaction. The middle layer is 'Functionalities' containing Data collection, Model training/deployment, Policy management, and Security and privacy. The bottom layer is 'Networks' containing Wireless wide area network and Wireless short-range communication networks. Double-headed arrows connect Services to Functionalities and Functionalities to Networks. Below the 'Vehicle system' box, a 'Cloud server' and 'RSUs/vehicles/pedestrians/bicycles' are shown, connected to the Networks layer by zigzag lines.](54fabc351eda5228d2fa28cd9ba07971_img.jpg) + +The diagram illustrates the high-layer architecture for multimedia communication enabled vehicle systems using artificial intelligence. It is structured as follows: + +- Vehicle system** (enclosed in a dashed box): + - Services** (top layer): Environment perception service, Prediction service, Planning service, Human-machine interaction service. + - Functionalities** (middle layer): Data collection and analysis, Model training/deployment, Policy management, Security and privacy. + - Networks** (bottom layer): Wireless wide area network, Wireless short-range communication networks.Vertical double-headed arrows connect the Services layer to the Functionalities layer, and the Functionalities layer to the Networks layer. +- Below the Vehicle system box: + - Cloud server** and **RSUs/vehicles/pedestrians/bicycles** are shown. + - Zigzag lines connect the Cloud server and RSUs/vehicles/pedestrians/bicycles to the Networks layer of the Vehicle system. + +Figure 2: High-layer architecture for multimedia communication enabled vehicle systems using artificial intelligence. The diagram shows a layered structure within a 'Vehicle system' box. The top layer is 'Services' containing Environment perception, Prediction, Planning, and Human-machine interaction. The middle layer is 'Functionalities' containing Data collection, Model training/deployment, Policy management, and Security and privacy. The bottom layer is 'Networks' containing Wireless wide area network and Wireless short-range communication networks. Double-headed arrows connect Services to Functionalities and Functionalities to Networks. Below the 'Vehicle system' box, a 'Cloud server' and 'RSUs/vehicles/pedestrians/bicycles' are shown, connected to the Networks layer by zigzag lines. + +F.749.4(21)\_F02 + +**Figure 2 – High-layer architecture for multimedia communication enabled vehicle systems using artificial intelligence** + +The high-layer architecture for multimedia communication enabled vehicle systems using artificial intelligence covers three layers: services, functionalities and networks. A general description of these three blocks is presented below. + +- **Services:** Services are the software that implement the artificial intelligence-related automotive multimedia communication applications. Services include the environment perception service, the prediction service, the planning service, and the human-machine interaction service. +- **Functionalities:** Functionalities are the basic functional modules for supporting services and applications. Functionalities include data collection and analysis, model training/deployment, policy management, and security and privacy. +- **Networks:** Networks are the communication networks for data exchange between vehicle system and cloud server/RSUs/vehicles/pedestrians/bicycles. Networks include wireless wide area networks and wireless short-range communication networks. + +# 8 Requirements + +## 8.1 Services and network requirements + +### 8.1.1 Environment perception service requirements + +The artificial intelligence requirements for the environment perception service are as follows: + +- The vehicle system shall support one or more environment perception methods (e.g., camera, lidar, radar, etc.) to capture the driving environment data. +- The vehicle system shall support machine learning algorithms/models to process driving environment data. +- The vehicle system shall support object recognition, object detection and semantic segmentation of the driving environment using machine learning algorithms/models in different weather conditions. +- The vehicle system shall support recognition, detection and semantic segmentation of the driving environment objects using machine learning algorithms/models with low latency and high accuracy. +- The vehicle system should upload driving environment data to the cloud server. +- The vehicle system shall obtain driving environment information from other vehicles, RSUs and pedestrians/bicycles through wireless short-range communication networks. +- The vehicle system shall support high precision positioning according to the high-definition map. + +### 8.1.2 Prediction service requirements + +The artificial intelligence requirements for prediction service are as follows: + +- The vehicle system shall support the prediction of the trajectory using multiple environment data that are captured through camera, lidar, radar, etc. +- The vehicle system shall support the prediction of the trajectory of pedestrians, other vehicles and obstacles using machine learning algorithms/models with low latency and high accuracy. +- The vehicle system shall obtain motion information from other vehicles, RSUs and pedestrians/bicycles through wireless short-range communication networks. + +### 8.1.3 Planning service requirements + +The artificial intelligence requirements for the planning service are as follows: + +- The vehicle system shall support the planning of a moving path and moving speed using machine learning algorithms/models with low latency and high accuracy. +- The vehicle system shall support safety/smoothness/comfort of moving path/speed planning. + +### **8.1.4 Human-machine interaction service requirements** + +The artificial intelligence requirements for human-machine interaction service are as follows: + +- The vehicle system shall support speech recognition according to the voice of the driver/passengers. +- The vehicle system should support voiceprint recognition according to the voice of the driver/passengers. +- The vehicle system shall support natural language understanding based on the speech recognition. +- The vehicle system should support computer vision applications (e.g., face recognition, eyeball tracking, gesture recognition, etc.) using machine learning algorithms/models. +- The vehicle system should support speech recognition, voiceprint recognition, natural language understanding and computer vision applications through a cloud server. + +### **8.1.5 Network requirements** + +The artificial intelligence requirements for the network are as follows: + +- The vehicle system should support wireless wide area networks (WWANs) with high bandwidth and low latency (e.g., IMT-2020). +- The vehicle system should support wireless short-range communication networks (e.g., IEEE 802.11p or cellular-V2X (C-V2X)). + +## **8.2 Functional requirements** + +### **8.2.1 Data collection and analysis requirements** + +The requirements for data collection and analysis are as follows: + +- The vehicle system shall gather environment perception data, vehicle running data and human-machine interaction data. +- The vehicle system should provide data analysis functionality in order to produce intermediate information for further analysis. +- The vehicle system shall transmit necessary data to the cloud server through WWAN. + +### **8.2.2 Model training/deployment service requirements** + +The artificial intelligence requirements for the model training/deployment service are as follows: + +- The vehicle system should support multiple machine learning algorithms/models and deep learning frameworks. +- The vehicle system should support local artificial intelligence (AI) model training based on cooperation with other vehicles or RSUs. +- The vehicle system shall deploy/update the AI models from the cloud server through WWAN. + +### **8.2.3 Policy management** + +Policies shall be used to constrain the operation of the vehicle. The policies may be either a set of commands or recommendations, such as traffic laws or moral rules. + +The requirements for policy management are as follows: + +- The vehicle system shall be constrained by predefined policies when a decision is being made. +- The vehicle system shall use policies that are triggered by one or more events. +- When an event indicates that a policy violation is detected, the vehicle system shall perform appropriate actions. +- The vehicle system shall perform policy conflict resolution. +- The vehicle system shall update the latest policies timely. + +### **8.2.4 Security and privacy requirements** + +This clause lists the requirements related to security and privacy issues. For more details, see [ITU-T X.1371], [ITU-T X.1372], [ITU-T X.1373], [ITU-T X.805], [ITU-T X.1111] and [ITU-T X.1811]. The requirements for security and privacy are as follows: + +- The vehicle system and the cloud server shall provide secure data storage. +- The vehicle system and the cloud server shall provide secure data transmission through WWAN. +- The vehicle system shall provide secure data transmission between vehicles and RSUs. +- The vehicle system and the cloud server shall provide secure models deployment. +- The vehicle system and the cloud server shall provide privacy protection for the personal data. + +## **8.3 Non-functional requirements** + +### **8.3.1 Performance requirements** + +The performance requirements are as follows: + +- The vehicle system shall identify traffic signs, obstacles and status of surrounding vehicles/pedestrians/bicycles, etc. quickly and accurately under various natural conditions, and give alarms or perform safe and compliant actions. +- The vehicles system shall identify surrounding vehicles, pedestrians, bicycles, etc. quickly and accurately under various natural conditions, give alarms or plan appropriate paths, and pass various types of traffic intersections safely and compliantly. + +### **8.3.2 Regulatory requirements** + +The regulatory requirements are as follows: + +- The privacy of the users shall be properly protected during the collection, storage and analysis of the data in the vehicle's systems and the cloud server. +- It shall be possible to analyse the vehicle data without exposing its users. +- The vehicle system and the cloud server shall comply with the data protection regulations. + +# Appendix I + +## Use cases of multimedia communication enabled vehicle systems using artificial intelligence + +(This appendix does not form an integral part of this Recommendation.) + +## I.1 Use cases + +### I.1.1 Environment perception use cases + +#### I.1.1.1 Overview + +The environment perception mainly includes visual perception related scenarios. First, cameras, driving recorders and vehicle sensors are used to collect information from the surrounding environment. Then, the machine learning model is used to process the collected information. Finally, the vehicle's driving trajectory is determined to ensure the vehicle safety and efficient arrival at the destination. The specific environment perception scenario is shown in Figure I.1. + +![Figure I.1 – Scenario of environment perception. The diagram illustrates a vehicle's perception of its environment. A central vehicle icon is shown with arrows pointing to various environmental elements: 'Infrastructure/RSU' (represented by a tower icon), 'Traffic light/zebra crossing/traffic sign' (represented by a traffic light, a zebra crossing image, and a directional sign), and 'Pedestrian/non-motor vehicle/vehicle' (represented by icons of a person, a bicycle, a motorcycle, and a car). Dashed lines with arrows indicate 'Short-range communication' between the vehicle and the infrastructure/RSU, and between the vehicle and the other road users. Solid arrows labeled 'Perception' point from the vehicle to each of these elements. The diagram is labeled F.749.4(21)_FI.1 in the bottom right corner.](16c1175b5f05a4b55e6d396fc51b15b3_img.jpg) + +The diagram illustrates a vehicle's perception of its environment. A central vehicle icon is shown with arrows pointing to various environmental elements: 'Infrastructure/RSU' (represented by a tower icon), 'Traffic light/zebra crossing/traffic sign' (represented by a traffic light, a zebra crossing image, and a directional sign), and 'Pedestrian/non-motor vehicle/vehicle' (represented by icons of a person, a bicycle, a motorcycle, and a car). Dashed lines with arrows indicate 'Short-range communication' between the vehicle and the infrastructure/RSU, and between the vehicle and the other road users. Solid arrows labeled 'Perception' point from the vehicle to each of these elements. The diagram is labeled F.749.4(21)\_FI.1 in the bottom right corner. + +Figure I.1 – Scenario of environment perception. The diagram illustrates a vehicle's perception of its environment. A central vehicle icon is shown with arrows pointing to various environmental elements: 'Infrastructure/RSU' (represented by a tower icon), 'Traffic light/zebra crossing/traffic sign' (represented by a traffic light, a zebra crossing image, and a directional sign), and 'Pedestrian/non-motor vehicle/vehicle' (represented by icons of a person, a bicycle, a motorcycle, and a car). Dashed lines with arrows indicate 'Short-range communication' between the vehicle and the infrastructure/RSU, and between the vehicle and the other road users. Solid arrows labeled 'Perception' point from the vehicle to each of these elements. The diagram is labeled F.749.4(21)\_FI.1 in the bottom right corner. + +Figure I.1 – Scenario of environment perception + +#### I.1.1.2 Automatic identification of traffic environment scenario + +The vehicles mainly identify driving behaviours in traffic environment scenarios using computer vision technology. The main function of visual perception is that the vehicles use cameras, lidar/radars, driving recorders and other equipment to capture the surrounding environment of the vehicle from different angles. It identifies traffic conditions (such as lane conditions and traffic signs) according to the obtained images or video information and makes correct surrounding environment detection using a machine learning model (such as lane detection, identification of traffic signs and markings, identification of traffic lights, identification of driving status of vehicles ahead, obstacle identification, pedestrian and non-motorized vehicle identification, etc.). In the intelligent driving mode, the vehicles can recognize the surrounding environment according to the image/video stream that is captured by the cameras/lidar/radars in the vehicle, so that the vehicle can achieve lane keeping, car following, pull over, overtaking, merging, intersection passing, roundabout intersection passing, automatic emergency braking and other operations. + +#### **I.1.1.3 Obstacle detection and classification scenario** + +In the obstacle recognition and classification scenarios, there are mainly static obstacle and dynamic obstacle recognition/classification. Static obstacles include walls, trees, utility poles and buildings, etc., and the dynamic obstacles include pedestrians, non-motor vehicles, automobiles, etc. In this scenario, the vehicle needs to identify/classify the location of the obstacles. For example, the vehicle can determine the path and speed of its move based on the type of object being sensed: if the vehicle senses that there is a bicycle ahead, the vehicle can slow down or change lanes to safely drive past the bicycle. If the vehicle detects that there is another vehicle ahead (target vehicle), and predicts that the target vehicle in front will also travel at a speed close to the speed limit, then the vehicle can maintain its speed, distance and lane. In the traffic signal detection and classification scenario, the vehicle can locate the traffic lights in the captured images, and then classify the traffic lights according to the colour of the light display. This is based on computer vision technologies, and uses deep learning algorithms (such as convolutional neural networks and its variant algorithms) to detect/classify traffic lights or other obstacles. + +#### **I.1.1.4 Assisted environment perception scenario** + +This scenario includes vehicle-to-vehicle communication and vehicle-to-RSU communication scenarios. The vehicles and RSUs can transmit real-time information such as vehicle speed, position, braking information, driving direction and surrounding traffic conditions (such as blind spot detection information that is detected using machine learning technology) to nearby vehicles through wireless short-range communication networks (i.e., IEEE 802.11p, Cellular-V2X (C-V2X)) to prevent accidents (such as the early warning of intersection collision avoidance, the early warning of vehicles in front of vehicles, the early warning of road blind spots, motor vehicles, non-motor vehicles and pedestrian warnings, traffic light warnings, etc.). + +#### **I.1.1.5 High-definition map generation scenario** + +The vehicles collect point cloud data, road images and location information using cameras, lidar and other sensors. After overlaying the point cloud data, road images and location information, it can recognize and identify the road lines, curb, traffic signs and traffic lights, etc. using machine learning models. It can improve the efficiency and accuracy of high-definition map generation. + +### **I.1.2 Prediction use cases** + +The prediction scenarios include the motion prediction of the target lane, pedestrians, vehicles and obstacles. For example, if the vehicle detects a nearby target vehicle, the vehicle needs to predict the driving trajectory of the target vehicle. If the target vehicle will be overtaking, following or driving in parallel, lane prediction should be made, and then vehicle short-range communication technology with automatic driving technology could be combined to realize real-time target vehicle trajectory prediction. If the target vehicle is driving towards an intersection, and may make a left turn, a right turn, or a straight run, the autonomous vehicle may predict the path of the target vehicle in advance through a machine learning model. In another scenario, the vehicle at the intersection with a large flow of pedestrians could also predict the pedestrians' movement trajectories and perform deceleration or a steering operation in advance. + +### **I.1.3 Planning use cases** + +The planning scenarios are mainly for path planning and speed planning of vehicles in motion. For example, to avoid static or dynamic obstacles, path planning is required. The vehicles perceive the surrounding environment through cameras or radars, and use a machine learning model to generate the candidate paths. Next, using the cost function to evaluate each path, the factors affecting the cost function include smoothness, safety, deviation from the centre of the lane, etc. Then the paths are ranked by cost and the path with the lowest cost is selected. This is followed by speed planning, which needs to collect a series of speeds related to the path points, and then the data is processed by the machine learning model. It is possible to select the driving speed that is limited by various + +environments, thereby determining whether to change the speed or keep the original speed traveling on the path. Finally, the vehicle's trajectory could be constructed by combining path planning and speed planning. + +### I.1.4 Human-machine interaction use cases + +#### I.1.4.1 Overview + +By obtaining useful information through relevant equipment, and analysing it with machine learning models, the human-machine interaction scenarios could help drivers to focus on driving, reduce traffic accidents, and improve driving comfort. The specific human-machine interaction scenarios are shown in Figure I.2. + +![Diagram illustrating the scenario of human-machine interaction. A car icon is connected to a box labeled 'Human-machine interaction', which then branches out to four boxes: 'Gesture recognition', 'Speech recognition', 'Eye tracking', and 'Driver monitoring'. The text 'F.749.4(21)_FI.2' is visible at the bottom right of the diagram.](16152cf1d84aea10848758f51a91ff6a_img.jpg) + +``` +graph LR; Car[Car Icon] --> HMI(Human-machine interaction); HMI --> GR(Gesture recognition); HMI --> SR(Speech recognition); HMI --> ET(Eye tracking); HMI --> DM(Driver monitoring); +``` + +F.749.4(21)\_FI.2 + +Diagram illustrating the scenario of human-machine interaction. A car icon is connected to a box labeled 'Human-machine interaction', which then branches out to four boxes: 'Gesture recognition', 'Speech recognition', 'Eye tracking', and 'Driver monitoring'. The text 'F.749.4(21)\_FI.2' is visible at the bottom right of the diagram. + +Figure I.2 – Scenario of human-machine interaction + +#### I.1.4.2 Gesture recognition scenario + +The gesture recognition scenarios mainly include static gesture recognition and dynamic gesture recognition. The human-machine interaction interface of the smart cockpit controls vehicle navigation, information entertainment and other systems by capturing gestures performed by the drivers. It can then capture gestures and detect/return gestures or actions in pictures or videos. Then, it can parse the user's behaviour information, thus solving common gesture operations, such as determining, clicking, sliding left and right, opening/closing, etc. + +#### I.1.4.3 Speech recognition scenario + +In the voice control scenario, the driver performs voice communication through the human-machine interaction interface in the smart cockpit. Through voice communication, drivers can make phone calls, play music, switch navigation routes, remote control of smart home appliances, etc. It could be combined with natural language processing technology to identify driver's identity information and analyse driver's emotions, etc. + +#### I.1.4.4 Eye tracking scenario + +In the eyeball tracking scenario, the sensors first recognize the feature points of the human eyes to establish a model. Next, through the camera and other devices, the rotation motion is captured during the rotation of the eyeball. Finally, the machine learning model is used to analyse the state of the human eyes, the position of the gaze, etc. Therefore, driver distraction and low vigilance are prevented during driving. + +#### I.1.4.5 Driver monitoring scenario + +In the driver monitoring scenario, the cameras in the smart cockpit monitor the driver's driving state in real time using face recognition and eyeball tracking technology, which can promptly remind the driver to ensure the correctness of the driver's driving operation in the state of distraction, fatigue, bowing, chatting, etc. + +### **I.1.5 Model training/deployment use cases** + +#### **I.1.5.1 Model training scenario** + +Artificial intelligence technology includes two steps: model training step and inference step. The model training step should generate a proper model based on mass data, mass storage and enormous computing power. In the model training scenario, the vehicles should upload mass data (e.g., video, audio and vehicle running data) to the cloud server through a wireless wide area network (WWAN). The cloud server trains and optimizes the models and algorithms continuously to adapt to the changing road circumstances based on the mass data that was uploaded from the vehicles. + +#### **I.1.5.2 Model deployment scenario** + +The cloud server should deploy previously tested and optimized machine learning models to each vehicle through a WWAN. The vehicles have the ability of inference and can implement environment perception, prediction, planning and human-machine interaction in vehicles through the downloaded machine learning models. + +# Bibliography + +- [b-ITU-T F.749.1] Recommendation ITU-T F.749.1 (2015), *Functional requirements for vehicle gateways*. +- [b-ITU-T F.749.2] Recommendation ITU-T F.749.2 (2017), *Service requirements for vehicle gateway platforms*. +- [b-ITU-T H.550] Recommendation ITU-T H.550 (2018), *Architecture and functional entities of vehicle gateway platforms*. +- [b-ITU-T Y.2091] Recommendation ITU-T Y.2091 (2011), *Terms and definitions for next generation networks*. +- [b-ITU-T Y.3172] Recommendation ITU-T Y.3172 (2019), *Architectural framework for machine learning in future networks including IMT-2020*. +- [b-ITU-R handbook] ITU-R Handbook (2006), Land Mobile (including wireless access) – Volume 4: *Intelligent Transport Systems*. +- [b-ISO/IEC 2382] ISO/IEC 2382:2015, *Information technology – Vocabulary*. + +- [b-ISO/IEC TR 29119-11] ISO/IEC TR 29119-11: 2020, Software and systems engineering – Software testing – Part 11: *Guidelines on the testing of AI-based systems*. + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.749.7-202512-I_PDF-E/562f471e8153729557e6a4ee6343c32c_img.jpg b/marked/F/T-REC-F.749.7-202512-I_PDF-E/562f471e8153729557e6a4ee6343c32c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..4a45c03e9a81e84136464ef8f817f27073cdb0a7 --- /dev/null +++ b/marked/F/T-REC-F.749.7-202512-I_PDF-E/562f471e8153729557e6a4ee6343c32c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b794e884c38f8a27bb556a2a0b76674a6c7fa24ad7ee119b71c639409c25305e +size 105580 diff --git a/marked/F/T-REC-F.749.7-202512-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.749.7-202512-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..abda52cbb1047d6649f89e40c898a1c33bed7267 --- /dev/null +++ b/marked/F/T-REC-F.749.7-202512-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6f3b7e07231c0f3e116c99c493bdda98ecd52313b6735faa6cea23815cf3162b +size 7190 diff --git a/marked/F/T-REC-F.749.7-202512-I_PDF-E/raw.md b/marked/F/T-REC-F.749.7-202512-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..504495f06ba9b36b2ab0934534e29aeaf076cf3c --- /dev/null +++ b/marked/F/T-REC-F.749.7-202512-I_PDF-E/raw.md @@ -0,0 +1,480 @@ + + +Recommendation + +**ITU-T F.749.7 (12/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +**Requirements for vulnerable road user +protection services using a vehicle gateway** + +![ITU logo](84a1d09fb489061482111515543b60dc_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +# ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.749.7 + +## Requirements for vulnerable road user protection services using a vehicle gateway + +## Summary + +Recommendation ITU-T F.749.7 specifies the requirements for vulnerable road user (VRU) protection services using a vehicle gateway. The protection of vulnerable traffic participants has a positive impact on road traffic safety. By establishing harmonized functional requirements, this framework enables vehicle systems to more effectively detect and protect VRUs, thereby helping to avoid traffic accidents and reduce the severity and frequency of serious injuries. In addition, the adoption of uniform requirements across manufacturers supports cost efficiency in system development and facilitates interoperability among different automotive platforms. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.749.7 | 2025-12-14 | 21 | 11.1002/1000/15754 | + +## Keywords + +Cloud server, RSU, vulnerable road user. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2026 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|--------------|----------------------------------------------------------------------------|---| +| 1 | Scope..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 1 | +| 4 | Abbreviations and acronyms ..... | 2 | +| 5 | Conventions ..... | 2 | +| 6 | Introduction..... | 2 | +| 7 | Requirements of the vehicle system ..... | 3 | +| 7.1 | Perception requirements of the vehicle system ..... | 3 | +| 7.2 | Positioning requirements of the vehicle system ..... | 4 | +| 7.3 | Mission decision-making requirements of the vehicle system..... | 4 | +| 7.4 | HMI requirements of the vehicle system..... | 4 | +| 8 | Requirements of the VRU device ..... | 4 | +| 8.1 | Communication requirements of the VRU device ..... | 4 | +| 8.2 | Positioning requirements of the VRU device..... | 4 | +| 8.3 | Mission decision-making requirements of the VRU device..... | 4 | +| 8.4 | HMI requirements of the VRU device ..... | 5 | +| 9 | Requirements of road infrastructure ..... | 5 | +| 9.1 | Sensing requirements of road infrastructure..... | 5 | +| 9.2 | Computing requirements of road infrastructure ..... | 5 | +| 9.3 | Communication requirements of road infrastructure ..... | 5 | +| 9.4 | Positioning requirements of road infrastructure ..... | 5 | +| 10 | Requirements of cloud server ..... | 6 | +| 10.1 | General computing requirements of cloud server ..... | 6 | +| 10.2 | Data fusion and processing requirements of cloud server..... | 6 | +| 10.3 | Mission decision-making requirements of cloud server..... | 6 | +| 11 | Requirements of the communication network..... | 6 | +| 11.1 | General requirements of communication network ..... | 7 | +| 11.2 | Specific communication requirements of the communication network ..... | 7 | +| Appendix I | – Typical use cases of VRU protection services using a vehicle system..... | 8 | +| I.1 | Collision warning via direct communication ..... | 8 | +| I.2 | Collision warning via wireless wide area network communication ..... | 8 | +| I.3 | Collision warning with roadside perception ..... | 8 | +| I.4 | Collision warning with perception tag..... | 8 | +| I.5 | Collision warning with vehicle perception ..... | 8 | +| Bibliography | ..... | 9 | + + + +# Recommendation ITU-T F.749.7 + +## Requirements for vulnerable road user protection services using a vehicle gateway + +## 1 Scope + +This Recommendation specifies the requirements to support vulnerable road user (VRU) protection services, which enable the detection, warning and facilitation of assistance to protect VRUs (e.g., pedestrians, bicyclists, e-scooters) using a vehicle gateway, including functional requirements for the vehicle system, vulnerable road user device, communication network, road infrastructure and cloud server. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +- [ITU-T F.749.1] Recommendation ITU-T F.749.1 (2015), *Functional requirements for vehicle gateways*. +- [3GPP TS 22.185] 3GPP TS 22.185 V19.0.0 (2025-10), *3rd Generation Partnership Project; Technical Specification Group Services and System Aspects; Service requirements for V2X services; Stage 1*. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following term defined elsewhere: + +**3.1.1 vehicle gateway (VG)** [ITU-T F.749.1]: A VG is a device in a vehicle that enables communications between a device in the vehicle and another device which may be physically located either inside the vehicle or outside the vehicle (e.g., roadside station, cloud-based server, etc.). A VG provides standardized interfaces and protocols, communications across heterogeneous networks, optimized network selection based on application needs and network QoS, arbitration and integration of network communications, security and switching network connections to maintain service continuity. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 vulnerable road user protection service:** A service that detects, warns and helps protect vulnerable road users (VRUs) by avoiding or reducing the severity of collisions between vehicles and VRUs. + +NOTE – VRUs are vulnerable participants in road traffic, e.g., pedestrians, bicyclists, e-scooters, motorcyclists. + +**3.2.2 vulnerable road user device:** A physical device equipped with positioning and communication functions that enhances the safety of vulnerable road users (VRUs). + +NOTE – Devices vary widely and may include portable devices carried by pedestrians, or equipment installed on bicycles, motorcyclists, etc. + +**3.2.3 positioning function:** The capability to determine the geographical location information of objects. + +NOTE – Geographical location can be determined using the global navigation satellite system (GNSS), radio frequency identification (RFID), ultra-wideband (UWB) or cellular networks. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-------|----------------------------------------| +| C-V2X | Cellular Vehicle to Everything | +| GNSS | Global Navigation Satellite System | +| HMI | Human Machine Interface | +| IMT | International Mobile Telecommunication | +| RFID | Radio Frequency Identification | +| RSU | Roadside Unit | +| UWB | Ultra-Wideband | +| V2X | Vehicle-to-Everything | +| VG | Vehicle Gateway | +| VRU | Vulnerable Road User | +| WWAN | Wireless Wide Area Network | + +## 5 Conventions + +In this Recommendation: + +- The keywords "be required to" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. +- The keywords "can optionally" indicate an optional requirement which is permissible, without implying any sense of being recommended. This term is not intended to imply that the vendor's implementation must provide the option, and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +## 6 Introduction + +Road safety is a crucial aspect of mobility both for individuals and for policymakers. VRUs can be pedestrians, cyclists, e-scooters, motorcyclists, large animals and impaired participants in road traffic. Connectivity and automation offer a significant potential to reduce human errors and enhance the protection of VRUs. VRU protection services enhance the safety of the most vulnerable traffic participants. VRU protection services include the following actors: vehicle system, communication network, road infrastructure, VRU device and cloud server. The vehicle system cooperates with road infrastructure, cloud server and other vehicles to support different VRU protection services. + +Appendix I describes in detail five typical use cases considering the communication capability and functional entities involved in VRU protection services using a vehicle gateway [ITU-T F.749.1]. + +To support these use cases, the architecture of VRU protection services using vehicle gateway is abstracted and refined as shown in Figure 1, which includes vehicle system, VRU device, road infrastructure, cloud server and communication networks. + +1. **Vehicle system:** It ensures safe driving and collision prevention of vehicles. It has the functions of perception, mission decision-making, high-precision positioning, human machine interface (HMI) and driver alerts. +2. **VRU device:** It improves the safety of a VRU. It has the functions of positioning, communication, optional mission decision-making and HMI. +3. **Road infrastructure:** It enhances traffic safety by collecting and processing VRU objects. It has the functions of communication and optional local sensing, local computing and positioning. +4. **Cloud server:** It enhances traffic safety by processing and distributing VRU information to prevent collisions. It has the functions of data fusion and hierarchical computing, optional mission decision-making. +5. **Communication network:** It facilitates the reliable and prioritized exchange of messages between all modules. It has at least one of the communication functions of wireless short-range, wireless wide-area and wired connections. + +![Figure 1: Architecture of vulnerable road user protection services using a vehicle gateway. The diagram shows four main components: Cloud server, Vehicle system, Road infrastructure, and VRU. The Cloud server contains 'Data fusion and processing' (optional) and 'Mission decision-making' (optional). The Vehicle system contains a 'Vehicle gateway' (required) and 'Functionalities' (Perception, Mission decision-making, HMI, Positioning - all required). The Road infrastructure contains 'Sensing' (optional), 'Positioning' (optional), 'Computing' (optional), and 'Communication' (required). The VRU contains icons for a pedestrian, bicycle, and motorcycle. Communication links include WWAN between Cloud server and Road infrastructure, Wireless short-range between VRU and Vehicle system, and Wireless signal related position between VRU and Road infrastructure. A legend at the bottom defines the symbols for optional/required functions and communication connections.](562f471e8153729557e6a4ee6343c32c_img.jpg) + +F.749.7(25) + +Figure 1: Architecture of vulnerable road user protection services using a vehicle gateway. The diagram shows four main components: Cloud server, Vehicle system, Road infrastructure, and VRU. The Cloud server contains 'Data fusion and processing' (optional) and 'Mission decision-making' (optional). The Vehicle system contains a 'Vehicle gateway' (required) and 'Functionalities' (Perception, Mission decision-making, HMI, Positioning - all required). The Road infrastructure contains 'Sensing' (optional), 'Positioning' (optional), 'Computing' (optional), and 'Communication' (required). The VRU contains icons for a pedestrian, bicycle, and motorcycle. Communication links include WWAN between Cloud server and Road infrastructure, Wireless short-range between VRU and Vehicle system, and Wireless signal related position between VRU and Road infrastructure. A legend at the bottom defines the symbols for optional/required functions and communication connections. + +**Figure 1 – Architecture of vulnerable road user protection services using a vehicle gateway** + +## 7 Requirements of the vehicle system + +In the vehicle system, the positioning function, which is the capability to determine the geographical location information of objects, is used. + +### 7.1 Perception requirements of the vehicle system + +The requirements of perception are as follows: + +[VEH-PER-001] The vehicle system is required to support local perception to process the information collected by one type of local sensor (e.g., radar, camera) to detect and characterize VRUs. + +[VEH-PER-002] The vehicle system is recommended to support perception by data fusion when receiving sensor information from other vehicles, road infrastructure or cloud server to detect and characterize VRUs. + +[VEH-PER-003] The vehicle system is recommended to transmit the perception information of VRUs to other vehicles, cloud server or road infrastructure through a communication network. + +### **7.2 Positioning requirements of the vehicle system** + +The requirements of positioning are as follows: + +[VEH-POS-001] The vehicle system is required to support the positioning function. + +### **7.3 Mission decision-making requirements of the vehicle system** + +The requirements of mission decision-making are as follows: + +[VEH-MDM-001] The vehicle system is required to detect a risk of collision with VRUs to avoid such collisions when receiving from the VRU the position, velocity and heading of VRUs. + +[VEH-MDM-002] The vehicle system is required to detect a risk of collision with VRUs to avoid such collisions when receiving the position, velocity and heading of VRUs detected by road infrastructure. + +[VEH-MDM-003] The vehicle system is required to detect a risk of collision with VRUs to avoid such collisions when receiving the position, velocity and heading of VRUs detected by cloud server. + +### **7.4 HMI requirements of the vehicle system** + +The requirements of HMI are as follows: + +[VEH-HMI-001] The vehicle system is required to trigger a warning to alert the driver by one of the following methods: visual textual, or auditory. + +NOTE – Examples for visual warning is image and icon. Examples auditory warning is sound and voice. + +## **8 Requirements of the VRU device** + +### **8.1 Communication requirements of the VRU device** + +The requirements of communication are as follows: + +[VRU-COM-001] The VRU device is required to support wireless wide area communication or short-range communication. + +[VRU-COM-002] The VRU device is recommended to support both, wireless wide area communication and short-range communication. + +NOTE – Wide area communication includes IMT-Advanced [b-ITU-R M.2012-7], IMT-2020 [b-ITU-R M.2150-3] and short-range communication includes cellular vehicle to everything (C-V2X) direct communication [b-IEEE 802.11p]. + +### **8.2 Positioning requirements of the VRU device** + +The requirements of positioning are as follows: + +[VRU-POS-001] The VRU device is required to have positioning function. + +[VRU-POS-002] The VRU device can optionally have a positioning function by integrated through an auxiliary module, such as an ultra-wideband (UWB) tag or an RFID. + +### **8.3 Mission decision-making requirements of the VRU device** + +The requirements of mission decision-making are as follows: + +[VRU-MDM-001] The VRU device is recommended to detect a risk of collision with vehicles to avoid such collisions when receiving the information from vehicles, which includes the position, velocity, heading of vehicle, etc. + +[VRU-MDM-002] The VRU device is recommended to detect a risk of collision with vehicles to avoid such collisions when receiving the information of vehicles detected by road infrastructure, which includes the position, velocity, heading of detected vehicles, etc. + +[VRU-MDM-003] The VRU device is recommended to detect a risk of collision with vehicles to avoid such collisions when receiving the information of vehicles detected by a cloud server, which includes the position, velocity, heading of detected vehicles, etc. + +### **8.4 HMI requirements of the VRU device** + +The requirements of HMI are as follows: + +[VRU-HMI-001] The VRU device is recommended to trigger a warning to alert the VRUs by one or more of the following methods: visual textual, or auditory. + +NOTE – Examples of visual warning are image and icon. Examples of auditory warnings are sound and voice. + +## **9 Requirements of road infrastructure** + +### **9.1 Sensing requirements of road infrastructure** + +The requirements of sensing are as follows: + +[ROAD-SENS-001] The road infrastructure is recommended to support local sensing to collect information of VRUs by local sensors, such as radar, camera, etc. + +### **9.2 Computing requirements of road infrastructure** + +The requirements of computing are as follows: + +[ROAD-COMP-001] The road infrastructure is recommended to support local computing to detect and characterize VRUs and vehicles collected by local sensors. + +[ROAD-COMP-002] The road infrastructure is recommended to support computing of data fusion when receiving sensor information of detected VRUs and vehicles from vehicles, other road infrastructure or cloud server. + +[ROAD-COMP-003] The road infrastructure is recommended to transmit the computing information of VRUs to vehicles, cloud server or other road infrastructure through a communication network. + +[ROAD-COMP-004] The road infrastructure is recommended to transmit the computing information of vehicles to VRUs, cloud server or other road infrastructure through a communication network. + +### **9.3 Communication requirements of road infrastructure** + +The requirements of communication are as follows: + +[ROAD-COM-001] The road infrastructure is required to support dedicated short range communication technology, e.g., C-V2X direct communication or [b-IEEE 802.11p]. + +[ROAD-COM-002] The road infrastructure is recommended to have a wireless wide area communication function such as IMT-Advance [b-ITU-R M.2012-7], IMT-2020 [b-ITU-R M.2150-3] or a wired communication function, such as optical fibre communication. + +### **9.4 Positioning requirements of road infrastructure** + +The requirements of positioning are as follows: + +[ROAD-POS-001] The road infrastructure can optionally support positioning by integrated auxiliary module to identify the VRU's position, such as UWB anchor, RFID reader/writer, etc. + +## **10 Requirements of cloud server** + +### **10.1 General computing requirements of cloud server** + +The requirements of general computing are as follows: + +[Cloud-GCOMP-001] The cloud server is recommended to support hierarchical computing, such as three-layer architecture comprising edge, regional and central cloud server, or two-layer architecture combining edge and regional, or regional and central cloud server. + +[Cloud-GCOMP-002] The cloud server is required to transmit the computing information of VRUs, such as the position, velocity and heading of VRU, to vehicles and road infrastructure through a communication network. + +[Cloud-GCOMP-003] The cloud server is required to transmit the computing information of vehicles, such as the position, velocity and heading of vehicles, to VRUs and road infrastructure through a communication network. + +[Cloud-GCOMP-004] The cloud server is required to support to fuse data, detect and characterize VRUs when receiving information from vehicles and road infrastructure. + +[Cloud-GCOMP-005] The cloud server is required to support to fuse data, detect and characterize vehicles when receiving information from VRUs and road infrastructure. + +### **10.2 Data fusion and processing requirements of cloud server** + +The requirements of data fusion and processing of cloud server are as follows: + +[Cloud-DDF-001] The edge cloud server is required to detect and characterize VRUs when receiving sensor information from vehicles, road infrastructure or other cloud servers when the edge cloud server is deployed. + +[Cloud-DDF-002] The regional cloud server is required to detect and characterize VRUs when receiving sensor information from vehicles, road infrastructure or other cloud servers when the regional cloud is deployed. + +[Cloud-DDF-003] The central cloud server is required to detect and characterize VRUs when receiving the message from VRU, road infrastructure or other cloud servers when the central cloud server is deployed. + +### **10.3 Mission decision-making requirements of cloud server** + +The requirements of mission decision-making are as follows: + +[Cloud-MDM-001] The cloud server can optionally detect a risk of collision between VRUs and vehicles and send this information to relevant vehicles and VRUs. + +[Cloud-MDM-002] The cloud server can optionally make decisions of behaviour of vehicles to avoid collision between VRUs and vehicles and send these decisions to relevant vehicles. + +[Cloud-MDM-003] The cloud server can optionally make decisions of behaviour of VRUs to avoid collision between VRUs and vehicles and send these decisions to relevant VRUs. + +## **11 Requirements of the communication network** + +The requirements of the communication network include general requirements and specific communication requirements. + +### **11.1 General requirements of communication network** + +The general requirements of the communication network are as follows: + +[COMN-General-001] The communication network is required to prioritize message transmission among VRU devices. + +[COMN-General-002] The communication network is required to prioritize transmission of messages according to different applications. + +[COMN-General-003] The communication network is required to vary the transmission rate and range of the V2X communication based on service conditions. + +[COMN-General-004] The communication network is recommended to adjust the transmitting or receiving resources to minimize impact if the VRU device has limited resources, such as a battery. + +### **11.2 Specific communication requirements of the communication network** + +The specific communication requirements of the communication network are as follows: + +[COMN-SCR-001] The communication network is required to transfer messages between a VRU device and a vehicle with a maximum latency of 100 ms [3GPP TS 22.185]. + +[COMN-SCR-002] The communication network is required to transfer messages between a vehicle and the infrastructure with a maximum latency of 100 ms [3GPP TS 22.185]. + +[COMN-SCR-003] The communication network is required to transfer messages between a vehicle and the cloud server with a maximum latency of 100 ms [3GPP TS 22.185]. + +[COMN-SCR-004] The communication network is required to transfer messages between a VRU device and the cloud server with a maximum latency of 100 ms [3GPP TS 22.185]. + +[COMN-SCR-005] The communication network is required to support high reliability without requiring application-layer message re-transmissions. + +[COMN-SCR-006] The communication network is required to support a maximum frequency of 10 messages per second per transmitting VRU device. + +[COMN-SCR-007] The communication network is required to support a communication range sufficient to give the driver(s) ample response time, which is defined by application. + +[COMN-SCR-008] The communication network is recommended to support integrated sensing and communication, which can recognize VRUs either independently or in conjunction with other sensing technologies. + +## Appendix I + +## Typical use cases of VRU protection services using a vehicle system + +(This appendix does not form an integral part of this Recommendation.) + +### I.1 Collision warning via direct communication + +VRUs, such as pedestrians, bicycles, etc., equipped with VRU devices supporting dedicated short-range communication send information including VRU's position, speed, direction and moving intention to surrounding vehicles when crossing a road. After receiving the VRU's information, the vehicle determines its position relative to the VRU. If a risk of collision is detected, a warning will be sent to the driver or the automated driving system. + +### I.2 Collision warning via wireless wide area network communication + +VRUs, such as pedestrians, bicycles, etc., equipped with VRU devices supporting wireless wide area communication send information including VRU's position, speed, direction and moving intention to cloud server when crossing a road. The cloud server will determine VRU's position and send to surrounding VRU's vehicles via wireless wide area communication. After receiving the VRU's information, the vehicle determines its position relative to the VRU. If a risk of collision is detected, a warning message will be sent to the driver or the automated driving system. Optionally, the cloud server will make decisions of behaviour of vehicles to avoid collision between VRUs and vehicles after receiving VRU's information and send these decisions to the relevant driver or the automated driving system. + +### I.3 Collision warning with roadside perception + +VRUs, such as pedestrians, bicycles, etc., who are not equipped with a VRU device, may enter or may have entered the crosswalk at the intersection. Sensing devices deployed on the roadside identify vulnerable traffic participants that enter the warning area, identify the location, speed and direction information of VRU, and the road infrastructure will share the identified information of VRUs to vehicles around the intersection. Vehicles determine the relative position of vulnerable traffic participants. If a collision risk is detected, a warning will be sent to the driver or the automated driving system. + +### I.4 Collision warning with perception tag + +When vulnerable traffic participants, such as children and the elderly, equipped with a positioning perception tag (e.g., RFID tag or UWB tag), enter the intersection or sidewalk, the positioning perception system, such as RFID reader/writer, UWB anchor, deployed around the roadside automatically identifies their position, and will share the identified information to the roadside unit (RSU). Then the RSU broadcasts VRU information to the connected vehicles, and the vehicles determine the relative position to vulnerable traffic participants. If a risk of violations and collisions is detected, a warning is sent to the driver or the automated driving system. + +### I.5 Collision warning with vehicle perception + +VRUs, such as pedestrians, bicycles, etc., who are not equipped with a VRU device, may enter or may have entered the sensing range of the vehicle. Onboard sensing systems identify vulnerable traffic participants that enter the warning area, identify the location, speed and direction information of VRU, and the vehicle will share the identified information of VRUs to vehicles around. Nearby vehicles determine the relative position to vulnerable traffic participants. If a collision risk is detected, a warning will be sent to the driver or the automated driving system. + +## Bibliography + +- [b-ITU-R M.2012-7] Recommendation ITU-R M.2012-7 (2026), *Detailed specifications of the terrestrial radio interfaces of International Mobile Telecommunications-Advanced (IMT-Advanced)*. +- [b-ITU-R M.2150-3] Recommendation ITU-R M.2150-3 (2026), *Detailed specifications of the terrestrial radio interfaces of International Mobile Telecommunications-2020 (IMT-2020)*. +- [b-IEEE 802.11p] IEEE 802.11p-2010, *IEEE Standard for Information technology – Local and metropolitan area networks – Specific requirements Part 11: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) Specifications Amendment 6: Wireless Access in Vehicular Environments*. + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.749.8-202401-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/F/T-REC-F.749.8-202401-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..abda52cbb1047d6649f89e40c898a1c33bed7267 --- /dev/null +++ b/marked/F/T-REC-F.749.8-202401-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6f3b7e07231c0f3e116c99c493bdda98ecd52313b6735faa6cea23815cf3162b +size 7190 diff --git a/marked/F/T-REC-F.749.8-202401-I_PDF-E/9c6461e1e94afae4dec455e69a2ce152_img.jpg b/marked/F/T-REC-F.749.8-202401-I_PDF-E/9c6461e1e94afae4dec455e69a2ce152_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..36fd4b5f7a6f4e2971f7a963bba218e1e5c12d4f --- /dev/null +++ b/marked/F/T-REC-F.749.8-202401-I_PDF-E/9c6461e1e94afae4dec455e69a2ce152_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a50a46b5aad378fac413225a86ad6b3be272b190d07f2d3fbcd7e2dbb9b41be5 +size 49363 diff --git a/marked/F/T-REC-F.749.8-202401-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg b/marked/F/T-REC-F.749.8-202401-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..083ee5ce8bee42b246dfb7669057f3701c478418 --- /dev/null +++ b/marked/F/T-REC-F.749.8-202401-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bb0c229fb7bc7ffd7b1466d102c829252f08d268fe19df23389b136bb70f7628 +size 132845 diff --git a/marked/F/T-REC-F.749.8-202401-I_PDF-E/raw.md b/marked/F/T-REC-F.749.8-202401-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..440bacd0fda3da24c41a86e53a863c129e482674 --- /dev/null +++ b/marked/F/T-REC-F.749.8-202401-I_PDF-E/raw.md @@ -0,0 +1,1119 @@ + + +# Recommendation **ITU-T F.749.8 (01/2024)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +# --- **In-vehicle multimedia applets: Framework and functional requirements** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.749.8 + +# In-vehicle multimedia applets: Framework and functional requirements + +## Summary + +Recommendation ITU-T F.749.8 describes the in-vehicle multimedia applet (VMMA) concept, the VMMA framework, the functional requirements, the functional APIs and the reference parameters. + +Some detailed use cases and reference APIs are described in the appendix. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.749.8 | 2024-01-13 | 16 | 11.1002/1000/15762 | + +## Keywords + +Application programming interface, in-vehicle multimedia applet, reference parameters. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2024 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|-------|---------------------------------------|----| +| 1 | Scope | 1 | +| 2 | References | 1 | +| 3 | Definitions | 1 | +| 3.1 | Terms defined elsewhere | 1 | +| 3.2 | Terms defined in this Recommendation | 2 | +| 4 | Abbreviations and acronyms | 2 | +| 5 | Conventions | 2 | +| 6 | Overview of the VMMA | 3 | +| 7 | Reference framework of VMMA | 3 | +| 8 | Requirements | 5 | +| 8.1 | VMMA protocol requirements | 5 | +| 8.2 | VMMA basic interface requirements | 5 | +| 8.3 | Vehicle interface requirements | 8 | +| 8.4 | Navigation interface requirements | 9 | +| 8.5 | Speech interface requirements | 10 | +| 8.6 | Multimedia interface requirements | 11 | +| 8.7 | Display requirements | 13 | +| 8.8 | Performance monitoring requirements | 13 | +| | Appendix I – Error codes | 14 | +| | Appendix II – User interface examples | 15 | +| II.1 | POST request | 15 | +| II.2 | Login interface | 15 | +| II.3 | User information interface | 15 | +| II.4 | System information interface | 16 | +| II.5 | Login expire interface | 16 | +| II.6 | User setting interface | 16 | +| II.7 | Vehicle speed interface | 16 | +| II.8 | Vehicle fuel interface | 16 | +| II.9 | VIN interface | 16 | +| II.10 | Vehicle wheel rotation interface | 17 | +| II.11 | Vehicle information interface | 17 | +| II.12 | Navigation interface | 17 | +| II.13 | Play TTS interface | 17 | +| II.14 | Pause TTS interface | 17 | +| II.15 | Resume TTS interface | 17 | +| II.16 | TTS state change interface | 17 | +| II.17 | Stop TTS interface | 18 | + +| | Page | +|----------------------------------------------------------|-------------| +| II.18 Playlist interface | 18 | +| II.19 Display mode | 18 | +| Appendix III – Use case of in-vehicle multimedia applets | 19 | +| III.1 Navigation | 19 | +| III.2 Voice control | 19 | +| Bibliography | 20 | + +# **Recommendation ITU-T F.749.8** + +# In-vehicle multimedia applets: Framework and functional requirements + +# 1 Scope + +This Recommendation describes the concept of in-vehicle multimedia applet (VMMA) common framework characteristics and defines functional requirements, then brings a capability framework, the application programming interfaces and the reference parameters of the VMMA. + +The scope of this Recommendation includes: + +- Overview of the VMMA; +- Reference framework of the VMMA; +- Functional requirements of the VMMA; +- Application programming interfaces and reference parameters of the VMMA. + +In addition, the interface examples and the use cases of VMMA are provided in Appendices II and III, respectively. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +- [ITU-T F.746.3] Recommendation ITU-T F.746.3 (2015), *Intelligent question answering service framework*. +- [ITU-T H.625] Recommendation ITU-T H.625 (2017), *Architecture for network-based speech-to-speech translation services*. +- [ITU-T P.10] Recommendation ITU-T P.10/G.100 (2017) *Vocabulary for performance, quality of service and quality of experience*. +- [ECMA-262] ECMA-262 (2023), *ECMA Script 2023 Language Specification*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 vehicle multimedia system (VMS)** [b-ITU-T F.749.3]: The VMS consists of vehicle multimedia system inputs (VM I/P), vehicle multimedia unit (VMU) and vehicle multimedia system outputs (VM O/P). + +**3.1.2 natural language processing (NLP)** [ITU-T F.746.3]: A method that analyses text in natural languages through several processes such as part-of-speech recognition, syntactic analysis and semantic analysis. + +**3.1.3 text-to-speech synthesis (TTS)** [ITU-T P.10]: A TTS process generates a speech signal from text codes. It is usually composed of the two parts: + +- a language-dependent text-processing part (the high level processing part), which generates from the character string (by reading rules, vocabulary and semantic analysis) a set of phonetic, prosodic, etc., parameters which are used by: +- an acoustical signal generating part, the synthesizer itself, which generates the audible speech. + +**3.1.4 automatic speech recognition (ASR)** [ITU-T H.625]: A system that can recognize continuous speech, often having phoneme-sized references, using lexical, syntactic, semantic, and pragmatic knowledge, and reacts appropriately (therefore having interpreted the message and found the corresponding action to be taken). + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 in-vehicle multimedia applet:** In-vehicle multimedia applet (VMMA) is a new format of mobile application. VMMA not only provides a hybrid solution which relies on Web technologies (especially CSS and JavaScript) but also integrates with capabilities of native applications. + +**3.2.2 native applications:** Software applications that are developed for a specific platform or operating system using the native programming languages and tools provided by that platform. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|--------|-----------------------------------| +| ASR | Automatic Speech Recognition | +| CSS | Cascading Style Sheets | +| HMI | Human–Machine Interface | +| LAN | Local Area Network | +| LBS | Location-Based Service | +| NLP | Natural Language Processing | +| SDK | Software Development Kit | +| TTS | Text-to-Speech | +| TTS | Text-To-Speech | +| UI | User Interface | +| URL | Uniform Resource Locator | +| VIN | Vehicle Identification Number | +| VM I/P | Vehicle Multimedia System Inputs | +| VM O/P | Vehicle Multimedia System Outputs | +| VMMA | In-Vehicle Multimedia Applet | +| VMS | Vehicle Multimedia System | +| VMU | Vehicle Multimedia Unit | + +# 5 Conventions + +In this Recommendation: + +- The keywords "**is required to**" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "**is recommended**" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. +- The keyword "**may**" indicates an optional requirement which is permissible, without implying any sense of being recommended. This term is not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +In the tables with requirements, optionality is denoted as follows: M for mandatory, O for optional. + +# 6 Overview of the VMMA + +The in-vehicle multimedia applet (VMMA) is a new form of in-car application, leveraging both Web technologies (especially CSS and JavaScript [ECMAScript]) as well as capabilities of native applications. + +Without installation, the VMMA has a dynamic loading and update mechanism and can be greatly convenient for users to use for entertainment content. Combined with the intelligent voice system, the in-vehicle application will search for contents or services seamlessly. The whole process supports voice interaction, which effectively improves the application interaction experience in the vehicle. + +The VMMA enhances the service ecosystem in-vehicle, combines the scene perception, location-based service (LBS), social media etc., and provides more secure and nature scenerized services for the user. VMMA also provides uniform APIs and protocols compatible with different vehicle systems and hardware platforms. + +# 7 Reference framework of VMMA + +The VMMA is designed to be integrated into a vehicle's multimedia system. It offers a range of features and functionalities to enhance the driving experience while providing seamless connectivity with other infotainment services and platforms. + +The VMMA does not need to be installed. And the mechanism of dynamic loading and updating can greatly facilitate the user's behaviour and content consumption. Combined with the intelligent voice interaction in the whole processes, the VMMA can effectively improve the application interaction experience in the car. + +Figure 1 shows the reference framework of VMMA. + +![Figure 1 – Reference framework of VMMA. This is a block diagram showing the architecture of the VMMA system. On the left, the 'VMMA user' interacts with the 'Vehicle multimedia system'. The system is composed of several layers: 'User interoperate' (containing Microphone, Steering wheel button, and Touch screen), 'VMMA' (containing Management and integration (Vehicle interaction, OTA), VMMA client, Social media applet, VMMA controller, Voice SDK, LBS SDK, and Scenario SDK), 'VMMA framework' (containing HMI UI framework and Applet framework with In-vehicle multimedia applet engine, JS engine, and Browser kernel), and 'Vehicle OS'. Below the OS, inputs include Vehicle speed, Vehicle wheel rotation, Vehicle fuel, and Vehicle VIN. To the right, the 'VMMA cloud' contains VMMA management, VMMA services (Navigation, Music play, Speech recognition, LBB, Notification), and VMMA development kits. The 'VMMA developer' interacts with the cloud and the framework. Arrows indicate data flow between components.](a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg) + +Figure 1 – Reference framework of VMMA. This is a block diagram showing the architecture of the VMMA system. On the left, the 'VMMA user' interacts with the 'Vehicle multimedia system'. The system is composed of several layers: 'User interoperate' (containing Microphone, Steering wheel button, and Touch screen), 'VMMA' (containing Management and integration (Vehicle interaction, OTA), VMMA client, Social media applet, VMMA controller, Voice SDK, LBS SDK, and Scenario SDK), 'VMMA framework' (containing HMI UI framework and Applet framework with In-vehicle multimedia applet engine, JS engine, and Browser kernel), and 'Vehicle OS'. Below the OS, inputs include Vehicle speed, Vehicle wheel rotation, Vehicle fuel, and Vehicle VIN. To the right, the 'VMMA cloud' contains VMMA management, VMMA services (Navigation, Music play, Speech recognition, LBB, Notification), and VMMA development kits. The 'VMMA developer' interacts with the cloud and the framework. Arrows indicate data flow between components. + +F.749.8(24) + +**Figure 1 – Reference framework of VMMA** + +The framework of VMMA is composed of the user, the user interoperate module, the VMMA function module, the vehicle multimedia system, the VMMA cloud and the VMMA developer. + +The users can control VMMA through voice, steering wheel buttons and touch screens. The VMMA is recommended to be compatible with traditional vehicle interactive control modes. + +The VMMA is recommended to be based on the JavaScript framework, to be compatible with social media applets, and to include HMI UI framework and applet framework. The VMMA management module is required to upgrade configurations. The VMMA controller is required to support a variety of speech recognition technologies such as automatic speech recognition (ASR) [ITU-T H.625], text-to-speech synthesis (TTS) [ITU-T P.10] and natural language processing (NLP) [ITU-T F.746.3]. It is also required to support voice control SDK, location-based services software development kit (LBS SDK), and scenario SDK. A voice control SDK is a set of tools, libraries and documentation that enables developers to create, test and deploy applications or features that incorporate voice recognition, voice synthesis and other voice-related technologies. An LBS SDK is a collection of tools, libraries and documentation that enables developers to create, test and deploy applications or features that utilize location-based services. These services typically involve the use of GPS, wireless LAN ([b-IEEE 802.11a], [b-IEEE 802.11b], [b-IEEE 802.11g], [b-IEEE 802.11n], [b-IEEE 802.11ac], [b-IEEE 802.11ax]) or cellular network data to provide real-time location information, allowing developers to create location-aware applications, such as navigation, geofencing and location-based marketing or analytics. A scenario SDK is a set of tools, libraries and documentation that enable developers to create, test, and deploy scenario-based applications or features for specific platforms or environments. + +The VMMA framework is required to have an HMI UI framework, a JavaScript engine and a web browser kernel. A JavaScript engine is a software program or interpreter that processes and executes JavaScript code in web browsers. It translates the human-readable JavaScript code into machine- + +readable instructions, enabling the execution of interactive VMMA, and other JavaScript-based functionalities. + +The VMMA cloud is composed of the VMMA management platform, the VMMA services and the VMMA development kits. The VMMA management provides monitoring, operations and management functions of VMMA, including VMMA uploading, distribution and updating. The VMMA services provide the navigation, music play, speech recognition, LBS and notification functions. The VMMA development kits provide the development tool kits for VMMA developers. + +# 8 Requirements + +## 8.1 VMMA protocol requirements + +[PRO-001] The protocol of the vehicle applet is required to use HTTPS POST request with "Content-type: application/json" and "Content-encoding: UTF-8", and the request is returned as a Json object. + +The VMMA protocol parameters are shown in Table 1. + +**Table 1 – VMMA protocol parameter** + +| Element/attribute | Description | Necessity | +|-------------------|--------------------|-----------| +| appid | Applet id | M | +| sign | Signature | M | +| timestamp | Time stamp, second | M | +| nonce | Random string | M | +| code | Execution code | M | +| version | Protocol version | M | +| grant_type | Authorization code | O | + +## 8.2 VMMA basic interface requirements + +[BIR-001] The VMMA basic interface is required to have login interface, user information interface, system information interface, login expiration interface and user setting interface. + +### 8.2.1 Login + +User login interface is shown as below, and the parameters are shown in Table 2. + +``` + login(Object object) +``` + +**Table 2 – Login parameter** + +| Element/attribute | Type | Description | Necessity | +|-------------------|----------|-----------------------|-----------| +| timeout | number | Time out, millisecond | O | +| success | function | Return success | O | +| fail | function | Return fail | O | +| complete | function | Return complete | O | + +The login callback parameters are shown in Table 3. + +**Table 3 – Login callback parameter** + +| Element/attribute | Type | Description | +|-------------------|--------|------------------------| +| code | string | User login credentials | + +### 8.2.2 User information interface + +The getUserInfo object is used to retrieve the user information. + +The interface is shown below, and the parameters are shown in Table 4. + +``` +getUserInfo (Object object) +``` + +**Table 4 – User information parameter** + +| Element/attribute | Type | Description | Necessity | +|-------------------|----------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-----------| +| success | function | Return success | O | +| fail | function | Return fail | O | +| complete | function | Return complete | O | +| withCredentials | boolean | When withCredentials is true, it requires that login has previously been called and that the login status has not expired.
When withCredentials is false, login status is not required. | O | + +The user information callback parameters are shown in Table 5. + +**Table 5 – User information callback parameter** + +| Element/attribute | Type | Description | +|-------------------|--------|------------------------------------------------------------------------------------------| +| userInfo | number | The user information object does not contain sensitive information such as openid | +| rawData | number | The raw data string, excluding sensitive information, is used to calculate the signature | +| msignature | number | Used to verify user information | +| encryptedData | number | Encrypted data of complete user information | +| iv | number | The initial vector of the encryption algorithm | + +### 8.2.3 System information interface + +The getSystemInfo object is used to retrieve the vehicle system information. + +The interface is shown below, and the parameters are shown in Table 6. + +``` +getSystemInfo (Object object) +``` + +**Table 6 – System information parameter** + +| Element/attribute | Type | Description | Necessity | +|-------------------|----------|-----------------|-----------| +| success | function | Return success | O | +| fail | function | Return fail | O | +| complete | function | Return complete | O | + +The get system information interface callback parameters are shown in Table 7. + +**Table 7 – System information callback parameter** + +| Element/attribute | Type | Description | +|-------------------|--------|-------------------------------------------------------------------------------------| +| brand | string | Vehicle brand | +| model | string | Smartphone model | +| pixelRatio | number | Pixel ratio | +| screenWidth | number | Screen width | +| screenHeight | number | Screen height | +| windowWidth | number | Window width | +| windowHeight | number | Window height | +| language | string | System language | +| version | string | Version | +| system | string | Operation system | +| platform | string | Smartphone operation system | +| SDKVersion | string | SDK version | +| orientation | string | Screen display status: landscape-full, landscape-half, portrait-full, portrait-half | +| device | string | Device type | + +### 8.2.4 Login expiration interface + +The checkSession object is used to check if the login status has expired. + +The interface is shown below, and the parameters are shown in Table 8. + +``` +checkSession(Object object) +``` + +**Table 8 – Login expire parameter** + +| Element/attribute | Type | Description | Necessity | +|-------------------|----------|-----------------|-----------| +| success | function | Return success | O | +| fail | function | Return fail | O | +| complete | function | Return complete | O | + +### 8.2.5 User setting interface + +The getSetting object is used to retrieve the user's authorization settings. + +The interface is shown below, and the parameters are shown in Table 9. + +``` +getSetting(Object object) +``` + +**Table 9 – User setting parameter** + +| Element/attribute | Type | Description | Necessity | +|-------------------|----------|-----------------|-----------| +| success | function | Return success | O | +| fail | function | Return fail | O | +| complete | function | Return complete | O | + +The get user setting callback parameters are shown in Table 10. + +**Table 10 – User setting callback parameter** + +| Element/Attribute | Type | Description | +|-------------------|--------|----------------------------------| +| authSetting | string | Return user authorization result | + +## 8.3 Vehicle interface requirements + +[VIR-001] The vehicle interface is required to have vehicle speed interface, vehicle fuel interface, vehicle identification number (VIN) interface, vehicle wheel rotation interface and vehicle information interface. + +### 8.3.1 Vehicle speed interface + +The getSpeed object is used to retrieve the vehicle speed. + +The interface is shown below and the parameters are shown in Table 11. + +`getSpeed(Object object)` +**Table 11 – Vehicle speed parameter** + +| Element/attribute | Type | Description | Necessity | +|-------------------|----------|-----------------|-----------| +| success | function | Return success | O | +| fail | function | Return fail | O | +| complete | function | Return complete | O | + +### 8.3.2 Vehicle fuel interface + +The getRemainPower object is used to retrieve the vehicle fuel. + +The interface is shown below, and the parameters are shown in Table 12. + +`getRemainPower(Object object)` +**Table 12 – Vehicle fuel parameter** + +| Element/attribute | Type | Description | Necessity | +|-------------------|----------|-----------------|-----------| +| success | function | Return success | O | +| fail | function | Return fail | O | +| complete | function | Return complete | O | + +#### 8.3.3 Vehicle VIN interface + +The getVehicleId object is used to retrieve the VIN. + +The interface is shown below and the parameters are shown in Table 13. + +``` +getVehicleId(Object object) +``` + +**Table 13 – Vehicle VIN parameter** + +| Element/attribute | Type | Description | Necessity | +|-------------------|----------|-----------------|-----------| +| success | function | Return success | O | +| fail | function | Return fail | O | +| complete | function | Return complete | O | + +#### 8.3.4 Vehicle wheel rotation interface + +The getWheelRotationAngle object is used to retrieve the vehicle wheel rotation. + +The interface is shown below, and the parameters are shown in Table 14. + +``` +getWheelRotationAngle(Object object) +``` + +**Table 14 – Vehicle wheel rotation parameter** + +| Element/attribute | Type | Description | Necessity | +|-------------------|----------|-----------------|-----------| +| success | function | Return success | O | +| fail | function | Return fail | O | +| complete | function | Return complete | O | + +### 8.3.5 Vehicle information interface + +The addVehicleEventListener object is used to retrieve the vehicle authorization information. + +The interface is shown below and the parameters are shown in Table 15. + +``` +addVehicleEventListener(Object object, Function callback) +``` + +**Table 15 – Vehicle information parameter** + +| Element/attribute | Type | Description | +|-------------------|-------|--------------------| +| speed | array | Get speed change | +| keyCode | array | Get keycode change | + +## 8.4 Navigation interface requirements + +[NIR-001] The navigation interface is required to have the interface to provide the navigation map. + +The navigation interface is shown below, and the parameters are shown in Table 16. + +``` +navigateMap(Object object) +``` + +**Table 16 – Map parameter** + +| Element/attribute | Type | Description | +|-------------------|--------|-------------| +| destination | object | Destination | + +The destination parameters are shown in Table 17. + +**Table 17 – Destination parameter** + +| Element/attribute | Type | Description | Necessity | +|-------------------|--------|-------------|-----------| +| latitude | number | Latitude | M | +| longitude | number | Longitude | M | +| address | string | Address | M | +| name | string | Name | O | + +## 8.5 Speech interface requirements + +[SIR-001] The speech interface is required to have play TTS interface, pause TTS interface, resume TTS interface, TTS state change interface and stop TTS interface. + +### 8.5.1 Play TTS interface + +The play TTS interface is shown below, and the parameters are shown in Table 18. + +``` +playTTS(Object object) +``` + +**Table 18 – PlayTTS parameter** + +| Element/attribute | Type | Description | Necessity | +|-------------------|----------|--------------------------------------------------------------|-----------| +| connect | string | Play TTS | M | +| playAsMedia | boolean | True for long TTS broadcast
False for short TTS broadcast | O | +| volume | number | Volume | O | +| success | string | Return success | O | +| fail | function | Return fail | O | +| complete | function | Return complete | O | + +### 8.5.2 Pause TTS interface + +The pause TTS interface is shown below, and the parameters are shown in Table 19. + +``` +pauseTTS(Object object) +``` + +**Table 19 – Pause TTS parameter** + +| Element/attribute | Type | Description | Necessity | +|-------------------|----------|-----------------|-----------| +| success | function | Return success | O | +| fail | function | Return fail | O | +| complete | function | Return complete | O | + +### 8.5.3 Resume TTS interface + +The resume TTS interface is shown below, and the parameters are shown in Table 20. + +``` +resumeTTS(Object object) +``` + +**Table 20 – Resume TTS parameter** + +| Element/attribute | Type | Description | Necessity | +|-------------------|----------|-----------------|-----------| +| success | function | Return success | O | +| fail | function | Return fail | O | +| complete | function | Return complete | O | + +### 8.5.4 TTS state change interface + +The TTS state change interface is shown below, and the parameters are shown in Table 21. + +``` +onTTSStateChange(Function callback) +``` + +**Table 21 – TTS state change callback parameter** + +| Element/attribute | Type | Description | +|-------------------|--------|------------------------------------------------------------------------------------------| +| state | string | 'playing'
'paused'
'stopped'
'complete'
'error'
'pretrack'
'nexttrack' | + +### 8.5.5 Stop TTS interface + +The stop TTS interface is shown below and the parameters are shown in Table 22. + +``` +stopTTS(Object object) +``` + +**Table 22 – Stop TTS parameter** + +| Element/attribute | Type | Description | Necessity | +|-------------------|--------|----------------|-----------| +| msgid | string | Return playTTS | M | + +## 8.6 Multimedia interface requirements + +[MIR-001] The multimedia interface is required to have play music interface, get playlist interface and music play process interface. + +### 8.6.1 Play music interface + +The play music interface is shown below, and the parameters are shown in Table 23. + +``` +playmusic(Object object) +``` + +**Table 23 – Music parameter** + +| Element/attribute | Description | Necessity | +|-------------------|-------------------------------------------|-----------| +| operationType | "play", "pause", "stop", "resume", "seek" | M | +| src | Multimedia source | M | +| title | Multimedia title | M | +| coverImgUrl | Cover image URL | M | + +**Table 23 – Music parameter** + +| Element/attribute | Description | Necessity | +|-------------------|---------------------------------------------------------------------------------------------------|-----------| +| singer | Singer name | M | +| epname | Album name | M | +| currentTime | Time and location of media content (second) | M | +| state | "pause", "stop", "play", "timeupdate", "waiting",
"error", "occupied", "pretrack", "nexttrack" | M | +| errCode | Error code (Appendix I) | O | +| errMsg | Error message | O | +| paused | Paused | O | +| duration | Multimedia duration | O | +| currentTime | Current time | O | +| buffered | Buffer | O | + +### 8.6.2 Get playlist interface + +The get playlist interface is shown below, and the parameters are shown in Table 24. + +``` +getplaylist(Object object) +``` + +**Table 24 – Playlist parameter** + +| Element/attribute | Type | Description | Necessity | +|-------------------|--------|----------------------------------------------------------------------------------------------|-----------| +| title | string | Media content title | M | +| singer | string | Singer | M | +| coverImgUrl | string | Album URL | O | +| dataUrl | string | Media content source URL | M | +| options | string | Custom fields used by developers to store custom
information describing the current album | O | + +### 8.6.3 Music play process interface + +The message sequence among Vehicle OS, VMMA framework and VMMA client in-vehicle are shown in Figure 2. + +![Sequence diagram showing the VMMA message sequence for music play in-vehicle. Lifelines: Vehicle OS, VMMA framework, VMMA client. The sequence involves 8 steps: 1. VMMA client sends a multimedia playback control message to the VMMA framework. 2. The VMMA framework parses the message and generates a playback control request. 3. The VMMA framework sends the request to the Vehicle OS. 4. The Vehicle OS plays the multimedia content. 5. The VMMA client sends a multimedia state retrieval request to the VMMA framework. 6. The VMMA framework gets the real-time state of the multimedia. 7. The VMMA framework sends the real-time state to the VMMA client. 8. The VMMA client updates and displays the multimedia content playing state. The diagram is labeled F.749.8(24).](9c6461e1e94afae4dec455e69a2ce152_img.jpg) + +``` + +sequenceDiagram + participant Vehicle OS + participant VMMA framework + participant VMMA client + VMMA client->>VMMA framework: 1. Multimedia playback control message + Note over VMMA framework: 2. Parses multimedia playback control message +and generates playback control request + VMMA framework->>Vehicle OS: 3. Multimedia playback control request + Note over Vehicle OS: 4. Play multimedia content + VMMA client->>VMMA framework: 5. Multimedia state retrieval request + Note over VMMA framework: 6. Get the real-time state of multimedia + VMMA framework->>VMMA client: 7. Sends multimedia real-time state + Note over VMMA client: 8. Update and display multimedia +content playing state + Note right of VMMA client: F.749.8(24) + +``` + +Sequence diagram showing the VMMA message sequence for music play in-vehicle. Lifelines: Vehicle OS, VMMA framework, VMMA client. The sequence involves 8 steps: 1. VMMA client sends a multimedia playback control message to the VMMA framework. 2. The VMMA framework parses the message and generates a playback control request. 3. The VMMA framework sends the request to the Vehicle OS. 4. The Vehicle OS plays the multimedia content. 5. The VMMA client sends a multimedia state retrieval request to the VMMA framework. 6. The VMMA framework gets the real-time state of the multimedia. 7. The VMMA framework sends the real-time state to the VMMA client. 8. The VMMA client updates and displays the multimedia content playing state. The diagram is labeled F.749.8(24). + +**Figure 2 – The VMMA message sequence for music play in-vehicle** + +1. The VMMA client sends the music playback control message to the VMMA framework. +2. The VMMA framework parses the music playback control message and generates the music playback control request. +3. The VMMA framework sends the music playback control request to the vehicle operating system (OS). +4. The vehicle OS plays and controls the music according to the music control request. +5. The VMMA client sends a music state retrieval request to the VMMA framework. +6. The VMMA framework obtains the real-time status of music according to the music state retrieval request. +7. The VMMA framework sends the real-time state of music to the VMMA client. +8. The VMMA client updates and displays the multimedia real-time state. + +## 8.7 Display requirements + +[DIS-001] The VMMA is required to ensure the width-to-height ratio of the visible area. + +NOTE – The car head unit screen has a variety of sizes, and the screen orientation is divided into horizontal screen and vertical screen. + +## 8.8 Performance monitoring requirements + +[PMC-001] The VMMA is required to support the monitoring functions for the start-up time, memory usage and open failure rate statistics. + +# Appendix I + +## Error codes + +(This appendix does not form an integral part of this Recommendation.) + +Table I.1 shows the error code returned when the VMMA runs abnormally. + +**Table I.1 – Error codes** + +| errCode(string) | Description | +|------------------------|------------------------| +| -404 | Account error | +| -403 | Need to login | +| -402 | Illegal permissions | +| -401 | Denied authorization | +| -1 | Service error | +| 3 | Parameter error | +| 5 | Authorization error | +| 101 | User ID does not exist | +| 104 | Invalid package | +| 153 | Invalid session key | +| 250 | Car id does not exist | +| 317 | Logged in | +| 321 | Not login | +| 408 | Request timeout | + +# Appendix II + +## User interface examples + +(This appendix does not form an integral part of this Recommendation.) + +Clauses II.1 to II.19 show the user interface examples. + +## II.1 POST request + +``` +POST request: +{ + "appid":"60258", + "nonce":"abcde6789", + "timestamp":"1606286021", + "version":"1.0", + "code":"29225ba02ee811eba252b55ce8cf45c8", + "grant_type":"authorization_code", + "sign":"626802ae19bd45a33855594b3d2450df" +} +``` + +## II.2 Login interface + +``` +login({ + success (res) { + if (res.code) { + wx.request({ + url: 'https://example.com/onLogin', + data: { + code: res.code + } + }) + } else { + console.log('Login Failed' + res.errMsg) + } + } +}) +``` + +## II.3 User information interface + +``` +getUserInfo({ + success: function(res) { + var userInfo = res.userInfo + var nickName = userInfo.nickName + var avatarUrl = userInfo.avatarUrl + var gender = userInfo.gender + var province = userInfo.province + var city = userInfo.city + var country = userInfo.country + } +}) +encryptedData +{ + "openId": "OPENID", + "nickName": "NICKNAME", + "gender": GENDER, + "city": "CITY", + "province": "PROVINCE", + "country": "COUNTRY", + "avatarUrl": "AVATARURL", + "watermark": { + "appid": "APPID", +``` + +``` + "timestamp": TIMESTAMP + } +} +``` + +## **II.4 System information interface** + +``` +getSystemInfo({ + success (res) { + console.log(res.device) + } +}) +try { + const res = wx.getSystemInfoSync() + console.log(res.device) +} catch (e) { + +} +``` + +## **II.5 Login expire interface** + +``` +checkSession({ + success () { + //session_key + }, + fail () { + // session_key + login() + } +}) +``` + +## **II.6 User setting interface** + +``` +getSetting({ + success (res) { + console.log(res.authSetting) + res.authSetting = { + "scope.userInfo": true, + "scope.userLocation": true + } + } +}) +``` + +## **II.7 Vehicle speed interface** + +``` +getSpeed({ + success: function(res) { + console.log('get vehicle speed', res); + } +}) +``` + +## **II.8 Vehicle fuel interface** + +``` +getRemainPower({ + success: function(res) { + console.log('get vehicle fuel', res); + } +}) +``` + +## **II.9 VIN interface** + +``` +getVechicleId({ + success: function(res) { + console.log('get vehicle vin', res); + } +}) +``` + +## II.10 Vehicle wheel rotation interface + +``` +getWheelRotationAngle({ + success: function(res) { + console.log('get vehicle wheel rotation', res); + } +}) +``` + +## II.11 Vehicle information interface + +``` +addVehicleEventListener({keyCode: ['keycode'], speed: ['speed']}, +function(data) { + data.forEach(function(item) { + if (item.hasOwnProperty('speed')) { + console.log('vehicle speed', item.speed) + } + if (item.hasOwnProperty('keycode')) { + console.log('keycode', item.keycode) + } + }) +}) +``` + +## II.12 Navigation interface + +``` +const tamLatitude = 39.908692; +const tamLongitude = 116.397433; +const tamAddress = "XXXX"; + +navigateMap({ + destination: { + latitude: tamLatitude, + longitude: tamLongitude, + address: "xxxx" + } +}) +``` + +## II.13 Play TTS interface + +``` +playTTS({ + content: 'find the gas station', + volume: 50, + success (res) { } +}) +``` + +## II.14 Pause TTS interface + +``` +pauseTTS({ + success (res) { } +}) +``` + +## II.15 Resume TTS interface + +``` +resumeTTS({ + success (res) { } +}) +``` + +## II.16 TTS state change interface + +``` +onTTSStateChange(function(res) { + console.log(JSON.stringify(res.state)); + // Output: {"state":"COMPLETE"} +}) +``` + +## II.17 Stop TTS interface + +``` +stopTTS({ + msgId: this.data.msgId, + success: function (res) { + console.log(res) + } +}) +``` + +## II.18 Playlist interface + +``` +var curPlayList = [{ + + coverImgUrl:'http://image.biaobaiju.com/uploads/20180803/23/153 +3309822-GCcDphRmqw.jpg', + title: 'xxxx1', + singer: 'xxxx1', + dataUrl: +'http://music.163.com/song/media/outer/url?id=476592630.mp3', + options: JSON.stringify({ + epname: 'xxxx' + }) +}, +{ + coverImgUrl: +'https://qpic.y.qq.com/music_cover/ezXpob9biaedyoUWFJDttEQc9Hsq +SUpIiaPVnKrwkl94QrXna9xZ5wbg/300?n=1', + title: 'xxxx2', + singer: 'xxxx2', + dataUrl: 'https://cdn.kaishuhezi.com/audio/story/M-2-01- +20150408.mp3' +}, +{ + coverImgUrl: +'https://qpic.y.qq.com/music_cover/HvzXRphXYwDIM7tpX0XM4dCF2K6t +ulk38ts4H8sD5icicTtoeZPsM09g/300?n=1', + title: 'xxxx3', + singer: 'xxxx3', + dataUrl: 'https://cdn.kaishuhezi.com/audio/story/M-1-01- +20150408.mp3' +} +]; +const audioPlayer = wx.getBackgroundAudioManager() +audioPlayer.playInfo = { + playList:curPlayList, + context: JSON.stringify({key:1000,timeStamp:536760}) +} +audioPlayer.src = +"http://music.163.com/song/media/outer/url?id=476592630.mp3"; +``` + +## II.19 Display mode + +``` +
+ if (window.width > window.height) { + return 'horizontal' + } else { + return 'portrait' + } +
+``` + +# **Appendix III** + +## **Use case of in-vehicle multimedia applets** + +(This appendix does not form an integral part of this Recommendation.) + +## **III.1 Navigation** + +Distribution of in-vehicle multimedia applet is based on LBS geofence. + +The VMMA could support home page, search, navigation and map functions. The VMMA could reply to a real-time traffic information query and points of interest query for navigation. + +## **III.2 Voice control** + +Voice control of VMMA supports voice switch, voice control page up and down, and page return. + +Voice control of VMMA containing multimedia players supports play, pause, up and down, fast forward, fast back and other operations. + +# Bibliography + +- [b-ITU-T F.749.3] +Recommendation ITU-T F.749.3 (2020), *Use cases and requirements for vehicular multimedia networks.* +- [b-IEEE 802.11a] +Standard, IEEE 802.11a (1999), *Telecommunications and Information Exchange Between Systems – LAN/MAN Specific Requirements – Part 11: Wireless medium access control (MAC) and physical layer (PHY) specifications: High speed physical layer in the 5 GHz band.* +- [b-IEEE 802.11b] +Standard, IEEE 802.11b (2001), *Telecommunications and information exchange between systems – Local and Metropolitan networks – Specific requirements – Part 11. Cor-1: Wireless LAN medium access control (MAC) and physical layer (PHY) specifications: Higher speed physical layer (PHY) extension in the 2.4 GHz band – corrigendum 1.* +- [b-IEEE 802.11g] +Standard, IEEE 802.11g (2003), *Telecommunications and Information Exchange between Systems – Local and Metropolitan Area Networks – Specific Requirements – Part 11: Wireless LAN medium access control (MAC) and physical layer (PHY) specifications: Further higher data rate extension in the 2.4 GHz band.* +- [b-IEEE 802.11n] +Standard, IEEE 802.11n (2009), *Information technology – Local and metropolitan area networks – Specific requirements – Part 11: Wireless LAN medium access control (MAC) and physical layer (PHY) specifications Amendment 5: Enhancements for higher throughput.* +- [b-IEEE 802.11ac] +Standard, IEEE 802.11ac (2013), *Information technology – Telecommunications and information exchange between systems – Local and metropolitan area networks--Specific requirements – Part 11: Wireless LAN medium access control (MAC) and physical layer (PHY) specifications – amendment 4: Enhancements for very high throughput for operation in bands below 6 GHz.* +- [b-IEEE 802.11ax] +Standard, IEEE 802.11ax (2021), *Information Technology – Telecommunications and Information Exchange between Systems Local and Metropolitan Area Networks – Specific Requirements Part 11: Wireless LAN medium access control (MAC) and physical layer (PHY) specifications – Amendment 4: Enhancements for very high throughput for operation in bands below 6 GHz.* + + + +# SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.750-200502-I_PDF-E/1439cb942d9e363bbb3161b5540dd8c6_img.jpg b/marked/F/T-REC-F.750-200502-I_PDF-E/1439cb942d9e363bbb3161b5540dd8c6_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a8bbe4a93562f1624c3f62d4ac2567a990c53e0c --- /dev/null +++ b/marked/F/T-REC-F.750-200502-I_PDF-E/1439cb942d9e363bbb3161b5540dd8c6_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ecc24b4dda71678d497159a502d06364e4f08e81964b03ed9ab8e3652251c55c +size 37423 diff --git a/marked/F/T-REC-F.750-200502-I_PDF-E/19a5f0db57a21a0e82a7f326083e96fd_img.jpg b/marked/F/T-REC-F.750-200502-I_PDF-E/19a5f0db57a21a0e82a7f326083e96fd_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..6c1bcad4911e5f1b1a6a90b17d92ae810ad6544c --- /dev/null +++ b/marked/F/T-REC-F.750-200502-I_PDF-E/19a5f0db57a21a0e82a7f326083e96fd_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aefaab60beb41d87be1ad9029584ee4a95d16efe1db6f1f018533e88b52fa232 +size 65941 diff --git a/marked/F/T-REC-F.750-200502-I_PDF-E/5cab96b2d23174c25919840ecd50aa48_img.jpg b/marked/F/T-REC-F.750-200502-I_PDF-E/5cab96b2d23174c25919840ecd50aa48_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..60fa39696d46e3e7c4c0eaa4206b9c281c1aaf21 --- /dev/null +++ b/marked/F/T-REC-F.750-200502-I_PDF-E/5cab96b2d23174c25919840ecd50aa48_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:16de320d7a57cc2669da35cdfa10f552c8b33b3f2b66c78a86476484bcb05d11 +size 52556 diff --git a/marked/F/T-REC-F.750-200502-I_PDF-E/6f31cdb576d2f15c35c3f266e5f59211_img.jpg b/marked/F/T-REC-F.750-200502-I_PDF-E/6f31cdb576d2f15c35c3f266e5f59211_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..5bad8e3238fa4d7c3e6dab12092a567f80a33b5a --- /dev/null +++ b/marked/F/T-REC-F.750-200502-I_PDF-E/6f31cdb576d2f15c35c3f266e5f59211_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4e9d5cf283395dc8c798e6c404a26db6eca0993fcddf33027c0c25b46c6327c8 +size 41244 diff --git a/marked/F/T-REC-F.750-200502-I_PDF-E/70de8885bd7de15723aaad5eb5c5febf_img.jpg b/marked/F/T-REC-F.750-200502-I_PDF-E/70de8885bd7de15723aaad5eb5c5febf_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a5f98be5c9df9cd7f78ab889036b9be43088e3db --- /dev/null +++ b/marked/F/T-REC-F.750-200502-I_PDF-E/70de8885bd7de15723aaad5eb5c5febf_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c0fd1469af764a3fbfe322bd79f5b5fee6d543001d9424acece8dfa4d4c14dd8 +size 88542 diff --git a/marked/F/T-REC-F.750-200502-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.750-200502-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d28e17abc54484c3c7a0cb4d6e3600171d8fcc46 --- /dev/null +++ b/marked/F/T-REC-F.750-200502-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3b725e024f1a260e574fd481b7382c2eaf2eaf30908541c1191545a09717a6f6 +size 8421 diff --git a/marked/F/T-REC-F.750-200502-I_PDF-E/8d325fc12b494e42c9ea7ed2a7f327a6_img.jpg b/marked/F/T-REC-F.750-200502-I_PDF-E/8d325fc12b494e42c9ea7ed2a7f327a6_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..47f2b37df0f1402ea90e3f791771d954b83138b6 --- /dev/null +++ b/marked/F/T-REC-F.750-200502-I_PDF-E/8d325fc12b494e42c9ea7ed2a7f327a6_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a78f1b0d0d23da6e8009a67d888a48e0c4093c4484a5a07c3c5b8983c46ec7fb +size 16995 diff --git a/marked/F/T-REC-F.750-200502-I_PDF-E/9c6461e1e94afae4dec455e69a2ce152_img.jpg b/marked/F/T-REC-F.750-200502-I_PDF-E/9c6461e1e94afae4dec455e69a2ce152_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a90df22137f2ac5ba022634d547cb97f23decc68 --- /dev/null +++ b/marked/F/T-REC-F.750-200502-I_PDF-E/9c6461e1e94afae4dec455e69a2ce152_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b808f2ae61ec5c8154ffc48306f4bf0644978ca2e3ff2d5234e6e8ea273bec67 +size 40334 diff --git a/marked/F/T-REC-F.750-200502-I_PDF-E/9cd90f495b95ad2116ff780248c26d95_img.jpg b/marked/F/T-REC-F.750-200502-I_PDF-E/9cd90f495b95ad2116ff780248c26d95_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..61e9ff7e5ba6568884e29e3de8efa203ea01d050 --- /dev/null +++ b/marked/F/T-REC-F.750-200502-I_PDF-E/9cd90f495b95ad2116ff780248c26d95_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c7200146310dfd38b97fda946d89bc4ab92f7972e0080cfddd1d3b2e83f6c82f +size 61560 diff --git a/marked/F/T-REC-F.750-200502-I_PDF-E/a26e142d3df5bef41a84a9dd099d7825_img.jpg b/marked/F/T-REC-F.750-200502-I_PDF-E/a26e142d3df5bef41a84a9dd099d7825_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..da4a57f044f57d9d5e17e481e0e15eae38e49d00 --- /dev/null +++ b/marked/F/T-REC-F.750-200502-I_PDF-E/a26e142d3df5bef41a84a9dd099d7825_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:55dd1c9f75f4ec77523c9f7fb679c97149deb68816bc730b0e3b441da1826d91 +size 92927 diff --git a/marked/F/T-REC-F.750-200502-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg b/marked/F/T-REC-F.750-200502-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..25ed4732fde1d57beaee1b7c2be212f8f7710ee9 --- /dev/null +++ b/marked/F/T-REC-F.750-200502-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8ecbab690be10e683d304b5537eada831d71c391e5481e11cd7829a2f95475c2 +size 94224 diff --git a/marked/F/T-REC-F.750-200502-I_PDF-E/cfda9df1319e04207eb28bcefd1dab7b_img.jpg b/marked/F/T-REC-F.750-200502-I_PDF-E/cfda9df1319e04207eb28bcefd1dab7b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d5887aa93300da77cb040d4c9b43b8f4e6729944 --- /dev/null +++ b/marked/F/T-REC-F.750-200502-I_PDF-E/cfda9df1319e04207eb28bcefd1dab7b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ea2ace0b493895353123149302d4ff8f2ac40d8bd7b36fdc3e4a3243bfd1b619 +size 77958 diff --git a/marked/F/T-REC-F.750-200502-I_PDF-E/d4af765160d04ecef538e5066006dc77_img.jpg b/marked/F/T-REC-F.750-200502-I_PDF-E/d4af765160d04ecef538e5066006dc77_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1ec8cee2059b86b86e5dec2b87f926b4e0065456 --- /dev/null +++ b/marked/F/T-REC-F.750-200502-I_PDF-E/d4af765160d04ecef538e5066006dc77_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:177ded2f0c90f856aef5d32f0273935ee88fea87c4ee51549e9e2d07ed613574 +size 40347 diff --git a/marked/F/T-REC-F.750-200502-I_PDF-E/e69b9188aa2c14ec6b21c83f711fef65_img.jpg b/marked/F/T-REC-F.750-200502-I_PDF-E/e69b9188aa2c14ec6b21c83f711fef65_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..780b589b29a886dd0a70108fce318e67b9e531af --- /dev/null +++ b/marked/F/T-REC-F.750-200502-I_PDF-E/e69b9188aa2c14ec6b21c83f711fef65_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7d22af0544b3b5a7fe9846f13e7869b9eeb52bede4a38c01c87711f5766cc848 +size 102245 diff --git a/marked/F/T-REC-F.750-200502-I_PDF-E/ff0952ef692c9d960ce5f6708bcc9711_img.jpg b/marked/F/T-REC-F.750-200502-I_PDF-E/ff0952ef692c9d960ce5f6708bcc9711_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7633564f36cee5cf96b872a8a813594bb268de6e --- /dev/null +++ b/marked/F/T-REC-F.750-200502-I_PDF-E/ff0952ef692c9d960ce5f6708bcc9711_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3bf2ebf2caeaa65bb436d784d8513be64cf0a861dc74645a760ac79af3005e5b +size 79027 diff --git a/marked/F/T-REC-F.750-200502-I_PDF-E/raw.md b/marked/F/T-REC-F.750-200502-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..5d6a3b5d9be1a0b5baad1dfaf94b5aaf203f2ce4 --- /dev/null +++ b/marked/F/T-REC-F.750-200502-I_PDF-E/raw.md @@ -0,0 +1,794 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.750** + +(02/2005) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Audiovisual services + +# --- Metadata framework + +ITU-T Recommendation F.750 + +![ITU logo: A globe with a red lightning bolt, next to the text 'ITU International Telecommunication Union'.](84a1d09fb489061482111515543b60dc_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with a red lightning bolt striking across it. To the right of the globe, the text "ITU" is written in a large, bold, blue font, and below it, "International Telecommunication Union" is written in a smaller, blue font. + +ITU logo: A globe with a red lightning bolt, next to the text 'ITU International Telecommunication Union'. + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +## **TELEGRAPH SERVICE** + +| | | +|-----------------------------------------------------------------|-------------| +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | + +## **MOBILE SERVICE** + +| | | +|--------------------------------------------------------|-------------| +| Mobile services and multdestination satellite services | F.110-F.159 | +|--------------------------------------------------------|-------------| + +## **TELEMATIC SERVICES** + +| | | +|-------------------------------------------|-------------| +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | + +## **MESSAGE HANDLING SERVICES** + +F.400-F.499 + +## **DIRECTORY SERVICES** + +F.500-F.549 + +## **DOCUMENT COMMUNICATION** + +| | | +|--------------------------------------|-------------| +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | + +## **DATA TRANSMISSION SERVICES** + +F.600-F.699 + +## **AUDIOVISUAL SERVICES** + +**F.700-F.799** + +## **ISDN SERVICES** + +F.800-F.849 + +## **UNIVERSAL PERSONAL TELECOMMUNICATION** + +F.850-F.899 + +## **HUMAN FACTORS** + +F.900-F.999 + +*For further details, please refer to the list of ITU-T Recommendations.* + +# **ITU-T Recommendation F.750** + +# **Metadata framework** + +# **Summary** + +A metadata framework architecture with two layers structure is defined. The metadata gateway provides an integrated metadata retrieval capability across various metadata descriptions, and the policy-based service platform provides specific functions common to content delivery. This Recommendation provides a reference architecture of a metadata model for both content description and network control. + +## **Source** + +ITU-T Recommendation F.750 was approved on 13 February 2005 by ITU-T Study Group 16 (2005-2008) under the ITU-T Recommendation A.8 procedure. + +# FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications. The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure e.g. interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementors are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database. + +© ITU 2005 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +# CONTENTS + +| | Page | +|--------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 2.1 Normative references..... | 1 | +| 2.2 Non-normative references ..... | 1 | +| 3 Definition..... | 2 | +| 4 Abbreviations..... | 2 | +| 5 Advanced content delivery service architecture ..... | 3 | +| 5.1 Advanced content delivery ..... | 3 | +| 5.2 Network-related metadata ..... | 5 | +| 5.3 Role and application of metadata in network ..... | 6 | +| 6 Metadata framework..... | 7 | +| 6.1 Metadata gateway ..... | 7 | +| 6.2 Service platform using metadata ..... | 10 | +| Appendix I – Policy-based QoS control using metadata ..... | 11 | +| I.1 Model of policy-based QoS control using metadata ..... | 12 | +| I.2 Implementation of QoS control..... | 14 | +| Appendix II – Policy-based CDN management using metadata..... | 15 | +| II.1 Directory database as metadata retrieval infrastructure ..... | 15 | +| II.2 Metadata for CDN optimization..... | 16 | +| II.3 Policy descriptions in CDN management..... | 17 | +| II.4 Interoperability of metadata across different domains ..... | 19 | +| II.5 Service platform API..... | 19 | + + + +# Metadata framework + +# 1 Scope + +This Recommendation defines an ITU-T metadata framework architecture for policy-based content delivery over networks. Content includes multimedia Web content and digital broadcast programmes provided by real-time streaming, on-demand streaming, and downloading. In broadband ubiquitous networks, this architecture enables content adaptation to the usage environment and content-aware QoS control, which are provided as network services by the use of policies, rules and network-related metadata. The proposed metadata framework architecture is structured in two layers: a metadata gateway and a policy-based service platform each of which provides open APIs to the next higher layer. The metadata gateway provides an integrated metadata retrieval capability across various metadata descriptions and the policy-based service platform provides specific functions common to content delivery such as compatible content discovery, session control, QoS control, authentication and charging. As implementation examples of the metadata framework architecture, policy-based QoS control and CDN management are addressed. + +# 2 References + +## 2.1 Normative references + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +References: None. + +## 2.2 Non-normative references + +- [1] IETF RFC 2327 (1998), *SDP: Session Description Protocol*. +- [2] ETSI TS 102 822-3-1 (2004), *Broadcast and On-line Services: Search, select, and rightful use of content on personal storage systems ("TV-Anytime Phase 1"); Part 3: Metadata; Sub-part 1: Metadata schemas*. +- [3] ISO/IEC JTC1/SC29/WG11 2002, *N4980 MPEG-7 Overview (version 8)*. +- [4] W3C Recommendation (2004), *Composite Capabilities/Preference Profiles (CC/PP), Structure and Vocabularies*. +- [5] IETF RFC 2778 (2000), *A Model for Presence and Instant Messaging*. +- [6] IETF RFC 2779 (2000), *Instant Messaging/Presence Protocol Requirements*. +- [7] ISO/IEC 21000-7(2004), *Information technology – Multimedia framework (MPEG-21) – Part 7: Digital Item Adaptation*. +- [8] IETF RFC 1213 (1991), *Management Information Base for Network Management of TCP/IP-based internets: MIB-II*. +- [9] PAM Forum, . + +- [10] ITU-T Recommendation H.350 (2003), *Directory services architecture for multimedia conferencing*. +- [11] IETF RFC 3564 (2003), *Requirements for Support of Differentiated Services-aware MPLS Traffic Engineering*. +- [12] IETF RFC 3466 (2003), *A Model for Content Internetworking (CDI)*. +- [13] ETSI TS 102 822-4 (2004), *Broadcast and On-line Services: Search, select, and rightful use of content on personal storage systems ("TV-Anytime Phase 1"); Part 4: Content referencing*. + +# 3 Definitions + +This Recommendation defines the following terms: + +- 3.1 content delivery network (CDN):** A network optimized for delivering digital content. +- 3.2 differentiated service code point (DSCP):** A 6-bit bit-pattern replacing the three IP-Precedence bits and other ToS (Type of Service) bits for differentiated services. +- 3.3 label switch paths (LSP):** Virtual paths between pairs of edges prepared in MPLS. +- 3.4 metadata:** "Metadata" in this Recommendation refers to the attributes of not only the content but also the network. +- 3.5 resource description framework (RDF):** A general framework for describing a Web site's metadata, or the information about the information on the site. + +# 4 Abbreviations + +This Recommendation uses the following abbreviations: + +| | | +|--------|--------------------------------------------| +| 3GPP | 3rd Generation Partnership Project | +| API | Application Programming Interface | +| APL | Application Programming Language | +| CC | Call Control | +| CC/PP | Composite Capabilities/Preference Profiles | +| CDN | Content Delivery Network | +| CID | Content Identification | +| DSCP | Differentiated Services Code Point | +| DSL | Digital Subscriber Line | +| HTTP | HyperText Transfer Protocol | +| IMPP | Instant Messaging and Presence Protocol | +| LSP | Label Switch Paths | +| MIB | Management Information Base | +| MPLS | Multi Protocol Label Switching | +| OWL | Web Ontology Language | +| PAM | Presence and Availability Management | +| PBCDNM | Policy-based CDN Management | + +| | | +|------|-----------------------------------| +| PC | Personal Computer | +| PDA | Personal Digital Assistant | +| QoS | Quality of Service | +| RDB | Relational Database | +| RDF | Resource Description Framework | +| SDP | Session Description Protocol | +| SIP | Session Initiation Protocol | +| SLA | Service Level Agreement | +| SOAP | Simple Object Access Protocol | +| SQL | Structured Query Language | +| ToS | Type of Service | +| UI | User Interaction | +| UNI | User-Network Interface | +| W3C | World Wide Web Consortium | +| WSDL | Web Services Description Language | +| XML | eXtensible Markup Language | +| XQL | XML Query Language | + +# 5 Advanced content delivery service architecture + +## 5.1 Advanced content delivery + +Owing to rapid progress in broadband and wireless Internet access technologies, various kinds of digital content are being delivered to diverse terminal devices (PDAs, PCs, cellular phones, etc.) that are connected by diverse access methods (DSL, wireless, optical-fibre, etc.). Furthermore, digital broadcasting will soon arrive in such an environment. Figure 1 shows content delivery in a ubiquitous environment. + +In such a configuration, it is desirable that the network automatically detects the user's usage environment and adapts the content delivery to allow convenient viewing. It is also desirable that the network recognizes the user's preferences for delivery method and quality and adapts to them. To achieve such advanced content delivery, the network requires context understanding and content adaptation. Figure 2 shows a mechanism for advanced content delivery by the network. Conditions on the content side, user side, and network side are described in metadata. When a user requests content, the network gathers relevant metadata to understand the conditions. Then, for those conditions, it optimizes content delivery according to pre-defined policies. + +This Recommendation provides a standardized framework for achieving this mechanism. + +![Diagram of content delivery in a ubiquitous environment showing various access points and services.](a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg) + +This diagram illustrates the content delivery architecture in a ubiquitous environment. At the top, three service boxes are shown: **Digital broadcasting**, **Broadband services**, and **Internet services**. These connect to a central cloud labeled **Internet, etc.**, which is part of an **Access network**. Below the network, four access scenarios are depicted: + + +- Public area**: *Mobile access* (represented by a mobile phone and signal tower). +- Office**: *Fixed/wireless access* (represented by a building and computer). +- Airport, station, etc.**: *Nomadic wireless access (NWA)* (represented by a train and signal tower). +- Home**: *Fixed/wireless access* (represented by a house and computer). + + Below the diagram, two boxes highlight the diversity of the environment: + + +- Diverse terminals to be connected**: PC, PDA, cellular phone. +- Diverse access methods**: DSL, wireless, fibre. + + The label F.750\_F01 is present in the bottom right corner of the diagram area. + +Diagram of content delivery in a ubiquitous environment showing various access points and services. + +**Figure 1/F.750 – Content delivery in ubiquitous environment** + +![Flowchart of the mechanism for advanced content delivery.](cfda9df1319e04207eb28bcefd1dab7b_img.jpg) + +This flowchart details the mechanism for advanced content delivery. It begins with **Optimization in given conditions**, which leads to a **Policy** (cylinder icon). The **Policy** feeds into **Context understanding and content adaptation** (oval icon). This central process is supported by **Description of conditions in metadata** (rectangle icon). The **Context understanding** process leads to a **Consistency** check (diamond icon). This check involves three sides: + + +- Conditions on content side**: Stream attribute, File attribute. +- Conditions on user side**: Terminal attributes (type, capability, status), Access attributes (mode, contracted bandwidth), User preferences (delivery mode, quality). +- Conditions on network side**: SLA (Service Level Agreement), Service class, Available bandwidth. + + Arrows indicate the flow of information between these conditions and the central consistency check. The label F.750\_F02 is present in the bottom right corner of the diagram area. + +Flowchart of the mechanism for advanced content delivery. + +**Figure 2/F.750 – Mechanism for advanced content delivery** + +## 5.2 Network-related metadata + +Originally, metadata is defined for efficient content retrieval and for building up clear rules in secondary use of content across different providers and/or organizations. This metadata describes attribution, usage rules, and target user profiles of content for the consumer. + +In addition to such metadata for customer and/or content provider use, this Recommendation defines a new kind of metadata for network use, which will facilitate advanced content delivery. Called "network-related metadata", it specifies various conditions for content delivery over networks such as terminal device characteristics, access network characteristics, and user preferences. + +Several standardization bodies have already defined various attribute types of information for network services that are equivalent to network-related metadata: + +- a) *Content delivery metadata* + - i) Session Description Protocol (SDP) (IETF) [1]; + - ii) Instance description metadata (TV-Anytime Forum) [2]; + - iii) Multimedia description (MPEG-7) [3]. +- b) *User description metadata* + - i) Composite Capability and Preference Profiles (CC/PP) (W3C) [4]; + - ii) Consumer metadata (TV-Anytime Forum) [2]; + - iii) User interaction (MPEG-7) [3]; + - iv) Presence information (IETF) [5], [6]; + - v) Usage environment (MPEG-21) [7]. +- c) *Terminal description metadata* + - i) Composite Capability and Preference Profiles (CC/PP) (W3C) [4]; + - ii) Usage environment (MPEG-21) [7]. +- d) *Network description metadata* + - i) Service Level Agreement (SLA); + - ii) Management Information Base (MIB-II) (IETF) [8]. + +Except for SLA and MIB-II, these types of information are described by XML (or RDF) and some are interoperable. By using this information, 3GPP specifies CC/PP-based content adaptation service for cell phones, and IETF specifies SIP (Session Initiation Protocol) that establishes dynamic sessions to a target user by using presence information. + +As these types of metadata are defined by different organizations for different application domains, metadata translation and/or matching is needed to share metadata for specific services. The objective in defining this metadata framework is not to invent new metadata, but to extract existing metadata relevant to communication and make it interoperable. + +In a similar way to content-related metadata, network-related metadata should be interoperable within the same provider and/or across different providers. The requirements of metadata are that the total structure is clearly defined in terms of language, scheme, elements, and vocabulary and is interoperable at some levels of the metadata structure shown in Figure 3. + +![Figure 3/F.750 – Metadata structure diagram. On the left, a large cylinder labeled 'Function (search, rights expression)' is connected by lines to three smaller cylinders labeled 'Vocabulary', 'Element', and 'Scheme'. Below these is another large cylinder labeled 'Language (XML, list)'. On the right, a vertical stack of three cylinders is shown: 'Content' at the top, 'Metadata' in the middle, and 'Network' at the bottom. Arrows indicate a flow from 'Content' down to 'Metadata' and from 'Metadata' up to 'Network'. A line connects the 'Function' cylinder to the 'Content' cylinder. A label 'F.750_F03' is at the bottom right.](d4af765160d04ecef538e5066006dc77_img.jpg) + +Figure 3/F.750 – Metadata structure diagram. On the left, a large cylinder labeled 'Function (search, rights expression)' is connected by lines to three smaller cylinders labeled 'Vocabulary', 'Element', and 'Scheme'. Below these is another large cylinder labeled 'Language (XML, list)'. On the right, a vertical stack of three cylinders is shown: 'Content' at the top, 'Metadata' in the middle, and 'Network' at the bottom. Arrows indicate a flow from 'Content' down to 'Metadata' and from 'Metadata' up to 'Network'. A line connects the 'Function' cylinder to the 'Content' cylinder. A label 'F.750\_F03' is at the bottom right. + +**Figure 3/F.750 – Metadata structure** + +## 5.3 Role and application of metadata in network + +Figure 4 shows, as an example, a potential usage scenario of a metadata-based network service in terms of seamless content handover. This service allows easy handover of programmes between a wide range of terminals irrespective of the specific viewing site, terminal capability, or ownership, so that the user can continue to watch a programme even when he or she changes viewing terminal. In this example, the user can continue to watch the same TV programme that he was watching at home even if he is out on the street using his PDA, or in a NetCafe using a PC available there. + +![Figure 4/F.750 – Potential usage scenarios diagram. It shows a user at 'Home' watching TV, then 'On street' using a PDA, and finally at a 'Net Cafe' using a PC. Dashed arrows show the transition between these locations. Scenario 1 is labeled 'Request handover to PDA' and Scenario 2 is labeled 'Trigger for handover automatically obtained using wireless tag'. A label 'F.750_F04' is at the bottom right.](1439cb942d9e363bbb3161b5540dd8c6_img.jpg) + +Figure 4/F.750 – Potential usage scenarios diagram. It shows a user at 'Home' watching TV, then 'On street' using a PDA, and finally at a 'Net Cafe' using a PC. Dashed arrows show the transition between these locations. Scenario 1 is labeled 'Request handover to PDA' and Scenario 2 is labeled 'Trigger for handover automatically obtained using wireless tag'. A label 'F.750\_F04' is at the bottom right. + +Scenario 1: When the user leaves home, he continues to watch, on a PDA, the item of content he was watching on TV at home. + +Scenario 2: On entering a NetCafe, the user continues to view, on a PC, the item of content he was watching on a PDA. + +F.750\_F04 + +**Figure 4/F.750 – Potential usage scenarios** + +In order to provide this service, the network must automatically detect the user's usage environment and adapt the content delivery to allow continued viewing. Users need not identify the item of content or the interruption point in the new terminal to which the content delivery is handed over. + +To meet these requirements, various conditions related to content delivery over the network should be clearly described in metadata. Network metadata provides such descriptions so that the network can understand the context. Optimization of content delivery under given conditions is the scope of policy control. When there are contradictory conditions or when it is impossible to meet conditions simultaneously, resolving policies and policy enforcing schemes for network devices need to be provided. Policy-based control using metadata can provide advanced content delivery services. + +Besides seamless content handover, a metadata-based network service can provide a variety of context-aware content delivery services. + +# 6 Metadata framework + +This Recommendation describes a metadata framework to achieve advanced content delivery. The architecture of the metadata framework is shown in Figure 5. It is structured in two layers: namely, a metadata gateway and a service platform. The metadata gateway provides integrated metadata retrieval across different domains, and the policy-based service platform provides specific functions common to content delivery such as context-aware address resolution, QoS control, authentication and charging. + +![Figure 5/F.750 – Metadata framework architecture diagram. The diagram shows a hierarchical structure. At the top, three 'Application' ovals are connected to 'Service platform A', 'Service platform B', and 'Service platform C' respectively. Each service platform is connected to a 'Policy DB' (cylinder) via a 'Platform API' and to a central 'Metadata gateway' (grey bar) via a 'Metadata GW-API'. The 'Metadata gateway' is connected to a group of 'Metadata DBs' (cylinders) enclosed in a dashed box. Below the metadata gateway, the architecture branches into network components: a 'Backbone network' connected to a 'CDN' (oval), which is then connected to an 'ISP network'. This 'ISP network' connects to 'Fixed access' (oval) leading to a 'Server' (cylinder) and 'Terminal device' (square), and to 'Wireless access' (oval) leading to another 'Terminal device' (square). Another 'ISP network' connects to a second 'Fixed access' (oval) leading to another 'Server' (cylinder). The label 'F.750_F05' is in the bottom right corner.](ff0952ef692c9d960ce5f6708bcc9711_img.jpg) + +Figure 5/F.750 – Metadata framework architecture diagram. The diagram shows a hierarchical structure. At the top, three 'Application' ovals are connected to 'Service platform A', 'Service platform B', and 'Service platform C' respectively. Each service platform is connected to a 'Policy DB' (cylinder) via a 'Platform API' and to a central 'Metadata gateway' (grey bar) via a 'Metadata GW-API'. The 'Metadata gateway' is connected to a group of 'Metadata DBs' (cylinders) enclosed in a dashed box. Below the metadata gateway, the architecture branches into network components: a 'Backbone network' connected to a 'CDN' (oval), which is then connected to an 'ISP network'. This 'ISP network' connects to 'Fixed access' (oval) leading to a 'Server' (cylinder) and 'Terminal device' (square), and to 'Wireless access' (oval) leading to another 'Terminal device' (square). Another 'ISP network' connects to a second 'Fixed access' (oval) leading to another 'Server' (cylinder). The label 'F.750\_F05' is in the bottom right corner. + +**Figure 5/F.750 – Metadata framework architecture** + +When a user requests content, the metadata framework gathers relevant network-related metadata to understand the conditions. Then, for these conditions, it optimizes content delivery by using predefined policies. More specifically, the metadata gateway gathers relevant metadata and performs translations between different metadata descriptions when needed. On top of the metadata gateway, the service platform functions perform policy-based control functions such as QoS control, session control, authentication control, and charging control, to support content delivery applications. For this purpose, the service platform API and metadata gateway API are defined as open interfaces. + +## 6.1 Metadata gateway + +To make network-related metadata universally usable, we need a gateway function for retrieving required metadata and providing the application interface. This capability is provided by the metadata gateway. It is desired that network-related metadata such as content delivery attributes, user and terminal attributes and network attributes are stored in a metadata database and managed by a directory. The directory needs to include rules for selecting metadata that meet required + +conditions and evaluation criteria that determine the order in which candidate metadata should be provided. + +Furthermore, to enable metadata defined in a specific domain be usable in other domains, a function is required that absorbs the difference in metadata descriptions and translates description schema. This capability is also needed in the metadata gateway. + +To provide the above-mentioned capabilities, the use of PAM-API defined by the Presence, Availability and Management Forum (PAM Forum) [9] is envisaged. The PAM Forum defines an open interface for application development that integrates and/or associates various communication services by enabling the use of presence and/or availability information about networks through open APIs. + +PAM-API provides operations for identity (equivalent to user) or group identity, agent (equivalent to device), agent assignment, agent presence, and identity presence, as well as preference control, event notification and access control. Security protection is another significant feature of PAM-API. + +The metadata gateway architecture based on PAM-API is shown in Figure 6. It is composed of two parts. The platform part gathers and manages network-related metadata including presence and/or availability information. On top of the platform part, the application part provides various functions by manipulating metadata and/or integrating metadata from different domains through PAM-API. + +![Diagram of Metadata gateway architecture showing internal components and external interactions.](a26e142d3df5bef41a84a9dd099d7825_img.jpg) + +The diagram illustrates the Metadata gateway architecture. At the top, a box labeled 'Metadata gateway' contains two application components, 'APL#1' and 'APL#2', which connect to a 'Platform' layer. The 'Platform' layer connects to a 'DB' (database). The 'PAM API' is indicated on the left. Annotations on the right describe the functions: 'Manipulating metadata, Associating metadata across different domains' for the applications and 'Collecting and managing metadata' for the platform/DB. Below this, a central 'Metadata gateway' is shown within an 'Internet' cloud. It interacts with a 'Monitoring terminal' via 'Availability management PAM', with a 'User terminal (presenter)' via 'Metadata registration IMPP, SIP, HTTP', and with a 'User terminal (watcher)' via 'Metadata query IMPP, SIP, HTTP'. To the right, another 'Mobile network' cloud contains its own 'Metadata gateway', which connects to two 'User terminal' devices (mobile phones). The central 'Metadata gateway' and the 'Mobile network' gateway are connected via 'Metadata exchange & integration PAM'. The label 'F.750\_F06' is in the bottom right corner. + +Diagram of Metadata gateway architecture showing internal components and external interactions. + +Figure 6/F.750 – Functions of metadata gateway + +For the platform part, this metadata framework defines the following four metadata gateway functions: + +### a) *Metadata disclosure control* + +When notified metadata does not meet the application user's requirements in terms of item and/or format, this function requests the metadata provider to disclose new metadata items and/or metadata formats. It also reports candidate metadata that meet given conditions. + +### b) *Metadata storing and gathering* + +This stores static and/or dynamic metadata information. It enables metadata information to be manipulated and acquired according to given filtering conditions. + +### c) *Metadata transformation and sharing* + +This automatically transforms given metadata items of given subjects into given formats. + +### d) *Metadata distributed management* + +Based on the intentions of metadata providers, this function enables distributed management of metadata information according to its characteristics, usages, and objectives within servers and/or terminal devices. Furthermore, it enables dynamic changes in the location of stored metadata. + +Figure 7 shows a possible architecture for the metadata transformation and sharing function achieved by the metadata gateway. Integrated retrieval across different metadata is achieved by transformation rules stored in a repository in this example. The use of semantic transformation by using an ontology would also be possible. + +![Figure 7/F.750 – Metadata gateway architecture diagram showing two service platforms (A and B) connected to a central metadata gateway. The gateway contains a Metadata Sharing Function with XML translation and integrated retrieval. It is supported by a Repository (translation table/dictionary) and connected to various metadata databases (XML, RDB) via RDB-XML mapping and SQL/XQL queries. Service A's metadata includes XML and RDB databases, while Service B's metadata includes RDB and XML databases.](70de8885bd7de15723aaad5eb5c5febf_img.jpg) + +The diagram illustrates the metadata gateway architecture. At the top, two service platforms, 'Service platform A' and 'Service platform B', are shown. Each is connected to a central 'Metadata gateway' via a 'Metadata GW-API'. The 'Metadata gateway' is a large box containing three main components: 'Metadata Management Function' on the left, 'Metadata Sharing Function' in the center, and 'Metadata Management Function' on the right. The 'Metadata Sharing Function' is further divided into 'XML translation function' and 'Integrated retrieval function'. Below the 'Metadata Sharing Function' is a 'Repository (translation table/dictionary)'. The 'Metadata Management Functions' are connected to 'RDB-XML mapping' boxes. These boxes are then connected to 'Metadata DB' boxes. For 'Service "A" metadata', there are two databases: 'Metadata DB (XML)' connected via 'XQL' and 'Metadata DB (RDB)' connected via 'SQL'. For 'Service "B" metadata', there are also two databases: 'Metadata DB (RDB)' connected via 'SQL' and 'Metadata DB (XML)' connected via 'XQL'. The 'Metadata Sharing Function' is connected to the 'Repository' and to the 'RDB-XML mapping' boxes. The 'Repository' is also connected to the 'RDB-XML mapping' boxes. The label 'F.750\_F07' is in the bottom right corner. + +Figure 7/F.750 – Metadata gateway architecture diagram showing two service platforms (A and B) connected to a central metadata gateway. The gateway contains a Metadata Sharing Function with XML translation and integrated retrieval. It is supported by a Repository (translation table/dictionary) and connected to various metadata databases (XML, RDB) via RDB-XML mapping and SQL/XQL queries. Service A's metadata includes XML and RDB databases, while Service B's metadata includes RDB and XML databases. + +Figure 7/F.750 – Metadata gateway architecture + +To achieve common expression of metadata among different industry standard schemas, integration of attribute types, reference rules, and name definitions are needed. Figure 8 shows a possible mapping between various industry standard metadata and directory object classes. To harmonize metadata across domains, firstly an ontology language such as OWL needs to describe the relationship between metadata elements that are mutually referred to, and then these elements are mapped to the same attribute type in the directory database. + +![Figure 8/F.750 – Mapping from metadata to directory object classes. The diagram shows a hierarchy of directory object classes (User entry, Device entry, CDN component entry, Content entry) at the top, each with three sub-nodes. Below each class is a metadata instance (User profile metadata instance, Device capability metadata instance, CDN component metadata instance, Content metadata instance) with an upward arrow pointing to the class. A large arrow on the right points to the classes, labeled 'Ontology (such as OWL) RDF schema'. The label F.750_F08 is in the bottom right corner.](5cab96b2d23174c25919840ecd50aa48_img.jpg) + +Figure 8/F.750 – Mapping from metadata to directory object classes. The diagram shows a hierarchy of directory object classes (User entry, Device entry, CDN component entry, Content entry) at the top, each with three sub-nodes. Below each class is a metadata instance (User profile metadata instance, Device capability metadata instance, CDN component metadata instance, Content metadata instance) with an upward arrow pointing to the class. A large arrow on the right points to the classes, labeled 'Ontology (such as OWL) RDF schema'. The label F.750\_F08 is in the bottom right corner. + +**Figure 8/F.750 – Mapping from metadata to directory object classes** + +## 6.2 Service platform using metadata + +Utilizing network-related metadata, the service platform supports advanced content delivery by providing common functions through their application interfaces. The following platform functions are envisaged: + +### a) *Address resolution function* + +When a user retrieves and selects content logically, this function resolves the location of a physical content instance that meets the user's preference and usage environment as described by metadata. + +### b) *Quality of service (QoS) control function* + +This function controls network QoS based on the content delivery configuration and user preference as described by metadata. + +### c) *Session control function* + +This function establishes a connection from a server to a given user. It further enables simultaneous use of multiple terminals and/or seamless handover between terminals. + +### d) *Secure delivery function* + +This function prevents malicious access to the content delivery service by using user, terminal, and content authentication. Furthermore, it associates the user with upper-layer services by using single-sign-on authentication. + +### e) *Charging function* + +This function gathers relevant charging information needed to charge for a particular service and achieves bundled charging. It also enables associated charging of multiple services. + +These platform functions provide their services to application via an API called the network service API. + +### **6.2.1 Service platform based on the Parlay-API** + +The Parlay-API specifies open interfaces of network functions for applications over IP-based networks and/or mobile systems [10]. Parlay v4.0/3GPP OSA Release 5/ETSI ES 202 915 defines fourteen APIs including framework, call control (CC), user interaction (UI), mobility (user status/location), terminal capability, policy control, content-based charging, and presence and availability management (PAM). The framework provides access control to APIs as well as API instance creation, API retrieval, fault management, event notification and domain authentication, etc. PAM-API is a part of the Parlay-API. + +Since the Parlay-API allows flexible combination of network functions for application development, it is a realistic solution for a metadata framework that associates metadata and network functions. Parlay-API should be included in the service platform of the ITU-T metadata framework. + +### **6.2.2 Service platform based on Web service (SOAP)** + +W3C specified SOAP 1.2 as a W3C Recommendation for a framework of exchanging XML structured data between peers. Web services are associated by sharing data described in WSDL (Web Services Description Language). As these are language specifications, actual services need to be defined independently. If Parlay-based terminologies are defined, then a metadata framework could be built by using the Web service description. Parlay-X pursues aims for interoperability with Web services. This needs further study. + +Details about platform functions are also a subject for further study. Some details are discussed in the appendices. A QoS control function and address resolution function are included in Appendices I and II, respectively. + +# **Appendix I** + +## **Policy-based QoS control using metadata** + +This appendix addresses a scheme for policy-based QoS control in multimedia content delivery over a network, as an implementation example of the metadata framework. + +To utilize network QoS control functions effectively, user requirements need to be described and associated with network QoS control functions. In a policy-based QoS control scheme, user requirements and the usage environment are described in metadata, and policy control associates these requirements with network QoS capability. This scheme is provided as a guideline. + +As user requirements for content delivery are high-level and abstract, they need to be translated into an appropriate application QoS class and then mapped into an appropriate network QoS class, according to some rules. Once the mapping has been done, admission control needs to decide whether or not the selected network QoS class should be admitted considering relevant conditions such as the characteristics of transferred data and capability and/or environment of end-systems. The policy-based QoS control scheme performs admission control under the conditions described by metadata, according to predefined policies. + +A set of application QoS classes is defined in ITU-T Rec. G.1010 "End-user multimedia QoS categories". Various service level agreements (SLAs) define another type of application QoS classes. For network QoS , IETF DiffServ and ITU-T Rec. Y.1541 "Network performance objectives for IP-based services" are envisaged. ITU-T Rec. Y.1541 defines network QoS classes with UNI-to-UNI performance objectives. The policy-based QoS control scheme uses these + +definitions and provides mapping rules and mechanisms between these hierarchical QoS class definitions. + +Furthermore, although termination points of data flow are located at a server or terminal, edge-nodes cannot recognize these termination points in the current network configurations. The policy-based QoS control scheme introduces flag information within the packet header that is commonly used by end-systems (server, terminal) and edge-nodes to segregate the admitted flow. + +### I.1 Model of policy-based QoS control using metadata + +In the IP-based network, network QoS control is implemented by, for example, differentiated packet forwarding behaviour according to the assigned network QoS class. Since the network QoS class is defined on a UNI-to-UNI basis, QoS requirements for servers/terminals connected to edge-nodes are further categorized in QoS classes depending on the characteristics of transferred data as well as on the capabilities and/or environment of the end-system (server/terminal). In the policy-based QoS control scheme, QoS classes that the end-system provides to the user are defined as application QoS classes. Transferred data characteristics and end-system capabilities and/or environment are described by metadata. + +Firstly, user and/or application requirements are described in metadata. Secondly, these are translated into an application QoS class. Thirdly, they are mapped to a network QoS class. Translation and mapping are achieved by policy control. These relationships are shown in Figure I.1. + +![Hierarchical model of policy-based QoS control diagram](9cd90f495b95ad2116ff780248c26d95_img.jpg) + +The diagram illustrates the hierarchical model of policy-based QoS control. It shows the flow of information from user requirements to application QoS classes, then to network QoS classes, and finally into the network. On the left, an 'End system' (represented by a computer icon) sends 'User/application requirements' (indicated by a long double-headed arrow) to a 'Policy control' block. The 'Policy control' block performs 'Translate' and 'Mapping' operations, resulting in an 'Application QoS class (ITU-T Rec. G.1010)' and a 'Network QoS class (ITU-T Rec. Y.1541/ DiffServ)'. These are passed to the 'IP network cloud' (a large grey box containing two 'Network section' ovals) via a 'UNI' (User Network Interface). The 'Policy control' also sends 'Set-up' information to the network. On the right, another 'End system' (represented by a monitor icon) provides 'Metadata' (indicated by filmstrip icons) to the 'Policy control'. The 'IP network cloud' is connected to the right 'End system' via another 'UNI'. The label 'F.750\_F1.1' is in the bottom right corner. + +Hierarchical model of policy-based QoS control diagram + +**Figure I.1/F.750 – Hierarchical model of policy-based QoS control** + +Functional components that are required in the policy-based QoS control using metadata are shown in Figure I.2. They consist of the following five components: + +- 1) User and/or application requirement metadata; +- 2) Rules and mechanisms that map given metadata to an appropriate application QoS class; +- 3) Mechanism for querying the policy server about the admission of a selected application QoS class; +- 4) Network resource management database to provide information about network status, and an admission control mechanism equipped with set-up information notification; +- 5) Mechanism to configure set-up information via the network API. + +![Architectural model of policy-based QoS control using metadata. The diagram shows a Server (metadata -> SLA) (2) receiving (1) Metadata from Content sources. The Server sends (3) SLA request to a Policy server, which responds with (4) OK/NG and Set-up information. The Server then sends (5) Set information to Data blocks, which are then processed by an Edge router and delivered to a Client.](9c6461e1e94afae4dec455e69a2ce152_img.jpg) + +``` + +graph TD + subgraph Content + C1((Content)) + C2((Content)) + end + S[Server +(metadata -> SLA) (2)] + PS[Policy server] + D1[Data] + D2[Data] + ER[Edge router] + C[Client] + + C1 -- "(1) Metadata" --> S + C2 -- "(1) Metadata" --> S + S -- "(3) SLA request" --> PS + PS -- "(4) OK/NG +Set-up information" --> S + S -- "(5) Set" --> D1 + S -- "(5) Set" --> D2 + D1 --> ER + D2 --> ER + ER --> C + C -- "(1) Metadata" --> S + style Content fill:none,stroke:none + style C fill:none,stroke:none + style PS fill:none,stroke:none + style S fill:none,stroke:none + style ER fill:none,stroke:none + style D1 fill:none,stroke:none + style D2 fill:none,stroke:none + style C1 fill:none,stroke:none + style C2 fill:none,stroke:none + style F750_F12((F750_F12)) fill:none,stroke:none + F750_F12 -.-> ER + +``` + +Architectural model of policy-based QoS control using metadata. The diagram shows a Server (metadata -> SLA) (2) receiving (1) Metadata from Content sources. The Server sends (3) SLA request to a Policy server, which responds with (4) OK/NG and Set-up information. The Server then sends (5) Set information to Data blocks, which are then processed by an Edge router and delivered to a Client. + +**Figure I.2/F.750 – Architectural model of policy-based QoS control using metadata** + +#### I.1.1 Metadata mapping + +To describe user and/or application requirements, international standard metadata defined by MPEG, TV-Anytime and others could be used. As for content metadata, TV-Anytime content instance metadata that includes information about content delivery conditions such as data format, size, and delivery configuration (streaming or downloading), could be used as user requirements. As for the terminal profile, CC/PP defined by W3C could be used, which would enable the server to select a content instance or adjust the transmission bandwidth. Further information that is required in QoS control includes terminal type (PC, PDA, or cell-phone) and/or contracted bandwidth. These attributes are defined by MPEG21 DIA (Digital Item Adaptation) as usage environment metadata or by CC/PP. + +The metadata framework is used for sharing metadata coming from different organizations. + +#### I.1.2 Mapping rules and mechanisms + +The domain manager provides policy rules for translating user and/or application requirements to a particular application QoS class or SLA, by using metadata. One prerequisite is that at least one SLA is contracted. The application QoS class is specified by ITU-T Rec. G.1010 "End-user multimedia QoS categories". Parameters for determining service categories include delay, jitter, and information loss. It is recommended that the application QoS class should be categorized by a combination of these parameters and the allowable range in terms of delay, loss rate, etc. ITU-T Rec. G.1010 further categorizes audio, video, and data by their data characteristics and usage situations, to define assured delay and error rate; thus, it can be a basis for defining SLAs for content delivery over a network. + +For example, real-time streaming needed by broadcast content delivery could be classified into one-way video, and a download service could be classified into bulk data as recommended by ITU-T Rec. G.1010. For "one-way video service class", the delay is specified to be less than 150 ms and the maximum delay is 400 ms and admitted information loss is specified to be less than 1 %. The delay for "bulk data", is specified to be less than 15 s, and information loss is zero. These specifications are useful for defining SLAs between different organizations. + +#### **I.1.3 Mechanism for querying the policy server** + +In this component, the server asks the policy server about the admission of a data transmission by using the application QoS or SLA, which was selected by the mapping rules. The query information includes application QoS class or SLA, and content delivery attributes (required bandwidth, content size, etc.). The policy server provides the admission control function. It decides to grant or reject a request, after consulting a network resource management database. If the request is granted, the policy server sends back configuration information for setting a flag that indicates permitted data. Since the policy server and edge router can register the IP address and port number of the application server, the data stream from the permitted application server is differentiable from other data streams. If other terminals set the configuration information and send data, the edge router can refuse to forward the data. + +#### **I.1.4 Network configuration information** + +Network configuration information, such as DSCP, is provided by IETF to make guidelines for using Diffserv. For example, "one-way video" for broadcast should be set to AF31, AF32, AF33, or CS4 and "bulk data" should be set to AF11, AF12, AF13, or CS. These guidelines are stored in the policy server database and sent as a reply for configuration information to the application server. + +#### **I.1.5 Setting configuration information** + +Currently, even if the application server and policy server negotiate to reserve a connection for the application layer, there is no way to differentiate negotiated data from other data; hence, other data may be allowed to use the reserved connection. This means that the negotiated result is not reflected in network resource utilization. In this implementation example model, the application server gets configuration information from the policy server at the time of negotiation and puts it in packets as a flag by means of the network API provided by the operating system. As a result, the policy server and edge router can distinguish packets of a negotiated server from other packets. This kind of flag could be implemented in the DS field of IP packets and flow label defined by IPv6. + +### **I.2 Implementation of QoS control** + +Diffserv, MPLS, Wide Area Ethernet, and others could implement network QoS control. Diffserv differentiates defined service classes by using DSCP as a flag. MPLS prepares virtual paths between pairs of edges, which are called Label Switched Paths (LSPs). If an edge router recognizes a flow, then succeeding packets that belong to the flow are transmitted over the same LSP, which achieves service segregation and QoS control. IETF published an RFC about LSP switching algorithms using Diffserv DSCP (RFC 3564). When MPLS is used for QoS control, security technology such as intrusion detection, forensics, firewalls, honey pots, etc., should be deployed to make MPLS networks more secure. Wide Area Ethernet can set the IEEE 802.1p priority bit in the Ethernet frame to achieve multiple priority paths that have different priority levels. Since each IP-based network QoS is implemented as a multi-priority virtual path, it is easy to map service classes. + +# Appendix II + +## Policy-based CDN management using metadata + +The outstanding feature of the content delivery network (CDN) is the separation of the content pre-caching function and the content delivery function. A content request from a user is redirected to a surrogate (CDN edge node) close to where the actual content is delivered, which improves performance, scalability and availability. + +Address resolution plays a key role in CDN management. It helps to find a suitable content instance that meets not only the user's preference about content, but also the usage environment such as client device capabilities and access network characteristics. + +This Recommendation defines a CDN management optimization by use of the metadata framework. In particular, it optimizes the address resolution API managed by policy control taking into account various conditions described by shared metadata across different domains. Currently, metadata is defined by different organizations in different schemas for specific application domains. To adapt content to network environment and user profile, metadata must be shared across different domains. Policy-based CDN management provides metadata translation and matching across different domains and a policy scheme in terms of expression and control for CDN optimization. + +### II.1 Directory database as metadata retrieval infrastructure + +The benefits of a database are characterized by its standard retrieval language and repeatability of retrieval, propagation and tracking in renewal, availability, and robustness. For common expression among different industry standard schemas, integration of attribute types, reference rules, and name definitions are needed. For this purpose, mapping between various industry standard metadata and directory object classes as shown in Figure 8 in 6.1 is used. + +To harmonize metadata across domains, firstly an ontology language such as OWL needs to describe the relationship between metadata elements that are mutually referred to, and then these elements are mapped to the same attribute type in the directory database. A policy-based CDN management model (PBCDNM) using such a directory method is shown in Figure II.1. + +![Diagram of Policy-based CDN management architecture. At the top, a 'Content distribution policy' box sends a 'Policy' to a 'Service platform (PBCDNM)'. The 'Service platform' provides an 'Address resolution API' and sends a 'Query of context information' to a 'Metadata gateway'. The 'Metadata gateway' is connected to 'Content metadata' and a 'Metadata Directory'. It also sends 'Policy enforcement' to a central 'CDN' node. The 'CDN' node is connected to three 'CDN edge node' nodes. Each 'CDN edge node' is connected to a 'Device metadata' oval, which is further connected to a 'User metadata' oval. Each 'CDN edge node' is also connected to a 'Metadata Directory' cylinder icon. The diagram is labeled F.750_FII.1.](e69b9188aa2c14ec6b21c83f711fef65_img.jpg) + +Diagram of Policy-based CDN management architecture. At the top, a 'Content distribution policy' box sends a 'Policy' to a 'Service platform (PBCDNM)'. The 'Service platform' provides an 'Address resolution API' and sends a 'Query of context information' to a 'Metadata gateway'. The 'Metadata gateway' is connected to 'Content metadata' and a 'Metadata Directory'. It also sends 'Policy enforcement' to a central 'CDN' node. The 'CDN' node is connected to three 'CDN edge node' nodes. Each 'CDN edge node' is connected to a 'Device metadata' oval, which is further connected to a 'User metadata' oval. Each 'CDN edge node' is also connected to a 'Metadata Directory' cylinder icon. The diagram is labeled F.750\_FII.1. + +**Figure II.1/F.750 – Policy-based CDN management** + +### II.2 Metadata for CDN optimization + +The essentials of CDN technology are server pre-caching control that optimizes content instance allocation among surrogates, and redirection control that finds content instances of requested content having attributes that match to usage environment and selects the optimum surrogate, which stores such instances by address resolution. + +In pre-caching control, cost efficiency is a goal. The cost of a surrogate server depends on its capacity and delivery capability available to target users and target devices. The following metadata are envisaged: + +- Content delivery attributes; +- Surrogate delivery capability attributes and storage capability attributes for streaming and downloading; +- Delivery environment attributes of the access network between surrogate and client; +- Client device storage and rendering capability attributes (terminal profile); +- User profile; +- User preference attributes (statistics on content access between surrogate and client). + +As for content delivery attributes, TV-Anytime Forum (TVA) content instance description metadata (Programme Location: Audio/VideoCodingParameter (FileFormat, BitRate, etc.)) could be used. For the terminal profile, CC/PP (composite capability and preference profiles) could be used as it includes terminal hardware, software, and browser platform attributes. For the user profile, TVA targeting metadata (intended audience) could be used. + +For user preference attributes, statistical information, analysed by content viewing history and by access requests between surrogate and client, is required, where TVA content description metadata, + +(Programme (Group) Information: Title, Genre, CreditList, Keyword, ParentalGuideline, Language), and TVA instance description metadata, could be used. + +These sets of attributes are also used as context information for address resolution to achieve redirection control. Address resolution is performed to identify the location of a content instance by examining device capabilities, and available bandwidth in the access network between the surrogate and client to carry the content instance data. Among metadata listed above, content delivery attributes (TVA instance description metadata), access network delivery environment attributes, and terminal profile (CC/PP) could be used for this purpose. + +### II.3 Policy descriptions in CDN management + +This clause provides the policy description in CDN management for the optimization of pre-cache control and redirection control, based on the metadata as described in the previous clause. + +Broadcast and/or multimedia content are assigned unique names or identifiers independent of their locations on a network. Association between these identifiers and the location information of actual content instances is provided by the address resolution system. The role of policy-based CDN management (PBCDNM) is to resolve the location of requested content that matches client device capability, access network characteristics, and user preferences as described in metadata. As a result, a content request from a user is redirected to a surrogate that stores the content instance adapted to the usage environment. + +#### II.3.1 Optimization of pre-caching control + +To optimize pre-caching, surrogates need to store frequently accessed categories of content to improve the hit rate. The optimum pre-cache policy could be determined by analyzing surrogate user preferences and the statistics of content access frequency. For this purpose, metadata access frequency statistics stored at a client device need to be categorized according to metadata attributes. A surrogate stores access frequency statistics to a particular metadata attribute such as "Genre" in its directory entries. + +Figure II.2 shows an example, where two client devices store user preference metadata in their usage environment metadata entries, and a surrogate stores statistics of user preference metadata of client devices that frequently access the surrogate. The User-1 device classifies the ratio of Genre-A content and Genre-B content to be 4 to 3. Surrogate-1 stores Genre-A as a genre preference. + +![Diagram illustrating the Pre-cache policy. A central Surrogate 1 box contains 'PreferredGenre = A'. Below it, two client devices are shown. Client device 2 (associated with User 2) contains 'Genre = A weighting = 6' and 'Genre = B weighting = 1'. Client device 1 (associated with User 1) contains 'Genre = A weighting = 4' and 'Genre = B weighting = 3'. Lines connect the surrogate to both client devices.](6f31cdb576d2f15c35c3f266e5f59211_img.jpg) + +``` +graph TD; S[Surrogate 1 +PreferredGenre = A] --- U2[User 2]; S --- U1[User 1]; U2 --- CD2[Client device 2 +Genre = A weighting = 6 +Genre = B weighting = 1]; U1 --- CD1[Client device 1 +Genre = A weighting = 4 +Genre = B weighting = 3]; +``` + +F.750\_FII.2 + +Diagram illustrating the Pre-cache policy. A central Surrogate 1 box contains 'PreferredGenre = A'. Below it, two client devices are shown. Client device 2 (associated with User 2) contains 'Genre = A weighting = 6' and 'Genre = B weighting = 1'. Client device 1 (associated with User 1) contains 'Genre = A weighting = 4' and 'Genre = B weighting = 3'. Lines connect the surrogate to both client devices. + +**Figure II.2/F.750 – Pre-cache policy** + +The criteria for judging whether to cache a content instance or not in a given surrogate are stored in a surrogate entry of the directory database as a pre-caching policy. This policy characterizes the performance of the CDN service that a CDN provider provides to content providers. An example of a pre-cache policy is described below: + +``` + +IF (ContentInstance1.genre = Surrogate1.preferredGenre) +THEN + IF (ContentInstance1.fileSize < Surrogate1.availableStorageSize) + THEN + CacheIt(ContentInstance1) + ELSE + ThrowAway(ContentInstance1) +ELSE + ThrowAway(ContentInstance1) + +``` + +This policy evaluates the possibility of pre-caching content instance1 at Surrogate-1. Firstly, it evaluates whether the genre attribute of content1 matches the genre preference of Surrogate-1. If they match, then it evaluates whether the file size of content1 is within the storage size of Surrogate-1. As a result, it decides whether to cache content1 in Surrogate-1 or not. + +#### II.3.2 Optimization of redirection control + +A key technology for achieving redirection control is address resolution. The address resolution function redirects the content request to the content instance that matches the usage environment such as terminal capability and access network characteristics. + +Suppose that the following network-related metadata is described: + +- 1) Access network characteristics in terms of "assured maximum bit-rate"; +- 2) Terminal capability in terms of implemented "codec-type"; +- 3) Content delivery characteristics in terms of "coding method". + +The problem is how to select one instance from two content instances having different bit-rates, which are created from the same original content. The selection policy is described as follows: (see Figure II.3): + +![Diagram illustrating the surrogate selection policy. A User 1 (Client device1 with codecType = aaa) sends a request to Access Network 1 (AssuredBitRate = 20 Mbit/s). The request is resolved to Content ID = CID:123. From this ID, two Content Instances are identified: Content Instance 1 (URL = http://abc.def, BitRate = 10 Mbit/s, codingMethod = aaa) and Content Instance 2 (URL = http://uvw.xyz, BitRate = 1 Mbit/s, codingMethod = aaa). Arrows show the flow from the user through the network to the content instances.](19a5f0db57a21a0e82a7f326083e96fd_img.jpg) + +``` + +graph LR + subgraph User1 [User 1] + CD1[Client device1 +codecType = aaa] + end + AN1[Access Network 1 +AssuredBitRate = 20 Mbit/s] + CI1[Content Instance 1 +URL = http://abc.def +BitRate = 10 Mbit/s +codingMethod = aaa] + CI2[Content Instance 2 +URL = http://uvw.xyz +BitRate = 1 Mbit/s +codingMethod = aaa] + CID[Content ID = CID:123] + + CD1 --> AN1 + AN1 --> CI1 + AN1 --> CI2 + CI1 --- CID + CI2 --- CID + +``` + +Diagram illustrating the surrogate selection policy. A User 1 (Client device1 with codecType = aaa) sends a request to Access Network 1 (AssuredBitRate = 20 Mbit/s). The request is resolved to Content ID = CID:123. From this ID, two Content Instances are identified: Content Instance 1 (URL = http://abc.def, BitRate = 10 Mbit/s, codingMethod = aaa) and Content Instance 2 (URL = http://uvw.xyz, BitRate = 1 Mbit/s, codingMethod = aaa). Arrows show the flow from the user through the network to the content instances. + +F.750\_FII.3 + +**Figure II.3/F.750 – Surrogate selection policy** + +``` + +IF((ClientDevice1.codecType=ContentInstance1.codingMethod) AND +(ClientDevice1.codecType = ContentInstance2.codingMethod)) +THEN + dif1=ClientDevice1.accessNetwork1.assuredBitRate + - ContentInstance1.bitRate + dif2=ClientDevice1.accessNetwork1.assuredBitRate + - ContentInstance2.bitRate + IF((dif1 > 0) AND (dif2 > 0)) + THEN + IF (dif1 > dif2) + resolveTo(ContentInstance2.URL) + ELSE + resolveTo(ContentInstance1.URL) + +``` + +This selection policy is stored in the directory database. + +### II.4 Interoperability of metadata across different domains + +To achieve interoperability between metadata, we need to define mapping and/or translation methods between different metadata in terms of application domains or standardization bodies. In the above example, terminal capability attributes such as "codec-type" may be described by CC/PP, and content delivery attributes such as "coding method" may be described by TV-Anytime content instance metadata. + +To achieve interoperation of these types of metadata across different domains, the relationship between metadata elements needs to be described beforehand. For example, it should be pre-described that the codec-type of a client device and coding-method of a content instance have a mutually checkable relationship. Then, these elements need to be mapped into the same attribute type of a directory database and stored. In the above example, two metadata elements are stored as the codec-type attribute of a client device and coding-method attribute of content instance, respectively. Both have the attribute type "enumeration type". + +### II.5 Service platform API + +The service platform API of the address resolution service is defined as follows (see Figure II.4): + +![Diagram of the address resolution platform showing the interaction between a CID-API and a Metadata gateway. The CID-API sends a 'Retrieve request' to the Metadata gateway and receives a 'Locator' in response. The Metadata gateway contains a 'Directory database'.](8d325fc12b494e42c9ea7ed2a7f327a6_img.jpg) + +``` + +graph LR + CID-API[CID-API] -- "Retrieve request" --> Metadata-gateway[Metadata gateway] + Metadata-gateway -- "Locator" --> CID-API + subgraph Metadata-gateway + Directory-database[Directory database] + end + +``` + +F.750\_FII.4 + +Diagram of the address resolution platform showing the interaction between a CID-API and a Metadata gateway. The CID-API sends a 'Retrieve request' to the Metadata gateway and receives a 'Locator' in response. The Metadata gateway contains a 'Directory database'. + +**Figure II.4/F.750 – Address resolution platform** + +Content ID (CID) is an identifier of a content item that is independent of location. CID is resolved into a locator that shows the location of the content. By using the CID address resolution API that provides a network-transparent address resolution interface, various application services that need an address resolution system and/or protocol are implemented. The CID address resolution API accesses the directory database via a metadata gateway to obtain device metadata about the client device. Based on this device metadata, it accesses the directory database to obtain a content instance that matches the device characteristics, which resolves the optimum address. + + + + + +# SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|---------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | General tariff principles | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects and next-generation networks | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.751.0-202008-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.751.0-202008-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..5295e5f3b359f1cb7eb589492906ba42372ce460 --- /dev/null +++ b/marked/F/T-REC-F.751.0-202008-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b606969d95fd7f6efdad13ce9473bd814fc626737a78b4ace581548907b98c87 +size 7056 diff --git a/marked/F/T-REC-F.751.1-202008-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.751.1-202008-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9002ee68e2b9ebf63d27d39ba8c73bdafb0e6be6 --- /dev/null +++ b/marked/F/T-REC-F.751.1-202008-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8df701536c25ab8246d2868de3c20c8a4e25168bb52c5eb9078278ab1b5b4fd8 +size 6859 diff --git a/marked/F/T-REC-F.751.1-202008-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg b/marked/F/T-REC-F.751.1-202008-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..76c62769990bced3be107b3c342cab89dce7fd0c --- /dev/null +++ b/marked/F/T-REC-F.751.1-202008-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:14be625e2f0a246015d81ec1a3276f27715622b947ee1b6f04e0c49ee5b05df5 +size 132945 diff --git a/marked/F/T-REC-F.751.1-202008-I_PDF-E/raw.md b/marked/F/T-REC-F.751.1-202008-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..9e1cf43b8961ff601de41d6e3f3d38edbf50c679 --- /dev/null +++ b/marked/F/T-REC-F.751.1-202008-I_PDF-E/raw.md @@ -0,0 +1,710 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.751.1** + +(08/2020) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Assessment criteria for distributed ledger +technology platforms** + +Recommendation ITU-T F.751.1 + +ITU-T + +![ITU logo](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) features a stylized globe with a red lightning bolt striking through it, symbolizing global connectivity and telecommunications. To the right of the globe, the text "International Telecommunication Union" is written in a blue, sans-serif font, with "ITU" in a larger, bold font above it. + +ITU logo + +International +Telecommunication +Union + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.751.1 + +## Assessment criteria for distributed ledger technology platforms + +## Summary + +Recommendation ITU-T F.751.1 specifies assessment criteria for distributed ledger technology (DLT) platforms. The purpose of Recommendation ITU-T F.751.1 is to assist implementers in the evaluation and comparison of different platforms. The assessment framework includes a set of criteria for function, performance and other aspects. It can be used as a guideline for DLT platform assessment, as well as for information disclosure of a given DLT platform product. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.751.1 | 2020-08-13 | 16 | 11.1002/1000/14333 | + +## Keywords + +Assessment, blockchain, criteria, distributed ledger technology, DLT, ledger, test. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2020 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|------|-----------------------------------------------|----| +| 1 | Scope ..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 2 | +| 4 | Abbreviations and acronyms ..... | 3 | +| 5 | Conventions ..... | 3 | +| 6 | Overview ..... | 3 | +| 7 | Criteria for DLT core functions ..... | 4 | +| 7.1 | Account creation..... | 4 | +| 7.2 | Transaction processing ..... | 4 | +| 7.3 | Query ..... | 5 | +| 7.4 | Consensus mechanism effectiveness ..... | 5 | +| 7.5 | Private key management..... | 5 | +| 7.6 | Smart contract mechanism ..... | 6 | +| 7.7 | Security of cryptography ..... | 6 | +| 7.8 | Decentralization..... | 7 | +| 8 | Criteria for DLT application functions ..... | 7 | +| 8.1 | User authentication ..... | 7 | +| 8.2 | System stability ..... | 7 | +| 8.3 | Economic mechanism design ..... | 8 | +| 8.4 | Information privacy ..... | 8 | +| 8.5 | Application support functions ..... | 9 | +| 8.6 | Transaction origin..... | 9 | +| 9 | Criteria for DLT operation functions..... | 9 | +| 9.1 | Network management..... | 9 | +| 9.2 | Risk management and mitigation ..... | 10 | +| 9.3 | Data storage sustainability..... | 10 | +| 10 | Performance ..... | 10 | +| 10.1 | Metric definitions ..... | 11 | +| 10.2 | Preconditions for performance evaluation..... | 11 | +| 10.3 | Transaction ..... | 11 | +| 10.4 | Test tools ..... | 11 | +| 11 | Criteria for the DLT ecosystem ..... | 11 | +| 11.1 | Platform maturity..... | 11 | +| 11.2 | Open source ..... | 11 | +| 11.3 | Maintenance ..... | 12 | + +| | Page | +|----------------------------------------------------|-------------| +| 11.4    Availability of professionals..... | 12 | +| 11.5    Running cost of DLT systems ..... | 12 | +| 11.6    Avoid vendor lock-in..... | 12 | +| Annex A – How to use the assessment criteria ..... | 13 | +| Bibliography..... | 14 | + +# Recommendation ITU-T F.751.1 + +## Assessment criteria for distributed ledger technology platforms + +# 1 Scope + +This Recommendation specifies an assessment framework for distributed ledger technology (DLT) platforms, which includes a set of criteria for function, performance and other aspects. The framework is recommended for use as a guideline for DLT platform assessment, as well as information disclosure of a given DLT platform product. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 asset** [b-ITU-T F.751.0]: A representation of value. It can be a diamond, a unit of currency, items inside a shipping container, etc. An asset can be physical or virtual. + +**3.1.2 blockchain** [b-ITU-T F.751.0]: A type of distributed ledger that is composed of digitally recorded data arranged as a successively growing chain of blocks with each block cryptographically linked and hardened against tampering and revision. + +**3.1.3 consensus** [b-ITU-T F.751.0]: Agreement that a set of transactions is valid. + +**3.1.4 entity** [b-ITU-T F.751.0]: Anything that has a separately identifiable existence (e.g., organization, person, group, smart contract or device). An entity uses distributed ledger technology to solve the problem of its business or information systems. + +**3.1.5 entity address** [b-ITU-T F.751.0]: Identifier for one or more entities performing transactions or other actions in a blockchain or distributed ledger network. + +**3.1.6 ledger** [b-ITU-T F.751.0]: Information store that keeps final and definitive (immutable) records of transactions. + +**3.1.7 node** [b-ITU-T F.751.0]: Device or process that participates in a distributed ledger network. + +**3.1.8 permissionless** [b-ISO 22739]: Not requiring authorization to perform any particular activity. + +**3.1.9 permissionless distributed ledger system** [b-ITU-T F.751.0]: Distributed ledger system where permissions are not required to maintain and operate a node. + +**3.1.10 private distributed ledger technology system** [b-ISO 22739]: DLT system that is accessible for use only to a limited group of DLT users. + +**3.1.11 public DLT; public distributed ledger system; public distributed ledger technology system** [b-ISO 22739]: DLT system which is accessible to the public for use. + +**3.1.12 smart contract** [b-ITU-T F.751.0]: Program written on the distributed ledger system that encodes the rules for specific types of distributed ledger system transactions in a way that can be validated, and triggered by specific conditions. + +**3.1.13 token** [b-ITU-T F.751.0]: A digital representation of value on a shared distributed ledger that is owned and secured using cryptography to ensure its authenticity and prevent modification or tampering without the owner's consent. + +**3.1.14 transaction** [b-ITU-T F.751.0]: An incident or an operation that leads to a change in the status of a ledger, such as adding a record or equivalent exchange based on currency. + +## **3.2 Terms defined in this Recommendation** + +This Recommendation defines the following terms: + +**3.2.1 account:** Representation of an entity whose data is recorded on a distributed ledger. + +NOTE – Originally published in [b-ITU-T TS FG DLT D1.1]. + +**3.2.2 block:** Individual data unit of a blockchain, composed of a collection of transactions and a block header. + +NOTE – Originally published in [b-ITU-T TS FG DLT D1.1]. + +**3.2.3 Byzantine fault tolerance (BFT):** Property that enables a system to continue operating properly even if some of its components fail or existence of intentional bad actors. + +NOTE – Adapted from [b-ITU-T TS FG DLT D1.1]. + +**3.2.4 consensus mechanism:** Rules and procedures by which consensus is reached. + +NOTE – Originally published in [b-ITU-T TS FG DLT D1.1]. + +**3.2.5 crash fault tolerance (CFT):** Property that enables a system to continue operating properly even if some of its components fail. + +NOTE – Adapted from [b-ITU-T TS FG DLT D1.1]. + +**3.2.6 decentralized autonomous organization (DAO):** A digital entity that manages assets and operates autonomously in a decentralized system, but also relies on individuals tasked to perform certain functions that the automaton itself cannot. + +NOTE – Originally published in [b-ITU-T TS FG DLT D1.1]. + +**3.2.7 distributed ledger technology (DLT):** Technology enabling large groups of nodes in distributed ledger networks to reach agreement and record information without the need for a central authority. + +**3.2.8 token ecosystem:** Digital system or digital space where participants and users interact and co-ordinate with each other using tokens. + +NOTE – Originally published in [b-ITU-T TS FG DLT D1.1]. + +**3.2.9 tokenomics (token economic):** Economics of a DLT-based token. + +NOTE – Adapted from [b-ITU-T TS FG DLT D1.1]. + +**3.2.10 wallet:** Software or hardware used to generate, manage and store both private and public keys and addresses, which enable DLT users to transact. Some wallets are recommended to interact with smart contracts and allow single or multi-signature. + +NOTE – Adapted from [b-ITU-T TS FG DLT D1.1]. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|-----------------------------------------| +| API | Application Programming Interface | +| ASIC | Application-Specific Integrated Circuit | +| BFT | Byzantine Fault Tolerance | +| CFT | Crash Fault Tolerance | +| CPU | Central Processing Unit | +| DAO | Decentralized Autonomous Organization | +| DDoS | Distributed Denial of Service | +| DLT | Distributed Ledger Technology | +| ID | Identifier | +| SDK | Software Development Kit | +| SPV | Simplified Payment Verification | +| TPS | Transaction Per Second | + +# 5 Conventions + +This Recommendation uses the following conventions: + +- The keywords "**is required to**" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "**is recommended to**" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement needs not be present to claim conformance. +- The keywords "can optionally" indicate an optional requirement which is permissible, without implying any sense of being recommended. This term is not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator or service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with this Recommendation. + +# 6 Overview + +The assessment criteria framework established in this Recommendation consists of 27 assessment items. They are recommended for classification into three domains: core functions; application support functions; and operation functions of a DLT platform. + +This framework also includes a performance domain and an ecosystem domain, See Figure 1. + +![Figure 1 – Overview of the DLT assessment framework. The diagram is a hierarchical structure. At the top is a box labeled 'Ecosystem' containing six sub-boxes: 'Platform maturity', 'Open source', 'Maintenance', 'Availability of professionals', 'Running cost of DLT systems', and 'Avoid vendor lock-in'. Below this is a large box divided into three columns: 'Application', 'Operation', and 'Performance'. The 'Application' column is further divided into two sub-boxes: 'Application support' (containing 'User authentication', 'System stability', 'Economic mechanism design', 'Information privacy', 'Application support functions', and 'Transaction origin') and 'Core technology' (containing 'Account creation', 'Private key management', 'Transaction processing', 'Smart contract mechanism', 'Query', 'Security of cryptography', 'Consensus mechanism effectiveness', and 'Decentralisation'). The 'Operation' column contains three sub-boxes: 'Network management', 'Risk management and mitigation', and 'Data storage sustainability'. The 'Performance' column contains four sub-boxes: 'Metric definitions', 'Preconditions for performance evaluation', 'Transaction', and 'Test tools'. At the bottom is a box labeled 'Infrastructure'.](a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg) + +Figure 1 – Overview of the DLT assessment framework. The diagram is a hierarchical structure. At the top is a box labeled 'Ecosystem' containing six sub-boxes: 'Platform maturity', 'Open source', 'Maintenance', 'Availability of professionals', 'Running cost of DLT systems', and 'Avoid vendor lock-in'. Below this is a large box divided into three columns: 'Application', 'Operation', and 'Performance'. The 'Application' column is further divided into two sub-boxes: 'Application support' (containing 'User authentication', 'System stability', 'Economic mechanism design', 'Information privacy', 'Application support functions', and 'Transaction origin') and 'Core technology' (containing 'Account creation', 'Private key management', 'Transaction processing', 'Smart contract mechanism', 'Query', 'Security of cryptography', 'Consensus mechanism effectiveness', and 'Decentralisation'). The 'Operation' column contains three sub-boxes: 'Network management', 'Risk management and mitigation', and 'Data storage sustainability'. The 'Performance' column contains four sub-boxes: 'Metric definitions', 'Preconditions for performance evaluation', 'Transaction', and 'Test tools'. At the bottom is a box labeled 'Infrastructure'. + +F.751.1(20)\_F01 + +**Figure 1 – Overview of the DLT assessment framework** + +Vendors are required to reveal details about these metrics, which are recommended for verification by document review or functional testing. + +# 7 Criteria for DLT core functions + +## 7.1 Account creation + +This relates to the ability to create user accounts. Accounts contain public and private key pairs. The creation is recommended for launch by a client or by a smart contract or other automated functionality components. If an account name is recommended for customization, the account name is required to be unique in the system. + +## 7.2 Transaction processing + +This relates to the ability to process transaction(s). There are two types of transaction, but it is not necessary for a DLT platform to support both. The system is required to have all transactions timestamped. + +- An **asset transfer transaction** relates to transfer of a certain amount of an asset between accounts, ensuring the asset in the ledger is balanced. +- A **non-asset transfer transaction** (such as changing the configuration parameter of an account, modifying the status of a smart contract and other status modification operations within an account) relates to a transaction without transfer of any asset. + +Users can check the result from any node in the DLT system after a successful transaction. + +## 7.3 Query + +Users are recommended to get result(s) by information request(s). + +#### 7.3.1 Balance query + +Users are recommended to acquire their account balance with a searching condition. A DLT platform without an asset transfer function does not need to support this function. + +#### 7.3.2 Conditional query + +Users are recommended to search their historical information on a DLT platform with a searching condition, such as a time period specification or specified user account. + +## 7.4 Consensus mechanism effectiveness + +A consensus mechanism is a set of rules and procedures by which mutual agreement is reached. A data consistency mechanism is used on a DLT platform to achieve the necessary agreement on a single data value or a single state of the network among distributed processes or multi-agent systems. + +To ensure the effectiveness of the consensus mechanism, sufficient nodes to participate in the consensus process are required, aligned with the objectives of the DLT platform, e.g., some DLT platforms only require a specific number of masternode participants, while others require all token holders. + +### 7.4.1 Data consistency + +The data synchronization module ensures that the distributed ledger is consistent. The synchronization module also validates the synchronized data to ensure its correctness and consistency no matter how immediate or prolonged is the achievement of finality. + +### 7.4.2 Byzantine fault tolerance/crash fault tolerance (BFT/CFT) + +DLT systems are required to continue to function in spite of some nodes taking malicious action or system failure. The tolerance threshold for malicious or crashed nodes is determined by the consensus mechanism selected, as well as the economics design of the DLT platform. + +## 7.5 Private key management + +Public-key cryptography is a cryptographic system that uses pairs of keys: public keys (which are recommended for wide dissemination); and private keys (which are known only to the owner). + +Private key management for DLT use is an important function for user experience and security measures. It provides a reliable and safe way to keep users' keys private. Storage and usage of private keys are recommended to be kept separate. Users are required to have full control of their private key usage. There are two common wallet methods to store private keys: software; and hardware. + +NOTE – Separate private key storage and usage are recommended to lead to many other applications, such as that of custodian. + +#### 7.5.1 Software wallet + +A software wallet is recommended to be a software application (i.e., client application, mobile application) or a service (i.e., digital asset exchanges) to store the private key and to track asset ownership. + +In a deterministic wallet, a mnemonic sentence or a word seed is generated and that single root key is safe enough. In a non-deterministic wallet, automatic random generation of each key is recommended. Therefore, any backup of the wallet is required to store each private key used as an address or account password. + +#### **7.5.2 Hardware wallet** + +A service is recommended to create a private key offline and provide the user with a device or physical medium to store it. This service, called a hardware wallet, is required to have the ability to execute basic operations such as signing transactions. + +### **7.6 Smart contract mechanism** + +A DLT is recommended to support more complex transactions as the technology evolves. Some complex transactions are stored in DLT systems in the form of source code or bytecode programs, and are recommended for execution to deal with different business logics. These programs are called smart contracts. + +The smart contract mechanism includes language specification, compilation and execution of the code. Smart contracts for different DLT systems are recommended for implementation using simple interpreted scripts or programming languages. + +#### **7.6.1 Monitor ability of participants' status** + +This relates to a system's ability to monitor the status of nodes (i.e., allocation of computing resources, computing resources consumed, network resources used) that are participating in the execution of a smart contract. + +#### **7.6.2 Lifecycle management of smart contract** + +This relates to the availability of a smart contract state identifier (ID) and the availability of functions for its lifecycle management, such as create, deploy, activate, suspend and destroy. + +#### **7.6.3 Security of smart contract** + +This item relates to the capability to write high-quality smart contracts (i.e., low bug rate) by using a programming language, either from scratch or by using a template. + +#### **7.6.4 Smart contract data access control** + +A DLT system is required to disclose how developers specify the authorization and confidentiality of their smart contract from a technical viewpoint. + +Sharing of smart contract data is required among parties to a smart contract, e.g., via interfaces to represent smart contracts and query them. + +## **7.7 Security of cryptography** + +Security includes encryption, cryptography, crash tolerance and hack tolerance. A DLT system is required to ensure the highest security for the system and disclose its security measures. + +#### **7.7.1 Encryption declaration** + +A DLT platform is required to specify whether the encryption, is derived from an open source solution or through regulatory compliance. + +#### **7.7.2 Pluggable encryption algorithm** + +A DLT platform is recommended to use pluggable modular encryption and to switch to a specified encryption algorithm online or offline as required. + +#### **7.7.3 Efficiency of encryption algorithm** + +A DLT platform is required to use and to make available for use secure, sufficiently strong encryption with acceptable efficiency, depending on the objectives of the system. + +#### **7.7.4 Strength of encryption** + +A DLT platform is required to declare the security level of the cryptographic schemas used. Category and cipher strength of the encryption can be taken as metrics. In addition, quantum-resistant encryption algorithms can be taken into account. + +## **7.8 Decentralization** + +The system is required to have means to create a distributed and decentralized architecture. It is important to analyse whether there are recommended special node types or limitations that are likely to compromise the decentralization. It is also important to verify whether there is support for decentralized management of the network. + +## **8 Criteria for DLT application functions** + +## **8.1 User authentication** + +A DLT platform is required to have modules for user authentication and user access control management. An electronic signature is an effective way to authenticate a user. The platform is recommended to allow the creation of smart contracts to authenticate and control the access of users. + +#### **8.1.1 User account verification** + +This is the validation of information, such as key-store and password, or two-step verification. + +#### **8.1.2 Login status management** + +The platform is required to update the status after user login. + +#### **8.1.3 User classification and user management** + +This is to assign users into one of several types and manage their permissions. + +#### **8.1.4 Authorization** + +Users are recommended to grant authority to others to access or modify their private data. + +## **8.2 System stability** + +The platform is required to satisfy at least the requirements of clauses 8.2.1 to 8.2.6. + +#### **8.2.1 Stability for node management** + +The system is required to grant normal operations when some nodes join, leave or upgrade. + +### **8.2.2 Stability for cross-chain operation** + +The system is required to grant normal operations when co-operating with another DLT or cloud system. + +#### **8.2.3 Network latency** + +The system is required to remain stable after running $7 \times 24$ h with network latency. The tolerable extent of latency is based on the design of the system. + +#### **8.2.4 Memory utilization** + +The system is required to remain stable after running $7 \times 24$ h without memory exceptions. + +#### **8.2.5 Central processing unit utilization** + +The system is required to remain stable after running $7 \times 24$ h without central processing unit (CPU) exceptions. + +#### **8.2.6 Stability for concurrency** + +The system is required to remain stable with bursts of concurrent transactions. + +## **8.3 Economic mechanism design** + +Economic mechanisms are required to be in place to incentivize user participation. These include, but are not limited to, consensus protocol, resolution mechanisms, voting protocol, allocation mechanism, bargaining protocol, monetary policy of the token and transaction fees. + +Economics design focuses on both the economics of the DLT platform, as well as the economics of the tokens that are produced in the DLT system (when applicable). + +#### **8.3.1 Incentive mechanisms** + +Incentive mechanisms includes both financial and non-financial incentives. Non-financial incentives are recommended to include voting protocol, reputation mechanism and allocation mechanisms. Financial incentives are a more direct form of reward, like monetary policy of the token, transaction fees or platform activities (i.e., block rewards when a block is mined). The last is more applicable in permissionless DLT. + +Incentive mechanisms are ways to co-ordinate activities among decentralized participants, to achieve the objectives of the DLT system. The incentives (financial and non-financial) are in place to align the behaviours of decentralized participants with the DLT ecosystem as a whole. + +For example, Bitcoin's proof-of-work mechanism rewards users in financial ways (diminishing returns of block rewards and transaction fees) in return for the investment of electricity usage and special application-specific integrated circuits (ASICs). This incentive mechanism is required to be described and published. + +#### **8.3.2 Token economics disclosure** + +The token is a digital representation of value. Tokens are recommended to be fungible or non-fungible in nature. Fungible tokens are recommended to have different functions, namely security, utility and money. Depending on the token function, a tokenomics report describing token economics is required to be published. + +This is required to include, but not limited to, token policy (monetary policy of the token), valuation policy (for security tokens), platform activities (e.g., block rewards for miners), transaction fees, property rights and distribution of tokens. + +#### **8.3.3 Token lifecycle** + +If tokens are used, the distributed ledger system is required to support token issuance, token transfer, token withdrawal, token settlement and clearance, as well as token balance enquiries. + +A DLT platform is required to use a standard protocol for its tokens. The system is required to have functionalities to facilitate cross-chain or cross-DLT system operations, when applicable or required by regulations. + +## **8.4 Information privacy** + +Privacy of information is a key requirement for DLT platforms, useful in sectors like finance and healthcare. It relates to confidential generation, storage and transmission of data in a DLT system and safe storage of user private information. + +#### **8.4.1 Secure transmission** + +Transfer over a secure channel of confidential or proprietary information is required, which is achievable by a specified secure transmission protocol. + +#### **8.4.2 Restricted data access** + +Any confidential or personal information that is protected by law or policy requires the appropriate level of differential access control and security protection, whether in storage or in transit. + +#### **8.4.3 Privacy protection** + +A DLT platform is recommended to use privacy protection algorithm(s) such as zero knowledge proofs, ring signature, secure multi-party computation and homomorphic encryption to avoid privacy disclosure. + +### **8.5 Application support functions** + +A DLT platform is recommended to implement application support functions to improve user-friendliness (i.e., user experience and user interface). + +#### **8.5.1 User interface for query** + +A DLT platform is recommended to provide functionality (web page, app, browser plug-in, etc.) to perform a query, visualize the query result and show ledger status. + +#### **8.5.2 User interface for smart contract** + +A DLT platform is recommended to provide functionalities for the visualization of the deployment and invocation of smart contracts, as well as queries about smart contract data. + +#### **8.5.3 Multi-language software development kits** + +A DLT platform is required to provide at least one software development kit (SDK) and is recommended to translate it into other programming languages. + +## **8.6 Transaction origin** + +A DLT system is required to provide mechanisms to identify the origin of a transaction. + +#### **8.6.1 Transaction origin – node** + +A DLT system is required to provide mechanisms to identify the original node to dispatch a transaction to the network. + +#### **8.6.2 Transaction origin – account** + +A DLT system is required to provide mechanisms to segregate the account signing a transaction from the account dispatching the transaction to the network. + +# **9 Criteria for DLT operation functions** + +## **9.1 Network management** + +The management and monitoring of nodes within a DLT system, including status, configuration, node type and behaviour, are described in clauses 9.1.1 to 9.1.4. + +### **9.1.1 Node status monitoring** + +A DLT system is required to have the ability to monitor overall node status, such as number of nodes on- or offline, synchronization status and client version. + +#### **9.1.2 Multi type nodes** + +A DLT system is required to have the ability to classify nodes. For instance, a node is recommended for classification in two categories, either full or lightweight, according to whether a complete ledger copy is stored within it. With lightweight nodes, the simplified payment verification (SPV) method is required for verification of the correctness of shared ledgers. + +#### **9.1.3 Node configuration modification** + +A DLT system is required to have the ability to support hot or cold modification of the node's configuration parameter, such as the block size and node type. + +#### **9.1.4 Network fairness** + +The system is required to support a mechanism to balance the sharing of use of the network among nodes. + +## **9.2 Risk management and mitigation** + +A DLT system is required to have the ability to resist distributed denial of service (DDoS) attacks, Sybil attacks or dishonest nodes. If failure occurs after an attack, the DLT system is required to have the ability to recover its previously known state. + +### **9.2.1 Recovery mechanisms** + +A DLT system is required to recover from failure by downgrading recovery, security service, etc. A recovery solution is required to be flexible and problem oriented. + +### **9.2.2 Trouble shooting** + +A DLT system is required to have the ability to execute rapid trouble shooting and automatically send failure notifications. + +#### **9.2.3 Avoid single point of failure** + +The DLT system is required to be independent of any centralized system that might cause a single point of failure. + +## **9.3 Data storage sustainability** + +A DLT system is an append-only trusted ledger. However, mass data stored in DLT systems is likely to degrade the query performance. The system is required to provide sustainable ways to deal with data volume growth. + +### **9.3.1 Alternative solutions to storage** + +The system is required to provide a way to minimize the total size of data stored by the whole system. It is recommended to store data no longer used or whose activity level is below a threshold. Some examples are the transfer of data to independent storage or to store the data only on specific types of node. + +#### **9.3.2 Data query** + +A DLT system is required to have ability to query archived data by providing application programming interfaces (APIs) or tools. + +#### **9.3.3 Data recovery** + +A DLT system is required to have ability to recover archived data in some ways, and the ledger(s) is required to keep the same status as before it was archived. + +# **10 Performance** + +The throughput and resource usage of processing standard transactions. Environment and deployment reasons is recommended to affect performance, such as network topology and test environment (CPU, memory, disk, network). The number of transactions per second (TPS) is a standard performance indicator. When evaluating performance by TPS, indications of topology deployment and test environment are required. + +## 10.1 Metric definitions + +The number of TPS is given by + +$$\frac{n_{\text{proc}}}{t_{\text{proc}}}$$ + +where $n_{\text{proc}}$ is the number of processed transactions and $t_{\text{proc}}$ is the time taken to process transactions. The data used to calculate TPS needs to be gathered when the system is stable. Both system booting or shutting down and test environment instability are likely to lead to inaccuracy. + +The maximum transaction delay, $t_{\text{max del}}$ , is given by + +$$t_{\text{max del}} = \max (t_{\text{con}} - t_{\text{com}})$$ + +where $t_{\text{con}}$ is the transaction confirmed time and $t_{\text{com}}$ is the transaction committed time. + +The average transaction delay, $\bar{t}_{\text{del}}$ , is given by + +$$\bar{t}_{\text{del}} = \frac{\sum (t_{\text{con}} - t_{\text{com}})}{n}$$ + +where $n$ is the number of transactions. + +## 10.2 Preconditions for performance evaluation + +### 10.2.1 Test environment + +When evaluating performance of a DLT system, indications of the hardware such as CPU, memory, hard disk, network are required. + +### 10.2.2 Topology of the network + +When evaluating performance of a DLT system, an indication of network topology is required. + +### 10.2.3 Test system deployment + +When evaluating performance of a DLT system, an indication of the type of deployment is required. + +## 10.3 Transaction + +When evaluating performance of a DLT system, indications of the numbers of transactions is required. + +## 10.4 Test tools + +There are open source tools designed especially for DLT system to test performance, such as Hyperledger Caliper [b-Caliper] or TrustedBench [b-TB]. Script is also recommended for use to test performance. + +# 11 Criteria for the DLT ecosystem + +## 11.1 Platform maturity + +Platform maturity of DLT includes many factors, such as year of creation, launch of production version, deploying networks, numbers of production networks, numbers of applications and team expertise. Whether a network is permissioned or permissionless, a DLT vendor is required to disclose this information to all consumers. + +## 11.2 Open source + +DLT platforms are required to be open sourced (to users) and to announce the licence it is using. + +## **11.3 Maintenance** + +DLT platforms are required to be well maintained by either individuals, companies or non-profit organizations. Inclusion of regular updates on source repository, active discussion by the community regarding the DLT platform and the ease of updating it with respect to the decentralized applications and DAOs that exists on it are recommended. + +## **11.4 Availability of professionals** + +Different DLT platforms require expertise from various technical backgrounds. This indirectly limits the number of professionals available in the market. Sufficient talent supply is an important factor in the evaluation of DLT platforms. + +## **11.5 Running cost of DLT systems** + +It is essential for DLT systems to evaluate running costs, which include transaction fees payable, transaction confirmation time and cost of writing, as well as the reading and execution of smart contracts. In the future, with the importance of smart contract security, audit fees for smart contract and code review, as well as many other intermediate fees, will also be required to be taken into consideration. + +## **11.6 Avoid vendor lock-in** + +DLT system is required to have standardized APIs, such as service access, data dictionary, communication protocol, encryption algorithm and system testing, so that there is recommended to be multiple vendors that provide similar services for the customers to avoid vendor lock-in. + +# **Annex A** + +## **How to use the assessment criteria** + +(This annex forms an integral part of this Recommendation.) + +Assessment criteria concerning the economic mechanism (clause 8.3) and ecosystem (clause 11) are required to undergo testing by information disclosure. The tester conducts a comprehensive assessment of information about the DLT system and market. + +Criteria related to application (clause 8) and operation (clause 9) functions are required to undergo testing on DLT platforms in a production environment. + +Performance criteria (clause 10) are required to undergo assessment in a laboratory environment, i.e., a quality assurance environment, which is required to be stable and controllable [b-ISO/IEC 25000]. + +DLT core function criteria (clause 7) are required to undergo assessment in a test environment. These tests are recommended to ensure normal operation of a DLT platform. + +# Bibliography + +- [b-ITU-T F.751.0] Recommendation ITU-T F.751.0 (2020), *Requirements for distributed ledger systems*. +- [b-ITU-T TS FG DLT D1.1] ITU-T Technical Specification FG DLT D1.1 (2019), *DLT terms and definitions*. +- [b-ITU-T FG DLT 3.1] ITU-T Technical Specification FG DLT D3.1 (2019), *DLT reference architecture*. +- [b-ISO 22739] ISO 22739:2020, *Blockchain and distributed ledger technologies — Vocabulary*. +- [b-ISO/IEC 25000] ISO/IEC 25000:2014, *Systems and software engineering – Systems and software Quality Requirements and Evaluation (SQuaRE) – Guide to SQuaRE*. +- [b-Caliper] Linux Foundation Projects (Internet). *Hyperledger – Caliper*. San Francisco, CA: Linux Foundation. Available [viewed 2020-10-09] at . +- [b-TB] Github (Internet). *TrustedBlockchain/TrustedBench*. San Francisco, CA: Github. Available [viewed 2020-10-09] at: . + + + +# SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.751.12-202309-I_PDF-E/0f985b39edc1d52ba3600c438bc8f0a5_img.jpg b/marked/F/T-REC-F.751.12-202309-I_PDF-E/0f985b39edc1d52ba3600c438bc8f0a5_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d2df96940ef19aa8dbcbc47bc8376e4543cdc8bc --- /dev/null +++ b/marked/F/T-REC-F.751.12-202309-I_PDF-E/0f985b39edc1d52ba3600c438bc8f0a5_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:83a0ca63a39dbead690192faeca9d97152291f4328fa07440738f6a5fb04e916 +size 31892 diff --git a/marked/F/T-REC-F.751.12-202309-I_PDF-E/27b06ec9f42b5d727a2630f61a5f1861_img.jpg b/marked/F/T-REC-F.751.12-202309-I_PDF-E/27b06ec9f42b5d727a2630f61a5f1861_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..bfdc9dd498e7adf6ae69fd1338c6ee94777afd74 --- /dev/null +++ b/marked/F/T-REC-F.751.12-202309-I_PDF-E/27b06ec9f42b5d727a2630f61a5f1861_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:63eb7882741585caf42d036ca1b7dc2d2e29d31a92557d4d43adf5c9431b5319 +size 135500 diff --git a/marked/F/T-REC-F.751.12-202309-I_PDF-E/4ee27dbf5ef12e7b58b0ef0937bc5a5e_img.jpg b/marked/F/T-REC-F.751.12-202309-I_PDF-E/4ee27dbf5ef12e7b58b0ef0937bc5a5e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..5d3d6cf6a381338540be0fc929813428e077be50 --- /dev/null +++ b/marked/F/T-REC-F.751.12-202309-I_PDF-E/4ee27dbf5ef12e7b58b0ef0937bc5a5e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:977f17223554049789b380ef46a500acfe3027e50fed2f508e118bf6b80781cf +size 46262 diff --git a/marked/F/T-REC-F.751.12-202309-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.751.12-202309-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..25f8fbfaef317942ed3102c74b40a7d42f80608b --- /dev/null +++ b/marked/F/T-REC-F.751.12-202309-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:83cab7dcabfe939b223d97242d8523e461968d21f7c5432693c30a64f2d95521 +size 7050 diff --git a/marked/F/T-REC-F.751.12-202309-I_PDF-E/a83ba9e3e2c1e21dd69953a7b09e45b4_img.jpg b/marked/F/T-REC-F.751.12-202309-I_PDF-E/a83ba9e3e2c1e21dd69953a7b09e45b4_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3c7d3ab300d9ed0a6929e74deb835f4c69e10341 --- /dev/null +++ b/marked/F/T-REC-F.751.12-202309-I_PDF-E/a83ba9e3e2c1e21dd69953a7b09e45b4_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5a53de3f614e7a4cc1ef18e22a300291176b8afb1b5f569c7a8924c360a1f173 +size 76302 diff --git a/marked/F/T-REC-F.751.12-202309-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.751.12-202309-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..db298676fba294a81bcf25995b10ad7a4b1a2e52 --- /dev/null +++ b/marked/F/T-REC-F.751.12-202309-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3f592ad96381d53d1f92b0d9757179de34c9d4d7fb59047f78faed450d9a6e89 +size 113482 diff --git a/marked/F/T-REC-F.751.12-202309-I_PDF-E/f4fdd410cdb84df81274da55721e56fb_img.jpg b/marked/F/T-REC-F.751.12-202309-I_PDF-E/f4fdd410cdb84df81274da55721e56fb_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..6e33d57d135a97bd21b8bd1f30e7d273d6c0ba10 --- /dev/null +++ b/marked/F/T-REC-F.751.12-202309-I_PDF-E/f4fdd410cdb84df81274da55721e56fb_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dff1f867bf99979f9b47f012dd8e7a63b166f1949a5a9a3f522274b630ae8b17 +size 97328 diff --git a/marked/F/T-REC-F.751.12-202309-I_PDF-E/ff0952ef692c9d960ce5f6708bcc9711_img.jpg b/marked/F/T-REC-F.751.12-202309-I_PDF-E/ff0952ef692c9d960ce5f6708bcc9711_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f659747de68763beb19ab5ed0d02364f3b0f2de6 --- /dev/null +++ b/marked/F/T-REC-F.751.12-202309-I_PDF-E/ff0952ef692c9d960ce5f6708bcc9711_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6cd6b3f3ee51d6b089ce14e4c2df967546e52fee917c5aa30b12c1bc460d09e8 +size 68709 diff --git a/marked/F/T-REC-F.751.12-202309-I_PDF-E/raw.md b/marked/F/T-REC-F.751.12-202309-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..d226c58de46ff324a1f468a6b9d5f64a7a535f5f --- /dev/null +++ b/marked/F/T-REC-F.751.12-202309-I_PDF-E/raw.md @@ -0,0 +1,453 @@ + + +Recommendation + +**ITU-T F.751.12 (09/2023)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +**Formal verification framework for smart +contract on distributed ledger technology** + +![ITU logo](84a1d09fb489061482111515543b60dc_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +# ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.751.12 + +# Formal verification framework for smart contract on distributed ledger technology + +## Summary + +Smart contracts can be used to reduce complex business contracts by directly enforcing the contract's payment methods and paybacks, and by automating the process of contract execution and verification into the network, without the intervention and cost of the person checking the contract's performance. However, smart contracts are a series of program codes generated on distributed ledger technology (DLT) and problems may occur in the process of executing the smart contract. As a method to solve problems that occur in the program execution environment, there is a formal verification. Recommendation ITU-T F.751.12 specifies the formal verification framework for smart contract on DLT, its overview, requirement and architecture in its framework, as well as the main technical direction of its formal method component. This Recommendation can be used as a guideline for a smart contract developer to build systems. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.751.12 | 2023-09-13 | 16 | 11.1002/1000/15626 | + +## Keywords + +Distributed ledger technologies, security, smart contract, verification. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2023 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 1 | +| 5 Conventions ..... | 2 | +| 6 Overview of formal verification for smart contracts ..... | 2 | +| 6.1 Smart contract..... | 2 | +| 6.2 Formal method..... | 3 | +| 7 Requirements of formal verification for smart contracts..... | 4 | +| 7.1 DApp requirements ..... | 4 | +| 7.2 Smart contract mechanism requirements..... | 5 | +| 7.3 Formal method requirements..... | 5 | +| 8 Architecture of formal verification for smart contracts ..... | 5 | +| 8.1 Overview ..... | 5 | +| 8.2 Formal verification ..... | 7 | +| Appendix I – Use case of verification for smart contract ..... | 9 | +| I.1 Overall procedure for formal verification ..... | 9 | +| I.2 IoT environment ..... | 9 | +| Appendix II – Related projects on verification for smart contract ..... | 11 | +| Bibliography..... | 12 | + + + +# Recommendation ITU-T F.751.12 + +## Formal verification framework for smart contract on distributed ledger technology + +# 1 Scope + +This Recommendation provides a formal verification framework of smart contract on distributed ledger technology (DLT). It also includes: + +- Overview of formal verification for smart contract; +- Requirements of formal verification for smart contract; +- Architecture of formal verification for smart contract. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T F.751.2] Recommendation ITU-T F.751.2 (2020), *Reference framework for distributed ledger technologies*. + +[ITU-T Y.3320] Recommendation ITU-T Y.3320 (2014), *Requirements for applying formal methods to software-defined networking*. + +# 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 consensus protocol** [b-ITU-T X.1401]: Process for determining what blocks get added to the chain and what the current state is. + +**3.1.2 distributed ledger** [b-ISO 22739]: Ledger that is shared across a set of DLT nodes and synchronized between the DLT nodes using a consensus mechanism. + +**3.1.3 smart contract** [b-ITU-T F.751.0]: Program written on the distributed ledger system that encodes the rules for specific types of distributed ledger system transactions in a way that can be validated, and triggered by specific conditions. + +### 3.2 Terms defined in this Recommendation + +None. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +DApp Decentralized Application + +DLT Distributed Ledger Technology + +## 5 Conventions + +In this Recommendation: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. +- The keywords "can optionally" indicate an optional requirement which is permissible, without implying any sense of being recommended. This term is not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator or service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with this Recommendation. + +## 6 Overview of formal verification for smart contracts + +### 6.1 Smart contract + +Smart contracts are a series of program codes generated on the distributed ledger platform. It ensures that the contents of the contract are automatically executed on the assets contained in the distributed ledger when predetermined conditions are achieved. Smart contracts can be used to reduce complex business contracts by directly enforcing the contract's payment methods and paybacks, and by automating the process of contract execution and verification into the network, without the intervention and cost of the person checking the contract's performance. It is characterized by the ability to safely execute contracts based on low cost and agreement trust. Thus, it is possible to perform record storage, cash flow, and contract fulfilment on one platform, thereby increasing work efficiency. + +Smart contracts reduce contract execution costs and the potential for disputes by automatically fulfilling contracts upon fulfilment of conditions. However, the cost of the contract writing phase for programming languages does not decrease. Since the contract is automatically executed, it gives the other party the confidence that the contract will be fulfilled. When combined with the disclosed transaction information, it reduces the risk of the transaction and reduces the uncertainty of the contract. It reduces the ambiguity of contract interpretation because it is written in program code. However, there is a possibility of inconsistencies in interpretation and writing errors in the writing phase when programming contracts in program languages. + +![Figure 1 – High-level conceptual architecture of DLT (ITU-T F.751.2). The diagram shows a layered architecture for DLT. On the left, a vertical bar labeled 'Operation and maintenance' spans the middle three layers. The top layer is 'Application', containing 'DApp(s)' and three sub-components: 'Runtime', 'Life cycle', and 'Devtool'. The second layer contains 'Governance and compliance' (left), 'Protocol' (center), and 'Utils' (right). The 'Protocol' layer is further divided into 'Account/DID', 'Consensus', and 'Messaging' in the first row, and 'Right management', 'Ledger management', and 'Crypto' in the second row. The third layer contains 'Node management' (left), 'Resources' (center), and 'Data integrity and privacy util' (right). Below 'Resources' are 'Storage management' and 'Network management'. To the right of the main stack, 'External interaction management' is at the top, and 'External resource management' is at the bottom. On the far right, a dashed box labeled 'Extensions' contains 'Side chain', 'Off chain', 'Multi chain', 'Internal extensions', and '...'.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +Figure 1 – High-level conceptual architecture of DLT (ITU-T F.751.2). The diagram shows a layered architecture for DLT. On the left, a vertical bar labeled 'Operation and maintenance' spans the middle three layers. The top layer is 'Application', containing 'DApp(s)' and three sub-components: 'Runtime', 'Life cycle', and 'Devtool'. The second layer contains 'Governance and compliance' (left), 'Protocol' (center), and 'Utils' (right). The 'Protocol' layer is further divided into 'Account/DID', 'Consensus', and 'Messaging' in the first row, and 'Right management', 'Ledger management', and 'Crypto' in the second row. The third layer contains 'Node management' (left), 'Resources' (center), and 'Data integrity and privacy util' (right). Below 'Resources' are 'Storage management' and 'Network management'. To the right of the main stack, 'External interaction management' is at the top, and 'External resource management' is at the bottom. On the far right, a dashed box labeled 'Extensions' contains 'Side chain', 'Off chain', 'Multi chain', 'Internal extensions', and '...'. + +F.751.12(23) + +**Figure 1 – High-level conceptual architecture of DLT (ITU-T F.751.2)** + +Figure 1 illustrates the high-level architecture constrains the highly abstract hierarchy of distributed ledger technologies. The smart contract mechanism is an optional in protocol component, but useful, component for DLT and most decentralized applications (DApps) are written by smart contracts. Smart contracts are programs written in a programming language that automatically execute contracts at specific times or events as defined by the parties. Therefore, smart contracts have the following problems. + +- Since smart contracts are software code written in a programming language, they have the same vulnerabilities as overflow problems and exception handling problems. If, in the process of implementing the contract as a program, logic is omitted or wrong, there may be a vulnerability, or the scope of the variable may be specified incorrectly. +- Smart contracts have a property called contract, which causes dependency problems. In order for the contract to be confirmed as a block, a proof of work process is performed. In this proof of work, the contract order is changed, which causes a problem. +- Since smart contracts are recorded on the distributed ledger, it is difficult to apply patches when the smart contract is recorded on the distributed ledger. Since most smart contract developers create based on existing smart contract code, once the vulnerable code is raised, not only is it impossible to fix, but it also causes a problem that the error is expanded and reproduced. +- Vulnerabilities exist when writing smart contracts using data outside the distributed ledger. In the case of data outside the distributed ledger, it is difficult to check whether the data is forged or has been tampered with. + +### 6.2 Formal method + +Figure 2 is modified from [ITU-T Y.3320] and illustrates the overall flow from the informal description of the conceptual model, smart contract implementation and domain specific distributed ledger through to the implementation of a system using formal methods. Formal methods are those that specify or verify hardware/software systems in a way that is based on mathematics and logic. Characteristics to specify the system using mathematical symbols are also described using logical + +expressions to verify whether the system satisfies the specific conditions required by the user using mathematical properties to reduce the ambiguity and uncertainty inherent in natural language. In other words, it is possible to verify whether a complex system satisfies a specific condition and to have a reliable verified system. + +![Figure 2 – Overall flow of formal methods. The diagram shows the flow from informal descriptions to formal specifications and then to analysis and implementation. It distinguishes between modified modules (grey) and original modules (white).](f4fdd410cdb84df81274da55721e56fb_img.jpg) + +The diagram illustrates the overall flow of formal methods. At the top, four boxes represent different types of informal descriptions: 'Informal description Conceptual model' (white), 'Informal description Smart contract implementation' (grey), 'Informal description Domain specific distributed ledger description' (grey), and 'Informal description DApp implementation' (grey). These boxes point to two boxes representing formal specifications: 'Formal Specification Textual representation' (white) and 'Formal specification Graphical representation' (white). Both formal specification boxes point to a central box labeled 'Analysis' which includes '- Validation', '- Verification', and '- Conformance testing'. This central box is connected to a box labeled 'Implementation' (white) via a double-headed arrow. To the left of the 'Analysis' box, three boxes point to it with dashed lines: 'Smart contract and DApp developers' (grey), 'Domain experts' (white), and 'Formal methods experts' (white). To the right of the 'Analysis' box, three boxes are connected to it with dashed lines: 'Theorem proving' (white), 'Model checking' (white), and 'Static analysis' (white). A legend at the bottom indicates that grey boxes represent 'Modified module' and white boxes represent 'Original module'. + +Figure 2 – Overall flow of formal methods. The diagram shows the flow from informal descriptions to formal specifications and then to analysis and implementation. It distinguishes between modified modules (grey) and original modules (white). + +F.751.12(23) + +**Figure 2 – Overall flow of formal methods** + +There are two types of formal methods: formal specification and formal verification. Formal specification is based on the environment in which the system will operate using formal logic or mathematical logic. It describes the requirements that the system must meet and the system design to fulfil the requirements. The formal specification can be further divided into requirements specification and design specification. A requirements specification is a specification that defines what the system must satisfy, and a design specification shows how the system is constructed. Formal verification is a technique that verifies the inconsistency and completeness of the system by analysing the formal specification, using the verification method provided by the formal logic or mathematical logic, or verifying whether the design satisfies the requirements in a given assumption. This describes methods that are working for verification of smart contract models. + +- Theorem proving is that involves encoding a system and its desired properties into a particular mathematical logic. Then, it attempts to derive a formal proof of satisfaction of these properties based on the axioms and inference rules of the formal system. +- Model checking is a technique for automatically verifying a system model against its specification. +- Static analysis is any offline computation that inspects code and produces opinions about the code. + +## 7 Requirements of formal verification for smart contracts + +### 7.1 DApp requirements + +This clause identifies a set of DApp requirements that must be supported by DLT. + +- It is required that a DApp component provide an open interface in DLT. Through this, verification request and reply of smart contracts are performed. +- It is required that a DApp component provide the ability to build smart contracts and interact with the smart contract mechanism component. + +### 7.2 Smart contract mechanism requirements + +This clause identifies a set of smart contract mechanism requirements that must be supported by DLT. + +- It is required that the smart contract mechanism component provide an open interface in DLT. Through this, the verification request and reply of smart contracts are performed. +- It is required that the smart contract mechanism component provide the ability to store a compiled execute file of the smart contract in the distributed ledger. +- It is required that the smart contract mechanism component provide the ability to store a related file of the smart contract in the distributed ledger. + +### 7.3 Formal method requirements + +This clause identifies a set of formal method requirements that must be supported by DLT. + +- It is required that formal method component provide open interface in the distributed ledger environment. Through this, the verification request and reply of smart contracts are performed. +- It is recommended that formal method component provide verification function for script language problems of the smart contract, problems occurring in the process of interworking with internal or external data in distributed ledger and problems caused by a contract order problem. +- It is recommended that the formal method component allow access by authorized users. +- It is recommended that the formal method component provide formal syntax, interpret the meaning of scripting language and support distributed ledger external API and protocol interworking. +- It is recommended that the formal method component provide a function to check whether there are no logical errors between the existing contract in the distributed ledger and the smart contracts to be verified. +- It is recommended that the formal method component check whether there is an error in the order between smart contracts in the block to be created. +- The formal method component can optionally provide for verification of the user who composes or participates in a smart contract. +- The formal method component can optionally provide for verification of external data in a smart contract. + +# 8 Architecture of formal verification for smart contracts + +### 8.1 Overview + +In the high-level conceptual architecture of DLT, smart contracts and formal method components are configured as shown in Figure 3, modified from [ITU-T F.751.2]. The high-level conceptual architecture of DLT requires smart contract and smart contract mechanism modules to build smart contracts. Formal method component for verification is also needed. + +DApps are software that communicates with the DLT. The user interface of the distributed application is similar to a website or mobile application. The smart contract represents the core logic of a + +decentralized application. Therefore, these are executed through the interworking of smart contract mechanisms. + +Smart contract mechanisms are integral building blocks of the distributed ledger that process information from external sensors or events and help the distributed ledger manage. By using smart contract mechanisms for event processing and expressing logic, many complex applications can be built on DLT. Smart contract mechanisms are used in the process of configuring and executing decentralized applications. However, some DLT can only be composed of a DLT without decentralized application. In this environment, smart contract writing and execution is done on the DLT. + +A formal method component is configured to prevent errors in the process of writing and executing a smart contract. The formal method component has an interface to connect with the DLT and distributed applications. The formal method component is composed of a parser, a model builder and a model checker for smart contract and transaction data. + +![High-level DLT architecture for formal verification diagram. The diagram shows a layered architecture for a Distributed Ledger Technology (DLT). The central part is a large rectangle divided into several horizontal layers. From top to bottom: 'Application' (containing 'DApp(s)' and 'Smart contract'), 'Runtime', 'Life cycle', and 'Devtool'. Below these is a dashed box containing 'Governance and compliance' and 'Utils'. The next layer is 'Protocol', which contains 'Smart contract mechanism' (shaded), 'Account/DID', 'Consensus', 'Messaging', 'Right management', 'Ledger management', and 'Crypto'. Below 'Protocol' is a dashed box containing 'Resources', 'Data integrity and privacy util', 'Node management', 'Storage management', and 'Network management'. To the right of the central rectangle are two boxes: 'External interaction management' and 'External resource management'. To the far right is a vertical stack of boxes: 'Extensions' (containing 'Side chain', 'Off chain', 'Multi chain', 'Internal extensions', and '...'), 'Formal method for smart contract' (containing 'Parser for smart contract and transaction data', 'Model builder for smart contract and transaction data', and 'Model checker for smart contract and transaction data'), and 'F.751.12(23)'. A legend at the bottom left shows a white box for 'Original module' and a shaded box for 'Additional module'.](27b06ec9f42b5d727a2630f61a5f1861_img.jpg) + +The diagram illustrates the high-level DLT architecture for formal verification. It is structured as follows: + +- Central Architecture:** + - Application Layer:** Contains DApp(s) and Smart contract (Additional module). + - Runtime, Life cycle, Devtool:** Original modules. + - Governance and compliance and Utils:** Enclosed in a dashed box. + - Protocol Layer:** Contains Smart contract mechanism (Additional module), Account/DID, Consensus, Messaging, Right management, Ledger management, and Crypto. + - Resources, Data integrity and privacy util, Node management, Storage management, Network management:** Enclosed in a dashed box. +- External Management:** + - External interaction management + - External resource management +- Extensions:** Side chain, Off chain, Multi chain, Internal extensions, ... +- Formal method for smart contract:** + - Parser for smart contract and transaction data + - Model builder for smart contract and transaction data + - Model checker for smart contract and transaction data +- Legend:** + - Original module (White box) + - Additional module (Shaded box) +- Reference:** F.751.12(23) + +High-level DLT architecture for formal verification diagram. The diagram shows a layered architecture for a Distributed Ledger Technology (DLT). The central part is a large rectangle divided into several horizontal layers. From top to bottom: 'Application' (containing 'DApp(s)' and 'Smart contract'), 'Runtime', 'Life cycle', and 'Devtool'. Below these is a dashed box containing 'Governance and compliance' and 'Utils'. The next layer is 'Protocol', which contains 'Smart contract mechanism' (shaded), 'Account/DID', 'Consensus', 'Messaging', 'Right management', 'Ledger management', and 'Crypto'. Below 'Protocol' is a dashed box containing 'Resources', 'Data integrity and privacy util', 'Node management', 'Storage management', and 'Network management'. To the right of the central rectangle are two boxes: 'External interaction management' and 'External resource management'. To the far right is a vertical stack of boxes: 'Extensions' (containing 'Side chain', 'Off chain', 'Multi chain', 'Internal extensions', and '...'), 'Formal method for smart contract' (containing 'Parser for smart contract and transaction data', 'Model builder for smart contract and transaction data', and 'Model checker for smart contract and transaction data'), and 'F.751.12(23)'. A legend at the bottom left shows a white box for 'Original module' and a shaded box for 'Additional module'. + +**Figure 3 – High-level DLT architecture for formal verification** + +In order to provide formal verifications to the smart contract, additional functions comprised in the formal methods component are needed to process the parser, model builder and model checker for the smart contract. The parser for the smart contract function provides an open application interface for DApps or a smart contract mechanism to express syntactic analysis. The model checker for a smart contract function provides an open application interface for the DApp or smart contract mechanism to check whether new semantics can cause errors or operational conflicts within the DLT. + +### 8.2 Formal verification + +Smart contracts are expressed in various types of logic. They are composed of metadata and events constituting the contract. Problems can arise due to smart contracts built or participated in by malicious users. Therefore, previously written transactions and smart contracts are analysed to identify malicious users. Malicious user verification can be classified using reinforcement learning and deep learning methods. Based on this, it is possible to verify a user who composes or participates in a smart contract. + +For smart contracts built or participated in by normal users, the verification application is used for verification. Verifying a smart contract consists of a formal verification in the form of graphical and textual representation. A process of transmitting and parsing related information through an external interface is required. It has an architecture in which a verification model is created and verified based on parsed information. The relationship between the verification module and the smart contract for this is shown in Figure 4. + +![Diagram showing the relationship between the formal methods component and the DLT. The DLT side includes Application (DApp(s) with Smart contract), Protocol (Smart contract mechanism), and Resources. The Formal method side includes Parser, Model builder, and Model checker for smart contract and transaction data, plus External interaction management (External resource management). Interfaces connect the Application and Protocol to the Formal method component, and Resources to External resource management.](ff0952ef692c9d960ce5f6708bcc9711_img.jpg) + +The diagram illustrates the architecture for formal verification of smart contracts on a DLT. On the left, a large box labeled 'DLT' contains three nested boxes: 'Application' (which contains 'DApp(s)' and 'Smart contract'), 'Protocol' (containing 'Smart contract mechanism'), and 'Resources'. On the right, a large box labeled 'Formal method for smart contract' contains three stacked boxes: 'Parser for smart contract and transaction data', 'Model builder for smart contract and transaction data', and 'Model checker for smart contract and transaction data'. Below this is another box labeled 'External interaction management' containing 'External resource management'. Dashed double-headed arrows represent interfaces: 'Application formal methods interface' between the 'Smart contract' and the 'Parser'; 'Smart contract formal methods interface' between the 'Smart contract mechanism' and the 'Model checker'; and a direct connection between 'Resources' and 'External resource management'. A small label 'F.751.12(23)' is at the bottom right. + +Diagram showing the relationship between the formal methods component and the DLT. The DLT side includes Application (DApp(s) with Smart contract), Protocol (Smart contract mechanism), and Resources. The Formal method side includes Parser, Model builder, and Model checker for smart contract and transaction data, plus External interaction management (External resource management). Interfaces connect the Application and Protocol to the Formal method component, and Resources to External resource management. + +**Figure 4 – Relationship between the formal methods component and the DLT** + +Formal verification is performed in the following procedure: + +- The smart contract or smart contract mechanism module requests verification to the formal method component. +- The formal method component requests and receives smart contract code, transaction data, and external data. +- The parser module parses smart contract and transaction data. +- The parser module verifies whether any of the smart contract author and participants is a malicious user. +- The parser module verifies the validity of external data. +- The model builder module creates a formal verification model based on the parsed data. +- The model checker performs verification based on the created model and replies with the result. + +Data used to build smart contracts in DLT can be divided into external data and internal data. Internal data refers to data managed by DLT. External data is data provided from outside DLT. In the case of internal data in DLT, the data can be verified during the smart contract verification process. + +In order to use external data in a smart contract, a smart contract for external data verification or a provision site is required. Through this, external data is authenticated and accessed. A flow chart for this is shown in Figure 5. When a smart contract execution request is received from a user, the smart contract checks whether external data is used. If external data is used, it checks whether there is a related data verification smart contract and executes it. External data verification is performed in the procedure shown in Figure 5. + +![Flow chart for smart contract based external data verification](4ee27dbf5ef12e7b58b0ef0937bc5a5e_img.jpg) + +``` +graph TD; A[Smart contract user] --> B[Smart contract request]; B --> C{External data access}; C -- No --> D[Smart contract execution]; C -- Yes --> E{Smart contract for data verification}; E -- No --> F[Create a smart contract for data verification]; F --> E; E -- Yes --> G[Smart contract execution for data verification]; G --> H[Smart contract execution for user]; H --> D; +``` + +The flowchart illustrates the process for smart contract based external data verification. It begins with a 'Smart contract user' box, which points down to a 'Smart contract request' box. This box points down to a decision diamond labeled 'External data access'. If the answer is 'No', the flow proceeds down to a 'Smart contract execution' box. If the answer is 'Yes', the flow proceeds right to a decision diamond labeled 'Smart contract for data verification'. From this diamond, a 'No' path leads down to a 'Create a smart contract for data verification' box, which then loops back up to the 'Yes' path of the 'Smart contract for data verification' diamond. A 'Yes' path from the 'Smart contract for data verification' diamond leads right to a 'Smart contract execution for data verification' box. This box points down to a 'Smart contract execution for user' box, which then loops back up to the 'Smart contract execution' box at the end of the process. + +Flow chart for smart contract based external data verification + +F.751.12(23) + +**Figure 5 – Flow chart for smart contract based external data verification** + +The procedure is as follows: + +- The user requests smart contract execution in DLT. +- The system checks whether external data is used. +- The system checks that a smart contract has been created to verify the use of external data. +- The system executes smart contracts for external data usage and smart contracts requested by users. + +# Appendix I + +## Use case of verification for smart contract + +(This appendix does not form an integral part of this Recommendation.) + +### I.1 Overall procedure for formal verification + +A smart contract is written using the program language supported in the distributed ledger and stored in the ledger in the form of an executable file. A smart contract is executed according to the event conditions. There are two ways to write these smart contracts, through a smart contract module or through a DApp. The procedure for writing and verifying smart contracts through a DApp is as follows. + +- Building a smart contract through DApp. +- The smart contract written by DApp is delivered to the formal method component. +- The parser classifies smart contracts written in a programming language. +- A model builder generates a verification model from the parsed smart contract. +- The model checker is used to verify, and the results are delivered to the DApp. +- The DApp decides whether to execute a smart contract based on the verification results. +- The smart contract module is applied to the DApp according to whether the contract is executed or not. + +### I.2 Internet of Things (IoT) environment + +Various distributed ledger platforms have emerged for the Internet of Things (IoT) environment. In addition, various distributed ledger applications that support the IoT environment have been developed and applied to services. In this distributed ledger platform and application for the IoT environment, smart contracts that support events between IoT devices are very important. + +**Table I.1 – Distributed ledger platforms or applications for IoT** + +| | | +|--------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Platforms | Distributed ledger for IoT, which overcomes the inefficiencies of current distributed ledger designs and introduces a new way of reaching consensus in a decentralized peer-to-peer system | +| | A consortium exploring the role distributed ledger may have in providing security for the IoT. | +| | A decentralized network for IoT powered by a privacy-centric distributed ledger | +| Applications | E-charging mobility: Share & Charge was a "distributed-ledger-based mobility product for EV users, allowing users to share charging stations across a decentralized platform without middlemen." | +| | Energy: Combining IoT devices (smart meters) with DLT simplified the process of tracking and supplying renewable energy to end users | +| | Wireless network: A decentralized machine network simplifies connecting anything to the Internet through a distributed ledger, wireless network and open-source software | +| | Location data sharing: Provides infrastructure for the transparent and compliant exchange of location data; foot traffic sensor readings, store visit information and commute routes | + +#### I.2.1 Smart contract verification for resource sharing service + +Resource sharing services support IoT device resource sharing, such as for bicycles. Consumer, providers and service providers are required to provide resource sharing services. Figure I.1 shows a distributed-ledger-technology-based resource sharing platform. + +![Figure I.1: DLT based resource sharing service diagram. A Consumer sends a 'Payment' to a 'Service provider (Resource sharing for IoT device)' and receives a 'Utilize' response. The Service provider sends a 'Provide' to a 'Provider' and receives a 'Reward' response. All three entities are connected to a 'Distributed ledger technology' block. The diagram is labeled F.751.12(23) in the bottom right corner.](0f985b39edc1d52ba3600c438bc8f0a5_img.jpg) + +Figure I.1: DLT based resource sharing service diagram. A Consumer sends a 'Payment' to a 'Service provider (Resource sharing for IoT device)' and receives a 'Utilize' response. The Service provider sends a 'Provide' to a 'Provider' and receives a 'Reward' response. All three entities are connected to a 'Distributed ledger technology' block. The diagram is labeled F.751.12(23) in the bottom right corner. + +**Figure I.1 – DLT based resource sharing service** + +The consumer builds a smart contract and requests a service to utilize the resource sharing service. It is necessary to check if there is an error in the smart contract written by the consumer, so verification is performed before a smart contract is executed. The overall process for this is shown in Figure I.2. + +![Figure I.2: Overall process of smart contract verification for resource sharing service sequence diagram. The sequence involves Consumer, Distributed ledger platform (Object contract), Dapp (Access control) with Booking details (Device: From to, Time: Start end, Pay:), Verification, Distributed ledger platform (Payment system), and Provider. The process includes message actions, rent transactions, verification requests, transaction data requests/replies, verification results, object transactions, payments, access requests/replies, and confirmations. The diagram is labeled F.751.12(23) in the bottom right corner.](a83ba9e3e2c1e21dd69953a7b09e45b4_img.jpg) + +Figure I.2: Overall process of smart contract verification for resource sharing service sequence diagram. The sequence involves Consumer, Distributed ledger platform (Object contract), Dapp (Access control) with Booking details (Device: From to, Time: Start end, Pay:), Verification, Distributed ledger platform (Payment system), and Provider. The process includes message actions, rent transactions, verification requests, transaction data requests/replies, verification results, object transactions, payments, access requests/replies, and confirmations. The diagram is labeled F.751.12(23) in the bottom right corner. + +**Figure I.2 – Overall process of smart contract verification for resource sharing service** + +## Appendix II + +## Related projects on verification for smart contract + +(This appendix does not form an integral part of this Recommendation.) + +### **VeriDis: Verification of Distributed Systems** + +- This project aims to exploit and further develop the advances and integration of interactive and automated theorem proving applied to the area of concurrent and distributed systems. The goal of this project is to assist algorithm and system designers to carry out formally proved developments, where proofs of relevant properties as well as bugs can be found completely automatically. +- Reference site: + +### **Certik: Building Fully Trustworthy SmartContracts and distributed ledger Ecosystems** + +- This platform is envisaged to be a formal verification framework for building fully trustworthy smart contacts and distributed ledger ecosystems. Different from the traditional testing approaches to detect bugs, this platform attempts to mathematically prove that distributed ledger ecosystems are bug-free. +- Reference site: + +# Bibliography + +- [b-ITU-T F.751.0] Recommendation ITU-T F.751.0 (2020), *Requirements for distributed ledger systems*. +- [b-ITU-T X.1401] Recommendation ITU-T X.1401 (2019), *Security threats to distributed ledger technology*. +- [b-ISO 22739] ISO 22739:2020, *Blockchain and distributed ledger technologies — Vocabulary*. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.751.14-202406-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.751.14-202406-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..38bdd7bc1d3581fbdc3cf6b035f618693dbe3846 --- /dev/null +++ b/marked/F/T-REC-F.751.14-202406-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:75efa8e3513a799767c04a0bb7bcd87a550f2378f5880629068581808c1c9f7f +size 7353 diff --git a/marked/F/T-REC-F.751.14-202406-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg b/marked/F/T-REC-F.751.14-202406-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..6be4886c40aa1a6ab0ada440caf40f60d006eed1 --- /dev/null +++ b/marked/F/T-REC-F.751.14-202406-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:06b1916829ab691f240d1b18f4c641560ad427bc7aed9136adc6de41a3cf68bd +size 85275 diff --git a/marked/F/T-REC-F.751.14-202406-I_PDF-E/b8661c6c54f72ecc7ff6cb05e47b2891_img.jpg b/marked/F/T-REC-F.751.14-202406-I_PDF-E/b8661c6c54f72ecc7ff6cb05e47b2891_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..b9a506161c1ff4a15d61ab8588389485e023198e --- /dev/null +++ b/marked/F/T-REC-F.751.14-202406-I_PDF-E/b8661c6c54f72ecc7ff6cb05e47b2891_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:387808e6eb4a850dfd53d75f67f9d4af72d56edc30ead76c1705c0b172f9c4a3 +size 113337 diff --git a/marked/F/T-REC-F.751.14-202406-I_PDF-E/raw.md b/marked/F/T-REC-F.751.14-202406-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..a8639679857bd0661072ca8c9ae017336c85a039 --- /dev/null +++ b/marked/F/T-REC-F.751.14-202406-I_PDF-E/raw.md @@ -0,0 +1,499 @@ + + +# Recommendation **ITU-T F.751.14 (06/2024)** + +SERIES F: Non-telephone telecommunication services +Multimedia services + +--- + +## **Reference architecture for information tracing of renewable energy consumption based on distributed ledger technology** + +![ITU logo](390120de4fe440c42fea8154fcaad334_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with white lines representing latitude and longitude, and the letters 'ITU' in a bold, blue, sans-serif font superimposed on the globe. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.751.14 + +## Reference architecture for information tracing of renewable energy consumption based on distributed ledger technology + +## Summary + +Recommendation ITU-T F.751.14 defines the relevant participants, tracing process and reference architecture for the renewable energy consumption information traceability (RECT) system based on distributed ledger technology (DLT). It provides guidance for users and demanders, who choose DLT for renewable energy consumption information traceability. + +## History\* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.751.14 | 2024-06-13 | 16 | 11.1002/1000/15919 | + +## Keywords + +DLT, reference architecture, renewable energy consumption, traceability technology. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2024 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|----------------------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Participants ..... | 2 | +| 6.1 Business-related parties ..... | 2 | +| 6.2 Regulatory agencies..... | 3 | +| 6.3 System service providers ..... | 3 | +| 7 Traceability process of DLT-based RECT system ..... | 3 | +| 8 Reference architecture of the DLT-based RECT system ..... | 4 | +| 9 Requirements for infrastructure layer of DLT-based RECT system ..... | 5 | +| 10 Requirements for the core technology layer of the DLT-based RECT system ..... | 5 | +| 11 Requirements for data management layer of DLT-based RECT system ..... | 6 | +| 11.1 Data integration ..... | 6 | +| 11.2 Data pre-processing ..... | 6 | +| 11.3 Data on-chain..... | 6 | +| 11.4 Data security ..... | 6 | +| 12 Requirements for application service layer of DLT-based RECT system ..... | 7 | +| 12.1 User management ..... | 7 | +| 12.2 Data interface services..... | 7 | +| 12.3 Renewable energy consumption information on-chain..... | 7 | +| 12.4 Renewable energy consumption information traceability ..... | 9 | +| Bibliography..... | 11 | + + + +# Recommendation ITU-T F.751.14 + +## Reference architecture for information tracing of renewable energy consumption based on distributed ledger technology + +## 1 Scope + +This Recommendation defines the requirements and reference architecture for the renewable energy consumption information traceability (RECT) system based on distributed ledger technology (DLT) including: + +- Relevant participants; +- Traceability process of the DLT-based RECT system; +- Reference architecture of the DLT-based RECT system; and +- Requirements for the DLT-based RECT system. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T F.751.2] Recommendation ITU-T F.751.2 (2020), *Reference framework for distributed ledger technologies*. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 blockchain** [b-ITU-T F.751.0]: A type of distributed ledger that is composed of digitally recorded data arranged as a successively growing chain of blocks with each block cryptographically linked and hardened against tampering and revision. + +**3.1.2 distributed ledger technology (DLT)** [b-ISO/TC 307]: Technology that enables the operation and use of distributed ledgers. + +**3.1.3 consensus mechanism** [b-ITU-T X.1400]: Rules and procedures by which consensus is reached. + +**3.1.4 on-chain** [b-ITU-T X.1400]: Located, performed or run inside a blockchain system. + +**3.1.5 off-chain** [b-ITU-T X.1400]: Related to a blockchain system, but located, performed or run outside that blockchain system. + +**3.1.6 power grid** [b-ITU-T F.751.5]: The whole consisting of substations of various voltages and transmission and distribution lines in the power system. + +**3.1.7 renewable energy** [b-ITU-T L.1201]: This is mainly non-fossil fuel converted into electricity (e.g., solar energy, wind, water flow, and biomass). + +**3.1.8 smart contract** [b-ITU-T X.1400]: A program written on a distributed ledger system which encodes the rules for specific types of distributed ledger system transactions in a way that can be validated, and triggered by specific conditions. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 renewable energy consumption:** When power users purchase renewable energy certificates (RECs) or participate in renewable energy trading or spontaneous self-use of renewable energy to meet green-low-carbon development needs. + +**3.2.2 renewable energy consumption information:** The relevant data of the entire process of renewable energy consumption, including renewable energy generation data, renewable energy transmission data, renewable energy transaction data, renewable energy sales data and other proving-consumption data. + +NOTE 1 – For those involving RECs, it also includes RECs data and RECs transaction data. + +NOTE 2 – For those involving spontaneous-self-use, it also includes its power generation data, power usage data, and other grid connected data. + +**3.2.3 renewable energy consumption information traceability:** The process of tracing relevant data and information related to the production, trading, and consumption of renewable energy, which starts from the consumer side. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +DLT            Distributed Ledger Technology + +REC            Renewable Energy Certificate + +RECT          Renewable Energy Consumption Information Traceability + +## 5 Conventions + +The following conventions are used in this Recommendation: + +- The keywords "**is required to**" indicate a requirement which must be strictly followed and from which no deviation is permitted, if conformance to this Recommendation is to be claimed. +- The keywords "**is recommended**" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. + +## 6 Participants + +### 6.1 Business-related parties + +The business-related parties of the DLT-based RECT system includes power generation users, power grid enterprises, renewable energy trading institutions, renewable energy sales companies and renewable energy users. + +- **Renewable energy generation users:** They are integrated into the power grid for operation and engage in the production and sales of power from non-fossil fuel or renewable energy. +- **Power grid enterprises:** They provide power supply services for renewable energy users, including installation, metering, meter reading, settlement and charging. + +- **Renewable energy trading institutions:** They are responsible for organizing renewable energy trading, providing relevant settlement basis, calculating the renewable energy demand of generation enterprises and users, and allocating trading power within the trading area. +- **Renewable energy sales companies:** They provide renewable energy trading agency services for small-scale users and demanders. +- **Renewable energy users:** They purchase renewable energy or RECs. Some users may have distributed resources that are spontaneously self-used or partial surplus grid connection. + +### 6.2 Regulatory agencies + +Regulatory agencies provide regulatory audits services of the RECT process. + +### 6.3 System service providers + +The system service providers provide services such as DLT-based RECT system construction, deployment, optimization, operation and maintenance. + +## 7 Traceability process of DLT-based RECT system + +The traceability process of DLT-based RECT system is recommended to include the following steps, also shown in Figure 1: + +- **Step 1:** Data integration. It is required to integrate data from renewable energy generation users, power grid enterprises, renewable energy trading institutions, renewable energy sales companies and renewable energy users. The data is required to include renewable energy generation data, RECs data, renewable energy transmission data, renewable energy transaction data, RECs transaction data, renewable energy sales and proving-consumption data. +- **Step 2:** Data pre-processing. Data is required to undergo pre-processing such as cleaning, standardization and normalization. +- **Step 3:** Data on-chain. After pre-processing, the data is required to be stored on the blockchain. +- **Step 4:** Data verification. To ensure data consistency and accuracy, it is required to support the comparison and verification of the on-chain data and the source business data. +- **Step 5:** Data traceability. Renewable energy users and regulatory agencies is recommended to obtain and query traceability information on demand. + +![](a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg) + +| Renewable energy generation users | Power grid enterprises | Renewable energy trading institutions | Renewable energy sales companies | Renewable energy users | DLT-based RECT system | Regulatory agencies | +|--------------------------------------------|------------------------------------|----------------------------------------|----------------------------------|--------------------------|------------------------|---------------------| +| Renewable energy generation data/RECs data | Renewable energy transmission data | Renewable energy/RECs transaction data | Renewable energy sales data | Proving-consumption data | 1. Data integration | | +| | | | | | 2. Data pre-processing | | +| | | | 3. Data on-chain | | | | +| | | 4. Data verification | | | | | +| | 5. Data traceability | Traceability information query | | | | | + +F.751.14(24) + +**Figure 1 – The traceability process of the DLT-based RECT system** + +## 8 Reference architecture of the DLT-based RECT system + +This Recommendation defines a reference architecture for DLT-based RECT systems; see Figure 2. + +![](b8661c6c54f72ecc7ff6cb05e47b2891_img.jpg) + +| | | | | +|---------------------------|---------------------------------------------------|-------------------------------------------------------|----------------------------------| +| Application service layer | | | | +| User management | Renewable energy consumption information on-chain | Renewable energy consumption information traceability | | +| Data interface service | Storage service | Traceability service | Traceability information query | +| | Verification service | Traceability information analysis | Traceability information display | +| Data management layer | | | | +| Data integration | Data pre-processing | Data on-chain | Data security | +| Core technology layer | | | | +| Node management | Smart contract | Consensus mechanism | Cross chain technology | +| Infrastructure layer | | | | +| Computing resource | Network resource | Storage resource | | + +F.751.14(24) + +**Figure 2 – Reference Architecture for DLT-based RECT system** + +The reference architecture for the DLT-based RECT system is composed of infrastructure layer, core technology layer, data management layer and application service layer. + +- The infrastructure layer provides the software environment and hardware facilities required for the system's regular operation, including computing resources (CPU, GPU, ASIC, etc.), network resources (switches, routers, network bandwidth, etc.) and storage resources (hard disks, clouds, virtual hosts, etc.). +- The core technology layer provides the technical support required for the data management layer and application service layer, including node management, smart contract, consensus mechanism and cross-chain technology. +- The data management layer provides the ability to collect and standardize data, supporting the renewable energy consumption information on-chain and traceability, including data integration, data pre-processing, data on-chain and data security. +- The application service layer provides management and services for specific businesses, including user management, data interface services, renewable energy consumption information on-chain, and renewable energy consumption information traceability. + +## 9 Requirements for infrastructure layer of DLT-based RECT system + +The infrastructure layer provides the DLT platform construction resources. Its requirements refer to [ITU-T F.751.2], as follows: + +- **Computing resource:** It is required to have compatibility with renewable energy consumption information on-chain and with the traceability business requirement. +- **Network resource:** It is required to support efficient and secure communication between nodes, and possess authentication and encryption capabilities. +- **Storage resource:** It is required to possess storage for strategy documents, including storage management and data backup mechanisms. In addition, it is required to meet the storage needs required for renewable energy consumption on-chain and for traceability. + +## 10 Requirements for the core technology layer of the DLT-based RECT system + +The core technology layer is required to meet the following requirements: + +- **Node management:** It is required to support the ability to start, stop, and restart blockchain nodes and networks, and adjust the nodes' resource occupancy. It is also required to support the deployment of blockchain nodes and networks across organizations and regions. +- **Smart contract:** It is required to provide programming language support, development environment and deployment testing tools required for the RECT system. It is also required to configure different smart contract response rules according to different scenarios and situations, including renewable energy consumption information on chain contracts, calculating contracts, power summary contracts, data verification and information push smart contracts. The calculating smart contracts are configured using data analysis methods such as correlation, classification and regression. +- **Consensus mechanism:** The selection of the consensus algorithm is required to follow [ITU-T F.751.2]. +- **Cross-chain technology:** It is required to support multiple cross-chain interaction modes, and both isomorphic and heterogeneous cross-chain, and possess the ability of cross-chain mutual authentication and cross-chain interoperability. In addition, it is required to support cross-chain demanders to coordinate and negotiate based on the blockchain communication standards of different countries or regions. + +## **11 Requirements for data management layer of DLT-based RECT system** + +### **11.1 Data integration** + +Data integration is required to meet the following requirements: + +- Support the integration of various data types, including renewable energy generation data, renewable energy sales data, renewable energy transmission data, renewable energy transaction data, proving-consumption data, REC data and REC transaction data. +- Support structured data, unstructured data and semi-structured data. +- Support data integration through access methods such as databases, service interface calls and internet data-collection. +- Support data integration through methods such as associated business systems, file transfer, data extraction and content crawling. +- Support monitoring data access channel operational status and rationality by data resource monitoring technology. +- Provide multidimensional analysis and result viewing such as logs and abnormal alarms. + +### **11.2 Data pre-processing** + +Data pre-processing is required to meet the following requirements: + +- Support data extraction, cleaning and specification, handling invalid values, filling in missing values and filtering out data that does not meet requirements. +- Support data normalization and standardization processing, and convert it into a data form suitable for calculation according to business needs and on-chain technical requirements. + +### **11.3 Data on-chain** + +Data on-chain is required to meet the following requirements: + +- Send an on-chain request to the blockchain node after data encapsulation. +- Possess the ability to verify the consistency of the data on-chain and the original data. +- Support the on-chain process credibility by using privacy computing technology. +- Provide a unique identity verification method for the data owners after data on-chain and build a trusted digital identity. + +### **11.4 Data security** + +Data security is required to meet the following requirements: + +- Sensitive data collected is required to be authorized by data ownership before being encrypted and circulated according to power data's classification and grading regulations of different countries. +- Set permission management based on the participants' demand to prevent unauthorized node accessing and calling. +- Support identity authentication of the accessed data source system to prevent data source impersonation. +- Support data backup management to avoid data omission and loss during operation. +- Support the full process logging of data management, including data source related logs, process operation logs and system configuration change logs. + +## **12 Requirements for application service layer of DLT-based RECT system** + +### **12.1 User management** + +User management is required to meet the following requirements: + +- Provide decentralized user identity verification using blockchain. +- Support services including user login, user permission management, user information modification, password reset, addition of roles and permissions for users. + +### **12.2 Data interface services** + +The system provides data interface services, including user interface, renewable energy consumption information interface, management interface and cross-chain interface. The data interface service is recommended to meet the following requirements: + +- The user interface provides participants with information permission configuration, verification, management and sharing capabilities. +- The renewable energy consumption information interface integrates renewable energy consumption information, supporting the integration of related business systems. +- The management interface is oriented towards regulatory authorities and system service providers. +- The cross-chain interface provides interface configurations and specifications that are suitable for business needs in countries and regions with cross-chain requirements. Nodes in the blockchain interact with off-chain systems. + +### **12.3 Renewable energy consumption information on-chain** + +#### **12.3.1 Storage service** + +##### **12.3.1.1 Renewable energy generation data** + +The storage requirements for renewable energy power generation data are as follows: + +- Renewable energy generation user information is required to include the power generation users' name, contact information, etc. +- Renewable energy generation information is required to include power generation type, installed capacity, power generation units and geographical location, etc. +- Operational data is required to include real-time online and cumulative online power quantity. +- The data configuration items are required to be adjusted according to actual needs. +- It is recommended that original data be uploaded either directly or in batches, depending on its reliability. + +NOTE – The renewable energy generation users involved in spontaneous-self-use are required to also include power generation data, power usage data and other grid connected data. + +##### **12.3.1.2 Renewable energy transmission data** + +The storage requirements for renewable energy transmission are as follows: + +- Power grid enterprises information is required to include the enterprise's name, etc. +- Transmission information is required to include transmission provinces, transmission electricity quantity, transmission electricity price, power grid loss rate and grid loss discount. +- It is recommended to choose the original data on-chain or hash value on-chain according to the trust of the original data. + +##### **12.3.1.3 Renewable energy transaction data** + +The storage requirements for renewable energy transaction data are as follows: + +- Renewable energy trading institutions information is required to include the institution name, geographical area covered by the power markets, etc. +- Renewable energy trading contract information is required to include contract number, contract name, contract type, contract power quantity, contract start time, contract end time, contract power price and power quantity type. +- Settlement information is required to include settlement period, type of settlement power, settlement power quantity, settlement power price, settlement power fee and corresponding contract number. +- The data configuration items are required to be adjusted according to actual needs. +- It is recommended to choose the original data on-chain or hash value on-chain according to the trust of the original data. + +##### **12.3.1.4 Renewable energy sales data** + +The storage requirements for renewable energy sales data are as follows: + +- Renewable energy sales companies' information is required to include the company name, etc. +- Power sales information is required to include power generation type and the geographical location of the seller. +- It is recommended to choose the original data on-chain or hash value on-chain according to the trust of the original data. + +##### **12.3.1.5 Proving-consumption data** + +The storage requirements for renewable energy proving data are as follows: + +- Renewable energy users' information is required to include the username, the geographical area and location, and the user's cumulative power quantity of renewable energy consumption. +- It is required to include power transaction contracts and settlement documents signed by renewable energy users when purchasing green power. +- It is required to include RECs transaction contracts and settlement documents signed by renewable energy users when purchasing RECs. +- It is recommended to choose the original data on-chain or hash value on-chain according to the trust of the original data. + +##### **12.3.1.6 REC data** + +The storage requirements for REC data are as follows: + +- REC information is required to include renewable energy generation enterprise, country, geographical location, power generation type, power generation quantity and REC number. +- REC issuance information is required to include the issuing institution and date. + +##### **12.3.1.7 RECs transaction data** + +The storage requirements for REC transaction data are as follows: + +- RECs transaction contracts information is required to include RECs sellers (name, country, and geographical location), RECs buyers (name, country, geographical location), the amount of RECs transaction, and contract signing time. +- RECs trading information is required to include trading institutions and trading methods. + +#### **12.3.2 Verification services** + +The verification service is recommended to meet the following requirements: + +- Provide real-time verification and automated comparison services for on-chain information. +- Set the threshold for abnormal storage with functions such as data anomaly warning. +- Use regular testing or manual triggering to verify the consistency between the current business data and the original on-chain data. +- The verification results include the data source system, access path, data type, hash value on-chain, block height, timestamp, other information and the number of verification times. + +### **12.4 Renewable energy consumption information traceability** + +#### **12.4.1 Traceability service** + +The traceability service is required to meet the following requirements: + +- Possess the ability to verify the integrity of traceability data. +- Establish rules for the flow of traceability information and share them with participants according to the rules. +- Provide traceability services for data from different stages of renewable energy generation, sales, transmission, trading and consumption. +- Support bidirectional traceability, that is, tracing renewable energy consumption information forward or backward through a particular stage of information. + +#### **12.4.2 Traceability information query** + +The traceability information query is required to meet the following requirements: + +- The traceability information includes renewable energy generation data, renewable energy transmission data, renewable energy transaction data, renewable energy sales data, renewable energy proving data, RECs and RECs transaction data. +- Support to query traceability information based on different time dimensions such as month, quarter and year. +- Support to query traceability information based on the type and location of renewable energy. +- Support to customize the traceability information query module according to their business needs. + +#### **12.4.3 Traceability information analysis** + +The traceability information analysis is recommended to meet the following requirements: + +- Support the statistical analysis of renewable energy generation, renewable energy consumption, renewable energy generation forecasting, renewable energy demand forecasting, etc. The statistical analysis is conducted by month, quarter, year and other cycles, using year-on-year, month-on-month and other methods, or by region and user type, and automatically generate analysis reports. +- Analyse user traceability and query behaviour by region, user type, market entity, etc., and set up cross-provincial and cross-regional analysis function. The management functions include transaction analysis, regional consumption characteristics analysis and user type analysis. + +#### **12.4.4 Traceability information display** + +The traceability information display is recommended to meet the following requirements: + +- Support the presentation of traceability information in various ways such as diagrams, tables, text and traceability code. +- Support the display of energy and data flows in critical links. + +## Bibliography + +- [b-ITU-T F.751.0] Recommendation ITU-T F.751.0 (2020), *Requirements for distributed ledger systems.* +- [b-ITU-T F.751.5] Recommendation ITU-T F.751.5 (2022), *Requirements for distributed ledger technology-based power grid data management.* +- [b-ITU-T L.1201] Recommendation ITU-T L.1201 (2014), *Architecture of power feeding systems of up to 400 VDC.* +- [b-ITU-T X.1400] Recommendation ITU-T X.1400 (2020), *Terms and definitions for distributed ledger technology.* +- [b-ISO/TC 307] ISO DIS 22739, *Blockchain and distributed ledger technologies – Terminology.* + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.751.15-202406-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/F/T-REC-F.751.15-202406-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d4d8969f805414689ab79cfd08c333d454d482b3 --- /dev/null +++ b/marked/F/T-REC-F.751.15-202406-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:636c3839edff955a5b6208530b05fa1c983ab153d90f00216517a7a99478b988 +size 7074 diff --git a/marked/F/T-REC-F.751.15-202406-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.751.15-202406-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..991ab340accc56dcea32dcb198ea697c6016fde7 --- /dev/null +++ b/marked/F/T-REC-F.751.15-202406-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7d12b86a3882c9baf4f6d1bc0709606eac194bc4df8da93a7a34fc98140fb3e2 +size 184372 diff --git a/marked/F/T-REC-F.751.15-202406-I_PDF-E/raw.md b/marked/F/T-REC-F.751.15-202406-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..fa449502f2bef68c11ffc308a4911e670794dfc3 --- /dev/null +++ b/marked/F/T-REC-F.751.15-202406-I_PDF-E/raw.md @@ -0,0 +1,531 @@ + + +# Recommendation **ITU-T F.751.15 (06/2024)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +## --- **Assessment methods for distributed ledger technology (DLT) management service platforms** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with white lines representing latitude and longitude, and the letters 'ITU' in a bold, blue, sans-serif font superimposed on the globe. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.751.15 + +## Assessment methods for distributed ledger technology (DLT) management service platforms + +## Summary + +Recommendation ITU-T F.751.15 provides the assessment methods for distributed ledger technology management service platform (DLT-MSP) and the assessment criteria cover the distributed ledger technology (DLT) layer, management layer and business layer. The description of each test case is composed of test purpose, test workflows and expected results. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.751.15 | 2024-06-13 | 16 | 11.1002/1000/15920 | + +## Keywords + +Assessment method, DLT, DLT-MSP, management service platform. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2024 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|-----|----------------------------------------------|----| +| 1 | Scope ..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 2 | +| 4 | Abbreviations and acronyms ..... | 2 | +| 5 | Conventions ..... | 2 | +| 6 | Overview of assessment methods ..... | 2 | +| 7 | Assessment method for DLT layer ..... | 3 | +| 7.1 | Cryptographic algorithm ..... | 3 | +| 7.2 | Consensus mechanism..... | 3 | +| 7.3 | P2P network..... | 4 | +| 7.4 | Smart contract..... | 4 | +| 7.5 | Data storage ..... | 5 | +| 8 | Assessment method for management layer ..... | 5 | +| 8.1 | User management ..... | 5 | +| 8.2 | Node management ..... | 6 | +| 8.3 | DLT management..... | 6 | +| 8.4 | Smart contract management ..... | 8 | +| 8.5 | Log management ..... | 8 | +| 9 | Assessment method for business layer ..... | 9 | +| 9.1 | User key management ..... | 9 | +| 9.2 | Interface management ..... | 9 | +| 9.3 | DLT browser ..... | 9 | +| 9.4 | Contract template..... | 10 | +| | Bibliography..... | 11 | + + + +# Assessment methods for distributed ledger technology (DLT) management service platforms + +# 1 Scope + +This Recommendation specifies the assessment methods for distributed ledger technology management service platform (DLT-MSP), including the distributed ledger technology (DLT) layer, management layer and business layer. The description of each test case is composed of the test purpose, test workflows and expected results. This Recommendation can be used as a guideline for DLT management service platform assessment. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T F.751.16] Recommendation ITU-T F.751.16 (2024), *Reference framework for distributed ledger technology (DLT) management service platforms*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 distributed ledger** [b-ITU-T F.751.0]: A type of ledger that is shared, replicated, and synchronized in a distributed and decentralized manner. + +**3.1.2 peer-to-peer network** [b-ITU-T F.751.7]: A computer network comprised of nodes with equal control and operation capabilities. + +**3.1.3 consensus** [b-ITU-T F.751.0]: Rules and procedures by which consensus among DLT nodes is reached. + +**3.1.4 smart contract** [b-ITU-T F.751.0]: A program written on a distributed ledger system which encodes the rules for specific types of distributed ledger system transactions in a way that can be validated and triggered by specific conditions. + +**3.1.5 entity** [b-ITU-T F.751.0]: Anything that has a separately identifiable existence (e.g., organization, person, group, smart contract or device). An entity uses distributed ledger technology to solve the problem of its business or information systems. + +**3.1.6 transaction** [b-ITU-T F.751.0]: An incident or an operation that leads to a change in the status of the ledger changed, such as adding a record to the ledger or equivalent exchange based on currency. + +**3.1.7 Byzantine fault tolerance** [b-ITU-T X.1400]: Property that enables a system to continue to operate properly even if some of its components fail or the existence of intentional bad actors. + +**3.1.8 crash fault tolerance** [b-ITU-T X.1400]: Property that enables a system to continue operating properly even if some of its components crash. + +**3.1.9 node** [b-ITU-T F.751.1]: Device or process that participates in a distributed ledger network. + +**3.1.10 blockchain as a service (BaaS)** [b-ITU-T X.1400]: A cloud service category in which the capabilities provided to the cloud service customer are to deploy and manage a blockchain network in order to enable the abilities of consensus, smart contract, transaction, crypto engine, block record storage, peer-to-peer connectivity and management using blockchain. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following term: + +**3.2.1 distributed ledger technology management service platform (DLT-MSP)**: A management and service platform that helps users to create, manage and maintain distributed ledger technology (DLT) systems and applications. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|---------|-----------------------------------------------------------| +| BFT | Byzantine Fault Tolerance | +| CFT | Crash Fault Tolerance | +| DAPP | Decentralized Application | +| DLT | Distributed Ledger Technology | +| DLT-MSP | Distributed Ledger Technology Management Service Platform | +| P2P | Peer-to-Peer | +| SDK | Software Development Kit | + +# 5 Conventions + +None. + +# 6 Overview of assessment methods + +The distributed ledger technology management service platform (DLT-MSP) assessment framework is shown in Figure 1, referencing [ITU-T F.751.16]. The assessment criteria cover the distributed ledger technology (DLT) layer, management layer and business layer, and the aim of this Recommendation is to define specific test methods and test cases for DLT-MSP assessment. + +For each test case, the following three aspects are proposed: + +- test purpose, which clarifies the purpose of this test case; +- test workflow, which specifies the test execution steps of this test case for the testers; and +- expected results, which specifies the expected test consequences of this test case. + +![Reference framework of DLT-MSP diagram showing layers: Application, Business, Management, DLT, and Resource, with a Security layer on the right.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +The diagram illustrates the reference framework of DLT-MSP, organized into five horizontal layers and a vertical security layer on the right. The layers are: + +- Application layer:** Supply chain finance, Copyright verification, Deposit evidence, Other application. +- Business layer:** User key management, Interface management, Browser, Contract templates. +- Management layer:** + - User management: User creation, Information management, Permission management, User notification. + - Node management: Node deployment, Resource configuration, Information query, Node monitor. + - DLT management: Cryptographic management, Consensus management, Network management, Storage management. + - Smart contract management: Contract deployment, Contract upgrade, Contract freeze, Contract unfreeze. + - Log management: Node log, User log, System log, Other log. +- DLT layer:** Cryptographic algorithm, Consensus mechanism, P2P network, Smart contract, Data storage. +- Resource layer:** Computing resource, Network resource, Storage resource. + +The **Security layer** on the right side includes: Application security, Business security, User security, Node security, Smart contract security, Log security, DLT security, and Resource security. + +F.751.15(24) + +Reference framework of DLT-MSP diagram showing layers: Application, Business, Management, DLT, and Resource, with a Security layer on the right. + +**Figure 1 – Reference framework of DLT-MSP [ITU-T F.751.16]** + +# **7 Assessment method for DLT layer** + +## **7.1 Cryptographic algorithm** + +Table 1 describes the assessment method for the cryptographic algorithm. + +**Table 1 – Cryptographic algorithm** + +| | | +|-------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Test case 01 | Cryptographic algorithm | +| Reference | Clause 7.1 of [ITU-T F.751.16] | +| Test purpose | To validate the cryptographic algorithms supported by the DLT managed within the DLT-MSP. | +| Test workflow | Step 1. Display the default supported cryptographic algorithms, including hash algorithm, symmetric encryption algorithm and asymmetric algorithm;
Step 2. Check the cryptographic algorithm, including the key codes and configuration files. | +| Expected results | 1. The key codes and configuration files were valid. | + +## **7.2 Consensus mechanism** + +Table 2 describes the assessment method for the consensus mechanism. + +**Table 2 – Consensus mechanism** + +| | | +|-------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Test case 02 | Consensus mechanism | +| Reference | Clause 7.2 of [ITU-T F.751.16] | +| Test purpose | To validate the consensus mechanisms supported by the DLT managed within the DLT-MSP. | +| Test workflow |

Step 1. Display the consensus mechanisms supported by the DLT managed within the DLT-MSP, such as crash fault tolerance (CFT) consensus or byzantine fault tolerance (BFT) consensus;

Step 2. Check the relevant codes and configuration files of the consensus mechanisms;

Step 3. Construct and submit a transaction to verify the consensus mechanisms supported in the system.

| +| Expected results |
  1. 1. The relevant codes and configuration files of the consensus mechanisms supported by the system were displayed successfully;
  2. 2. The transaction was sent successfully and the supported consensus mechanisms were verified correctly.
| + +## 7.3 P2P network + +Table 3 describes the assessment method for the peer-to-peer (P2P) network. + +**Table 3 – P2P network** + +| | | +|-------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Test case 03 | Peer-to-peer (P2P) network | +| Reference | Clause 7.3 of [ITU-T F.751.16] | +| Test purpose | To validate the P2P network protocols supported by the DLT managed within the DLT-MSP. | +| Test workflow |

Step 1. Display the P2P network protocols in the DLT-MSP;

Step 2. Check the P2P network protocols, including the key codes and configuration files;

Step 3. Verify the synchronization status of newly launched nodes through DLT-MSP.

| +| Expected results |
  1. 1. The key codes and configuration files were valid;
  2. 2. The new node was synchronized.
| + +## 7.4 Smart contract + +Table 4 describes the assessment method for the smart contract. + +**Table 4 – Smart contract** + +| | | +|-------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Test case 04 | Smart contract | +| Reference | Clause 7.4 of [ITU-T F.751.16] | +| Test purpose | To validate the functionalities of smart contracts supported by the DLT managed within the DLT-MSP. | +| Test workflow |

Step 1. Display the smart contract languages and engines;

Step 2. Compile and deploy a smart contract through DLT-MSP;

Step 3. Invoke the smart contract through the DLT-MSP.

| +| Expected results |
  1. 1. The smart contract languages and engines were valid;
| + +**Table 4 – Smart contract** + +| | | +|--|-------------------------------------------------------------------------------------------------------------------------------------| +| | 2. The smart contract was compiled and deployed successfully;
3. The smart contract was invoked successfully by the transaction. | +|--|-------------------------------------------------------------------------------------------------------------------------------------| + +## 7.5 Data storage + +Table 5 describes the assessment method for data storage. + +**Table 5 – Data storage** + +| | | +|-------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Test case 05 | Data storage | +| Reference | Clause 7.5 [ITU-T F.751.16] | +| Test purpose | To validate data storage ability supported by the DLT managed within the DLT-MSP. | +| Test workflow | Step 1. Construct and submit a transfer transaction through DLT-MSP to validate the persistent storage capability;
Step 2. Construct and submit a query transaction through DLT-MSP to validate the data query capability. | +| Expected results | 1. The transfer transaction was executed successfully;
2. The query transaction was executed successfully. | + +# 8 Assessment method for management layer + +## 8.1 User management + +Table 6 describes the assessment method for user management. + +**Table 6 – User management** + +| | | +|-------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Test case 06 | User management | +| Reference | Clause 8.1 of [ITU-T F.751.16] | +| Test purpose | To validate DLT-MSP user management ability. | +| Test workflow | Step 1. Display the user management method supported by DLT-MSP, including user creation, information management, permission management and user notification;
Step 2. Verify the user login and logout capability of DLT-MSP;
Step 3. Verify the user information modification capability of DLT-MSP;
Step 4. Verify the user permission management capability of DLT-MSP;
Step 5. Verify the user notification capability of DLT-MSP. | +| Expected results | 1. The user creation method, information management method, permission management method and user notification method were valid;
2. The user account was successfully created and the user was successfully logged in and logged out;
3. The user information was successfully modified and the user modification information was saved;
4. The user permission management strategies and the user permission configuration schemes were valid and the user permission management was executed successfully;
5. The user triggered the notification condition successfully and received the notification. | + +## 8.2 Node management + +Table 7 describes the assessment method for node management. + +**Table 7 – Node management** + +| | | +|-------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Test case 07 | Node management | +| Reference | Clause 8.2 of [ITU-T F.751.16] | +| Test purpose | To validate DLT-MSP node management ability. | +| Test workflow | Step 1. Display the node management method supported by DLT-MSP, including node deployment, resource configuration, information query and node monitor;
Step 2. Verify the node deployment capability of DLT-MSP;
Step 3. Verify the resource configuration capability of DLT-MSP;
Step 4. Verify the information query capability of DLT-MSP;
Step 5. Verify the node monitor capability of DLT-MSP. | +| Expected results | 1. The node deployment, resource configuration, information query and node monitor method were valid;
2. The node deployment process was executed successfully and the node synchronized the latest block successfully;
3. The node resource configuration process was executed successfully;
4. The information query was executed successfully;
5. The faulty node was monitored in real time. | + +## 8.3 DLT management + +### 8.3.1 Cryptographic management + +Table 8 describes the assessment method for cryptographic management. + +**Table 8 – Cryptographic management** + +| | | +|-------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Test case 08 | Cryptographic management | +| Reference | Clause 8.3 of [ITU-T F.751.16] | +| Test purpose | To validate DLT-MSP cryptographic management ability. | +| Test workflow | Step 1. Display the cryptographic algorithms supported by DLT, such as hash algorithms, symmetric encryption algorithms and asymmetric encryption algorithms;
Step 2. Check the relevant codes and configuration files of the cryptographic algorithms;
Step 3. Check the cryptographic algorithm management capabilities supported by DLT-MSP;
Step 4. Construct and submit a transaction by the default algorithm, then check the transaction execution result. | +| Expected results | 1. The cryptographic algorithms were valid;
2. The transaction was constructed and submitted successfully, and the transaction was executed successfully. | + +### 8.3.2 Consensus management + +Table 9 describes the assessment method for consensus management. + +**Table 9 – Consensus management** + +| | | +|-------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Test case 09 | Consensus management | +| Reference | Clause 8.3 of [ITU-T F.751.16] | +| Test purpose | To validate DLT-MSP consensus management ability. | +| Test workflow | Step 1. Display the consensus mechanism supported by the DLT-MSP;
Step 2. Check the relevant codes and configuration files of the consensus mechanisms;
Step 3. Check the management capability of the consensus mechanism supported by DLT-MSP;
Step 4. Deploy a consensus algorithm through DLT-MSP. | +| Expected results | 1. The consensus mechanism was valid;
2. The relevant codes and configuration files of the consensus mechanisms supported by the system were displayed successfully;
3. The consensus was deployed successfully. | + +### **8.3.3 Network management** + +Table 10 describes the assessment method for network management. + +**Table 10 – Network management** + +| | | +|-------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Test case 10 | Network management | +| Reference | Clause 8.3 of [ITU-T F.751.16] | +| Test purpose | To validate DLT-MSP network management ability. | +| Test workflow | Step 1. Display the P2P network protocols supported by DLT-MSP;
Step 2. Check the relevant codes and configuration files of the network protocol;
Step 3. Check the network protocol management capabilities supported by DLT-MSP;
Step 4. Verify the node joining ability of DLT-MSP. | +| Expected results | 1. The P2P network protocols were valid;
2. The relevant codes and configuration files of the network protocol supported by the system were displayed successfully;
3. The node was joined to the DLT network successfully. | + +### **8.3.4 Storage management** + +Table 11 describes the assessment method for storage management. + +**Table 11 – Storage management** + +| | | +|----------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Test case 11 | Storage management | +| Reference | Clause 8.3 of [ITU-T F.751.16] | +| Test purpose | To validate DLT-MSP storage management ability. | +| Test workflow | Step 1. Display the storage mechanism supported by DLT-MSP;
Step 2. Check the relevant codes and configuration files of the storage mechanism;
Step 3. Check the storage mechanism management capabilities supported by DLT-MSP;
Step 4. Deploy the databases through DLT-MSP. | + +**Table 11 – Storage management** + +| | | +|-------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Expected results |
  1. 1. The storage management was valid;
  2. 2. The relevant codes and configuration files of the storage mechanisms supported by the system were displayed successfully;
  3. 3. The database was deployed successfully.
| +|-------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| + +## 8.4 Smart contract management + +Table 12 describes the assessment method for smart contract management. + +**Table 12 – Smart contract management** + +| | | +|-------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Test case 12 | Smart contract management | +| Reference | Clause 8.4 of [ITU-T F.751.16] | +| Test purpose | To validate DLT-MSP smart contract management ability. | +| Test workflow |

Step 1. Display the smart contract management method supported by DLT-MSP, including contract deployment, contract upgrade, contract freeze and contract unfreeze;

Step 2. Verify the contract deployment capability of DLT-MSP;

Step 3. Verify the contract upgrade capability of DLT-MSP;

Step 4. Verify the contract freeze capability of DLT-MSP;

Step 5. Verify the contract unfreeze capability of DLT-MSP.

| +| Expected results |
  1. 1. The smart contract deployment, contract upgrade, contract freeze and contract unfreeze method were valid;
  2. 2. The contract was compiled and deployed successfully and the contract was invoked successfully by the transaction;
  3. 3. The contract was upgraded successfully and the smart contract invoked successfully;
  4. 4. The contract was frozen successfully and the frozen contracts cannot be invoked normally;
  5. 5. The contract was unfrozen successfully and the unfrozen contracts can be invoked normally.
| + +## 8.5 Log management + +Table 13 describes the assessment method for log management. + +**Table 13 – Log management** + +| | | +|-------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Test case 13 | Log management | +| Reference | Clause 8.5 of [ITU-T F.751.16] | +| Test purpose | To validate DLT-MSP log management ability. | +| Test workflow |

Step 1. Display the log type supported by DLT-MSP, including node log, user log, system log and other log;

Step 2. Verify the log management capability of DLT-MSP.

| +| Expected results |
  1. 1. The log query was executed successfully through DLT-MSP, with logs including node log, user log, system log and other log;
  2. 2. Log level management and hierarchical authorization.
| + +# 9 Assessment method for business layer + +## 9.1 User key management + +Table 14 describes the assessment method for user key management. + +**Table 14 – User key management** + +| | | +|-------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Test case 14 | User key management | +| Reference | Clause 9.1 of [ITU-T F.751.16] | +| Test purpose | To validate the ability of user key management by the DLT-MSP. | +| Test workflow | Step 1. Generate a key pair through DLT-MSP;
Step 2. Verify the key pair storage capability of DLT-MSP;
Step 3. Verify the key pair backup and recovery capability of DLT-MSP;
Step 4. Verify the key authorization access capability of DLT-MSP;
Step 5. Verify the functionality of key pair usage through DLT-MSP. | +| Expected results | 1. The key pair was generated successfully;
2. The private key was stored by the user's wallet or through other storage methods;
3. DLT-MSP can provide mechanisms for users to recover and backup their keys in case of loss or damage;
4. DLT-MSP can ensure that only authorized users can use specific keys for transactions and operations;
5. The user can sign and send transactions through DLT-MSP using private keys, and the transactions are successfully executed. | + +## 9.2 Interface management + +Table 15 describes the assessment method for interface management. + +**Table 15 – Interface management** + +| | | +|-------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Test case 15 | Interface management | +| Reference | Clause 9.2 of [ITU-T F.751.16] | +| Test purpose | To validate the interface management capabilities of DLT-MSP. | +| Test workflow | Step 1. Display the types of user interfaces provided by the DLT-MSP;
Step 2. Invoke the interface, submit a transaction and verify the transaction execution result. | +| Expected results | 1. The interface management was valid;
2. The interface was invoked successfully. | + +## 9.3 DLT browser + +Table 16 describes the assessment method for DLT browser. + +**Table 16 – DLT browser** + +| | | +|----------------------|----------------------------------------------------------------------| +| Test case 16 | DLT browser | +| Reference | Clause 9.3 of [ITU-T F.751.16] | +| Test purpose | To validate the basic information query function by the DLT browser. | +| Test workflow | Step 1. Check the current block height through a DLT browser; | + +**Table 16 – DLT browser** + +| | | +|-------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| | Step 2. Check the original data of a specific block through a DLT browser;
Step 3. Check the block information by block hash or block height through a DLT browsers. | +| Expected results | 1. The basic information was obtained successfully, including block height and block data. | + +## **9.4 Contract template** + +Table 17 describes the assessment method for contract template. + +**Table 17 – Contract template** + +| | | +|-------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------| +| Test case 17 | Contract template | +| Reference | Clause 9.4 of [ITU-T F.751.16] | +| Test purpose | To validate the contract template management capabilities of DLT-MSP. | +| Test workflow | Step 1. Display the contract template types supported by DLT-MSP;
Step 2. Deploy a contract utilizing a contract template through DLT-MSP. | +| Expected results | 1. The contract template was valid;
2. The contract was deployed successfully. | + +# Bibliography + +- [b-ITU-T F.751.0] Recommendation ITU-T F.751.0 (2020), *Requirements for distributed ledger system.* +- [b-ITU-T F.751.1] Recommendation ITU-T F.751.1 (2020), *Assessment criteria for distributed ledger technology (DLT) platforms.* +- [b-ITU-T F.751.2] Recommendation ITU-T F.751.2 (2020), *Reference framework for distributed ledger technologies.* +- [b-ITU-T F.751.7] Recommendation ITU-T F.751.7 (2022), *Functional assessment methods for distributed ledger technology platforms.* +- [b-ITU-T X.1400] Recommendation ITU-T X.1400 (2020), *Terms and definitions for distributed ledger technology.* + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.751.16-202406-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/F/T-REC-F.751.16-202406-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1761a7470e8a9e428785eb00e74296e25ebd9a42 --- /dev/null +++ b/marked/F/T-REC-F.751.16-202406-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d5e42f1d16b75bfee442ee7219f85c74294e811f7de5183a80a7c9d1342b0ccd +size 7289 diff --git a/marked/F/T-REC-F.751.16-202406-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.751.16-202406-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..2d8b944a3d7f97d0f9e6e07b9f0447e34a03a88a --- /dev/null +++ b/marked/F/T-REC-F.751.16-202406-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3652e112072168d08077b83e64c512d6476e1e702bb73ac6bc91d08cc72fa105 +size 181087 diff --git a/marked/F/T-REC-F.751.17-202406-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/F/T-REC-F.751.17-202406-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ac295b1b0c7f14b1c5a33ef6682b1b017c812477 --- /dev/null +++ b/marked/F/T-REC-F.751.17-202406-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37aed4fc114a39843438175326e0b7e1d09a9d6075723d82fea32040fd560ee4 +size 7192 diff --git a/marked/F/T-REC-F.751.17-202406-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.751.17-202406-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..cd4817d3060ea51621398fbbb17b692c3e53ef28 --- /dev/null +++ b/marked/F/T-REC-F.751.17-202406-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fd0e80a304cca90ca67b1f354432a1733776ddef7ea307ad06d082531335fdd8 +size 24776 diff --git a/marked/F/T-REC-F.751.18-202406-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.751.18-202406-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c1629394c02f697f60ae954ad36bf274fa8082cb --- /dev/null +++ b/marked/F/T-REC-F.751.18-202406-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3763acd7a0ed56aa30f3f90468c1fb0cb54573eb9045bca3eddf8976b4f0f41d +size 7219 diff --git a/marked/F/T-REC-F.751.18-202406-I_PDF-E/4ee27dbf5ef12e7b58b0ef0937bc5a5e_img.jpg b/marked/F/T-REC-F.751.18-202406-I_PDF-E/4ee27dbf5ef12e7b58b0ef0937bc5a5e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..78adf019735d88ad8a481eac5b57964198527579 --- /dev/null +++ b/marked/F/T-REC-F.751.18-202406-I_PDF-E/4ee27dbf5ef12e7b58b0ef0937bc5a5e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0ebbabf27837af1f725032763a8f11602d122b225a54e06be33b7ef41e8a44ed +size 36239 diff --git a/marked/F/T-REC-F.751.18-202406-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg b/marked/F/T-REC-F.751.18-202406-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..46835a83a05c3adde693a8de7452a4bab2fe9935 --- /dev/null +++ b/marked/F/T-REC-F.751.18-202406-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:924bc085993324a60e52a60bd78de1aa2871df9f07824fb167272ecb12b7d543 +size 96312 diff --git a/marked/F/T-REC-F.751.18-202406-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg b/marked/F/T-REC-F.751.18-202406-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1e39f25e3380e113ced2fb75abd9c0b2a8640783 --- /dev/null +++ b/marked/F/T-REC-F.751.18-202406-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8c0ca89f9748bcf4b359dd8fe34da66255651008c9b29cc5402b9151ddcd0d6d +size 98990 diff --git a/marked/F/T-REC-F.751.19-202406-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.751.19-202406-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c1629394c02f697f60ae954ad36bf274fa8082cb --- /dev/null +++ b/marked/F/T-REC-F.751.19-202406-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3763acd7a0ed56aa30f3f90468c1fb0cb54573eb9045bca3eddf8976b4f0f41d +size 7219 diff --git a/marked/F/T-REC-F.751.19-202406-I_PDF-E/5a4e62bead259c258d069fd3663ea670_img.jpg b/marked/F/T-REC-F.751.19-202406-I_PDF-E/5a4e62bead259c258d069fd3663ea670_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..fc9301269d59b7ace55869932b614e054b6a7fd1 --- /dev/null +++ b/marked/F/T-REC-F.751.19-202406-I_PDF-E/5a4e62bead259c258d069fd3663ea670_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8922612709e7ddefe319f1637c5463c393daf51a73342afa8025e3cf25635835 +size 32173 diff --git a/marked/F/T-REC-F.751.19-202406-I_PDF-E/7a0db9703b68b3d06cdaeefc084c0006_img.jpg b/marked/F/T-REC-F.751.19-202406-I_PDF-E/7a0db9703b68b3d06cdaeefc084c0006_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ad25b7bd441bba08ec144ed2d14aed499c75cf2a --- /dev/null +++ b/marked/F/T-REC-F.751.19-202406-I_PDF-E/7a0db9703b68b3d06cdaeefc084c0006_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:25a59bc065094edd763995f716a6c541d8fcb420e6d238b3b321dd1c636a9754 +size 117686 diff --git a/marked/F/T-REC-F.751.19-202406-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg b/marked/F/T-REC-F.751.19-202406-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9c6493c0a2e61ae1d48f2643955c5f1dfd2ca0b4 --- /dev/null +++ b/marked/F/T-REC-F.751.19-202406-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:082c804f0daf33f7fba8aa30c14b356224880ae9d120d95353fc1202278f679e +size 15615 diff --git a/marked/F/T-REC-F.751.2-202008-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.751.2-202008-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9002ee68e2b9ebf63d27d39ba8c73bdafb0e6be6 --- /dev/null +++ b/marked/F/T-REC-F.751.2-202008-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8df701536c25ab8246d2868de3c20c8a4e25168bb52c5eb9078278ab1b5b4fd8 +size 6859 diff --git a/marked/F/T-REC-F.751.2-202008-I_PDF-E/3ad00ce93ad9dea9ee0f47535e5355e6_img.jpg b/marked/F/T-REC-F.751.2-202008-I_PDF-E/3ad00ce93ad9dea9ee0f47535e5355e6_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..75033e34e142dc49d554c79ac5864d2fdb420007 --- /dev/null +++ b/marked/F/T-REC-F.751.2-202008-I_PDF-E/3ad00ce93ad9dea9ee0f47535e5355e6_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aedef5e555ff1101ef598f4f03b3b9593df4629f02a2554c2a6c83732eb8f28f +size 33447 diff --git a/marked/F/T-REC-F.751.2-202008-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg b/marked/F/T-REC-F.751.2-202008-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..bf0768442aab85c0c0696ffcd547dcd5b1806a91 --- /dev/null +++ b/marked/F/T-REC-F.751.2-202008-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c633fb41ab72f277dc9e47ee9503fb5f2738ed9c9443edc2044404b7fd3b94b3 +size 97360 diff --git a/marked/F/T-REC-F.751.2-202008-I_PDF-E/af7916c89a458fdab6c3f443217388ae_img.jpg b/marked/F/T-REC-F.751.2-202008-I_PDF-E/af7916c89a458fdab6c3f443217388ae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a11c7578b53aaf7cd6d9fb6d117d981479e050db --- /dev/null +++ b/marked/F/T-REC-F.751.2-202008-I_PDF-E/af7916c89a458fdab6c3f443217388ae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:99a40ccd17b84c36d7af93fe315ee31e7f4f1e1b74e64d6c168a08f20ffc3719 +size 159434 diff --git a/marked/F/T-REC-F.751.2-202008-I_PDF-E/raw.md b/marked/F/T-REC-F.751.2-202008-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..305652ce608e9498a46fc6b79d20e64e516d766d --- /dev/null +++ b/marked/F/T-REC-F.751.2-202008-I_PDF-E/raw.md @@ -0,0 +1,854 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.751.2** + +(08/2020) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Reference framework for distributed ledger +technologies** + +Recommendation ITU-T F.751.2 + +ITU-T + +![ITU logo: A globe with a red lightning bolt, next to the text 'International Telecommunication Union'.](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with a red lightning bolt striking across it. To the right of the globe, the text "International Telecommunication Union" is written in a blue, sans-serif font. + +ITU logo: A globe with a red lightning bolt, next to the text 'International Telecommunication Union'. + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.751.2 + +## Reference framework for distributed ledger technologies + +## Summary + +Recommendation ITU-T F.751.2 specifies the reference architecture for distributed ledger technology (DLT), its hierarchical relationship and specific functions, important modules and specific functions in its structure, as well as the main technical route and direction of its core module. Recommendation ITU-T F.751.2 can be used as a guideline for DLT service providers to build systems and for organizations to select and use a DLT platform. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.751.2 | 2020-08-13 | 16 | 11.1002/1000/14334 | + +## Keywords + +Architecture, blockchain, components, DLT, distributed ledger technology, functions, ledger, platforms, reference architecture. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2020 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-----------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 3 | +| 6 Architecture overview ..... | 3 | +| 6.1 Overview ..... | 3 | +| 6.2 Resource and infrastructure functions ..... | 4 | +| 6.3 Protocol or governance and compliance functions..... | 4 | +| 6.4 Application functions ..... | 5 | +| 6.5 Operation and maintenance functions ..... | 5 | +| 6.6 External interaction management functions ..... | 6 | +| 6.7 Extension functions ..... | 6 | +| 7 Functional components ..... | 6 | +| 7.1 Core layer ..... | 7 | +| 7.2 Service layer ..... | 13 | +| 7.3 Application service platform ..... | 15 | +| 7.4 DLT applications ..... | 17 | +| 7.5 DLT extendable interfaces and external services ..... | 17 | +| Bibliography..... | 18 | + + + +## Recommendation ITU-T F.751.2 + +## Reference framework for distributed ledger technologies + +# 1 Scope + +This Recommendation specifies the reference architecture for distributed ledger technology (DLT), its hierarchical relationship and specific functions, important modules and specific functions in its structure, as well as the main technical route and direction of its core module. + +This Recommendation can be used as a guideline for DLT service providers to build systems and for organizations to select and use a DLT platform. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 asset** [b-ITU-T F.751.0]: A representation of value. It can be a diamond, a unit of currency, items inside a shipping container, etc. An asset can be physical or virtual. + +**3.1.2 blockchain** [b-ITU-T F.751.0]: A type of distributed ledger that is composed of digitally recorded data arranged as a successively growing chain of blocks with each block cryptographically linked and hardened against tampering and revision. + +**3.1.3 consensus** [b-ITU-T F.751.0]: Agreement that a set of transactions is valid. + +**3.1.4 entity** [b-ITU-T F.751.0]: Anything that has a separately identifiable existence (e.g., organization, person, group, smart contract or device). An entity uses distributed ledger technology to solve the problem of its business or information systems. + +**3.1.5 entity address** [b-ITU-T F.751.0]: Identifier for one or more entities performing transactions or other actions in a blockchain or distributed ledger network. + +**3.1.6 ledger** [b-ITU-T F.751.0]: Information store that keeps final and definitive (immutable) records of transactions. + +**3.1.7 node** [b-ITU-T F.751.0]: Device or process that participates in a distributed ledger network. + +**3.1.8 permissionless distributed ledger system** [b-ITU-T F.751.0]: Distributed ledger system where permissions are not required to maintain and operate a node. + +**3.1.9 public DLT; public distributed ledger system; public distributed ledger technology system** [b-ISO 22739]: A public DLT has transaction records that are readable by anyone. + +**3.1.10 private distributed ledger technology system** [b-ISO 22739]: DLT system that is accessible for use only to a limited group of DLT users. + +**3.1.11 smart contract** [b-ITU-T F.751.0]: Program written on the distributed ledger system that encodes the rules for specific types of distributed ledger system transactions in a way that can be validated, and triggered by specific conditions. + +**3.1.12 token** [b-ITU-T F.751.0]: A digital representation of value on a shared distributed ledger that is owned and secured using cryptography to ensure its authenticity and prevent modification or tampering without the owner's consent. + +**3.1.13 transaction** [b-ITU-T F.751.0]: An incident or an operation that leads to a change in the status of a ledger, such as adding a record or equivalent exchange based on currency. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 balance model:** Model that keeps track of the balance of each account as a global state. The balance of an account is checked to make sure it is greater than or equal to the spending transaction amount. + +NOTE – Adapted from [b-ITU-T TS FG DLT D3.1]. + +**3.2.2 event model:** Model where consensus takes place upon event (transaction). + +NOTE – Originally published in [b-ITU-T TS FG DLT D3.1]. + +**3.2.3 state model:** Model where consensus takes place upon state (result). + +NOTE – Originally published in [b-ITU-T TS FG DLT D3.1]. + +**3.2.4 unspent transaction output (UTXO) model:** Model where transaction spends output from prior unspent transactions and generates new outputs that can be spent in the future. The unspent transactions are kept in each node. + +NOTE – Adapted from [b-ITU-T TS FG DLT D3.1]. + +**3.2.5 trust endorsement:** An endorsement for entities inside DLT systems to trust each other, e.g., for one node to trust other nodes. Instances can be a programmable governance model, or contracts or agreements signed by node owners. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +AAA      Authentication, Authorization and Accounting + +API        Application Programming Interface + +BFT        Byzantine Fault Tolerance + +CA          Certificate Authority + +DApp      Decentralized Application + +dBFT      Delegated Byzantine Fault Tolerance + +DID        Decentralized Identity + +DLT        Distributed Ledger Technology + +DPoS      Delegated Proof of Stake + +EVM      Ethereum Virtual Machine + +IBE        Identity-Based Encryption + +| | | +|------|------------------------------------------------------| +| JVM | Java Virtual Machine | +| P2P | Peer to Peer | +| PKI | Public Key Infrastructure | +| PoS | Proof of Stake | +| PoW | Proof of Work | +| RPC | Remote Procedure Call | +| SDK | Software Development Kit | +| TEE | Trust Execution Environment | +| TXO | Transaction Output | +| UTXO | Unspent Transaction Output | +| VBFT | Byzantine Fault Tolerance with Verifiable Randomness | +| VM | Virtual Machine | + +# 5 Conventions + +In this Recommendation: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. +- The keywords "can optionally" indicates an optional requirement which is permissible, without implying any sense of being recommended. This term is not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with this Recommendation. + +# 6 Architecture overview + +## 6.1 Overview + +The high-level architecture constrains the highly abstract hierarchy of distributed ledgers. The high-level architecture can cover almost all distributed ledgers, including public chains represented by Ethereum [b-ethe] and Bitcoin [b-bitc], private chains represented by Hyperledger Fabric and non-blockchain distributed ledgers systems. See Figure 1. + +![Figure 1 – High-level conceptual architecture of DLT. The diagram shows a layered architecture. On the left, a vertical bar labeled 'Operation and maintenance'. The central part consists of three main layers: 'Application' at the top (containing DApp(s), Runtime, Lifecycle, Devtool); 'Protocol' in the middle (containing Governance and compliance, Consensus, Ledger management, Messaging, Crypto, Account/DID, Right management); and 'Resources' at the bottom (containing Node management, Storage management, Network management, Data integrity and privacy util). To the right, 'External interaction management' and 'External resource management' are shown. On the far right, a dashed box labeled 'Extensions' contains 'Side chain', 'Off chain', 'Multi chain', 'Internal extensions', and '.....'. The text 'F.751.2(20)_F01' is in the bottom right corner.](a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg) + +Figure 1 – High-level conceptual architecture of DLT. The diagram shows a layered architecture. On the left, a vertical bar labeled 'Operation and maintenance'. The central part consists of three main layers: 'Application' at the top (containing DApp(s), Runtime, Lifecycle, Devtool); 'Protocol' in the middle (containing Governance and compliance, Consensus, Ledger management, Messaging, Crypto, Account/DID, Right management); and 'Resources' at the bottom (containing Node management, Storage management, Network management, Data integrity and privacy util). To the right, 'External interaction management' and 'External resource management' are shown. On the far right, a dashed box labeled 'Extensions' contains 'Side chain', 'Off chain', 'Multi chain', 'Internal extensions', and '.....'. The text 'F.751.2(20)\_F01' is in the bottom right corner. + +**Figure 1 – High-level conceptual architecture of DLT** + +## 6.2 Resource and infrastructure functions + +The infrastructure provides the operating environment and basic components required for the normal operation of the distributed ledger system. The base layer includes network management functions, storage management functions, utility functions and node management functions. This layer is the resource that most software systems rely on and is the foundational support of the distributed ledger system. + +- Network management functions – Each DLT system is built upon a network hypothesis, which leads to the distribution model of the system. For example, in the study of bitcoin, each node inside bitcoin network has the same privileges, thus a peer-to-peer (P2P) network model is used. +- Storage management functions – Each DLT system has a standard storage component to persist data and ensure data integrity and privacy. In particular, based on the cost of distributed storage, storage management may need to provide solutions for on-chain business to balance cost and data protection. +- Utility functions – DLT has utility functions to protect data integrity – not only raw data, but also a runtime channel for data transfer. +- Node management functions – Each node inside a DLT system is maintained by a node owner or operators. Node management is a component to manage the resource of a single node inside a DLT system. + +## 6.3 Protocol or governance and compliance functions + +In DLT systems, blockchain systems in particular, each node can have its own implementations based on the system's technical specifications. The protocol layer is a conceptual layer to serve the technical specification across nodes inside a DLT system. + +The protocol layer includes: governance and compliance; consensus; ledger management; and utility. Furthermore, the governance and compliance includes: node management; and authentication, authorization and accounting (AAA) management; (account management and right management). The governance and compliance functions are to support the management of system governance (based on trust endorsement hypothesis) and AAA functions of other components. The utility includes: messaging functions; and cryptographic functions. The utility functions are used to support data privacy and integrity protection. + +### **6.3.1 Consensus mechanism functions** + +A consensus mechanism is the core component of a distributed ledger, especially a decentralized one, and is used to ensure the consensus of all nodes on the data. The consensus mechanism contains consensus algorithms, data validation, data distribution and synchronization. By use of the consensus mechanism, the distributed ledger system sets up a trust mechanism upon the network hypothesis. Any trust endorsement module, e.g., incentive mechanisms, is built upon that. Consensus mechanisms can maintain public data inside a chain and data based on various distributed ledger-partitioning mechanisms. + +### **6.3.2 Ledger management functions** + +The ledger provides basic data management of distributed ledgers and distribution management of ledger data on the network. It determines the local data storage methods of ledgers and the synchronization mechanism between nodes, and responds to rights management by use of the consensus mechanism. + +## **6.4 Application functions** + +Based on the runtime management, decentralized applications (DApps) are built to serve different business requirements in a distributed network environment. + +Furthermore, web applications with a combination of both off- and on-chain services can be a solution for businesses. + +### **6.4.1 Smart contract mechanism functions** + +Most DApps are written by smart contracts. + +Smart contracts are programs written on the distributed ledger system that encode the rules for specific types of distributed ledger system transactions in a way that can be validated and triggered by specific conditions. + +The smart contract mechanism includes language specification, compilation and execution of the code. Smart contracts for different DLT systems can be implemented using simple interpreted scripts or programming languages. + +The smart contract mechanism is an optional, but useful, component for DLT. + +### **6.4.2 DApp management functions** + +DApp management functions provide a tool set for DApp development, manage the runtime of DApp and control the lifecycle of DApp. + +An open DApp management layer is middleware that connects the DLT network with the DApp business. + +The DApp management middleware contains a DApp framework for DApp developers to create and maintain DApps, and a smart contract management mechanism for DApp hosts to manage their DApps easily. + +Furthermore, the DApp framework can provide a series of interfaces for DApp developers. The interface enables the use of distributed ledgers. DApp users access DLT services through the interface. The interface can usually have an application programming interface (API), software development kit (SDK), remote procedure call (RPC), and so on. + +## **6.5 Operation and maintenance functions** + +Operation and maintenance functions include various libraries, such as log, monitoring, node or network management and scaling libraries. + +## 6.6 External interaction management functions + +Each DLT system has its own network hypothesis, trust endorsement hypothesis and governance model. Thus, a DLT system with an open-network hypothesis is able to interact or interoperate with external system(s). + +The network hypothesis describes the node scalability of the business, e.g., whether a network is open and node counts. + +Trust endorsement judges whether a chain is permissioned or permissionless. + +The governance model guides the management of the behaviour of individual nodes. + +These three components determine whether the DLT system is able to support external interactions. + +In most cases, external interaction management is the runtime engine of external resource management. + +## 6.7 Extension functions + +The extension component of a DLT platform aims to resolve different requirements of data interoperability. Extension functions include a series of protocols or specifications for data interoperations of external systems, e.g., multi-chain, side-chain, off-chain, or internal systems, e.g., child-chain, sharding1 [b-shard1], [b-shard2]. + +### 6.7.1 Internal system extensions + +Each DLT system has one governance model. The internal system extension aims to resolve the problem of scalability for one DLT ecosystem with the same governance model. + +#### Child chain + +Child chains are individual ledgers with their own native tokens responsible for operational transactions, such as deploying smart contracts, issuing assets, voting on polls and sending messages. All child chains receive consensus from, and share the same source code as, the network's main (parent) chain; therefore, all child chains on the network are interoperable. + +### 6.7.2 External system extensions + +In most business cases, the data interoperability is a cross-system requirement. A DLT system can access external systems to satisfy business requirements. + +#### Off-chain system(s) functions + +In the study of DLT, interaction or interoperation with off-chain systems are scenarios in most use cases. + +Furthermore, a series of layer 2 (L2) solutions, a combination of on-chain and off-chain techniques, are used to optimize the performance, as well as the scalability, of DLT systems. + +# 7 Functional components + +The different distributed ledger platforms are highly consistent on the top-level architecture, but the components in the detailed architecture are different. Clauses 7.1 to 7.5 explain in detail the components and functions of the detailed architecture, which varies slightly from the top-level architecture described in clause 6.1. See Figure 2. + +--- + +1 Sharding has many solutions, some share the same chain protocol, some have their own extensions, e.g., message queue, side-chain with the same governance model. + +![Figure 2 – Schematic diagram of the detailed architecture. This diagram shows a layered architecture for a DLT system. The top layer is 'Application service platform' containing 'DLT application(s)', 'DApp development I/F' (with IDE, Build management, Test management), 'DLT data I/F' (with Data storage, Data synchronization, Data privacy), 'DLT account I/F' (with AAA, Data exchange, Exchange protocol), and 'External service(s)' (with External Mgmt and extenitons in Fig.1, External resources, Permissioned agreement). The 'Service layer' contains 'Internal governance (Public BLC)' (with Security and privacy, Access management, Policy configuration, Monitoring and auditing, SC registration, SC template, SC compiler, VM runtime, Decentralization ID, Authentication, Authorization, Delegation, Global configuration, Governance control, Supervisory support). The 'Core layer' is divided into 'Protocol layer in Fig.1' (with Data integrity, Trusted storage, Ledger, Smart contract mechanism, Account management, Consensus mechanism, System management, Node Admin, Admin I/F) and 'Resources layer in Fig.1' (with Storage service, P2P network, Extendable protocol communication, Exchange protocol-core, Safe hardware, Network and infrastructure). External components include 'Admin App(s)' and 'Homogeneous BLCT cross chain resources'.](af7916c89a458fdab6c3f443217388ae_img.jpg) + +Figure 2 – Schematic diagram of the detailed architecture. This diagram shows a layered architecture for a DLT system. The top layer is 'Application service platform' containing 'DLT application(s)', 'DApp development I/F' (with IDE, Build management, Test management), 'DLT data I/F' (with Data storage, Data synchronization, Data privacy), 'DLT account I/F' (with AAA, Data exchange, Exchange protocol), and 'External service(s)' (with External Mgmt and extenitons in Fig.1, External resources, Permissioned agreement). The 'Service layer' contains 'Internal governance (Public BLC)' (with Security and privacy, Access management, Policy configuration, Monitoring and auditing, SC registration, SC template, SC compiler, VM runtime, Decentralization ID, Authentication, Authorization, Delegation, Global configuration, Governance control, Supervisory support). The 'Core layer' is divided into 'Protocol layer in Fig.1' (with Data integrity, Trusted storage, Ledger, Smart contract mechanism, Account management, Consensus mechanism, System management, Node Admin, Admin I/F) and 'Resources layer in Fig.1' (with Storage service, P2P network, Extendable protocol communication, Exchange protocol-core, Safe hardware, Network and infrastructure). External components include 'Admin App(s)' and 'Homogeneous BLCT cross chain resources'. + +F.751.2(20)\_F02 + +NOTE – Dashed borders indicate optional components. + +**Figure 2 – Schematic diagram of the detailed architecture** + +## 7.1 Core layer + +Functions map to resources, protocol and operation and maintenance in Figure 1. + +The DLT system aims to execute a transaction; however, it receives events that will not always result in transactions. Figure 3 presents two options of how DLT systems process events of incoming transactions and store the result in the ledger. + +![Figure 3 – Two options for typical flow of DLT systems. This diagram shows two flow options for processing events. Option 1: Event -> Consensus -> VM (dashed) -> Execute event -> state -> Cache (dashed) -> Store Event/state/block. Option 2: Event -> VM pre-execute env. (dashed) -> Execute event -> state -> Consensus -> Cache (dashed) -> Store State/block.](3ad00ce93ad9dea9ee0f47535e5355e6_img.jpg) + +Figure 3 – Two options for typical flow of DLT systems. This diagram shows two flow options for processing events. Option 1: Event -> Consensus -> VM (dashed) -> Execute event -> state -> Cache (dashed) -> Store Event/state/block. Option 2: Event -> VM pre-execute env. (dashed) -> Execute event -> state -> Consensus -> Cache (dashed) -> Store State/block. + +F.751.2(20)\_F03 + +NOTE – Dashed borders indicate optional components. + +**Figure 3 – Two options for typical flow of DLT systems** + +The standard process involves: + +- event – to gather event(s) from client(s); +- virtual machine (VM) – to prepare the environment for event execution; +- execute – to execute the transaction(s) inside the event producing the state result – the state result stands for the status of the ledger; +- store – to store data (state or event). + +In decentralized systems, a consensus mechanism is required to solve data consistency between different nodes. + +There are two modes for approaching this in DLT systems: + +- state mode: consensus upon states, mostly used for pre-execution, UTXO models. +- event mode: consensus upon events, mostly used for post execution, balance models. + +### **7.1.1 Network and infrastructure** + +As an IT solution, DLT nodes use the same network solutions of typical distributed system solutions (including cloud solutions). + +##### **7.1.1.1 Safe hardware** + +One option to enable DLT systems to support data privacy enhancement with high performance is to use trusted or /safe hardware, e.g., a trust execution environment (TEE). + +### **7.1.2 Extendable protocol communication** + +Scalable protocol communication module based on network hypothesis across distributed systems. + +The communication component aims to give the capability of data exchange across different chain networks with L7-filtering2 for a homogeneous DLT network, if possible. + +### **7.1.3 Network (P2P network) management** + +A DLT system is based on a network hypothesis, thus network management is required. Especially in permissionless DLT systems, each node has the same privileges, the node owner is able to decide the contribution of the node itself, network management is to provide basic capability to control nodes in the network. + +#### **7.1.3.1 Network discovery** + +In a distributed ledger system, there are usually many nodes, especially in public permissionless DLT systems. Each node needs to discover neighbour nodes through a network discovery protocol and establish a link with neighbour nodes. According to different permissioned distributed ledger architectures, the network discovery protocol also needs to identify and authenticate the node identity to prevent attacks such as a Sybil attack. + +#### **7.1.3.2 Data transceiver** + +After a node connects to a neighbour node through the network discovery protocol, data exchange (e.g., transaction broadcast, consensus message and data synchronization) with other nodes is completed by the data transceiver module. According to the architecture of different distributed ledgers, the design of the data transceiver needs to consider requirements, e.g., confirmation and encryption. + +#### **7.1.4 Storage service** + +#### **7.1.4.1 Data persistence** + +Distributed ledgers need to persistently store a variety of data, such as block data, transaction data, status data and private data of a local account. Depending on the type of data and the design of the distributed ledger, different storage modes can be used. Storage modes include relational databases such as MySQL [b-mysql], non-relational databases such as LevelDB [b-leveldb] and self-organizing files. + +--- + +2 ISO application layer data filter. + +#### 7.1.5 Consensus mechanism + +#### 7.1.5.1 Data synchronization + +The data synchronization module ensures that the distributed ledger has a consistent ledger. The data synchronization module transmits the new part of the ledger between different nodes. The synchronization module also validates the synchronized data to ensure the correctness and consistency of the synchronized data. + +Different types of nodes can have different synchronization methods, which include synchronizing all transactions, blocks and status data, synchronizing all transactions and block data and synchronizing partial transaction data. + +#### 7.1.5.2 Legitimate validation + +Consensus is the core of distributed ledgers. The purpose of consensus is to make many nodes involved in accounting jointly maintain the same ledger. It should be noted the validation refers only to algorithms that agree on the transaction order, and not verification of the transaction data itself. + +Consensus algorithm design should ensure the following: + +- consistency: eventual consensus node agreement on the data; +- timeliness: completion by consensus nodes of data consensus in as short a time as possible; +- security: a substantial cost barrier to the undermining of consistency and to provide protection against attack. + +In theory, all kinds of algorithms can meet the requirements in the previous paragraph and be applied in a given setting as the consensus algorithm of a distributed ledger. + +#### 7.1.5.3 Verification + +The module contains a transaction buffer pool for distributed ledgers and basic verification of transactions. Distributed ledgers typically verify the sender's balance of the transaction, the sender's transaction number, etc. Once the transaction is validated, the transaction is saved in the DLT system. The verification module can effectively prevent various attacks and ensure the stable operation of the distributed ledger. + +#### 7.1.5.4 Consensus algorithm + +**Proof of work (PoW)** [b-pow], commonly known as mining, can be simply understood as providing proof that a certain amount of work has been done. PoW requires the node to carry out a certain amount of computation to obtain the accounting right, which means that it takes a certain amount of computation resource, which results in energy consumption by the computer and calculates a verifiable result through a mathematical operation. The node sends the data that needs to be recorded in this round. After verification, the other nodes in the whole network store the data. + +The main feature of the PoW system is that the prover needs to do more work to get the result, while the verifier can easily verify whether the prover has done the corresponding work through the result. A core feature of this approach is asymmetry: work is more difficult for a prover and easier for approver (such as hash algorithm). + +**Series of Byzantine fault tolerance algorithm.** Byzantine fault tolerance (BFT) is a common solution to achieve efficient fault tolerance. The system comes from the Byzantine general problem. A system based on the BFT algorithm can reach valid consensus when the number of failed or fraudulent nodes is less than the number of fault-tolerant nodes. Generally speaking, the number of fault-tolerant nodes is less than one-third of the total number of nodes. + +At present, there are many implementations of BFT series algorithms. The typical implementation method is the state machine copy replica algorithm. Under the control of the master node, all nodes confirm the status of other participants through the three-phase protocol and determine the accounting + +according to the status of all participants' content. BFT series algorithms currently include the synchronized PBFT algorithm [b-pbft], asynchronous Honey Badger BFT [b-hbft] algorithm and open channel BFT algorithm. There are also some BFT algorithms that support dynamic changes in the number of nodes. + +**Proof of stake (PoS).** PoS allows so-called token holders to replace the miners. The accountant is the holder of relevant tokens and the accounting right is their stake. A typical PoS reduces the difficulty of mining in equal proportion to the percentage and time tokens that each node occupies in order to find a verifiable result and determine the ownership of accounting rights faster. + +**Delegated proof of stake (DPoS)** [b-dpos]. Each token holder determines the accounting right of a DLT system by voting, similar to the election of a board of directors. All nodes whose votes exceed the agreed votes become system trustees, forming a so-called board of directors and alternately signing blocks. There are incentives for encouraging directors never to miss signing a block, but if it occurs, other nodes on the network may vote for a new director to take their place. + +**Traditional consensus algorithms.** Traditional consensus algorithms are based on traditional distributed consensus techniques, including Paxos [b-Lamport], Raft [b-raft], and others. There are many similarities between the traditional and other consensus algorithms, all of which aim to solve data inconsistency caused by failure of network or hardware and data loss in the system. The traditional consistency algorithm pays more attention to performance and has higher requirements on the environment. At present, the traditional consistency algorithm is also widely used in DLT systems. + +**Hybrid consensus algorithm:** The current consensus algorithms have their strengths and weaknesses, so there are some cases where consensus algorithms are mixed, including: + +BFT-RAFT [b-Clow], BFT combined with the high performance of Raft and support for fraud nodes; delegated Byzantine fault tolerance (dBFT) [b-dbft], authorized BFT algorithm, integrated BFT with election and authorization mechanism for more node scenes; + +Byzantine fault tolerance with verifiable randomness (VBFT) [b-vbft], achieves chain scalability by consensus node selection with virtual routing and forwarding, anti-attack ability by randomness and PoS, and fast state finality with BFT. + +#### 7.1.6 Smart contract mechanism + +Smart contracts are programs written on the distributed ledger system, which encode the rules for specific types of distributed ledger system transactions in a way that can be validated and triggered by specific conditions. + +The smart contract mechanism handles trusted data processing, including smart contract lifecycle management, contract registration, pre-authorization, deployment, upgrading, iteration and cancellation. + +The smart contract mechanism usually includes the contract execution engine, contract code management and contract data management. + +Contract code management is responsible for the operation of the availability, deployment and storage of the contract code. + +Contract data management maintains contract data and provides an access interface for the contract engine. + +The contract execution engine can execute the contract code and maintain the contract data according to the contract code. + +#### 7.1.6.1 Language and compiler + +For very limited business requirements, you can use a script to implement a contract, which is fast, secure and low in resource consumption. + +For lightweight businesses, an inline contract engine is recommended that is highly efficient and reduces the complexity of deployment. + +For heavyweight businesses, an external execution environment is recommended that has more resources and higher execution efficiency. These are generally compatible with mainstream programming languages and are easy to develop. + +Language and compiler provide the grammar specification of the contract, as well as the compilation specification to ensure that the transaction can be processed by the execution engine. + +##### 7.1.6.2 Execution engine + +**External contract execution environment:** It uses the external execution environment to execute the contract code, and usually uses an external container (such as Docker [b-dock]) or virtual machine to ensure a consistent execution environment for all nodes. The maintenance of contract data is accomplished by the distributed ledger process communicating with a process in the external environment, a typical representative includes the realization of the contract in the Docker container of Hyperledger Fabric. + +The advantages of this contractual technology are that they can be Turing complete, utilize easy-to-master programming languages and offer efficient contract development. The heavyweight business is more efficient due to the permanent contractual execution environment. The disadvantages are slow deployment of contract code, weak external engine attack resistance, large resource usage, large inter-process communication overhead and unsuitability for short contract codes. + +**Inline contract engine:** A contract engine is embedded in the process, and the contract codes are interpreted during execution or implemented after being compiled. The engine can use open third-party engines, such as the Java virtual machine (JVM) [b-jvm], LUA [b-lua], JS [b-js] and other scripts, or be implemented by the DLT platform, such as the Ethereum virtual machine (EVM) [b-vm]. At present, most distributed ledgers use this technology, e.g., Ethereum. + +The advantages of inline contract engines include that they can be Turing complete, offer fast contract deployment, speed up data access, harden security of the engine, and provide convenient management and customization of the computing resources of the contract engine. The disadvantages are that the contract is generally not resident in memory, and the engine efficiency is difficult to compare with the external engine, making it less efficient to perform heavyweight business. + +**Scripting or finite state machine:** A set of execution instructions is pre-set in the distributed ledger, this sequence forms a piece of contract code, which is interpreted by the state machine. Due to the complexity of the instructions and the size of the code, the script generally has a weak function and is not Turing complete, meaning it only performs basic operations. Many current virtual currency ledgers, including Bitcoin, use this technology. + +Advantages include that scripting technology is easy to implement, its script functions are effectively controlled and there is strong anti-attack mitigation. Disadvantages include complex script development, Turing incompleteness, and the scale of scripts is small, resulting in limited functions. + +#### 7.1.7 Ledger + +The ledger stores all the transaction data and contract data. The ledger needs to be able to complete the transaction and the processing,3 indexing, and storage of contracts. + +Ledger management contains two components: + +- ledger mechanism – based on consensus mechanism, relying on P2P networking and extended storage services; +- ledger storage – extended storage services for ledger, e.g., databases. + +--- + +3 For those transactions that do not involve smart contract. + +Consensus algorithm design should ensure the following: + +- there are complete definitions of the various data types in the ledger, such as transactions, blockchain, assets and accounts; +- the transaction data can be summed up using algorithms, such as the Merkle tree, the blocks can be continuously added, and successive blocks are chain linked to ensure the integrity of the consensus; +- the design of the ledger can be used for data synchronization and verification, and the ledgers should support large-scale data storage and high concurrency. + +#### 7.1.7.1 Transaction record + +**Account:** The account employs asymmetric cryptography to create a public and private key pair that controls access to the account, which can authenticate account ownership. + +**Assets:** Assets can be expressed in the form of account and balance models, or in the form of a transaction output (TXO; e.g., UTXO [b-utxo]) model, or a combination of the two. Assets can also be specified in the contract data independent of the underlying assets on the ledger. + +**Ledger storage:** The ledger storage is usually implemented as a custom file or as an embedded database. The ledger store is divided into two parts: one part is the transaction history, usually the block data, which contains the original transaction; the other part is the result of the transaction or contract execution, usually stored outside the block. + +**Block structure:** A block consists of the transactions, a block header, a variety of relevant data, and its own block digest. The digest of the block is usually calculated from the transactions in the block, the digest of the previous block, and various other related data according to an algorithm such as the Merkle tree. It is a type of data structure to keep transactions in a DLT. + +#### 7.1.7.2 Status + +Most distributed ledgers, in addition to saving blocks and transaction data, also hold some data or results of the transaction execution, which can be called status. The status of different nodes of the distributed ledger should be consistent. For a simple distributed ledger like Bitcoin, the status may contain only a list of unspent assets, and so on. For distributed ledgers like Ethereum and Hyperledger Fabric with Turing-complete intelligent contract engines, the status preserves more complex data, generated during virtual machine execution. + +By maintaining status data, distributed ledgers can continue to process a series of complex and continuous transactions. + +#### 7.1.8 Data integrity and trusted storage + +DLT platforms should provide data integrity capabilities for data reading from and writing to ledgers. They should use trusted storage to ensure that data is tamperproof. + +### 7.1.9 Account management + +Account management manages entities, including users and nodes, and data identifiers in a DLT system, which can be expanded to different modules for different services. + +### 7.1.10 System management + +System management manages nodes and the DLT network, which can be expanded to different modules for different services. + +Node management of DLT system management provides an operation and maintenance feature, including node deployment and node administration. + +The deployment of distributed ledger refers to the installation and use of distributed ledger services for different scenarios and users, with different node permissions and service modes. According to + +the type, distributed ledgers are divided into permissionless and permissioned DLTs, where their deployment methods are not the same. + +#### **Permissionless DLT** + +Permissionless DLTs generally do not make any restrictions on node access and are less demanding on the operating environment. The ledger nodes are relatively simple. All nodes are free to participate in consensus and read and write data. + +#### **Permissioned DLT** + +In such distributed ledgers, consensus processes can only be involved with authorized customer nodes. Authorized nodes can participate in the consensus and data read and write process according to the rules. + +The deployment of DLTs can occur in the following ways. + +- Process-based deployment – Operations staff personnel deploy nodes on the host or virtual machine to complete the configuration. This deployment is flexible. However, due to the complexity of the configuration file, the deployment is inefficient and prone to problems caused by inconsistent node environments. +- Container-based deployment – This involves ensuring a uniform environment within the container and encapsulating the distributed ledger in the container, then deploying the container by the operations staff. This approach is simpler to deploy than process-based deployment and is more reliable. +- Cloud service-based deployment – Integration of distributed ledger and cloud services that facilitates rapid deployment of distributed ledger services through the cloud platform, while providing different levels of services, such as platform as a service, backend as a service. +- Hybrid-based deployment –An approach that can combine cloud service-based deployment with any of the two other ways mentioned here to address cloud perspectives and possible on-premises compliance. + +### **7.1.11 Utility** + +Core layer components share the same technical utility modules. Utility functions map to utility in Figure 1. + +##### **7.1.11.1 Cryptographic library** + +Distributed ledgers use a variety of cryptographic algorithms. The cryptographic algorithm library provides basic cryptographic algorithm support for each component, including various commonly used encoding algorithms, such as hash algorithms, signature algorithms and privacy protection algorithms. The cryptographic algorithm library also provides functions such as maintenance and storage of secret keys. + +##### **7.1.11.2 Messaging** + +The message module provides message notification services between different components within the distributed ledger and between different nodes. For example, after a successful transaction, the customer usually needs to track the results of the execution of the transaction and even some records during the execution of the transaction. The message module can complete the generation, distribution, storage and other functions of the message to meet the needs of the distributed ledger. + +## **7.2 Service layer** + +Functions map to protocol and governance in Figure 1. + +The service layer is middleware between the core and application service platform layers. Their services are an extension from four basic services in the core layer; data integrity; smart contract mechanism; account management; and system management. + +The services described in the service layer are optional. + +### **7.2.1 Expandable services for account management** + +A distributed account system serves all kinds of entities and data in a DLT system. The account management component controls addresses and identities. + +- Address is bound to trust data. Especially in public chain projects, a large part of trust data is usually digital asset. +- Identity is bound to entity. + +Based on decentralized identity (DID) and multi-dimensional authentication protocol, the account management component provides the identity management of node operators, trust endorsement regulatory roles, DApp operators, end users and extended Internet of things access entities. + +The account management of identity usually utilizes the following technologies: + +- certificate authority (CA): the certificates are issued through a centralized CA for various applications in the system, and both the identity and authority management are certified and confirmed by these certificates; +- identity-based encryption (IBE): the identities are confirmed by IBE; +- public key infrastructure (PKI): the identities are confirmed by addresses or accounts based on PKI; +- third party identity authentication: the identities are confirmed by the third party. + +#### **7.2.1.1 Decentralized identity** + +DID establishes a cryptographic-based digital identity for entities (nodes, users, data, things, etc.) in a DLT system. The digital identity is based on DLT and is not subject to any centralized organization. It is potentially controlled by multiple entities and is both secure and trustworthy. + +#### **7.2.1.2 Authentication and authorization** + +AAA solutions support the full use of account management. + +Approaches to authority management differ since a variety of distributed ledgers have different application scenarios. + +The existing business expansions can interface with the existing authentication and authority management. + +The more concentrated authority management business can be complemented by using CA/IBE. + +For instance, the public chains usually adapt PKI technology rather than central authority management like CA/IBE. + +#### **7.2.1.3 Delegation** + +In DLT account management, non-user entities can also use distributed identities. The use of non-user entities shall enable ID delegation, where entity owners can make full use of them. + +### **7.2.2 Expandable services for system management** + +System management includes secure communication, trusted data transmission, and related services of node operation and network governance. + +#### **7.2.2.1 Global configuration** + +Global configuration for nodes inside a DLT system includes network configuration, communication management configuration, node system configuration, etc. + +##### **7.2.2.2 Governance control** + +Governance control is required for individual nodes and a DLT system via multiple nodes, including network status, communication channel monitoring, alarm and tracking, trust endorsement and node failure monitoring. + +#### **7.2.2.3 Supervisory support** + +Under this heading comes component support for the monitoring and supervision of malicious events in a DLT system. + +#### **7.2.3 Expandable services for smart contract mechanism** + +In the service layer, the smart contract mechanism provides the component for contract registration, contract template, contract compiler and VM runtime. + +#### **7.2.4 Expandable services for data integrity** + +#### **7.2.4.1 Policy configuration** + +The policy configuration of trusted data includes access templates, privacy templates, and monitoring and auditing strategies. + +#### **7.2.4.2 Access control** + +Under this heading comes trusted data access control and dynamic data operation control. + +#### **7.2.4.3 Data security and privacy management** + +Under this heading comes security management of data storage and privacy management for data affirmative easement. + +#### **7.2.4.4 Data monitoring and auditing** + +Tracking and monitoring of data, data processing, review and audit of special data events. + +### **7.2.5 Trust endorsement management** + +According to the different trust endorsement approaches of distributed systems, trust endorsement management is used to meet the governance model, e.g., the legal endorsement of off-chain governance in chains that unite nodes with agreements or contracts; the tokenomics endorsement of on-chain governance in public chains. + +#### **7.2.5.1 Incentive module** + +The incentive module is used for accounting incentives for partially distributed ledgers, such as typical public chains. Most public chains use consensus algorithms for targeted token distribution and for motivating the accounting nodes to account. In general, the incentive module is very closely related to the consensus algorithm and governance model. + +## **7.3 Application service platform** + +Functions map to application in Figure 1. + +### **7.3.1 DApp service interfaces** + +DApp services include interfaces that support DApp development, DLT data management and DLT account management. + +The upper layer interface provided by the distributed ledger gives external systems efficient access to distributed ledger data, to external applications to integrate distributed ledgers or to other distributed ledgers for mutual access, usually including RPC, API and SDK. The interface layer mainly completes data synchronization, processing, transferring and exchange, etc. + +The RPC interface connects peripheral components with the distributed ledger nodes over the network and to access the services provided by the distributed ledger. SDK provides a development package for other components to integrate part of the functionality of a distributed ledger. + +RPC and SDK should observe the following rules: + +- completely functional: the transactions of distributed ledger can be completed and maintained, and an intervention strategy and privilege management are operational; +- portable: it can be used in a variety of applications and environment, and is not limited to one absolute software or hardware platform; +- extensible and compatible: it should be as forward and backward compatible as possible, and try not to modify or minimize changes as extending functions; +- easy to use: the structured design and good naming methods should be used to reduce the cost of development. + +Common implementation techniques include call control, serialized objects, and network components. There are various architectures that can be used, such as CORBA [b-corb], JsonRPC [b-jrpc], gRPC [b-grpc], Thrift [b-thrft], RestAPI [b-rapi] and XMLRPC [b-xrpc]. + +### 7.3.2 Accounting, authorization and authentication + +The AAA system manages the DApp user's ability to access data, process data and perform data exchange based on transactions. + +The AAA management shall focus on three parts: + +- access control for DApp users to submit their transaction; +- access control for nodes to access the chain network (permission control), usually being applied by a permissioned chain; +- privilege control for DApp users to operate the data and function calls via transactions. + +### 7.3.3 Data privacy + +Cryptographic techniques are used to protect on-chain user data and to meet data privacy requirements for applications (result in DApps). + +Privacy has always been one obstacle to the application of distributed ledgers. A successful means of satisfying regulatory requirements while not infringing data privacy is key to the distributed ledger industry. + +Privacy protection should therefore meet the following requirements: + +- anonymity controls: to ensure that users can set the transaction, so it is not visible to unrelated parties; +- high performance: privacy-protected design must still be able to meet performance requirements; +- transparent supervision: to satisfy the requirement that privacy protection should not evade the regulatory functions of regulatory agencies – technically, this feature is optional. + +### 7.3.4 Data storage and synchronization + +Useful tools for end users and DApps to facilitate DApp data storage, synchronization, operation and credentials. + +## **7.4 DLT applications** + +Functions map to application in Figure 1. + +Multiple applications based on DLT systems are called DApps. DApps are not part of DLT systems, they are consumers instead. + +## **7.5 DLT extendable interfaces and external services** + +Functions map to external interaction management and extensions in Figure 1. + +The reference architecture of DLT, especially a blockchain system, shall meet the requirement to balance the needs of security, decentralization and scalability. Furthermore, decentralized systems focus on resolving trust issues in competitive business environments, therefore, not all business cases shall use decentralized systems. + +A hybrid system combines DLT with typical IT systems and can satisfy most business requirements. From a DLT perspective, external services provide solutions to cooperate with external systems. + +### **7.5.1 Administration interfaces** + +Administration interfaces provide capability for external systems to create applications to communicate with DLT services and manage DLT nodes if possible. + +### **7.5.2 External governance interfaces** + +For DLT systems that use trust endorsement for nodes externally, they should provide external governance interfaces, e.g., permissioned agreements or contracts signed by nodes. + +### **7.5.3 Exchange protocol and extensional resources** + +Exchange protocols define the way DLT system interact with external resources in non-DLT systems, third party DLT systems or L2 blockchain technology. + +Many extensions are provided to implement exchange protocols. + +The main purpose of L2 blockchain technology is to scale blockchain transaction capacity while retaining the benefits decentralization brings to a distributed protocol. Solving the scalability problem will significantly help with blockchain's mainstream adoption. Layer 2 blockchain technology systems are those that connect to and rely on blockchain systems as a base layer of security and finality. + +L2 solutions include plasma, state channel, sharding, raiden network and lightening network. + +#### **External interaction or interoperation management** + +In L2 solutions, the DLT system is able to interact or interoperate with L2 systems. + +#### **External resource management** + +To cooperate with non-DLT systems and third party DLT systems, mostly data or resource exchange transactions, resource management is required. + +# Bibliography + +- [b-ITU-T F.751.0] Recommendation ITU-T F.751.0 (2020), *Requirements for distributed ledger systems*. +- [b-ITU-T TS FG DLT D3.1] Technical Specification ITU-T FG DLT D3.1 (2019), *Distributed ledger technology reference architecture*. +- [b-ISO 22739] ISO 22739:2020, *Blockchain and distributed ledger technologies – Vocabulary*. +- [b-Clow] Clow, J., Jiang, Z. (2017). *A Byzantine fault tolerant raft*. Stanford, CA: Stanford University. 6 pp. Available [viewed 2020-10-14] at: [http://www.scs.stanford.edu/17au-cs244b/labs/projects/clow\\_jiang.pdf](http://www.scs.stanford.edu/17au-cs244b/labs/projects/clow_jiang.pdf) +- [b-Lamport] Lamport, L. (1998). The part-time parliament. *ACM T. Comput. Syst.*, **16**(2), pp. 133-169. Updated version available [viewed 2020-10-14] at: +- [b-bitc] Bitcoin (Internet). *Bitcoin is an innovative payment network and a new kind of money*. Available [viewed 2020-10-14] at: +- [b-corb] CORBA (Internet). *Common Object Request Broker Architecture*. Milford, MA: CORBA. Available [viewed 2020-10-14] at: +- [b-dbft] NEO (Internet). *Consensus mechanism*. Available [viewed 2020-10-14] at: +- [b-dock] Docker (Internet). *Get started with Docker*. Palo Alto, CA: Docker Inc. Available [viewed 2020-10-14] at: +- [b-dpos] Bitshares (Internet). *Delegated proof-of-stake consensus*. Blacksburg, VA: Bitshares. Available [viewed 2020-10-14] at: +- [b-ethe] Ethereum (Internet). *Ethereum is a global, open-source platform for decentralized applications*. Zug: Ethereum Foundation. Available [viewed 2020-10-14] at: +- [b-vm] Chinchilla, C., editor (2020). *Ethereum virtual machine (EVM) awesome list*. San Francisco, CA: Github. Available [viewed 2020-10-14] at: [https://github.com/ethereum/wiki/wiki/Ethereum-Virtual-Machine-\(EVM\)-Awesome-List](https://github.com/ethereum/wiki/wiki/Ethereum-Virtual-Machine-(EVM)-Awesome-List) +- [b-grpc] Cloud Native Computing Foundation (Internet). *About gRPC*. San Francisco, CA: Linux Foundation. Available [viewed 2020-10-14] at: +- [b-hbft] Miller, A. (2020) *The honey badger of BFT protocols*. San Francisco, CA: Github. Available [viewed 2020-10-14] at: +- [b-jrpc] Morley, M. (Internet). *JSON-RPC*. MPCM Technologies LLC. Available [viewed 2020-10-14] at: +- [b-js] Pluralsight (Internet). *Ready to try JavaScript?*. javascript.com. Available [viewed 2020-10-14] at: +- [b-jvm] Tyson, M. (2018). *What is the JVM? Introducing the Java virtual machine*. Framingham, MA: IDG Communications. Available [viewed 2020-10-14] at: + +- [b-leveldb] Ghemawat, S., Dean, J. (2020). *LevelDB*. San Francisco, CA: Github. Available [viewed 2020-10-14] at: +- [b-lua] Ierusalimschy, R., Celes, W., de Figueiredo, L.H. (Internet). *Lua*. Available [viewed 2020-10-14] at: +- [b-mysql] Oracle (Internet). *MySQL*. Redwood City, CA: Oracle. Available [viewed 2020-10-14] at: +- [b-raft] Raft (Internet). *The Raft consensus algorithm*. San Francisco, CA: Github. Available [viewed 2020-10-14] at: +- [b-rapi] RESTfulAPI.net (2020). *What is REST?* Available [viewed 2020-10-14] at: +- [b-shard1] Ethereum (2020). *Sharding FAQs*. San Francisco, CA: Github. Available [viewed 2020-10-14] at: +- [b-shard2] Ontology (2020). *Ontology Sharding PDF*. San Francisco, CA: Github. Available [viewed 2020-10-14] at: +- [b-thrift] Apache Software Foundation (2017). *Apache Thrift*. Wakefield, MA: Apache Software Foundation. Available [viewed 2020-10-14] at: +- [b-utxo] Bitcoin Project (2009-2020). Unspent transaction output (UTXO). In: *Transactions*. Arlington, VA: Bitcoin. Available [viewed 2020-10-14] at: +- [b-vbft] Ontology (2020). *VBFT introduction*. San Francisco, CA: Github. Available [viewed 2020-10-14] at: [https://ontio.github.io/documentation/vbft\\_intro\\_en.html](https://ontio.github.io/documentation/vbft_intro_en.html) +- [b-xrpc] XML-RPC (Internet). *What is XML-RPC?* San Francisco, CA: Github. Available [viewed 2020-10-14] at: + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.751.20-202406-I_PDF-E/raw.md b/marked/F/T-REC-F.751.20-202406-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..83a71c272ff8bd76930a291e64eeb3f270bfbcc9 --- /dev/null +++ b/marked/F/T-REC-F.751.20-202406-I_PDF-E/raw.md @@ -0,0 +1,459 @@ + + +# Recommendation **ITU-T F.751.20 (06/2024)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +## **Reference architecture for distributed ledger technology (DLT)-based multimedia data delivery management systems** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.751.20 + +# Reference architecture for distributed ledger technology (DLT)-based multimedia data delivery management systems + +## Summary + +Recommendation ITU-T F.751.20 specifies the reference architecture for distributed ledger technology (DLT)-based multimedia data delivery management systems including three subsystems and the related high-level requirements to solve the problems of quality assurance and trusted sharing of multimedia data, support the construction of a traceable value and revenue distribution system, and the corresponding multimedia data applications and services. + +### History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.751.20 | 2024-06-13 | 16 | 11.1002/1000/15925 | + +### Keywords + +DLT, data delivery, multimedia. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2024 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|-----|----------------------------------------------------------------------------------------------------|----| +| 1 | Scope ..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation..... | 2 | +| 4 | Abbreviations and acronyms ..... | 2 | +| 5 | Conventions ..... | 2 | +| 6 | Overview of reference architecture for DLT-based multimedia data delivery management systems ..... | 2 | +| 7 | DLT management subsystem and related high-level requirements..... | 4 | +| 7.1 | Smart contract function ..... | 4 | +| 7.2 | Node management function..... | 4 | +| 7.3 | Privacy protection function ..... | 4 | +| 7.4 | Consensus mechanism function ..... | 5 | +| 7.5 | DLT monitoring function ..... | 5 | +| 7.6 | User management function..... | 5 | +| 8 | Multimedia data delivery service management subsystem and related high-level requirements ..... | 5 | +| 8.1 | Data registration function ..... | 5 | +| 8.2 | Incentive management function ..... | 6 | +| 8.3 | Traceability service function ..... | 6 | +| 8.4 | Data request function..... | 6 | +| 8.5 | Revenue management function ..... | 6 | +| 8.6 | Service log function..... | 7 | +| 8.7 | Data response delivery function ..... | 7 | +| 8.8 | Data verification function..... | 7 | +| 8.9 | Service evaluation function ..... | 7 | +| 9 | Multimedia data management subsystem and related high-level requirements ..... | 7 | +| 9.1 | Data type function ..... | 8 | +| 9.2 | Data quality function ..... | 8 | +| 9.3 | Data authentication function..... | 8 | +| 9.4 | Data encryption function ..... | 8 | +| 9.5 | Data identification function ..... | 8 | +| 9.6 | Data storage function..... | 8 | +| | Bibliography..... | 10 | + + + +## Reference architecture for distributed ledger technology (DLT)-based multimedia data delivery management systems + +# 1 Scope + +This Recommendation specifies the reference architecture for distributed ledger technology (DLT)-based multimedia data delivery management systems. It provides three subsystems and the related high-level requirements in the reference architecture: + +- DLT management subsystem and related high-level requirements. +- DLT-based multimedia data delivery service management subsystem and related high-level requirements. +- Multimedia data management subsystem and related high-level requirements. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +# 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 consensus** [b-ITU-T X.1400]: Agreements to confirm the correctness of the blockchain transaction. + +**3.1.2 distributed ledger technology** [b-ISO 22739]: Technology that enables the operation and use of distributed ledgers. + +**3.1.3 delivery** [b-ITU-T X.609]: The procedures and means employed to provide a user with the required archived material for reuse. + +**3.1.4 distributed ledger** [b-ITU-T X.1400]: A type of ledger that is shared, replicated, and synchronized in a distributed and decentralized manner. + +**3.1.5 incentive** [b-ISO 22739]: Offering reward for some activities concerned with the operation of a distributed ledgers technology system. + +**3.1.6 identification** [b-ISO/IEC 29100]: Set of attributes which make it possible to identify the personally identifiable information principal. + +**3.1.7 log** [b-ISO 22739]: Information created, received and maintained as evidence and as an asset by an organization or person, in pursuit of legal obligations or in the transaction of business. + +**3.1.8 multimedia data** [b-ISO/IEC 15938-5]: A representation of multimedia in a formalized manner suitable for communication, interpretation, or processing by automatic means. + +**3.1.9 privacy protection** [b-ISO/TR 18638]: Capacity to control when, how and to what degree information about oneself is communicated to others. + +**3.1.10 revenue** [b-ISO 15663]: Income generated from normal business activities. + +**3.1.11 smart contract** [b-ITU-T X.1400]: A program written on the distributed ledger system which encodes the rules for specific types of distributed ledger system transactions in a way that can be validated and triggered by specific conditions. + +**3.1.12 traceability** [b-ISO 6707-2]: Ability to trace the history, application or location of that which is under consideration. + +### **3.2 Terms defined in this Recommendation** + +This Recommendation defines the following term: + +**3.2.1 cultural service industry**: Service departments specializing in various cultural tasks, including photography, entertainment services, sports, education, art, publishing, library, mass culture, cultural property, cultural brokerage and agency, broadcasting, film, television, and other cultural service industries that do not fall into the above categories. + +## **4 Abbreviations and acronyms** + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|-------------------------------------| +| AES | Advanced Encryption Standard | +| DLT | Distributed Ledger Technology | +| ECC | Elliptic Curve Cryptography | +| GUID | Globally Unique Identifier | +| MD5 | Message Digest Algorithm 5 | +| PBFT | Practical Byzantine Fault Tolerance | +| POS | Proof of Stake | +| POW | Proof of Work | +| SHA | Secure Hash Algorithm | +| TPS | Transactions Per Second | + +## **5 Conventions** + +In this Recommendation: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. + +# **6 Overview of reference architecture for DLT-based multimedia data delivery management systems** + +There are three main parts in the reference architecture for DLT-based multimedia data delivery management systems, including data requesters, multi-party data sources and the DLT-based multimedia data delivery management systems. + +Under this architecture shown in Figure 1, the interaction between data sources and data requesters is mediated by DLT, and the entire process is controlled through the DLT-based multimedia data delivery management system. A typical interaction process is as follows: + +- 1) After uploading or updating multimedia data to the various centralized or decentralized data storage systems, data source registers the scattered and decentralized multimedia data on the DLT-based multimedia data delivery management systems with data identification (including data unique identifier, data fingerprint, and data source user ID) and storage address index, to get the revenues that they deserve. +- 2) Data requesters get data identification after sending a data request to DLT-based multimedia data delivery management systems and receive the requested multimedia data from data storage systems. They then verify the data and send the verified result back to let DLT-based multimedia data delivery management systems know whether they get the right data they want or not. +- 3) DLT-based multimedia data delivery management systems store data identification and the storage address index of the corresponding multimedia data in the DLT and allow the data storage systems to send the inquired multimedia data to the data requesters after querying, according to the storage address index of the multimedia data. In this way, the whole lifecycle of the multimedia data including data source, data delivery, data update and data integration can be trusted recorded and traced, as well as the value contribution, distribution and the changes. So, in this way the rationality of data value and revenue distribution can be guaranteed. The revenue matching their contribution makes the multi-party data sources more willing to upload/update multimedia data through this system, which further promotes the establishment and positive development of the system. + +![Figure 1: Overview of reference architecture for DLT-based multimedia data delivery management systems. The diagram is divided into three main horizontal sections. The top section shows three management subsystems: 'DLT management subsystem' (containing Smart contract, Node management, Privacy protection, Consensus mechanism, DLT monitoring, User management), 'Multimedia data delivery service management subsystem' (containing Data registration, Incentive management, Traceability service, Data request, Revenue management, Service log, Data response delivery, Data verification, Service evaluation), and 'Multimedia data management subsystem' (containing Data type, Data quality, Data authentication, Data encryption, Data identification, Data storage). The middle section shows the 'DLT-based multimedia data delivery management systems' which include a 'Data identification, address index table', an 'Index update' process leading to an 'Index table', and several 'DLT' blocks connected by a 'Smart contract'. The bottom section shows 'Data source' (represented by icons of a game controller, laptop, camera, headphones, and a person) sending 'Data incentive information', 'Data revenue', 'Multimedia data registration', and 'Multimedia data update/upload' to the DLT system. On the right, 'Data requester' (represented by icons of a smartphone, tablet, and laptop) sends 'Multimedia data request' and receives 'Data identification', 'Verify result', and 'Multimedia data delivery'. All three sections point towards a central 'Storage' block at the bottom.](562f471e8153729557e6a4ee6343c32c_img.jpg) + +Figure 1: Overview of reference architecture for DLT-based multimedia data delivery management systems. The diagram is divided into three main horizontal sections. The top section shows three management subsystems: 'DLT management subsystem' (containing Smart contract, Node management, Privacy protection, Consensus mechanism, DLT monitoring, User management), 'Multimedia data delivery service management subsystem' (containing Data registration, Incentive management, Traceability service, Data request, Revenue management, Service log, Data response delivery, Data verification, Service evaluation), and 'Multimedia data management subsystem' (containing Data type, Data quality, Data authentication, Data encryption, Data identification, Data storage). The middle section shows the 'DLT-based multimedia data delivery management systems' which include a 'Data identification, address index table', an 'Index update' process leading to an 'Index table', and several 'DLT' blocks connected by a 'Smart contract'. The bottom section shows 'Data source' (represented by icons of a game controller, laptop, camera, headphones, and a person) sending 'Data incentive information', 'Data revenue', 'Multimedia data registration', and 'Multimedia data update/upload' to the DLT system. On the right, 'Data requester' (represented by icons of a smartphone, tablet, and laptop) sends 'Multimedia data request' and receives 'Data identification', 'Verify result', and 'Multimedia data delivery'. All three sections point towards a central 'Storage' block at the bottom. + +F.751.20(24) + +**Figure 1 – Overview of reference architecture for DLT-based multimedia data delivery management systems** + +Corresponding to the above three parts, three management subsystems are defined. + +- 1) DLT management subsystem focuses on managing the specific DLT systems that support trusted multimedia data delivery service. The related management functions and requirements of DLT systems include smart contract, node management, privacy protection, consensus mechanisms, DLT monitoring, and user management. +- 2) Multimedia data delivery service management subsystem focuses on managing the essential factors and steps of the whole lifecycle of DLT-based multimedia data delivery service to standardize the supply and use of multimedia data delivery service. The related management functions and requirements include data registration, incentive management, traceability service, data request, revenue management, service log, data response delivery, data verification and service evaluation. +- 3) Multimedia data management subsystem focuses on managing the characteristics and processing requirements of multimedia data to ensure the quality of the delivered multimedia data. The related management functions and requirements include data type, data quality, data authentication, data encryption, data identification and data storage. + +# **7 DLT management subsystem and related high-level requirements** + +DLT management subsystem contains the functions of smart contract, node management, privacy protection, consensus mechanism, DLT monitoring and user management. + +### **7.1 Smart contract function** + +The smart contract provides functions to generate automatically executed program patterns of incentive management and revenue management. The functions included are as follows. + +1. It is required to support formulating trigger conditions of smart contract and corresponding rules of performing contracts in advance. +2. It is required to support flexible adjustment of contract patterns based on the changes in the incentive mechanism. +3. It is required to support trigger and execution of incentive management function during data delivery. +4. It is required to support trigger and execution of revenue management function after data verification results are received. + +### **7.2 Node management function** + +The node management provides functions to manage node behaviours and node attributes. The functions included are as follows. + +1. It is required to support managing the entrance or exit of nodes by providing a secure admittance authentication mechanism and exit mechanism. +2. It is required to support managing the function and identity of nodes, such as selecting nodes to participate in consensus based on historical contribution, reputation value, and other factors. + +### **7.3 Privacy protection function** + +The privacy protection provides functions to protect the identity privacy and data privacy during data delivery. It is recommended to adopt the following mechanisms. + +1. Information hiding technology, which provides invisibility, correctness and verifiability of data by cryptography encryption and concomitant credentials. For example, zero knowledge proof, ring signature, and homomorphic encryption are recommended. + +2. Channel isolation technology, which constructs isolated channels located on or off the chain between different nodes, thereby isolating data from the network layer to restrict the data visibility outside the channel. Typical use cases include Raiden Network, Hyperledger channel, etc. + +### **7.4 Consensus mechanism function** + +The consensus mechanism provides functions to achieve the agreement on the validity of ledger transaction records using a prescribed consensus algorithm. The functions included are as follows. + +1. It is required to support selection of the appropriate consensus algorithm based on the data delivery scenarios and requirements, including proof of work (POW), proof of stake (POS), practical byzantine fault tolerance (PBFT), etc. +2. It is recommended to support customization of the consensus mechanism parameters, including ledger packaging interval, the amount of data contained in a single ledger, etc. + +### **7.5 DLT monitoring function** + +The DLT monitoring provides functions to monitor and record the DLT changes and related indicators. The functions included are as follows: + +1. It is required to support monitoring and recording the transactions per second (TPS), response delay, data hit ratio, and other key performance indicators of the ledger. +2. It is required to support monitoring and recording the illegal operations of nodes, including unauthorized access, vicious requests, etc. +3. It is required to support monitoring and recording the network access information and the changes in the ledger. + +### **7.6 User management function** + +The user management provides functions for user attribute creation, modification, deletion and query. The functions included are as follows: + +1. It is required to support specifying the scope of user attributes, including user ID, user registration information, user category, user permission, etc. +2. It is required to support creating uses instance and recording corresponding user attribute information, including user ID, contract account, signature information, address information, etc. +3. It is required to support modifying and deleting specific user attribute information and return processing results. +4. It is required to support querying specific user attribute information. + +## **8 Multimedia data delivery service management subsystem and related high-level requirements** + +Multimedia data delivery service management subsystem contains the functions of data registration, incentive management, traceability service, data request, revenue management, service log, data response delivery, data verification and service evaluation. + +### **8.1 Data registration function** + +The data registration provides functions to store and manage multimedia data uploaded by data sources. The functions included are as follows. + +1. It is required to support encryption and release of the data registration information to the distributed ledger. The registration information should include the following: + +- a. Data unique identifier, which can be generated by a message digest algorithm 5 (MD5), globally unique identifier (GUID), snowflake, and other identification methods. + - b. Data fingerprint, which can be generated by message digest, secure hash algorithm (SHA), and other hash functions. + - c. Data description, which can be composed of data content and structure, data abstract, data owner, etc. + - d. Data source user ID, a unique user identity generated by the user management function. + - e. Multimedia data addresses. +2. It is required to support encryption and storage of the uploaded data. The storage mode can be free storage or on-chain storage. + +### **8.2 Incentive management function** + +The incentive management provides functions to establish and manage the incentive mechanism, including the incentive measurement criteria and incentive rules, to calculate the revenue for data owners. The functions included are as follows. + +1. It is required to support the establishment and management of incentive measurement criteria, including data value, data quality, user feedback, etc. +2. It is required to support the establishment and management of incentive rules, including the forms based on games, auctions, etc. +3. It is recommended to support the updating and evaluation of the incentive mechanism. + +### **8.3 Traceability service function** + +The traceability service provides functions to collect and record information about the use, evolution, and annotation process of multimedia data, in order to locate the data source to assist the revenue distribution. The functions included are as follows. + +1. It is required to support tracing and collecting process information of the evolution from source data to current data. +2. It is recommended to support tracing, describing and recording all the factors that can affect the data state in the evolution process from source data to current data. +3. It is required to support querying the source of data through the annotation method, reverse query method and other methods. + +### **8.4 Data request function** + +The data request provides functions to find or seek the required data and release the incentive information to the data source according to the request information uploaded by the data requester. The following data request information are required to be included. + +1. Data description, including data type, data size, content keywords, etc. +2. User ID, which is used to identify the requesting user and clarify the data destination. +3. Expected payment, as an important parameter of the incentive information. + +### **8.5 Revenue management function** + +The revenue management provides functions to manage and distribute the revenue of multiple data sources participating in the multimedia data delivery. After the evaluation results given by the incentive management, service evaluation and legal audit, the revenue of each data source user will be distributed through a prescribed payment method. The revenue distribution result is recorded in the distributed ledger synchronously. + +### **8.6 Service log function** + +The service log provides functions to record information such as data registration, request, incentive release, data verification, revenue distribution, etc., to facilitate traceability of the lifecycle of data delivery. The following service log types are required to be included. + +1. Data registration log, which records the registration time, data identifier, user ID and other information in the data register process. +2. Data request log, which records the request time, request user ID, request data identifier, request status, service evaluation and other information in the data request process. +3. Incentive release log, which records the incentive release time, incentive data source identification, incentive information and other information in the incentive release process. +4. Data verification log, which records the verification time, fingerprint information, data verification results and other information in the data verification process. +5. Revenue distribution log, which records the distribution time, revenue information, distribution results and other information in the revenue distribution process. + +### **8.7 Data response delivery function** + +The data response delivery provides functions to manage the process of delivering the stored data to the requester. The functions included are as follows. + +1. It is required to support storage data query. The data storage module queries the data in the multimedia database according to the address information in the data registration. +2. It is required to support storage data delivery. The data storage module encrypts the data and delivers the encrypted data to the designated data requester after querying the corresponding multimedia data. + +### **8.8 Data verification function** + +The data verification provides functions to check the consistency between the data fingerprint in the data identification and the original multimedia data. The function can generate a data fingerprint according to the hash algorithm used in the system after obtaining multimedia data, compare the generated data fingerprint with the data fingerprint in the data identification, and verify the integrity and correctness of the data. + +### **8.9 Service evaluation function** + +The service evaluation provides functions to manage data requesters' evaluation for multimedia data delivery service. After obtaining multimedia data, data requesters can evaluate the quality, integrity and service response time of multimedia data delivery. The following evaluation indicators are recommended to be included. + +1. Objective indicators, including objective multimedia data quality, multimedia data integrity, information redundancy, service response time, etc. +2. Subjective indicators, including description accuracy, data readability, user satisfaction, etc. + +## **9 Multimedia data management subsystem and related high-level requirements** + +Multimedia data management subsystem contains the functions of data type, data quality, data authentication, data encryption, data identification and data storage. + +### **9.1 Data type function** + +The data type provides the function to manage various digital type information of multimedia data, such as text, image, graph, audio, video, etc., and supports the identification and access of prevailing multimedia file formats. + +### **9.2 Data quality function** + +The data quality provides functions to judge the registered multimedia data by various metrics, including but not limited to: + +1. Clarity, including image pixels, video resolution, audio signal fidelity, etc. +2. Integrity, including the loss of data amount and data information package, etc. +3. Fluency, including video bitrate and frame rate, etc. + +### **9.3 Data authentication function** + +The data authentication provides functions to reach a consensus on the source and authenticity of data based on digital signature and digital watermarking technology. The following technologies are recommended to be adopted: + +1. Digital signature. The sender of multimedia data attaches a digital signature to the original data by means of private key encryption, etc., so that the receiver can confirm the source and authenticity of the data by means of the corresponding public key decryption. +2. Digital watermarking. The specific digital signals are embedded into multimedia data to protect data integrity and supports data traceability. + +### **9.4 Data encryption function** + +The data encryption provides functions to encrypt data to prevent information leakage and ensure the tamper resistance of data. Suggested encryption algorithms include but are not limited to: + +1. Symmetric encryption algorithm, including advanced encryption standard (AES), etc. +2. Asymmetric encryption algorithm, including elliptic curve cryptography (ECC), etc. +3. Hash algorithm, including MD5, SHA, etc. + +### **9.5 Data identification function** + +The data identification provides functions to generate and manage the identification information generated for the registered data, which provides an overview of the data for the data requester. The functions included are as follows. + +1. It is required to support combining the information that comprehensively and exactly describes the character of the data, such as data unique identifier, data fingerprint, data source user ID, etc., to identify specific data in detail. +2. It is required to support addition, deletion and query for the identification information. +3. It is required to support consequently updating the relevant items of the identification once the data changes. + +### **9.6 Data storage function** + +The data storage provides functions to support on-chain storage or free storage modes of multimedia data and management of the data access. + +1. For on-chain storage mode, it is recommended to support multimedia data storage and access on the chain according to the index in the distributed ledger. + +2. For free storage mode, it is recommended to support multimedia data storage off the chain, data access according to the address information stored on-chain, and the management and monitoring of data access. + +## Bibliography + +- [b-ITU-T X.609] Recommendation ITU-T X.609 (2015), *Managed peer-to-peer (P2P) communications: Functional architecture*. +- [b-ITU-T X.1400] Recommendation ITU-T X.1400 (2020), *Terms and definitions for distributed ledger technology*. +- [b-ISO 6707-2] ISO 6707-2:2017, *Buildings and civil engineering works – Vocabulary – Part 2: Contract and communication terms*. +<> +- [b-ISO 15663] ISO 15663:2021, *Petroleum, petrochemical and natural gas industries – Life cycle costing*. +<> +- [b-ISO 22739] ISO 22739:2024, *Blockchain and distributed ledger technologies – Vocabulary*. +<> +- [b-ISO/IEC 15938-5] ISO/IEC 15938-5:2003, *Information technology – Multimedia content description interface – Part 5: Multimedia description schemes*. +<> +- [b-ISO/IEC 29100] ISO/IEC 29100:2024, *Information technology – Security techniques – Privacy framework*. +<> +- [b-ISO/TR 18638] ISO/TR 18638:2017, *Health informatics – Guidance on health information privacy education in healthcare organizations*. +<> + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.751.21-202406-I_PDF-E/raw.md b/marked/F/T-REC-F.751.21-202406-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..9748b397a5927d4e5e89671d7490c22412929fbc --- /dev/null +++ b/marked/F/T-REC-F.751.21-202406-I_PDF-E/raw.md @@ -0,0 +1,542 @@ + + +# Recommendation **ITU-T F.751.21 (06/2024)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +## **Technical requirements on inter-chain interoperability for permissioned distributed ledger technologies** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.751.21 + +# Technical requirements on inter-chain interoperability for permissioned distributed ledger technologies + +## Summary + +In order for data exchanging and sharing in different distributed ledger technology (DLT) systems, this Recommendation specifies general flows of the inter-chain interoperability between different DLT systems, functional requirements of the inter-chain interoperability and general requirements of the inter-chain interoperability for the DLT systems. + +Recommendation ITU-T F.751.21 may be used for guiding the design and implementation of inter-chain interoperability for DLT systems. + +## History\* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.751.21 | 2024-06-13 | 16 | 11.1002/1000/15926 | + +## Keywords + +Distributed ledger technology, inter-chain, interoperability. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2024 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|------|------------------------------------------------------------------------------------------------------|----| +| 1 | Scope..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 3.1 | Terms defined elsewhere ..... | 1 | +| 3.2 | Terms defined in this Recommendation ..... | 2 | +| 4 | Abbreviations and acronyms ..... | 2 | +| 5 | Conventions ..... | 2 | +| 6 | Overview..... | 2 | +| 7 | General flows of inter-chain interoperability for permissioned DLT ..... | 2 | +| 8 | Functional requirements of inter-chain interoperability for permissioned DLT ..... | 5 | +| 8.1 | The overview of functional models in the inter-chain interoperability component ..... | 5 | +| 8.2 | Functional requirements of network model..... | 5 | +| 8.3 | Functional requirements of storage model ..... | 5 | +| 8.4 | Functional requirements of verification model ..... | 6 | +| 8.5 | Functional requirements of certification model ..... | 6 | +| 8.6 | Functional requirements of permission management model..... | 6 | +| 8.7 | Functional requirements of transaction management model..... | 6 | +| 8.8 | Functional requirements of consensus management model ..... | 6 | +| 9 | General requirements of inter-chain interoperability for permissioned DLT ..... | 6 | +| 9.1 | Consistency requirements of inter-chain interoperability for permissioned DLT ..... | 6 | +| 9.2 | Traceability requirements of inter-chain interoperability for permissioned DLT ..... | 6 | +| 9.3 | Performance requirements of inter-chain interoperability for permissioned DLT ..... | 7 | +| 9.4 | Availability requirements of inter-chain interoperability for permissioned DLT ..... | 7 | +| 9.5 | Information security requirements of inter-chain interoperability for permissioned DLT ..... | 7 | +| | Appendix I – Use cases of the inter-chain interoperability for distributed ledger technologies ..... | 9 | +| I.1 | Inter-chain interoperability for joint tracing..... | 9 | +| I.2 | Inter-chain interoperability for DLT-based invoices..... | 9 | +| I.3 | Inter-chain interoperability for digital forensics of legal contracts ..... | 9 | +| | Appendix II – Use cases..... | 10 | +| II.1 | Enabling to resolve blockchain performance bottlenecks ..... | 10 | +| II.2 | Enabling eliminating silos and increasing data exchange& sharing ..... | 10 | +| | Bibliography..... | 11 | + + + +# Recommendation ITU-T F.751.21 + +## Technical requirements on inter-chain interoperability for permissioned distributed ledger technologies + +## 1 Scope + +This Recommendation specifies general flows, functional requirements of the inter-chain interoperability component and general requirements of the inter-chain interoperability for the permissioned distributed ledger technology (DLT) systems. + +This Recommendation may be used for guiding the design and implementation of inter-chain interoperability for DLT system. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 blockchain** [b-ITU-T F.751.0]: A type of distributed ledger that is composed of digitally recorded data arranged as a successively growing chain of blocks with each block cryptographically linked and hardened against tampering and revision. + +**3.1.2 consensus** [b-ITU-T F.751.0]: Agreement that a set of transactions is valid. + +**3.1.3 distributed ledger** [b-ITU-T F.751.0]: A ledger that is shared across a set of distributed nodes and synchronized between the nodes using a consensus mechanism. + +**3.1.4 distributed ledger technology (DLT)** [b-ITU-T F.751.1]: Technology enabling large groups of nodes in distributed ledger networks to reach agreement and record information without the need for a central authority. + +**3.1.5 distributed ledger system** [b-ITU-T F.751.0]: A system that implements a distributed ledger. + +**3.1.6 permissioned distributed ledger system** [b-ITU-T F.751.0]: Distributed ledger system in which permissions are required to maintain and operate a node. + +**3.1.7 transaction** [b-ITU-T F.751.0]: An incident or an operation that leads to a change in the status of a ledger, such as adding a record or equivalent exchange based on currency. + +**3.1.8 interoperability** [b-ITU-T F.751.2]: The ability of two or more systems or applications to exchange and use information. + +**3.1.9 inter-chain interoperability** [b-ITU-T F.751.2]: Interoperability between different distributed ledger system systems regarding permissioned distributed ledger system refers to the ability to exchange information between different DLT system instances and use the exchanged information, which can be called cross-chain or east-west interoperability. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following term: + +**3.2.1 permissioned distributed ledger technology (DLT)**: Technology enabling large groups of nodes in distributed ledger networks to reach agreement and record information without the need for a central authority, where each node of the groups of nodes are required to be maintained and operated by permissions. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviation and acronym: + +DLT Distributed Ledger Technology + +## 5 Conventions + +In this Recommendation: + +- The keywords "**is required to**" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The keywords "**is recommended**" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. +- The keywords "**can optionally**" indicate an optional requirement which is permissible, without implying any sense of being recommended. This term is not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator or service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with this Recommendation. + +## 6 Overview + +With the development of the DLT ecosystem, different DLT service and system is based on different architectures, protocols, consensus algorithm data formats, etc. each DLT system becomes a DLT island. How to bridge between different DLT systems and guarantee their interconnection has become an important issue. + +In order to bridge and achieve the interconnection between different permissioned DLT systems, it is important to ensure trusted message transmission between them and guarantee the consistency of information exchanged between different permissioned DLT systems. + +This Recommendation describes the general flows of the inter-chain interoperability between the different permissioned DLT systems, functional requirements of the inter-chain interoperability, and general requirements of the inter-chain interoperability for the permissioned DLT systems. Based on this, data exchange and sharing in different permission DLTs systems could be achieved. + +## 7 General flows of inter-chain interoperability for permissioned DLT + +Inter-chain interoperability for permissioned DLT system involves three key roles: DLT system, inter-chain interoperability component and client. A DLT system is responsible for transaction processing to ensure that the data is traceable and cannot be tampered with. An inter-chain + +interoperability component is responsible for the transmission of inter-chain interoperability transactions, as well as the sending and receiving of inter-chain messages. A client generally refers to the other information technology system that interacts with the DLT system. + +The general flows of inter-chain interoperability for permissioned DLT system are shown in Figure 1. For the convenience of explaining, the initiator's DLT system, the initiator's inter-chain interoperability component as well as the initiator's client are referred to as DLT system A, inter-chain interoperability component A and client A for short; and the destination's DLT, the destination's inter-chain interoperability component and the destination's clients are referred to as DLT B, inter-chain interoperability component B and client B for short. + +![Sequence diagram illustrating the general flows for inter-chain interoperability for DLT. The diagram shows interactions between Client A, DLT system A, Component A, Component B, DLT system B, and Client B. The process involves initiating a cross-chain transaction, invoking pre-invoked contracts, verifying and pre-writing the transaction, returning results and proofs, committing or rolling back, notifying the result, writing or rolling back, and finally querying the transaction to return the result.](562f471e8153729557e6a4ee6343c32c_img.jpg) + +``` + +sequenceDiagram + participant Client A + participant DLT system A + participant Component A + participant Component B + participant DLT system B + participant Client B + + Client A->>Component A: 1. Initiate cross-chain transaction + Component A-->>Client A: 2. Response of cross-chain transaction + Component A->>DLT system A: 3. Invoke the pre-invoked cross-chain contract of blockchain A + DLT system A-->>Component A: 4. Return result + Component A->>Component B: 5. Invoke cross-chain transaction + Component B->>DLT system B: 6. Verify and pre-write cross-chain transaction + DLT system B-->>Component B: 7. Return result and proof + Component B-->>Component A: 8. Return result and proof + Component A->>DLT system A: 9. Commit or rollback transaction result + DLT system A-->>Component A: Return OK + Component A->>Component B: 10. Notify result of cross-chain transaction + Component B-->>Component A: Return OK + Component B->>DLT system B: 11. Write or rollback cross-chain transaction + DLT system B-->>Component B: Return OK + Component B->>Client B: 12. Query cross-chain transaction + Client B-->>Component B: Return result + Component A-->>Client A: 13. Response of cross-chain transaction + +``` + +Sequence diagram illustrating the general flows for inter-chain interoperability for DLT. The diagram shows interactions between Client A, DLT system A, Component A, Component B, DLT system B, and Client B. The process involves initiating a cross-chain transaction, invoking pre-invoked contracts, verifying and pre-writing the transaction, returning results and proofs, committing or rolling back, notifying the result, writing or rolling back, and finally querying the transaction to return the result. + +**Figure 1 – General flows for inter-chain interoperability for DLT** + +The general flows for inter-chain interoperability for DLT consists of the following steps: + +Step 1: The inter-chain interoperability component A receives the cross-chain transaction request. The request should include: + +- Globally unique cross-chain transaction identifier; + - The name of DLT system A; + - The identity of DLT system A; + - The name of DLT system B; + - The identity of DLT system B; + - The smart contract name and contract calling method of the blockchain; or the transaction carrier; +- NOTE – The transaction carrier refers to the transaction content information used to execute cross-chain transaction transactions. +- Time stamp of cross-chain transaction occurrence. + +Step 2: The inter-chain interoperability component A returns the response of the cross-chain transaction to the client A. + +Step 3: The inter-chain interoperability component A invokes the pre-invoked cross-chain contract of DLT system A. + +Step 4: DLT system A executes the transaction and returns the result to the inter-chain interoperability component A. wherein, optionally, the returned result includes documents of proof of source. Documents of proof of source include transaction version, transaction hash value, transaction type, timestamp, etc. + +Step 5: The inter-chain interoperability component A sends a message to initiate a cross-chain transaction operation to the inter-chain interoperability component B, and the message should include: + +- a) Cross-chain transaction identification; +- b) Type of cross-chain transaction operation (optional); +- c) Proof documents of source; +- d) Initiator identification; +- e) Destination identification; +- f) The transaction execution parameters or transaction carrier of the destination; +- g) Data credibility proof. + +Step 6: Inter-chain interoperability component B verifies the message and calls the pre-written contract of DLT system B. + +Step 7: DLT system B executes the pre-written contract and returns the proof file to the inter-chain interoperability component B. + +Step 8: The inter-chain interoperability component B sends a message to the inter-chain interoperability component A to notify the execution result of the cross-chain transaction. The message should include the following: + +- a) Cross-chain transaction identification; +- b) Type of cross-chain transaction operation (optional); +- c) Proof documents of source; +- d) Initiator identification; +- e) Destination identification; +- f) The transaction execution parameters or transaction carrier of the destination; +- g) Data credibility proof. + +Step 9: The inter-chain interoperability component A verifies the message and sends a request message to the DLT system A for the final submission or rollback of the request. + +Optionally, the request message may include cross-chain transaction identification; chain name or chain identifier; contract name; the method for the contract to confirm the cross-chain result; or the cross-chain execution result. + +Step 10: Based on the verification result of Step 8, the inter-chain interoperability component A initiates a transaction commit or rollback request to the inter-chain interoperability component B. The request should include the following contents: + +- a) Cross-chain transaction identification; +- b) Initiator identification; +- c) Destination identification; +- d) Data credibility proof; +- e) Type of cross-chain transaction operation (optional); + +- f) Extension fields (optional): cross-chain execution results; +- g) Extension fields (optional): transaction carrier. + +Step 11: Inter-chain interoperability component B sends a commit or rollback request to DLT system B according to the message in step 9 to complete the entire cross-chain transaction process. + +Step 12 (optional): The inter-chain interoperability component A returns the response of the cross-chain transaction to the client A, and the response includes the following contents: + +- a) Cross-chain transaction identification; +- b) Return information of cross-chain transaction operation. + +Step 13: Client A queries DLT system B for the execution of the cross-chain transaction and gets the return result. + +## 8 Functional requirements of inter-chain interoperability for permissioned DLT + +### 8.1 The overview of functional models in the inter-chain interoperability component + +An inter-chain interoperability component for permissioned DLT is composed of a network model, storage model, verification model, certificate model, authority model, transaction management model, and consensus management model, as shown in Figure 2. Models in Figure 2 is recommended to be deployed in one physical unit or be distributed in different physical units. + +![Diagram showing the functions of the inter-chain interoperability component. It consists of seven rectangular boxes arranged in a 3x3 grid within a larger frame. The top row contains 'Network model', 'Storage model', and 'Verification model'. The middle row contains 'Permission management model' and 'Certification model'. The bottom row contains 'Transaction management model' and 'Consensus management model'. The label F.751.21(24) is at the bottom right.](c78c2eefd86269d1740ab85a916f24f2_img.jpg) + +F.751.21(24) + +Diagram showing the functions of the inter-chain interoperability component. It consists of seven rectangular boxes arranged in a 3x3 grid within a larger frame. The top row contains 'Network model', 'Storage model', and 'Verification model'. The middle row contains 'Permission management model' and 'Certification model'. The bottom row contains 'Transaction management model' and 'Consensus management model'. The label F.751.21(24) is at the bottom right. + +**Figure 2 – Functions of the inter-chain interoperability component** + +### 8.2 Functional requirements of network model + +The functional requirements for network model are as follows: + +- a) It is required to support the function of the establishment of secure and encrypted network channel links. +- b) It is required to support data exchange between inter-chain interoperability component and the DLT system. +- c) It is required to support data exchange between the two different inter-chain interoperability components. + +### 8.3 Functional requirements of storage model + +The functional requirements for the storage model are as follows: + +- a) It is required to support encrypted storage of inter-chain interoperability transaction data in ledger mode, and the ledger is required to not be modified or deleted. + +### **8.4 Functional requirements of verification model** + +The functional requirements for the verification model are as follows: + +- a) It is required to support validity verification for inter-chain interoperability transactions, such as parameter check, permission check, relay signature verification, transaction execution result confirmation, etc. +- b) It is required to support the identity and validity verification of messages. + +### **8.5 Functional requirements of certification model** + +The functional requirements for the certification model are as follows: + +- a) It is required to support the generation of user certificates for clients. +- b) It is required to support managing user certificates. +- c) It is required to support the endorsement of validated inter-chain interoperability transactions. + +### **8.6 Functional requirements of permission management model** + +The functional requirements for the permission management model are as follows: + +- a) It is required to support the inter-chain interoperability permission addition and modification. +- b) It is required to support the inter-chain interoperability permission query. + +### **8.7 Functional requirements of transaction management model** + +The functional requirements for the transaction management model are as follows: + +- a) It is required to support the transaction parameter verification, including transactions from DLT and inter-chain interoperability components. +- b) It is required to support the executing transactions to write transactions to DLT. + +### **8.8 Functional requirements of consensus management model** + +The functional requirements for the consensus management model are as follows: + +- a) It is required to support the function of sorting different transactions. +- b) It is recommended to support the byzantine consensus algorithm. + +## **9 General requirements of inter-chain interoperability for permissioned DLT** + +### **9.1 Consistency requirements of inter-chain interoperability for permissioned DLT** + +The consistency requirements of inter-chain interoperability are as follows: + +- a) It is required to ensure the consistency of the inter-chain interoperability transactions in the entire process; +- b) It is required that the transaction performs rollback when it times out or does not meet the consistency requirements; +- c) It is required to support real-time or periodic transaction checking mechanism; +- d) It is required that transaction consistency is guaranteed through third-party intervention when transaction performs rollback fails. + +### **9.2 Traceability requirements of inter-chain interoperability for permissioned DLT** + +The traceability requirements of inter-chain interoperability are as follows: + +- a) It is required to ensure the transaction traceability of inter-chain interoperability transactions in the entire process; + +- b) It is required that inter-chain interoperability transactions are queried on both the initiator's DLT system and the destination's DLT system; +- c) It is required that inter-chain interoperability transactions are queried in the inter-chain interoperability component. + +### **9.3 Performance requirements of inter-chain interoperability for permissioned DLT** + +#### **9.3.1 Concurrent performance requirements of inter-chain interoperability for permissioned DLT** + +The concurrent performance requirements of inter-chain interoperability are as follows: + +- a) It is required that inter-chain interoperability component receives the DLT system's concurrent requests; +- b) It is required that inter-chain interoperability component receives concurrent requests from the other inter-chain interoperability component(s). + +#### **9.3.2 Network performance requirements of inter-chain interoperability for permissioned DLT** + +The network performance requirements of inter-chain interoperability are as follows: + +- a) It is required that the bandwidth occupied by inter-chain interoperability component is within the normal range; +- b) It is required that the inter-chain interoperability component is running properly when the quality of service requirements (round-trip delay, packet loss rate) of the network reaches the quality of Internet access. + +### **9.4 Availability requirements of inter-chain interoperability for permissioned DLT** + +The availability requirements of inter-chain interoperability are as follows: + +- a) It is required to support transaction performs automatic rollback after the failure of inter-chain interoperability component; +- b) It is required to avoid single node failure, and use multi-site or load balancing methods; +- c) It is required to support the automatic backup for the inter-chain interoperability transaction's data and its configuration; +- d) It is required to support data backup, and support quick restore of the business environment by using the backup data when the DLT system failures; +- e) It is required to support incremental or full backup of the ledger; +- f) It is required that the inter-chain interoperability component supports ledger backup and restores data for other nodes. + +### **9.5 Information security requirements of inter-chain interoperability for permissioned DLT** + +#### **9.5.1 Privilege requirements of inter-chain interoperability for permissioned DLT** + +The privilege requirements of inter-chain interoperability are as follows: + +- a) It is required to support transaction privilege setting, support separation of privileges for read and write, separation of privileges for ledger access and smart contract access; +- b) It is required to support the separation of privilege and domains, set management privilege according to user roles, and the principle of least privilege for user authorization; +- c) It is required to support identity authentication for both parties in inter-chain interoperability transactions. + +#### **9.5.2 Access security requirements of inter-chain interoperability for permissioned DLT** + +The security requirements of inter-chain interoperability are as follows: + +- a) It is required to support identity authentication through a user digital certificate; +- b) It is required to support the timeout disconnection mechanism, after the timeout of access, the user session is disconnected or the user identity is re-authenticated; +- c) It is required to prevent the user from re-initiating the authentication request after the user authentication fails for a specific number of times; +- d) It is required to support using dynamic password cards, USB keys, fingerprints, and smart card for identity authentication, and the supporting equipment and systems must comply with the relevant local national regulations; +- e) It is required to support two-factor authentication. + +#### **9.5.3 Data security requirements of inter-chain interoperability for permissioned DLT** + +The data security requirements of inter-chain interoperability are as follows: + +- a) It is required to support end-to-end integrity protection of inter-chain interoperability transaction(s); +- b) It is required to support encryption to protect end-to-end transmission of inter-chain interoperability communication; +- c) It is required to support encrypted storage for local data which is in the ledgers of the inter-chain interoperability component; +- d) It is required that the data of inter-chain interoperability transaction only be obtained by the destination DLT system and not be obtained by other parties; +- e) It is required to support that unauthorized data in the DLT system is not accessed by inter-chain interoperability component(s). + +## Appendix I + +## Use cases of the inter-chain interoperability for distributed ledger technologies + +(This appendix does not form an integral part of this Recommendation.) + +### I.1 Inter-chain interoperability for joint tracing + +Single distributed ledger technology (DLT) system or blockchain for source tracing are limited by the participating organizations of the alliance. It may be difficult to trace the entire product lifecycle. + +Manufacturers, processors, suppliers, and purchasers may build their own DLT platform for source tracing. Through inter-chain interoperability services, data flows between different DLT systems or different blockchains can be secured and trusted, and commodity information can be traced throughout the lifecycle, this could guarantee the commodity quality, and positioning of target customers more accurately. + +### I.2 Inter-chain interoperability for DLT-based invoices + +For DLT-based invoices, in ITU-T H.DLT-INV/[b-ITU-T F.751.4] "General framework for DLT-based invoices" specifies that the invoice can be issued, transferred and received in a structured electronic format over digital ledgers which allows for its automatic, electronic transactions based on the smart contracts. + +In the scenario of DLT-based invoices, for example, one DLT-based invoices system is used to support related service, especially for province A of the People's Republic of (China), and the other DLT-based invoices system is for province B of China. It is very important to exchange and share the data between the two systems regarding service cooperation between the two provinces. Based on this, data flows between different DLT-based invoices systems of different regions can be traced via inter-chain interoperability. + +### I.3 Inter-chain interoperability for digital forensics of legal contracts + +When any dispute occurs, the DLT system for supporting judicial can obtain evidence information from other DLT systems through inter-chain interoperability services. It could guarantee the evidence authenticity, improve the efficiency and effectiveness of evidence collection, and reduce judicial decision costs. + +## Appendix II + +### Use cases + +(This appendix does not form an integral part of this Recommendation.) + +### **II.1 Enabling to resolve blockchain performance bottlenecks** + +DLT itself has problems such as poor performance and insufficient capacity in a single DLT system architecture. + +Single chains or single DLT systems are limited by the trade-offs of decentralization, scalability, and security to support business applications with high transaction throughput and low latency. + +In addition, as the running time of the DLT increases, the storage capacity of the DLT gradually increases, and the data growth rate even exceeds the capacity of the single-chain storage media. The inter-chain interoperability technology can connect multiple dedicated DLT systems to a single composite network. The multi-layer multi-chain architecture with inter-chain interoperability technology is an ideal way to solve the performance bottleneck of DLT. + +### **II.2 Enabling eliminating silos and increasing data exchange& sharing** + +As the business form of blockchain and DLT becomes increasingly complex. + +The application scenarios are increasing, and the pain points of the lack of a unified interconnection mechanism between two DLT systems are magnified, which greatly limits the problem. The liquidity of digital asset values on DLT systems is increasing, and the demand for inter-chain interoperability is increasing. + +## Bibliography + +- [b-ITU-T F.751.0] Recommendation ITU-T F.751.0 (2020), *Requirements for distributed ledger systems*. +- [b-ITU-T F.751.1] Recommendation ITU-T F.751.1 (2020), *Assessment criteria for distributed ledger technology platforms*. +- [b-ITU-T F.751.2] Recommendation ITU-T F.751.2 (2020), *Reference framework for distributed ledger technologies*. +- [b-ITU-T F.751.4] Recommendation ITU-T F.751.4 (2022), *General framework for distributed ledger technology (DLT)-based invoices*. + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.751.23-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.751.23-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..25f8fbfaef317942ed3102c74b40a7d42f80608b --- /dev/null +++ b/marked/F/T-REC-F.751.23-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:83cab7dcabfe939b223d97242d8523e461968d21f7c5432693c30a64f2d95521 +size 7050 diff --git a/marked/F/T-REC-F.751.23-202503-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg b/marked/F/T-REC-F.751.23-202503-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7624adf51f51d480b6d1f3898fa43c1007fef973 --- /dev/null +++ b/marked/F/T-REC-F.751.23-202503-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b9d5e666f17bd952e9e7b163c723bd7e250f12fe65fc866eb5dd4d4e09b6f4ab +size 115200 diff --git a/marked/F/T-REC-F.751.23-202503-I_PDF-E/d4af765160d04ecef538e5066006dc77_img.jpg b/marked/F/T-REC-F.751.23-202503-I_PDF-E/d4af765160d04ecef538e5066006dc77_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..eb0990cb81ff655f4403b839e4c9cd4ffd04bf50 --- /dev/null +++ b/marked/F/T-REC-F.751.23-202503-I_PDF-E/d4af765160d04ecef538e5066006dc77_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:73c6ed934a8b081e5b36da399681e067a5c665d30e99622528c6f6a87679f8be +size 60922 diff --git a/marked/F/T-REC-F.751.23-202503-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.751.23-202503-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..80c08273419c92c2379a91cf424f60377aa0a855 --- /dev/null +++ b/marked/F/T-REC-F.751.23-202503-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aea5e9242d6f727284744a53cfa559918324ca18601381d96476ad2f19d402c8 +size 77887 diff --git a/marked/F/T-REC-F.751.23-202503-I_PDF-E/raw.md b/marked/F/T-REC-F.751.23-202503-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..bd16020fd6059e909de4b5218d6b695b601ef904 --- /dev/null +++ b/marked/F/T-REC-F.751.23-202503-I_PDF-E/raw.md @@ -0,0 +1,414 @@ + + +Recommendation + +**ITU-T F.751.23 (03/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +**Framework and requirements for distributed ledger technology interoperability** + +![ITU logo](84a1d09fb489061482111515543b60dc_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with white lines representing latitude and longitude, and the letters 'ITU' in a bold, blue, sans-serif font superimposed on the globe. + +ITU logo + +# ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.751.23 + +# Framework and requirements for distributed ledger technology interoperability + +## Summary + +Recommendation ITU-T F.751.23 specifies the framework and requirements for distributed ledger technology interoperability. The framework includes three types of interoperability, as well as technical requirements. The three types of interoperability are inter-chain interoperability, app-chain interoperability and off-chain interoperability. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.751.23 | 2025-03-16 | 21 | 11.1002/1000/16256 | + +## Keywords + +App-chain, cross chain, DLT, framework, inter-chain, interoperability, off-chain, requirements. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2025 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|----------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Overview of DLT interoperability framework ..... | 2 | +| 6.1 DLT interoperability framework ..... | 2 | +| 6.2 Interoperability between different DLT systems (inter-chain)..... | 3 | +| 6.3 Interoperability between application and DLT systems (App-chain) ..... | 3 | +| 6.4 Interoperability between DLT and off-chain systems (off-chain)..... | 3 | +| 7 Inter-chain interoperability requirements ..... | 3 | +| 7.1 Functional components..... | 3 | +| 7.2 User layer..... | 4 | +| 7.3 Interface layer..... | 4 | +| 7.4 Transaction layer ..... | 5 | +| 7.5 Core layer ..... | 5 | +| 7.6 Cross layer ..... | 5 | +| 8 App-chain interoperability requirements ..... | 5 | +| 8.1 Functional components..... | 5 | +| 8.2 Data structure..... | 6 | +| 8.3 Interface classification..... | 6 | +| 8.4 Error information..... | 7 | +| 9 Off-chain interoperability requirements ..... | 7 | +| Bibliography..... | 8 | + + + +# Recommendation ITU-T F.751.23 + +## Framework and requirements for distributed ledger technology interoperability + +## 1 Scope + +This Recommendation specifies the framework and requirements for distributed ledger technology interoperability, including: + +- Three types of distributed ledger technology (DLT) interoperability, which are inter-chain interoperability, app-chain interoperability, and off-chain interoperability; +- Technical requirements for distributed ledger technology interoperability. + +This Recommendation can be used as a guideline for development and assessment of DLT interoperability products. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[\[ITU-T F.751.2\]](#) Recommendation ITU-T F.751.2 (2020), *Reference framework for distributed ledger technologies*. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 Byzantine fault tolerance** [b-ITU-T F.751.1]: Property that enables a system to continue operating properly even if some of its components fail or existence of intentional bad actors. + +**3.1.2 consensus** [b-ITU-T F.751.0]: Agreement that a set of transactions is valid. + +**3.1.3 crash fault tolerance (CFT)** [b-ITU-T F.751.1]: Property that enables a system to continue operating properly even if some of its components crash. + +**3.1.4 distributed ledger** [b-ITU-T F.751.0]: A type of ledger that is shared, replicated, and synchronized in a distributed and decentralized manner. + +**3.1.5 entity** [b-ITU-T F.751.0]: Anything that has a separately identifiable existence (e.g., organization, person, group, smart contract or device). An entity uses distributed ledger technology to solve the problem of its business or information systems. + +**3.1.6 peer-to-peer (P2P)** [b-ITU-T F.751.0]: An instantiation of network architectures where all peers have equivalent authority and responsibility, differing completely from that of server and client system. + +**3.1.7 smart contract** [b-ITU-T F.751.0]: Program written on a distributed ledger system which encodes the rules for specific types of distributed ledger system transactions in a way that can be validated, and triggered by specific conditions. + +**3.1.8 transaction** [b-ITU-T F.751.0]: An incident or an operation that leads to a change in the status of the ledger, such as adding a record or equivalent exchange based on currency. + +**3.1.9 node** [b-ITU-T F.751.0]: Device or process that participates in a distributed ledger network. + +### **3.2 Terms defined in this Recommendation** + +This Recommendation defines the following terms: + +**3.2.1 app-chain interoperability:** Interoperability between application and distributed ledger technology (DLT) systems refers to the ability to exchange information between the application system and the dependent DLT system, and use the exchanged information. + +**3.2.2 inter-chain interoperability:** Interoperability between different distributed ledger technology (DLT) systems refers to the ability to exchange information between different DLT systems and use the exchanged information, which can be called cross-chain or east-west interoperability. + +**3.2.3 off-chain interoperability:** Interoperability between distributed ledger technology (DLT) and off-chain systems refers to the ability to exchange information between the off-chain system and DLT system, and use the exchanged information. + +## **4 Abbreviations and acronyms** + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|-----------------------------------| +| API | Application Programming Interface | +| App | Application | +| CFT | Crash Fault Tolerance | +| CLI | Command-Line Interface | +| DApp | Decentralized Application | +| DID | Decentralized Identifier | +| DLT | Distributed Ledger Technology | +| P2P | Peer-to-Peer | +| SDK | Software Development Kit | +| VM | Virtual Machine | + +## **5 Conventions** + +The following conventions are used in this Recommendation: + +- The keywords "**is required to**" indicate a requirement which must be strictly followed and from which no deviation is permitted, if conformance to this Recommendation is to be claimed. +- The keyword "**functions**" is defined as a collection of functionalities. + +## **6 Overview of DLT interoperability framework** + +### **6.1 DLT interoperability framework** + +The framework of DLT interoperability includes three types of interoperability, inter-chain interoperability, app-chain interoperability, and off-chain interoperability. The framework of DLT interoperability is shown in Figure 1. + +![Figure 1 – Framework of DLT interoperability. The diagram shows three layers: Application systems (top), DLT systems (middle), and Off-chain systems (bottom). Application systems include Financial application, Government application, Industrial application, and Other application. DLT systems include DLT system A (with Node 1, Node 2, Node 3) and DLT system B (with Node 1, Node 2, Node 3). Off-chain systems include Financial data system, Government data system, Industrial data system, and Other data system. Arrows indicate App-chain interoperability between Application systems and DLT systems, Inter-chain interoperability between DLT system A and DLT system B, and Off-chain interoperability between DLT systems and Off-chain systems. A small text 'F.751.23(25)' is in the bottom right corner.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +Figure 1 – Framework of DLT interoperability. The diagram shows three layers: Application systems (top), DLT systems (middle), and Off-chain systems (bottom). Application systems include Financial application, Government application, Industrial application, and Other application. DLT systems include DLT system A (with Node 1, Node 2, Node 3) and DLT system B (with Node 1, Node 2, Node 3). Off-chain systems include Financial data system, Government data system, Industrial data system, and Other data system. Arrows indicate App-chain interoperability between Application systems and DLT systems, Inter-chain interoperability between DLT system A and DLT system B, and Off-chain interoperability between DLT systems and Off-chain systems. A small text 'F.751.23(25)' is in the bottom right corner. + +**Figure 1 – Framework of DLT interoperability** + +### 6.2 Interoperability between different DLT systems (inter-chain) + +Interoperability between different DLT systems refers to the ability to exchange information between different DLT systems, which can be called cross-chain or inter-chain interoperability. Inter-chain interoperability includes interoperability between homogeneous DLT systems or heterogeneous DLT systems. + +### 6.3 Interoperability between application and DLT systems (App-chain) + +Interoperability between applications and DLT systems refers to the ability to exchange information between the application system and the dependent DLT system. It includes: + +- a) Data transferring and value sharing between different applications through the DLT systems; +- b) Interaction between the application system and DLT system. + +### 6.4 Interoperability between DLT and off-chain systems (off-chain) + +Interoperability between DLT and off-chain systems refers to the ability to exchange information between the off-chain system and DLT systems. It focuses on the process of security interaction and trust between off-chain systems and DLT systems. + +## 7 Inter-chain interoperability requirements + +### 7.1 Functional components + +Functional components of inter-chain interoperability are shown in Figure 2. These functional components include core functions, inter-connection functions, transaction functions, interface functions, user functions and cross layer functions. + +![](a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg) + +| | | | | | | +|-------------------------|------------------------|--------------------------|-------------------------|----------------------------|--------------------| +| User | Self-service functions | Business functions | DApp management | Identity management | Cross layer | +| Interface | Management interface | Development SDK | Smart contract template | Cross-DLT middleware | | +| Transaction | Dispatcher framework | Transaction consistency | Transaction isolation | Recovery | | +| Inter-connection | Block verification | Transaction verification | DLT alignment | Resource allocate protocol | | +| Core | Resource management | Cryptographic | Storage | Network | | + +**Cross layer** + +Governance + +Operation + +Security + +F.751.23(25) + +**Figure 2 – Functional components of inter-chain interoperability** + +The framework of inter-chain interoperability includes five layers and a group of cross-layer functions: + +- User layer components enable data interoperability cross systems for users; +- Interface layer defines the data interoperability interfaces for DApps; +- Transaction layer specifies the transaction protocols and components that enable interoperability across DLT systems; +- Interconnection layer includes components for resource allocation, data consistency, and data integrity across DLT systems; +- Core layer defines the essential components of DLT systems that provide foundational capabilities for data interoperability at higher layers; +- Cross-layer functions include functions of security, operation, and governance. + +### 7.2 User layer + +The user layer includes components of self-service, business operations, DApp management, and identity management: + +- Self-service, business operations and DApp management facilitate users interaction with inter-chain systems, which can be achieved through command-line interfaces (CLIs), graphical user interfaces (GUIs), or application programming interfaces (APIs); +- User identity manages the principal identity of operational users, and is required to support the identities mutual recognition of different DLTs, including identity parsing, mapping and conversion. + +### 7.3 Interface layer + +The interface layer includes components of management interfaces, software development kits (SDKs), smart contract templates, and cross-DLT mid-ware. Key functional components are required to include cross-DLT transaction interfaces, cross-DLT application management interfaces, and governance management interfaces: + +- Cross-DLT transaction interfaces are required to support the generation and submission of inter-chain transactions, deployment and invocation of inter-chain contracts, and query of execution results; + +- Cross-DLT application management interfaces are required to support the development, deployment, and invocation of inter-chain applications; +- Governance management interfaces are required to support inter-chain proposal initiation, voting, airfare, challenge, voting, and other functions. + +### **7.4 Transaction layer** + +The transaction layer includes components of the dispatcher framework, transaction consistency, transaction isolation, and recovery: + +- The dispatcher framework is required to schedule transactions, including transaction sorting and priority allocation, transaction scheduling and execution, and concurrency control; +- The transaction consistency is required to keep a consistent ledger state for all nodes, which includes ledger consistency, sequence consistency, and result consistency; +- The transaction isolation is required to ensure that the execution processes and results between transactions are independent; +- The recovery requirement is required to ensure data and node consistency when faults occur. + +### **7.5 Core layer** + +The core layer includes resource management, cryptographic, storage and network components. The basic requirements of the core layer are presented in clause 7.1 of [ITU-T F.751.2]. Additional requirements are proposed as follows: + +- The resource management component is required to manage resources, including cross-chain smart contracts, and cross-chain addresses; +- The cryptography component is required to ensure the security and reliability of cross-chain networks; +- The data storage component is required to ensure the integrity and consistency of cross-chain transactions, execution results, and states. + +### **7.6 Cross layer** + +The cross layer includes components of governance, operation and security: + +- Cross-chain governance is required to support different entities to vote on cross-chain operations, rules and services; +- Cross-chain operation is required to support the joining and exiting of cross-chain networks, and support operation and maintenance management; +- Cross-chain security is required to support cross-chain network status monitoring and risk alerts, support privacy protection of sensitive information, and support auditing of key operations. + +## **8 App-chain interoperability requirements** + +### **8.1 Functional components** + +The interface is the interaction hub between the application and underlying chain, the carrier of direct interaction between application developers and users. Different DLT systems have different interfaces, which include basic data structures, interface classifications, and error information. This document standardizes the interface layer from three aspects: data structure, interface classification and error information. Functional components of app-chain interoperability are shown as Figure 3. + +![Figure 3 – Functional components of app-chain interoperability. The diagram is divided into three main sections: Interface classification, Data structure, and Error information. Interface classification includes Contract interface, Transaction interface, Block interface, Account interface, Network interface, and Event interface. Data structure includes Block, Transaction, Contract, Account, Configuration, and Error. Error information includes Client error, Network error, System error, and VM error.](d4af765160d04ecef538e5066006dc77_img.jpg) + +| Interface classification | | | Error information | +|--------------------------|-----------------------|-----------------|-------------------| +| Contract interface | Transaction interface | Block interface | Client error | +| Account interface | Network interface | Event interface | Network error | +| Data structure | | | System error | +| Block | Transaction | Contract | VM error | +| Account | Configuration | Error | | + +F.751.23(25) + +Figure 3 – Functional components of app-chain interoperability. The diagram is divided into three main sections: Interface classification, Data structure, and Error information. Interface classification includes Contract interface, Transaction interface, Block interface, Account interface, Network interface, and Event interface. Data structure includes Block, Transaction, Contract, Account, Configuration, and Error. Error information includes Client error, Network error, System error, and VM error. + +**Figure 3 – Functional components of app-chain interoperability** + +### 8.2 Data structure + +As the basic data organization mode in DLT systems, basic data structure is closely related to data communication, consensus mechanisms and interaction interfaces. Basic data structure is composed of six data structures which are block, transaction, contract, account, configuration and error: + +- Block structure includes block height, block hash, previous block hash, block timestamp, block initiator, block signature, version information, transaction list, multi-signature, random number, transaction count, Merkle root, gas fee, consensus information, and other data; +- Transaction structure includes transaction hash, version information, transaction timestamp, initiator identifier, recipient identifier, initiator signature, transaction content data, transaction random number, transaction index, transaction ID, block height, block hash, Merkle root, gas information, and other data; +- Contract structure includes contract identifier, contract hash, version information, code hash, code information, contract type, application binary interface (ABI) description, contract status, contract name, contract initiator, and timestamp; +- Account structure includes ordinary accounts, contract accounts and node accounts, their data elements include account identifier, account public key, account private key, account algorithm type, account status, version information, account role type, and certificate information; +- Configuration structure includes node IP, node port, peer-to-peer (P2P) port, node name, node hash identifier, node public key, node version, and interface version; +- Error structure refers to the error information returned when an interface call encounters an exception. It includes error code, error name, and error description. + +### 8.3 Interface classification + +As the communication hub between the application layer and DLT systems, the interface is the main carrier for the implementation of the DLT application. The interface includes contract interface, transaction interface, block interface, account interface, network interface, and event interface: + +- Contract interface includes functions deploying contracts, initiating contracts, upgrading contracts, stopping contracts, and querying contracts; +- Transaction interface includes transaction assembly, transaction sending, transaction verification, and transaction querying; +- Block interface includes block signing, block packaging, block transmission, block verification, and block querying; + +- Account interface includes query account information, retrieve detailed information of a specific account, create a new account, modify specific account information, and delete an account; +- Network interface includes query the list of DLT networks in the system and their detailed information; +- Event interface includes query the list of available events for subscription, subscribe to an event, and cancel event subscription. + +### 8.4 Error information + +The error information is the error feedback information when an exception occurs. Different DLTs have different error information, which increases the difficulty and workload of application debugging for application developers. Error information includes client error, network error, chain system error, and virtual machine error: + +- Client error includes invalid data format, host unreachable, permission denied, and unknown error; +- Network error includes address in use, address not available, and connection aborted; +- Chain error includes service message convert failed, service run local transaction failed, and service query no result; +- Virtual machine error includes VM generic failed, VM invalid instruction, VM stack overflow, VM stack underflow, and VM out of gas. + +## 9 Off-chain interoperability requirements + +Off-chain interoperability mainly refers to trusted interaction between on-chain and off-chain. + +Off-chain interoperability is required to provide following functions: + +- Object abstraction functions define the basic data organization mode in off-chain systems, and support diverse object abstractions in terms of semantic and syntactic data, the representation, metadata description and coding; +- Syntactic mediate functions define the standard syntax for communication of off-chain data, including data formats, syntax, and coding methods, and enable DLT system calls to heterogeneous off-chain systems to generate data that are stored and used in different formats; +- Utility functions define shared functions been used by off-chain systems to enable data privacy and data protection for data operating on-chain, and to provide audit functions to satisfy regulation and compliance requirements. + +## Bibliography + +- [b-ITU-T F.751.0] Recommendation ITU-T F.751.0 (2020), *Requirements for distributed ledger systems*. +- [b-ITU-T F.751.1] Recommendation ITU-T F.751.1 (2020), *Assessment criteria for distributed ledger technology platforms*. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.751.24-202503-I_PDF-E/07b17a620c75522d53916a11e12d1bff_img.jpg b/marked/F/T-REC-F.751.24-202503-I_PDF-E/07b17a620c75522d53916a11e12d1bff_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7758f9dd96060ad7cd5b66bdb3d58e1b5fbd45ac --- /dev/null +++ b/marked/F/T-REC-F.751.24-202503-I_PDF-E/07b17a620c75522d53916a11e12d1bff_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4bf2e41cd6d91c3b152fe011cff28ec0168838ba0cfc8326f24a95dca4abf308 +size 59165 diff --git a/marked/F/T-REC-F.751.24-202503-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg b/marked/F/T-REC-F.751.24-202503-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3967096e8ab3b202695ca9fe29ada4eae56bcf00 --- /dev/null +++ b/marked/F/T-REC-F.751.24-202503-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0ff7b329815de4d0d58984820329ae8378b7cd3d189bb41eea23b45752c6e92b +size 51257 diff --git a/marked/F/T-REC-F.751.24-202503-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.751.24-202503-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c1629394c02f697f60ae954ad36bf274fa8082cb --- /dev/null +++ b/marked/F/T-REC-F.751.24-202503-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3763acd7a0ed56aa30f3f90468c1fb0cb54573eb9045bca3eddf8976b4f0f41d +size 7219 diff --git a/marked/F/T-REC-F.751.24-202503-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg b/marked/F/T-REC-F.751.24-202503-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..fbfbf7f747fae1648716bdd547f6681c0259a950 --- /dev/null +++ b/marked/F/T-REC-F.751.24-202503-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:92675f3ad4fdd89d008a6220813964e333d99088deca90ccebf034aaf32be2c9 +size 47930 diff --git a/marked/F/T-REC-F.751.24-202503-I_PDF-E/7e670a2b556b53ea9002dfff3a420e08_img.jpg b/marked/F/T-REC-F.751.24-202503-I_PDF-E/7e670a2b556b53ea9002dfff3a420e08_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..684236551d6e291d59f50127e2f0eb5f81b6ad3c --- /dev/null +++ b/marked/F/T-REC-F.751.24-202503-I_PDF-E/7e670a2b556b53ea9002dfff3a420e08_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:52948e990ae8affe5fa809159d03098857c45d900960ad04dd099bef5c321a6f +size 53077 diff --git a/marked/F/T-REC-F.751.24-202503-I_PDF-E/af7916c89a458fdab6c3f443217388ae_img.jpg b/marked/F/T-REC-F.751.24-202503-I_PDF-E/af7916c89a458fdab6c3f443217388ae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..137f5b294fcecb4e1c46a9d465ce77108460f6c1 --- /dev/null +++ b/marked/F/T-REC-F.751.24-202503-I_PDF-E/af7916c89a458fdab6c3f443217388ae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:41907c9c4c4c9657110a6d0906c4ad1a8a94ea58ad4f5e7732bc06a7295083ae +size 48154 diff --git a/marked/F/T-REC-F.751.24-202503-I_PDF-E/b8661c6c54f72ecc7ff6cb05e47b2891_img.jpg b/marked/F/T-REC-F.751.24-202503-I_PDF-E/b8661c6c54f72ecc7ff6cb05e47b2891_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d858764de1c53a139cdf8719bb401ed6845ea726 --- /dev/null +++ b/marked/F/T-REC-F.751.24-202503-I_PDF-E/b8661c6c54f72ecc7ff6cb05e47b2891_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ebb26653914774bf10a82fd57916f59b0030a41f0faa895123382e0ee8dcb128 +size 77076 diff --git a/marked/F/T-REC-F.751.24-202503-I_PDF-E/raw.md b/marked/F/T-REC-F.751.24-202503-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..cbaace0558196d388e24a9b26034c10c3b889517 --- /dev/null +++ b/marked/F/T-REC-F.751.24-202503-I_PDF-E/raw.md @@ -0,0 +1,484 @@ + + +# Recommendation **ITU-T F.751.24 (03/2025)** + +SERIES F: Non-telephone telecommunication services +Multimedia services + +--- + +## **Framework and requirements for authorization services based on distributed ledger technology** + +![ITU logo](390120de4fe440c42fea8154fcaad334_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.751.24 + +# Framework and requirements for authorization services based on distributed ledger technology + +## Summary + +In recent years, with the increasing number of online services, authorization has become necessary for the collection of information and the provision of services. In some cases, there may be default or mandatory authorization for using personal information without the issuer's explicit awareness during online interactions. + +Recommendation ITU-T F.751.24 outlines the framework and requirements for authorization services based on distributed ledger technology (DLT), aiming to enhance the management, control, and traceability of the authorization process between issuers and recipients. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.751.24 | 2025-03-16 | 21 | 11.1002/1000/16257 | + +## Keywords + +Authorization service, DLT, framework, requirement. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2025 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|---------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Framework of DLT-based authorization service..... | 2 | +| 7 Functional requirements ..... | 2 | +| 7.1 Functional requirements of DLT layer ..... | 2 | +| 7.2 Functional requirements of authorization base layer..... | 3 | +| 7.3 Functional requirements of authorization service layer ..... | 3 | +| Appendix I – Use cases of the DLT-based authorization service system..... | 4 | +| Appendix II – Use cases of TEE from client device..... | 5 | +| Appendix III – Authorization and verification reference procedures ..... | 6 | +| III.1 Stockholders ..... | 6 | +| III.2 Authorization and verification procedure..... | 6 | +| Bibliography..... | 10 | + + + +# Recommendation ITU-T F.751.24 + +# Framework and requirements for authorization services based on distributed ledger technology + +## 1 Scope + +This Recommendation specifies the framework and requirements for authorization services based on distributed ledger technology (DLT). It includes framework and functional requirements of a DLT-based authorization service. + +This Recommendation is intended to serve as a reference for interested parties regarding authorization services based on DLT. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[[ITU-T F.751.0](#)] Recommendation ITU-T F.751.0 (2020), *Requirements for distributed ledger systems*. + +[[ITU-T F.751.2](#)] Recommendation ITU-T F.751.2 (2020), *Reference framework for distributed ledger technologies*. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 distributed ledger** [ITU-T F.751.0]: A type of ledger that is shared, replicated, and synchronized in a distributed and decentralized manner. + +**3.1.2 consensus** [ITU-T F.751.0]: Agreement that a set of transactions is valid. + +**3.1.3 distributed ledger technology (DLT)** [b-ITU-T F.751.1]: Technology enabling large groups of nodes in distributed ledger networks to reach agreement and record information without the need for a central authority. + +**3.1.4 smart contract** [ITU-T F.751.0]: Program written on the distributed ledger system that encodes the rules for specific types of distributed ledger system transactions in a way that can be validated, and triggered by specific conditions. + +**3.1.5 decentralized identifier (DID)** [b-ITU-T X.1403]: A globally unique identifier that does not require a centralized registration authority because it is registered with distributed ledger technology (DLT) or other form of decentralized systems. + +### 3.2 Terms defined in this Recommendation + +None. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-----|-------------------------------| +| DID | Decentralized Identifier | +| DLT | Distributed Ledger Technology | +| TEE | Trusted Execution Environment | +| ZKP | Zero-Knowledge Proof | + +## 5 Conventions + +- The keywords "**is required to**" indicate a requirement which must be strictly followed and from which no deviation is permitted, if conformance to this Recommendation is to be claimed. +- The keywords "**is recommended**" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. + +## 6 Framework of DLT-based authorization service + +The overall framework of DLT-based authorization service is shown in Figure 1. + +![Figure 1: Framework of DLT-based authorization service. The diagram shows three stacked layers. The top layer, 'Authorization services layer', contains six modules arranged in a 2x3 grid: Authorization cancelling, Trusted enterprise list management, Authorization duration management (top row); Verifiable credential issuance, Verifiable credential storage, Authorization verification (bottom row). The middle layer, 'Authorization basic layer', contains two modules: Verifiable credential management and Decentralized identifier management. The bottom layer is a single block labeled 'DLT layer'.](07b17a620c75522d53916a11e12d1bff_img.jpg) + +| | | | | +|------------------------------|----------------------------------|------------------------------------|-------------------------------------| +| Authorization services layer | Authorization cancelling | Trusted enterprise list management | Authorization duration management | +| | Verifiable credential issuance | Verifiable credential storage | Authorization verification | +| Authorization basic layer | Verifiable credential management | | Decentralized identifier management | +| | DLT layer | | | + +Figure 1: Framework of DLT-based authorization service. The diagram shows three stacked layers. The top layer, 'Authorization services layer', contains six modules arranged in a 2x3 grid: Authorization cancelling, Trusted enterprise list management, Authorization duration management (top row); Verifiable credential issuance, Verifiable credential storage, Authorization verification (bottom row). The middle layer, 'Authorization basic layer', contains two modules: Verifiable credential management and Decentralized identifier management. The bottom layer is a single block labeled 'DLT layer'. + +F.751.24(25) + +**Figure 1 – Framework of DLT-based authorization service** + +The framework primarily consists of three layers: DLT layer, authorization basic layer and authorization services layer. + +- The DLT layer provides DLT basic capacity. The DLT framework can be found in [ITU-T F.751.2]. +- The authorization basic layer includes a verifiable credential management module and a decentralized identifier management module. +- The authorization service layer includes an authorization cancelling module, a trusted enterprise list management module, an authorization duration management module, a verifiable credential issuance module, a verifiable credential storage module, and an authorization verification module. + +## 7 Functional requirements + +### 7.1 Functional requirements of DLT layer + +The requirements for the DLT layer can be found in [ITU-T F.751.0]. + +- For smart contract transactions, the DLT-based authorization service is required to provide at least two kinds of smart contracts, one for decentralized identifier management and the other for verifiable credential management. + +### **7.2 Functional requirements of authorization base layer** + +#### **7.2.1 Decentralized identifier management** + +Functional requirements of the decentralized identifier management module are as follows: + +- It is required to be equipped with the capability of decentralized identifier registration; +- It is required to have the ability to store decentralized identifier related information; +- It is required to have the capability for decentralized identifier revocation, allowing users to revoke their decentralized identifier. + +#### **7.2.2 Verifiable credential management** + +Functional requirements of the verifiable credential management module are as follows: + +- It is required to provide support for importing verifiable credentials; +- It is required to support verifiable credential querying; +- It is required to provide support for deleting verifiable credentials. + +### **7.3 Functional requirements of authorization service layer** + +Functional requirements of the authorization service layer are as follows: + +- It required to support the authorization duration management; users can check and adjust their authorization duration; +- It required to be able to support users to revoke the authorization; +- It required to have the ability to manage the list of trusted enterprises, so that users can manage the list of trusted enterprises; +- It required to support verifiable credential issuance; +- It required to support verifiable credential storage; +- It required to verify the verifiable credential, with the verification content including the signature of the verifiable credential, the status of the verifiable credential, and other the essential elements; +- A user-friendly interface is recommended to be provided, with operations kept as simple as possible; +- Some new technologies such as zero-knowledge proof (ZKP) and biometric technology are recommended to be used for protecting individual privacy information; +- Identity verification is recommended to be based on a trusted environment, such as trusted execution environment (TEE) or secure element (SE) on the terminal, to ensure the trustworthiness of the identity verification computation process. Generally, individuals use a client device such as a smart phone to interact with the authorization service. It is very important to make sure that the client device can provide a security environment, such as TEE, that is commonly included in smart phone devices. + +## Appendix I + +### Use cases of the DLT-based authorization service system + +(This appendix does not form an integral part of this Recommendation.) + +Large-scale projects often involve multi-party participation, including core institutions and downstream companies, and others. If a core institution wants to expand their business and relate to consumers, downstream companies and independent software vendors may be necessary. + +However, core institutions often do not trust small marginal companies, and it is therefore difficult to establish a trust chain between enterprises and consumers. The DLT-based authorization system can help to solve this problem of trust. The specific procedures are shown in Figure I.1. + +![Diagram of User authorization flow showing interactions between User, Authorization APP, Downstream companies (ISV), Core institutions, and a DLT-based system.](b8661c6c54f72ecc7ff6cb05e47b2891_img.jpg) + +The diagram illustrates the user authorization flow. At the top, a 'User' (represented by a person icon) sends an '1. Apply for service' request to 'Downstream companies' (represented by two house icons, one labeled 'Downstream companies' and the other 'ISV'). The User also interacts with an 'Authorization APP' (represented by a smartphone icon) via '4.1. Confirm' and '4.3. Verifiable credential transfer'. The Authorization APP sends '4.2. Storage of verifiable credential hash' and '3. Initiate authorization request' to the 'DLT-based user self-sovereign authorization system' (represented by a chain of blocks). The Downstream companies send '2. Apply for verifiable credential for user' to the DLT-based system. The DLT-based system sends '5. Verifiable credential transfer and service request' to 'Core institutions' (represented by a building icon). The Core institutions send '6. Verify credential' to the DLT-based system. Finally, the Core institutions send '7. Provide service' to the User. + +``` +graph TD; User((User)) -- "1. Apply for service" --> DC[Downstream companies / ISV]; User <--> "4.1. Confirm" | "4.3. Verifiable credential transfer" --> AA[Authorization APP]; AA -- "4.2. Storage of verifiable credential hash" --> DLT[DLT-based user self-sovereign authorization system]; DLT -- "3. Initiate authorization request" --> AA; DC -- "2. Apply for verifiable credential for user" --> DLT; DLT -- "5. Verifiable credential transfer and service request" --> CI[Core institutions]; CI -- "6. Verify credential" --> DLT; CI -- "7. Provide service" --> User; +``` + +F.751.24(25) + +Diagram of User authorization flow showing interactions between User, Authorization APP, Downstream companies (ISV), Core institutions, and a DLT-based system. + +Figure I.1 – User authorization flow + +Users can authorize downstream companies by issuing verifiable credentials to those companies through an authorization application. Based on verifiable credentials, downstream companies can apply for services from core institutions, thereby establishing a trust chain. Through the authorization system, core institutions could provide services to consumers without the need to trust the downstream companies. + +## Appendix II + +### Use cases of TEE from client device + +(This appendix does not form an integral part of this Recommendation.) + +Security is the foundation of trust in any system. Both issuers and receivers rely on the system to ensure that authorization is unique and secure. Generally, individuals use a client device such as a smartphone to interact with the whole system, it is very important to make sure the client device can provide a security environment. + +Trusted execution environment (TEE) is commonly included in smartphones. The identity information and keys should be stored inside the TEE and there should be an authorization application running in the rich operating system (OS). This authorization application provides the basic interface for issuing, transferring, and revoking verifiable credentials. The digital signature is generated by the trusted application inside the TEE. With the help of the client device, the signed verifiable credential is immutable and can be stored safely. Figure II.1 shows a client device TEE. + +![Diagram of TEE architecture in a client device showing Normal and Trusted execution environments.](7e670a2b556b53ea9002dfff3a420e08_img.jpg) + +The diagram illustrates the architecture of a Trusted Execution Environment (TEE) within an application client device. The device is divided into two main execution environments: the 'Normal application environment' and the 'Trusted execution environment'. The 'Normal application environment' contains an 'Authorization application' and a 'Reading application', both connected to a 'Rich OS'. The 'Trusted execution environment' contains a 'Trusted application' and 'Trusted DRM', both connected to a 'TEE kernel'. Both the 'Rich OS' and the 'TEE kernel' are connected to the 'Hardware platform'. The 'TEE kernel' is also connected to 'Trusted UI storage/NFC/Keys'. + +``` +graph TD + subgraph Device [Application client device] + subgraph Normal [Normal application environment] + App1[Authorization application] + App2[Reading application] + OS[Rich OS] + App1 <--> OS + App2 <--> OS + end + subgraph Trusted [Trusted execution environment] + App3[Trusted application] + App4[Trusted DRM] + Kernel[TEE kernel] + App3 <--> Kernel + App4 <--> Kernel + end + OS <--> Kernel + OS <--> HW[Hardware platform] + Kernel <--> HW + Kernel <--> Storage[Trusted UI storage/NFC/Keys] + end +``` + +Diagram of TEE architecture in a client device showing Normal and Trusted execution environments. + +F.751.24(25) + +Figure II.1 – TEE from client device + +## Appendix III + +### Authorization and verification reference procedures + +(This appendix does not form an integral part of this Recommendation.) + +### III.1 Stockholders + +#### III.1.1 Authorization issuer + +The authorization issuer issues verifiable credentials about a specific subject. Authorization issuers intend to authorize the authorization receiver, and then consume the corresponding service. + +#### III.1.2 Authorization receiver + +The authorization receiver stores credentials on behalf of a subject. Authorization receivers will need the authorization issuer's authorization before providing service. Authorization receivers could be banks, logistics companies, insurance companies, real estate companies, education companies, game companies, etc. + +#### III.1.3 Authorization verification institution + +The authorization verification institution requests a profile of the subject. A profile contains a specific set of credentials. + +The authorization verification institution will check whether the authorization receivers have user's authorization to provide service or use user's information and whether the credentials provided in the profile are fit for purpose. Authorization verification institutions could be third-party trusted organizations, public security organizations, etc. + +### III.2 Authorization and verification procedure + +#### III.2.1 Overview of the procedure + +The procedure of authorization and verification usually includes three stages: a registration stage, an authorization issuance stage and an authorization verification stage, details of each stage are presented in clauses III.2.2 to III.2.4. + +#### III.2.2 Registration stage + +Authorization issuers and authorization receivers all need to perform identity registration before the authorization and verification process. + +It is optional for authorization verification institutions to register an identity, since distributed ledger can provide a verifiable credential query for non-identity participants. + +The identity registration procedure is illustrated in Figure III.1. + +- Authorization issuers and authorization receivers make an identity registration and generation request to the DLT-based authorization service; +- The personal information parameters are sent to the system for review. After the review is passed, the identity document is created and stored and the unique identification is assigned as an identifier. Registration steps are carried out; +- The system feeds back the identity registration information. + +![Diagram of Identity registration flow. A Stockholder icon is at the top. Below it is a horizontal bar representing a Decentralized identity management based on DLT, containing 12 block-like icons. Arrows show the flow: 1. Registration and identity generation request from Stockholder to DLT; 2. Identity document is created and stored from DLT to Stockholder; 3. Feed back the created identity from Stockholder to DLT.](af7916c89a458fdab6c3f443217388ae_img.jpg) + +**Identity registration** + +``` + +graph TD + Stockholder((Stockholder)) -- "1. Registration and identity generation request" --> DLT[Decentralized identity management based on DLT] + DLT -- "2. Identity document is created and stored" --> Stockholder + Stockholder -- "3. Feed back the created identity" --> DLT + +``` + +**Decentralized identity management based on DLT** + +F.751.24(25) + +Diagram of Identity registration flow. A Stockholder icon is at the top. Below it is a horizontal bar representing a Decentralized identity management based on DLT, containing 12 block-like icons. Arrows show the flow: 1. Registration and identity generation request from Stockholder to DLT; 2. Identity document is created and stored from DLT to Stockholder; 3. Feed back the created identity from Stockholder to DLT. + +**Figure III.1 – Identity registration flow** + +#### III.2.3 Authorization issuance stage + +When authorization receivers have an interaction with authorization issuers, they apply for an issuing authorization verifiable credential which contains the authorization issuers' authorization. Authorization issuers confer verifiable credentials and store them in a distributed ledger, then feed it back to the authorization receivers. + +The verifiable credential issuance procedure is illustrated in Figure III.2. + +- The authorization issuer has a business interaction with the authorization receiver, authorization receiver needs authorization from the authorization issuer about some project; +- The authorization receiver applies for authorization form the authorization issuer through the DLT-based authorization service; +- The system will initiate an authorization verifiable credential request to the authorization issuer; +- The authorization issuer verifies the authorization receiver's identity, and issues the authorization credential to the authorization receiver. The verifiable credential is also stored; +- The issued verifiable credential will be fed back to the authorization receiver. + +Authorization issuers can manage a trusted enterprises (authorization receivers) list and authorization duration, and can cancel the authorization within the authorization duration. + +![Diagram illustrating the flow of authorization issuance. It shows an Authorization issuer (person icon) and an Authorization receiver (building icon) interacting via a central DLT system (represented by a row of blocks). The flow consists of five steps: 1. Business interaction (Issuer to Receiver), 2. Apply for issuer to issue authorization verifiable credential (Receiver to DLT), 3. Initiate authorization verifiable credential request (DLT to Issuer), 4. Issue verifiable credential (Issuer to DLT), and 5. Feed back the issued verifiable credential (DLT to Receiver).](4801720824e4b5e2361a5564f91cfb70_img.jpg) + +``` + +graph TD + Issuer[Authorization issuer] -- "1. Business interaction" --> Receiver[Authorization receiver] + Receiver -- "2. Apply for issuer to issue authorization verifiable credential" --> DLT[DLT System] + DLT -- "3. Initiate authorization verifiable credential request" --> Issuer + Issuer -- "4. Issue verifiable credential" --> DLT + DLT -- "5. Feed back the issued verifiable credential" --> Receiver + +``` + +Diagram illustrating the flow of authorization issuance. It shows an Authorization issuer (person icon) and an Authorization receiver (building icon) interacting via a central DLT system (represented by a row of blocks). The flow consists of five steps: 1. Business interaction (Issuer to Receiver), 2. Apply for issuer to issue authorization verifiable credential (Receiver to DLT), 3. Initiate authorization verifiable credential request (DLT to Issuer), 4. Issue verifiable credential (Issuer to DLT), and 5. Feed back the issued verifiable credential (DLT to Receiver). + +F.751.24(25) + +**Figure III.2 – The flow of authorization issuance** + +#### III.2.4 Authorization verification stage + +When a question or dispute about authorization arises, the authorization verification institutions could verify the identity of the authorization issuers and authorization receivers through the DLT system. + +After completing authorization, the DLT system will provide authorization verifiable credential content to help achieve judgement. + +The verification procedure is illustrated in Figure III.3. + +- If a question or dispute regarding authorization arises between the authorization receiver and the authorization issuer, it is necessary to apply for authorization verification to the authorization verification institution; +- Firstly, the authorization verifier makes an identity verification request through the DLT-based authorization service to verify the identity of both parties; +- Then, the authorization verifier verifies the credential and confirms that the credential provided in the profile is fit for purpose; +- Finally, a judgement is made based on the verification results. + +![Diagram of the Authorization verification procedure flow showing interactions between receivers, issuers, a verification institution, and a blockchain.](33ed1f9b27c7c21c797aa928b0f06851_img.jpg) + +``` + +sequenceDiagram + participant R as Authorization receivers + participant I as Authorization issuers + participant V as Authorization verification institution + participant B as Blockchain + + I->>V: 1. Apply for authorization verification + V<<->>B: 2. Identity verification / 3. Authorization verification + V->>R: 4. Judgement according to verifiable credential content + +``` + +The diagram illustrates the authorization verification procedure flow. At the top, 'Authorization receivers' (represented by a building icon) and 'Authorization issuers' (represented by a person icon) are shown. A horizontal line connects them. Below this line, the 'Authorization verification institution' (represented by a building icon) is positioned. At the bottom, a horizontal bar contains a series of linked oval shapes, representing a blockchain. The flow is as follows: + +- An arrow points from the 'Authorization issuers' down to the 'Authorization verification institution', labeled '1. Apply for authorization verification'. +- A double-headed arrow connects the 'Authorization verification institution' down to the blockchain, labeled '2. Identity verification' and '3. Authorization verification'. +- An arrow points from the 'Authorization verification institution' up to the 'Authorization receivers', labeled '4. Judgement according to verifiable credential content'. + + +F.751.24(25) + +Diagram of the Authorization verification procedure flow showing interactions between receivers, issuers, a verification institution, and a blockchain. + +**Figure III.3 – Authorization verification procedure flow** + +## Bibliography + +- [b-ITU-T F.751.1] Recommendation ITU-T F.751.1 (2020), *Assessment criteria for distributed ledger technology platforms.* +- [b-ITU-T X.1403] Recommendation ITU-T X.1403 (2020), *Security guidelines for using distributed ledger technology for decentralized identity management.* + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.751.25-202503-I_PDF-E/053f1077d592e6622cd21dc4bb4cb366_img.jpg b/marked/F/T-REC-F.751.25-202503-I_PDF-E/053f1077d592e6622cd21dc4bb4cb366_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..b8344a495fba6081c6d63dc36ef8969a335e4441 --- /dev/null +++ b/marked/F/T-REC-F.751.25-202503-I_PDF-E/053f1077d592e6622cd21dc4bb4cb366_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f03fc16c56bca90d4ea9ef5ae0b631b14628550722b0b8424beee11de7a266e3 +size 70673 diff --git a/marked/F/T-REC-F.751.25-202503-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.751.25-202503-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c1629394c02f697f60ae954ad36bf274fa8082cb --- /dev/null +++ b/marked/F/T-REC-F.751.25-202503-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3763acd7a0ed56aa30f3f90468c1fb0cb54573eb9045bca3eddf8976b4f0f41d +size 7219 diff --git a/marked/F/T-REC-F.751.25-202503-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.751.25-202503-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a55f7606092b0cf397af47ca6dccb074c6a5dfc4 --- /dev/null +++ b/marked/F/T-REC-F.751.25-202503-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:14f0dfab84b0e12d4ce4e136cb0d44c083ee01f2e758a0b4fd11fc89ca84e7bd +size 71465 diff --git a/marked/F/T-REC-F.751.25-202503-I_PDF-E/raw.md b/marked/F/T-REC-F.751.25-202503-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..ee3cfcb6202b0a718647f7beb654ab057736faa2 --- /dev/null +++ b/marked/F/T-REC-F.751.25-202503-I_PDF-E/raw.md @@ -0,0 +1,436 @@ + + +# Recommendation **ITU-T F.751.25 (03/2025)** + +SERIES F: Non-telephone telecommunication services +Multimedia services + +--- + +## **Framework and requirements for distributed ledger technology-based demand response in smart grid** + +![ITU logo](390120de4fe440c42fea8154fcaad334_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with a white grid pattern, overlaid by the letters 'ITU' in a bold, blue, sans-serif font. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.751.25 + +# Framework and requirements for distributed ledger technology-based demand response in smart grid + +## Summary + +Recommendation ITU-T F.751.25 specifies framework and requirements for distributed ledger technology (DLT)-based demand response (DR) in smart grid, which aims at enhancing the efficiency of incentive mechanisms and automating responses while ensuring trust and accountability. It includes the stakeholders, the framework and the functional requirements of DLT-based DR. The business process of DLT-based DR is also provided in the appendix. This Recommendation can be used as guidance for the development and application of the DLT-based DR for smart grid. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.751.25 | 2025-03-16 | 21 | 11.1002/1000/16258 | + +## Keywords + +Demand response, DLT, smart grid. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +### NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2025 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Stakeholders of DLT-based demand response ..... | 2 | +| 7 Framework of DLT-based demand response..... | 4 | +| 7.1 DLT core layer ..... | 5 | +| 7.2 DR service layer ..... | 5 | +| 8 Functional requirements of DLT-based demand response ..... | 5 | +| 8.1 DLT core layer ..... | 5 | +| 8.2 DR service layer ..... | 6 | +| Appendix I – Business process of DLT-based demand response ..... | 8 | +| I.1 Overview ..... | 8 | +| I.2 DR preparation phase ..... | 8 | +| I.3 DR agreement phase..... | 8 | +| I.4 DR execution phase..... | 9 | +| I.5 DR verification phase ..... | 9 | +| I.6 DR settlement phase ..... | 10 | +| I.7 DR evaluation phase..... | 10 | +| Bibliography..... | 12 | + + + +# Recommendation ITU-T F.751.25 + +## Framework and requirements for distributed ledger technology-based demand response in smart grid + +# 1 Scope + +This Recommendation specifies the framework and requirements for distributed ledger technology (DLT)-based demand response (DR) in smart grid, including: + +- Stakeholders of DLT-based demand response; +- Framework of DLT-based demand response; +- Functional requirements of DLT-based demand response. + +This Recommendation provides guidance for the development and application of the DLT-based demand response for smart grid. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[[ITU-T F.751.0](#)] Recommendation ITU-T F.751.0 (2020), *Requirements for distributed ledger systems*. + +[[ITU-T F.751.2](#)] Recommendation ITU-T F.751.2 (2020), *Reference framework for distributed ledger technologies*. + +[[ITU-T F.751.9](#)] Recommendation ITU-T F.751.9 (2023), *Trusted execution environment based confidential computing on distributed ledger technology systems*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 aggregator** [b-ITU-T Y.2072]: A system in smart grid for collecting the energy generated by residences and micro grids. + +**3.1.2 consensus** [b-ITU-T X.1400]: Agreement that a set of transactions is valid. + +**3.1.3 demand response (DR)** [b-ITU-T Y.2072]: A smart grid feature that allows consumers to reduce or change their electrical use patterns during peak demand, usually in exchange for a financial incentive. Mechanisms and incentives for utilities, business, industrial, and residential customers to cut energy use during times of peak demand or when power reliability is at risk. Demand response is necessary for optimizing the balance of power supply and demand. + +**3.1.4 distributed ledger technology (DLT)** [b-ITU-T X.1400]: Technology that enables the operation and use of distributed ledgers. + +**3.1.5 DLT system** [b-ISO/TC 307]: A system that implements a distributed ledger. + +**3.1.6 off-chain** [b-ISO/TC 307]: Related to a blockchain system, but located, performed or run outside that blockchain system. + +**3.1.7 on-chain** [b-ISO/TC 307]: Located, performed or run inside a blockchain system. + +**3.1.8 smart contract** [b-ITU-T X.1400]: A program written on a distributed ledger system which encodes the rules for specific types of distributed ledger system transactions in a way that can be validated, and triggered by specific conditions. + +## 3.2 Terms defined in this Recommendation + +None. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-----|-------------------------------| +| DLT | Distributed Ledger Technology | +| DR | Demand Response | +| DRA | Demand Response Agreement | +| DRE | Demand Response Execution | +| DRS | Demand Response Settlement | +| DRV | Demand Response Verification | +| TEE | Trusted Execution Environment | + +# 5 Conventions + +In this Recommendation: + +- The keywords "**is required to**" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The keywords "**is recommended**" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement needs not be present to claim conformance. + +# 6 Stakeholders of DLT-based demand response + +DR provides an opportunity for consumers to play a significant role in the operation of the smart grid by reducing or shifting their electricity usage during peak periods in response to time-based rates or other forms of financial incentives. By introducing DLT, traditional DR can be enhanced through improved transparency and traceability of transactions, increased security against fraud and the automation of smart contracts. Additionally, DLT-based DR enables better demand forecasting and load management, leading to more efficient coordination between supply and demand. The stakeholders involved in DLT-based DR are shown in Figure 1. + +![Diagram illustrating the stakeholders of DLT-based demand response. The diagram shows a central 'DLT-based demand response' area surrounded by 'Power grid', 'Power consumers', 'Load aggregators', and a 'Regulatory body'. These stakeholders are connected to a network of 'DLT node' units, each associated with a 'Distributed ledger'. A legend defines the symbols: a stack of papers for 'Distributed ledger', a cylinder for 'DLT node', and a rectangle for 'Stakeholder'.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +The diagram illustrates the stakeholders of DLT-based demand response. At the center is a circle labeled "DLT-based demand response". Surrounding this central area are several "DLT node" units, represented by cylinders. Each cylinder has a small icon of a stack of papers next to it, representing a "Distributed ledger". These nodes are interconnected by dashed lines. Above the central area, a rectangle labeled "Power grid" is connected to one of the nodes. To the left, a stack of rectangles labeled "Power consumers" is connected to two nodes. To the right, a stack of rectangles labeled "Load aggregators" is connected to two nodes. Below the central area, a rectangle labeled "Regulatory body" is connected to one of the nodes. A legend at the bottom left identifies the symbols: a stack of papers for "Distributed ledger", a cylinder for "DLT node", and a rectangle for "Stakeholder". The text "F.751.25(25)" is in the bottom right corner. + +Diagram illustrating the stakeholders of DLT-based demand response. The diagram shows a central 'DLT-based demand response' area surrounded by 'Power grid', 'Power consumers', 'Load aggregators', and a 'Regulatory body'. These stakeholders are connected to a network of 'DLT node' units, each associated with a 'Distributed ledger'. A legend defines the symbols: a stack of papers for 'Distributed ledger', a cylinder for 'DLT node', and a rectangle for 'Stakeholder'. + +**Figure 1 – Stakeholders of DLT-based demand response** + +The DLT-based DR is composed of multiple DLT nodes, each hosting a DLT system and running decentralized applications. These nodes are contributed by various stakeholders, covering: + +- Grid operator: The grid operator, represented by utility companies or power grid companies, plays a central role in the DR process. It is responsible for managing the generation, transmission and distribution of electricity. In the context of DR, the grid operator issues a signal to request changes in electricity consumption from participating consumers or aggregators. It also monitors and evaluates the effectiveness of DR actions in maintaining grid stability and reliability. +- Power consumers: Power consumers represent residential, commercial and industrial entities that consume electricity from the grid. They have the option to participate voluntarily in DR programs to reduce their energy consumption during peak periods or in response to grid signals. Consumers may receive financial incentives or other benefits for participating in DR, such as lower electricity bills or enhanced grid reliability. They play a crucial role in contributing to overall grid flexibility and resilience by adjusting their energy usage in accordance with DR signals. +- Load aggregators: Load aggregators act as intermediaries between power consumers and the power grid, aggregating the electricity demand of multiple consumers to provide more flexible and efficient DR solutions. They often use smart meters and proprietary platforms to monitor and control energy usage across their aggregated portfolio of consumers. Load aggregators facilitate the participation of consumers in DR programs, optimize energy savings and incentives, and ensure compliance with grid requirements. They play a key role in coordinating and managing DR activities to maximize the overall benefits for both consumers and the power grid. +- Regulatory body: Regulatory bodies, such as government agencies or energy commissions, oversee the operation of the power grid and enforce rules and regulations related to energy markets and demand-side management. They establish policies, standards and incentives to encourage the adoption of DR programs and ensure fair and transparent market practices. + +Regulatory bodies may also provide guidelines for the implementation of smart contracts and DLT in DR initiatives to promote efficiency and compliance. + +The DLT-based DR enables stakeholders to collaborate and optimize their interactions through smart contracts: + +- It enables the regulatory body to establish rules, standards, and incentives for DR programs, ensuring fair and transparent participation. +- It enables the power grid to issue signals and instructions for DR actions, improving grid stability and reliability. +- It enables load aggregators to aggregate consumer demand, optimize response actions and manage energy portfolios efficiently, maximizing benefits for both consumers and the power grid. +- It enables power consumers to receive notifications, incentives and instructions for adjusting their energy consumption in response to grid conditions, contributing to grid flexibility and earning rewards. + +Overall, the DLT-based DR streamlines DR processes, enhances transparency and trust among stakeholders, and improves grid efficiency and resilience by using decentralized infrastructure and smart contract automation. + +# 7 Framework of DLT-based demand response + +The framework of DLT-based DR is shown in Figure 2. + +![Figure 2 – Framework of DLT-based demand response. The diagram shows two stacked layers. The top layer is the 'DR service layer' containing six boxes: DR preparation, DR agreement, DR execution, DR verification, DR settlement, and DR evaluation. The bottom layer is the 'DLT core layer' containing six boxes: P2P service, Distributed ledger, Consensus mechanism, Smart contract, Privacy computing, and On-chain and off-chain collaboration.](053f1077d592e6622cd21dc4bb4cb366_img.jpg) + +The diagram illustrates the framework of DLT-based demand response, consisting of two main layers: + +- DR service layer:** This layer contains six functional components arranged in a 2x3 grid: DR preparation, DR agreement, DR execution, DR verification, DR settlement, and DR evaluation. +- DLT core layer:** This layer contains six functional components arranged in a 2x3 grid: P2P service, Distributed ledger, Consensus mechanism, Smart contract, Privacy computing, and On-chain and off-chain collaboration. + +F.751.25(25) + +Figure 2 – Framework of DLT-based demand response. The diagram shows two stacked layers. The top layer is the 'DR service layer' containing six boxes: DR preparation, DR agreement, DR execution, DR verification, DR settlement, and DR evaluation. The bottom layer is the 'DLT core layer' containing six boxes: P2P service, Distributed ledger, Consensus mechanism, Smart contract, Privacy computing, and On-chain and off-chain collaboration. + +**Figure 2 – Framework of DLT-based demand response** + +The DLT-based DR is mainly composed of two layers: + +- **The DLT core layer.** Essential functionalities within this layer include P2P service, consensus mechanism, distributed ledger, privacy computing, smart contract and on-chain and off-chain collaboration. +- **The DR service layer.** Essential functionalities within this layer include user registration, DR agreement, DR execution, DR verification, DR settlement and DR evaluation. + +## 7.1 DLT core layer + +This layer serves as the foundational backbone, providing the necessary infrastructure for secure and efficient operations. It includes: + +- **P2P service:** This functionality enables peer-to-peer interactions among nodes in the network, facilitating direct transactions and communication without the need for intermediaries. +- **Consensus mechanism:** This functionality ensures that all nodes in the network agree on the state of the ledger. It is crucial for maintaining data integrity and preventing double-spending or fraudulent activities. +- **Distributed ledger:** This functionality is the central component of DLT, where all transactions and data are recorded in a secure and immutable manner. It allows for transparency and traceability of energy transactions. +- **Privacy computing:** This functionality protects sensitive data while allowing for its use in computations and transactions. Techniques such as trusted execution environment (TEE) can be employed to ensure privacy while maintaining data utility. +- **Smart contract:** This functionality is self-executing contract with the terms of the agreement directly written into code. Smart contracts automate processes such as executing transactions or triggering DR actions based on predefined conditions. +- **On-chain & off-chain collaboration:** This functionality allows for interactions between on-chain data (data recorded on the chain) and off-chain data (data stored outside the chain). This is important for integrating external data sources into the DR process. + +## 7.2 DR service layer + +This layer focuses on the specific functionalities that support DR activities. It includes: + +- **DR preparation:** This functionality allows participants to register for DR programs and provide necessary information such as energy consumption patterns and preferences. +- **DR agreement (DRA):** This functionality involves creating and managing agreements between the grid operator and participants, outlining the terms and conditions of participation in DR initiatives. +- **DR execution (DRE):** This functionality involves the actual implementation of DR strategies, where participants adjust their energy consumption based on signals or incentives provided by the grid operator. +- **DR verification (DRV):** This functionality ensures that the DR actions taken by participants are accurately recorded and verified against the agreed-upon terms. It may involve real-time monitoring of energy usage. +- **DR settlement (DRS):** This functionality involves the financial transactions related to DR participation, including payments for consumers or aggregators who successfully reduce their energy consumption during peak periods. +- **DR evaluation:** This functionality involves analysing the effectiveness of DR programs, assessing consumer participation and measuring the impact on grid stability and efficiency. + +# 8 Functional requirements of DLT-based demand response + +## 8.1 DLT core layer + +The functional requirements for the DLT core layer are as follows: + +- P2P service, consensus mechanism, smart contract and distributed ledger are required to comply with clauses 7 and 8 of [ITU-T F.751.0] and clause 7 of [ITU-T F.751.2]. + +- Privacy computing is required to adopt the TEE solution for the DLT system and is required to comply with clause 8 of [ITU-T F.751.9]. +- For on-chain & off-chain collaboration: + - It is required to provide the ability to integrate on-chain data with off-chain data to facilitate comprehensive analysis and decision-making. + - It is required to ensure synchronization between on-chain and off-chain data sources to maintain consistency and accuracy. + - It is required to implement secure APIs for accessing off-chain data to make sure that only authorized users and applications can access and interact with off-chain data. + - It is required to provide audit trail of interactions and data exchanges between on-chain and off-chain components for transparency and accountability. + - It is recommended to support interoperability with various off-chain data sources and systems, which ensures that the collaboration functionality can integrate with different platforms, databases or services that provide relevant off-chain data. + +## 8.2 DR service layer + +The functional requirements for the DR service layer are as follows: + +- For DR preparation: + - It is required to ensure the collection of essential user information and relevant information about energy consumption data and equipment specifications. + - It is required to provide a verification process to confirm the identity of new users. + - It is required to ensure compliance with related data protection regulations by implementing appropriate privacy and security measures for user data. +- For DR agreement: + - It is required to provide a standardized template for users to review and accept the terms and conditions of participation in DR programs. + - It is required to ensure that agreements are recorded on-chain for transparency and immutability. + - It is required to provide a mechanism for users to modify or terminate their agreements with appropriate notifications. + - It is required to ensure that users are informed about their rights and responsibilities under the agreement. +- For DR execution: + - It is required to provide automated processes for implementing DR actions based on predefined criteria or triggers. + - It is required to ensure real-time communication between on-chain and off-chain systems to facilitate immediate response actions. + - It is required to provide monitoring capabilities to track the performance of DR actions in real-time. + - It is required to ensure that actions taken are logged for future reference and accountability. +- For DR verification: + - It is required to provide mechanisms for validating the execution of DR actions against agreed-upon criteria. + - It is required to provide users with feedback on the outcomes of their DR participation. + +- It is required to ensure that discrepancies are identified and addressed promptly. +- It is recommended to ensure that independent third-party audits can be conducted to confirm the accuracy of DR performance. + +– For DR settlement: + +- It is required to provide a clear and transparent process for calculating settlements based on DR performance. +- It is required to ensure that financial transactions related to DR are recorded on-chain for transparency. +- It is required to provide users with detailed statements outlining their earnings or savings from participation in DR programs. +- It is required to ensure timely payments to users for their contributions to DR efforts. + +– For DR evaluation: + +- It is required to provide tools for assessing the effectiveness and efficiency of DR programs. +- It is recommended to ensure that user feedback is collected and analysed to improve future DR initiatives. +- It is recommended to provide reports summarizing the outcomes of DR activities, including energy savings and cost reductions. +- It is recommended to ensure that lessons learned are documented and shared with stakeholders for continuous improvement. + +# Appendix I + +## Business process of DLT-based demand response + +(This appendix does not form an integral part of this Recommendation.) + +### I.1 Overview + +The DLT-based DR business process is a streamlined approach designed to optimize energy management by engaging consumers and aggregators in DR programs, facilitated by the seamless collaboration between the DR service layer and the DLT core layer. It consists of six key phases: DR preparation, DR agreement, DR execution, DR verification, DR settlement and DR evaluation. + +- The whole process begins with DR preparation, where users finish registration. The DLT core layer securely stores user information and verifies identities, while the DR service layer provides a user-friendly interface for account creation. +- Next, in the DR agreement phase, standardized agreements are generated and recorded on-chain, ensuring immutability and transparency. The DR service layer facilitates communication, allowing users to review and accept their terms. +- During the DR execution phase, automated DR actions are implemented based on real-time data. The DLT core layer logs all actions while the DR service layer ensures real-time communication between on-chain and off-chain systems. +- Following execution, the performance of actions against agreed criteria is validated in the DR verification phase. The DLT core layer ensures the integrity of an immutable audit trail, while the DR service layer facilitates independent audits on service level. +- Then, financial outcomes are calculated in the DR settlement phase, ensuring all transactions are recorded transparently on the DLT core layer. The DR service layer facilitates timely payments and detailed statements for users. +- Finally, the DR program effectiveness is assessed in the DR evaluation phase using data from the DLT core layer, generating reports on energy savings and user participation. The DR service layer supports feedback collection and documentation of lessons learned for continuous improvement. + +### I.2 DR preparation phase + +The following are the steps for the preparation phase before the DR program starts: + +- **User registration:** Participants, such as power consumers or load aggregators, complete registration to participate in the DR program. This involves providing necessary identification and contact information for identity verification. +- **Information sharing:** Participants share relevant information such as energy consumption data, equipment specifications or availability information. This helps coordinators understand participant capabilities and resources for better DR planning and execution. + +### I.3 DR agreement phase + +In the DR agreement phase of the DLT-based DR process, smart contracts are utilized to establish the contractual framework between the demand-side participants and the program operators. This phase involves the following key steps: + +- **Smart contract development:** Smart contracts are created to encode the terms and conditions of the DRA. These contracts define the rights, responsibilities, and obligations of both parties, including the types of demand response actions permitted, response triggers, compensation mechanisms and dispute resolution procedures. + +- **Agreement formulation:** Demand-side participants engage in negotiations with program operators to define the terms of their involvement in the DR program. Once agreed upon, the terms are encoded into the smart contracts governing the DRA. Negotiation may involve response capabilities, contractual obligations, performance metrics and compensation structures. Agreements may be established through various means, such as: + - Pricing incentives: by providing a secure and transparent platform for adjusting electricity prices based on demand fluctuations and response effectiveness. + - Bidding and auctions: by soliciting bids on the platforms, allowing participants to submit offers based on their capabilities, ensuring fairness and efficiency in selecting the most suitable response solution. +- **Contract deployment:** Upon formulation, the smart contracts representing the DRA are deployed on the distributed ledger network. Digital signatures from all parties involved are required to execute the contracts, providing cryptographic proof of agreement and ensuring the authenticity and integrity of the contract execution process. + +### I.4 DR execution phase + +In the DR execution phase of the DLT-based DR process, smart contracts facilitate the execution of DR actions and the submission of response data by demand-side participants. This phase involves the following key steps: + +- **Response triggering:** When predefined conditions are met (e.g., high electricity demand, grid instability), smart contracts automatically trigger DR actions. These actions may include load curtailment, demand shifting or distributed energy resource utilization, depending on the capabilities and preferences of the demand-side participants. +- **Data submission:** Following the execution of DR actions, demand-side participants submit response data to the smart contracts deployed on the distributed ledger network. This data includes information on the magnitude, duration and timing of the response, as well as any other relevant metrics specified in the DRE contracts. It is also recommended to obtain the responsive metering data directly from another reliable sources. Refer to [b-ITU-T F.751.18] for more information on how the trustworthy metering data are recorded, transferred and shared over DLT-based energy metering data sharing. +- **Real-time monitoring:** Smart contracts monitor the incoming response data in real-time, validating its authenticity and accuracy against predefined criteria specified in the DRE contracts. This includes verifying that the response occurred within the designated time frames, met the required thresholds and complied with any contractual obligations. +- **Event logging:** All interactions and transactions related to DRE are logged and recorded on the distributed ledger, creating an immutable audit trail of events. This ensures transparency, traceability and accountability in the execution process, facilitating post-event analysis and auditing. + +### I.5 DR verification phase + +After the execution phase, where the demand-side participant submits response data to the smart contract, the process enters the verification phase. This phase is essential for validating the authenticity and accuracy of the reported DR actions against the actual grid conditions and the requirements specified in the DRA and DRE contracts. This phase involves the following key steps: + +- **Data validation:** The reported response data is analysed and compared against historical or real-time grid data. This ensures that the DR actions were executed as claimed and had the intended impact on the grid's demand-supply balance. + +- **Compliance checking:** The verification process includes a compliance check to ensure that the participant's actions align with the terms outlined in the DRA and DRE contracts. This includes verifying that the response met the required thresholds, occurred within the specified time frames and adhered to any other contractual obligations. +- **Anomaly detection:** Advanced analytics can be employed to detect anomalies or potential fraudulent activities. By analysing patterns and historical data, it can identify discrepancies that may indicate non-compliance or manipulation of the DR process. +- **Report generation:** Upon completing the verification process, a detailed report is generated and recorded on the chain. This report outlines the verification outcomes, including compliance status, discrepancies identified (if any) and recommendations for any corrective actions. +- **Smart contract triggering:** Based on the verification report, the DRV smart contract is triggered. If the verification is successful, the contract proceeds with the initiation of the settlement process. Conversely, if discrepancies or non-compliance issues are identified, the DRV contract may trigger dispute resolution mechanisms or adjust the settlement terms accordingly. + +## I.6 DR settlement phase + +Transitioning to the settlement phase, the DRS orchestrates the reconciliation of payment details between the power grid and the participant, aligning with the terms outlined in the executed DRA, DRE and DRV contracts. This phase involves the following key steps: + +- **Verification outcome integration:** Before initiating the settlement process, the DRS contract integrates the outcomes of the verification process conducted by the DRV contract. This ensures that only validated and compliant DR actions proceed to settlement, safeguarding against inaccuracies or discrepancies. +- **Payment processing:** Payment amounts, scheduled payment dates and other pertinent settlement particulars are meticulously processed and documented by the DRS contract, taking into account the verified DR actions. Payments are initiated by the power grid in accordance with the agreed-upon terms and conditions stipulated in the DRA and DRE contracts. +- **Transaction record maintenance:** Following the completion of the settlement process, comprehensive transaction records are securely stored and timestamped on the chain. These records provide transparency and enable audit trails for all parties involved, ensuring accountability and facilitating dispute resolution if necessary. + +## I.7 DR evaluation phase + +In the evaluation phase of the DLT-based DR process, the focus is on assessing the effectiveness of the DR to determine if it meets the intended objectives and to provide guidance for future improvements. This phase involves the following key steps: + +- **Data collection:** Gather data relevant to the operation of the DR, including energy market prices and response data, etc. Metrics related to the DLT system, including transaction records, smart contract execution logs, participant interactions and system performance are also collected and analysed. +- **Effectiveness assessment:** Utilizing the collected data, assess the effectiveness of the DLT-based DR in achieving its goals. This may involve cost–benefit analysis of DR actions and assessment of response capabilities. Evaluate factors such as the accuracy of smart contract execution, the timeliness of response triggering, the reliability of data submission and the overall impact on energy demand–supply balance are also considered. + +- **Improvement suggestions:** Based on the evaluation results, propose suggestions and optimization strategies for improving the DR program. This may involve refining smart contract logic, optimizing consensus algorithms, improving user interfaces or enhancing data privacy features, etc. + +# Bibliography + +- [b-ITU-T F.751.18] Recommendation ITU-T F.751.18 (2024), *Framework for distributed ledger technology (DLT)-based energy metering data sharing*. +- [b-ITU-T X.1400] Recommendation ITU-T X.1400 (2020), *Terms and definitions for distributed ledger technology*. +- [b-ITU-T Y.2072] Recommendation ITU-T Y.2072 (2018), *Framework for an energy-sharing and trading platform*. +- [b-ISO/TC 307] ISO DIS 22739, *Blockchain and distributed ledger technologies – Terminology*. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.751.26-202503-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/F/T-REC-F.751.26-202503-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c1629394c02f697f60ae954ad36bf274fa8082cb --- /dev/null +++ b/marked/F/T-REC-F.751.26-202503-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3763acd7a0ed56aa30f3f90468c1fb0cb54573eb9045bca3eddf8976b4f0f41d +size 7219 diff --git a/marked/F/T-REC-F.751.26-202503-I_PDF-E/daa4a6fa7e2ba1954258f86b4928eb32_img.jpg b/marked/F/T-REC-F.751.26-202503-I_PDF-E/daa4a6fa7e2ba1954258f86b4928eb32_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..76d11ffc799abc627d4abcff24519c9ecc1f430d --- /dev/null +++ b/marked/F/T-REC-F.751.26-202503-I_PDF-E/daa4a6fa7e2ba1954258f86b4928eb32_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:83d3c0d9f0a150dcd2d128823e9e00da8f0cdf9d40e922ca9aa28ae9a2cab481 +size 76414 diff --git a/marked/F/T-REC-F.751.26-202503-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.751.26-202503-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c04d6d07cefffdf7c1d82e3aaad3860a7684fe61 --- /dev/null +++ b/marked/F/T-REC-F.751.26-202503-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:96ed997ab67b3c3e17c6134be3332be460b7eb7bbfd1a2408b343df9fd0ef566 +size 71574 diff --git a/marked/F/T-REC-F.751.27-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.751.27-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..30d4bc9edd02d463a452fa930b82ff2dfd368c24 --- /dev/null +++ b/marked/F/T-REC-F.751.27-202503-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8bdea22ddbf01b83f16dbc58bc0c9e4b409b0baae46bf6dede68381ee2f7df83 +size 7130 diff --git a/marked/F/T-REC-F.751.27-202503-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.751.27-202503-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7bf293063065a4b77053b3dc7a227faf427b84e3 --- /dev/null +++ b/marked/F/T-REC-F.751.27-202503-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c0375bcaa043247d8a1421399f6e99cc81ae7ef55133a3305756627c9ed0aea2 +size 165255 diff --git a/marked/F/T-REC-F.751.28-202512-I_PDF-E/1033ab90cd19fd4c5901966aeec07a86_img.jpg b/marked/F/T-REC-F.751.28-202512-I_PDF-E/1033ab90cd19fd4c5901966aeec07a86_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a52777915da87079d6d5294ca7a136ae6383f057 --- /dev/null +++ b/marked/F/T-REC-F.751.28-202512-I_PDF-E/1033ab90cd19fd4c5901966aeec07a86_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9175bb5c66da27d6eef06c96302ea1153b45999a8428997a7d57e7a1c42dd81c +size 26686 diff --git a/marked/F/T-REC-F.751.28-202512-I_PDF-E/1439cb942d9e363bbb3161b5540dd8c6_img.jpg b/marked/F/T-REC-F.751.28-202512-I_PDF-E/1439cb942d9e363bbb3161b5540dd8c6_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..6d7fb0a32fc83d0cc3b48f411c701159727960e7 --- /dev/null +++ b/marked/F/T-REC-F.751.28-202512-I_PDF-E/1439cb942d9e363bbb3161b5540dd8c6_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c957e84a3a6ad3d62528d4519fe208de04a388749063ac5ca779c9ece3be103c +size 65540 diff --git a/marked/F/T-REC-F.751.28-202512-I_PDF-E/18442e4e239480f0c3c95b547aa8fde2_img.jpg b/marked/F/T-REC-F.751.28-202512-I_PDF-E/18442e4e239480f0c3c95b547aa8fde2_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..b40a05f4d48d2b47a360d8a5311e8477a32ed0f5 --- /dev/null +++ b/marked/F/T-REC-F.751.28-202512-I_PDF-E/18442e4e239480f0c3c95b547aa8fde2_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0d4422232d90a59a8acda30ea5612d972c094af9d95f501b36b80d967c28d7cb +size 19909 diff --git a/marked/F/T-REC-F.751.28-202512-I_PDF-E/2a23751710f7827065d2b99b6df588df_img.jpg b/marked/F/T-REC-F.751.28-202512-I_PDF-E/2a23751710f7827065d2b99b6df588df_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..807439be920495166d0039d4de75319cefc52044 --- /dev/null +++ b/marked/F/T-REC-F.751.28-202512-I_PDF-E/2a23751710f7827065d2b99b6df588df_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5db851ae76b06a3cb1d353806af7b04f0db4bf9b591b46e2b095764703a6b209 +size 25412 diff --git a/marked/F/T-REC-F.751.28-202512-I_PDF-E/33a8f3f01dfa8bce75d23017855a13c5_img.jpg b/marked/F/T-REC-F.751.28-202512-I_PDF-E/33a8f3f01dfa8bce75d23017855a13c5_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..4748c478a48fefec65170954a47760424209cba0 --- /dev/null +++ b/marked/F/T-REC-F.751.28-202512-I_PDF-E/33a8f3f01dfa8bce75d23017855a13c5_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:781adca338f4044f543806f417c7bd0f174291df3986c70a8cf481dae78082a6 +size 103399 diff --git a/marked/F/T-REC-F.751.28-202512-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg b/marked/F/T-REC-F.751.28-202512-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..14523e2841440c1e9067ddf8e15bc677afaa51d2 --- /dev/null +++ b/marked/F/T-REC-F.751.28-202512-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5498be52f16c6457144a80955b302569607b14726677740cf4ca56c5a71c55a0 +size 20932 diff --git a/marked/F/T-REC-F.751.28-202512-I_PDF-E/392a79ccd95e682ccd08f35ab2e64144_img.jpg b/marked/F/T-REC-F.751.28-202512-I_PDF-E/392a79ccd95e682ccd08f35ab2e64144_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..e7e611f60eb80a6c13d6da9327616f6424d36a52 --- /dev/null +++ b/marked/F/T-REC-F.751.28-202512-I_PDF-E/392a79ccd95e682ccd08f35ab2e64144_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6c237b8007b9062a8a940296dfc0e38a8c67c7cd9851118e2b404099a1f3de07 +size 8470 diff --git a/marked/F/T-REC-F.751.28-202512-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg b/marked/F/T-REC-F.751.28-202512-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d1a12974506486dcfc7945b90616d7c27df3fdb8 --- /dev/null +++ b/marked/F/T-REC-F.751.28-202512-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:932ce51b9224a6630cf928e79722bc435dfe016f6a188be8715af3be7c6df187 +size 31520 diff --git a/marked/F/T-REC-F.751.28-202512-I_PDF-E/7133ccf78043568ca62ecbcd43628a4a_img.jpg b/marked/F/T-REC-F.751.28-202512-I_PDF-E/7133ccf78043568ca62ecbcd43628a4a_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1e233af92b2af178ffeba44139ed56cd585abd8c --- /dev/null +++ b/marked/F/T-REC-F.751.28-202512-I_PDF-E/7133ccf78043568ca62ecbcd43628a4a_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fa607757fa6332fa723630fac46b5ddc13a603a9cb52d2aa1bbf69d1ce1ab1d3 +size 57128 diff --git a/marked/F/T-REC-F.751.28-202512-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.751.28-202512-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d4d8969f805414689ab79cfd08c333d454d482b3 --- /dev/null +++ b/marked/F/T-REC-F.751.28-202512-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:636c3839edff955a5b6208530b05fa1c983ab153d90f00216517a7a99478b988 +size 7074 diff --git a/marked/F/T-REC-F.751.28-202512-I_PDF-E/85b10db535b22e64c7d37f362705ddbf_img.jpg b/marked/F/T-REC-F.751.28-202512-I_PDF-E/85b10db535b22e64c7d37f362705ddbf_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..74ea40a063153fbaada64ba14898a2c1edc7839b --- /dev/null +++ b/marked/F/T-REC-F.751.28-202512-I_PDF-E/85b10db535b22e64c7d37f362705ddbf_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:90884518d5a9d604e1f712690a2d5ac969df3255513f78d82a15e3ebf345c9ce +size 28181 diff --git a/marked/F/T-REC-F.751.28-202512-I_PDF-E/9b686adccf125267a013fa25721231a3_img.jpg b/marked/F/T-REC-F.751.28-202512-I_PDF-E/9b686adccf125267a013fa25721231a3_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..915f69852e1ca3c4277301b29c92693b793120f5 --- /dev/null +++ b/marked/F/T-REC-F.751.28-202512-I_PDF-E/9b686adccf125267a013fa25721231a3_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f98e0e71a126c95698cc50807e04ee199473a1c62498afdf041bab320b1ac63a +size 16429 diff --git a/marked/F/T-REC-F.751.28-202512-I_PDF-E/a734898ce18e972938949637c32a34f4_img.jpg b/marked/F/T-REC-F.751.28-202512-I_PDF-E/a734898ce18e972938949637c32a34f4_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..20987c82ed502eeece119850b153dd566de60b0a --- /dev/null +++ b/marked/F/T-REC-F.751.28-202512-I_PDF-E/a734898ce18e972938949637c32a34f4_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0c74531e78c55d2227484633b9c1dc2b99d92fce3973aaa772af370a4044802d +size 40189 diff --git a/marked/F/T-REC-F.751.28-202512-I_PDF-E/ab846b81e78dbc8da2a6f9511e2f248a_img.jpg b/marked/F/T-REC-F.751.28-202512-I_PDF-E/ab846b81e78dbc8da2a6f9511e2f248a_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..5be2b00a6ebd1589d32b29d23d769cab92d60d5f --- /dev/null +++ b/marked/F/T-REC-F.751.28-202512-I_PDF-E/ab846b81e78dbc8da2a6f9511e2f248a_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e90ed57bc3ea8130987c836473d6136bf52d27a5f058b7c2628be9a36f28a1d4 +size 85379 diff --git a/marked/F/T-REC-F.751.28-202512-I_PDF-E/b8661c6c54f72ecc7ff6cb05e47b2891_img.jpg b/marked/F/T-REC-F.751.28-202512-I_PDF-E/b8661c6c54f72ecc7ff6cb05e47b2891_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f7a0d01c3e7ad9b9367439b2a8629cf029eb2fa6 --- /dev/null +++ b/marked/F/T-REC-F.751.28-202512-I_PDF-E/b8661c6c54f72ecc7ff6cb05e47b2891_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2141e3e956f8e4b85d68ee00a5b96566e2ea28635f3032d5cc8163564dad83c7 +size 37821 diff --git a/marked/F/T-REC-F.751.28-202512-I_PDF-E/d6a5f8423065d7cef55e11a0007014d9_img.jpg b/marked/F/T-REC-F.751.28-202512-I_PDF-E/d6a5f8423065d7cef55e11a0007014d9_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..8d93b8c356315d85c45eb9a45fcbbb0fb1bc02d5 --- /dev/null +++ b/marked/F/T-REC-F.751.28-202512-I_PDF-E/d6a5f8423065d7cef55e11a0007014d9_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ce2a25f0b2f1c3cdcc758b9cb4996bfed10c30231ecf1503c79e08cbfce917c6 +size 19642 diff --git a/marked/F/T-REC-F.751.28-202512-I_PDF-E/e69b9188aa2c14ec6b21c83f711fef65_img.jpg b/marked/F/T-REC-F.751.28-202512-I_PDF-E/e69b9188aa2c14ec6b21c83f711fef65_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..03185a58fe0916b068170cf0f463043f604735ed --- /dev/null +++ b/marked/F/T-REC-F.751.28-202512-I_PDF-E/e69b9188aa2c14ec6b21c83f711fef65_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a196ed8a5d70c0bcc5f4127ebab38c0f963355eb092a686cac48cf3150e52efd +size 106849 diff --git a/marked/F/T-REC-F.751.28-202512-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.751.28-202512-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..50c6661a2594c17ab2331aafdb8d4444a6d4162c --- /dev/null +++ b/marked/F/T-REC-F.751.28-202512-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0816020b9d273c33f256e42694d67a7761f13c558923c011fc481171c675ce27 +size 38915 diff --git a/marked/F/T-REC-F.751.28-202512-I_PDF-E/fa859e4e468bfb2710a94527f2c504af_img.jpg b/marked/F/T-REC-F.751.28-202512-I_PDF-E/fa859e4e468bfb2710a94527f2c504af_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3eb62498bb1ca5f7a615a0811358d1e9e2308563 --- /dev/null +++ b/marked/F/T-REC-F.751.28-202512-I_PDF-E/fa859e4e468bfb2710a94527f2c504af_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f58436893affd6f1b04d4466b423d115ed5accd51e9137be9681b7412139d23c +size 38093 diff --git a/marked/F/T-REC-F.751.3-202203-I_PDF-E/raw.md b/marked/F/T-REC-F.751.3-202203-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..65ab4d57817382dbd11f8a17cd4a1efddef3bbe6 --- /dev/null +++ b/marked/F/T-REC-F.751.3-202203-I_PDF-E/raw.md @@ -0,0 +1,346 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.751.3** + +(03/2022) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**Requirements for change management in +distributed ledger technology (DLT)-based +decentralized applications** + +Recommendation ITU-T F.751.3 + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +### **TELEGRAPH SERVICE** + +| | | +|-----------------------------------------------------------------|-------------| +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | + +### **MOBILE SERVICE** + +| | | +|--------------------------------------------------------|-------------| +| Mobile services and multdestination satellite services | F.110–F.159 | +|--------------------------------------------------------|-------------| + +## **TELEMATIC SERVICES** + +| | | +|-------------------------------------------|-------------| +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | + +## **MESSAGE HANDLING SERVICES** + +F.400–F.499 + +## **DIRECTORY SERVICES** + +F.500–F.549 + +### **DOCUMENT COMMUNICATION** + +| | | +|--------------------------------------|-------------| +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | + +### **DATA TRANSMISSION SERVICES** + +F.600–F.699 + +## **MULTIMEDIA SERVICES** + +**F.700–F.799** + +## **ISDN SERVICES** + +F.800–F.849 + +## **UNIVERSAL PERSONAL TELECOMMUNICATION** + +F.850–F.899 + +## **ACCESSIBILITY AND HUMAN FACTORS** + +F.900–F.999 + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.751.3 + +# Requirements for change management in distributed ledger technology (DLT)-based decentralized applications + +## Summary + +The development of applications using distributed ledger technology (DLT) enables the creation of new business models in various sectors of the economy and it has the potential to tackle, on a large scale, important challenges for our society, due to its ability to increase trust in the relationship between stakeholders. Technical immutability is key to building trust between stakeholders. On the other hand, real life introduces practical needs to update applications with smart contracts. This document defines some recommendations to tackle changes in applications using smart contracts. + +The discussion of whether DLT networks provide different levels of technical immutability is out of the scope of this Recommendation. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.751.3 | 2022-03-16 | 16 | 11.1002/1000/14965 | + +## Keywords + +Blockchain, change management, distributed ledger technologies. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2022 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|----------------------------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Background..... | 2 | +| 7 Technical requirements for change management in DLT-based decentralized applications ..... | 3 | +| 7.1 Facilitating the evolution of an application that uses smart contracts ..... | 3 | +| 7.2 Providing stakeholders with trust in the change management process ..... | 4 | +| 7.3 Ensuring that the application governance works appropriately ..... | 5 | +| Bibliography..... | 6 | + + + +# Recommendation ITU-T F.751.3 + +# Requirements for change management in distributed ledger technology (DLT)-based decentralized applications + +## 1 Scope + +This Recommendation is intended to propose requirements for change management in distributed ledger technology (DLT)-based decentralized applications considering the need for changes in the real-world environment (e.g., changes in user requirements or the need to adapt to new regulation) where the application is running or an error correction is in the code. Although a change management framework may be useful for the decentralized application as a whole, this Recommendation is focused on changes that impact deployed smart contracts. Also, the change management described aims to approach unanticipated changes, including data, data structure and business logic. This document does not include lifecycle management activities of DLT-based applications. So, it is not discussing topics like auditing, bug bounties, tests, etc. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following term defined elsewhere: + +**3.1.1 smart contract** [b-ITU-T X.1400]: A program written on a distributed ledger system which encodes the rules for specific types of distributed ledger system transactions in a way that can be validated, and triggered by specific conditions. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 application governance**: A subset of the stakeholders that act to ensure that the decentralized application code does exactly what it should. + +**3.2.2 current smart contracts**: Smart contracts that contain lines of code or data that motivated a change in an application that uses smart contracts. + +**3.2.3 change script**: A script that specifies a change in an application that uses smart contracts, e.g., deploy new contracts and change data of current smart contracts. + +**3.2.4 future smart contracts**: Smart contracts deployed or impacted in some way during the execution of a change in an application that uses smart contracts. + +**3.2.5 stakeholder**: Someone with an interest or concern in the decentralized application. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +DeFi Decentralized Finance + +DLT Distributed Ledger Technology + +## 5 Conventions + +In this Recommendation: + +The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. + +The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. + +## 6 Background + +The development of applications using DLT enables the creation of new business models in various sectors of the economy and it has the potential to tackle, on a large scale, important challenges for our society, due to its ability to increase trust in the relationship between stakeholders. Provided that there are no successful attacks or collusion between nodes, a DLT network, decentralized and synchronized by a consensus algorithm, will grant an immutable software execution environment. It is guaranteed to the stakeholders that the execution of the on-chain code will always correspond to what was specified in the smart contract code. As Kevin Werbach [b-Wer] discusses in his article, immutability is key to build trust between stakeholders, which rules out the need to involve centralized authorities, yet it does also increase relationship issues among these stakeholders with no easy conflict resolution mechanism. + +Technology alone cannot make sure that the code of the smart contract will really reflect what should be executed. For instance, an error in the code may lead its execution to mismatch the requirements that originated it. A change in the agreement of the stakeholders can also occur, for whatever reason, which may render it impossible to adapt the smart contract to the new reality. + +Application governance is what ensures that the decentralized application code does exactly what it should. The act of deploying one or more smart contracts is associated with an application governance decision. Any change in a smart contract is also a change in the original decision on how to code the smart contract and must therefore represent a new decision of the same group. + +Since the applications depend on the network to execute, the correct execution of the smart contracts also depends on the network governance. In some cases, especially when the permissioned networks are created to serve a specific purpose, the network's governance members can be the same members of the application governance, but this is merely a coincidence and does not necessarily occur all the time. + +Actions and decisions of the governance quite often occur off-chain. The stakeholders that possess authority jointly agree on a scope that must be built and they quite often hire or delegate to an executor the development and the deployment of the smart contract. In practical terms, this delegation is rooted in the application governance's trust towards the executor. + +Considering a more mature setting, application governance can occur through an automatized change management process in the DLT network itself. There are many ways to involve the members of the application governance and there are plenty of examples in both permissioned and permission-less DLT networks in how to manage group decisions. Some examples are smart contracts that require multiple signatures (multisig) to execute a transaction, voting mechanisms and governance token distribution (as common in some decentralized finance (DeFi) platforms). + +This idea of having an on-chain change management process allows minimizing trust in the change executor and sharing decision-making responsibilities. In this kind of setting, less involved stakeholders could trust that the changes in the smart contract keep the original high-level decisions that led to the development of the smart contract. + +Note that a change management solution does not necessarily imply a native support in the DLT platform, since it can also be implemented using smart contracts. In addition, a change management solution does not harm the technical immutability of DLT. It involves a set of trackable procedures applied in a specific period of time to change a DLT-based decentralized application. + +The use of a change management framework in DLT-based decentralized applications may cause relevant impact in the overall solution. For example, the framework may demand an additional amount of computational resources to run the smart contract code, may increase complexity of smart contract code or even avoid the use of existing smart contract standards. In order to decide when to use a change management framework in a specific project, it is necessary to consider the cost and benefits involved. + +## 7 Technical requirements for change management in DLT-based decentralized applications + +This Recommendation proposes three high-level requirements for an application change process as follows. All of them should be met through on-chain development, to guarantee the predictability and trust of the process. + +These requirements were grouped by three layers. The picture below presents the name and main goal of each layer. The bottom layer is very technical, related to software engineer requirements to solve the technical issue. The layers above address the relationship with the stakeholders and the application governance. All layers are detailed in the following clauses. + +![Diagram showing three layers of requirements for change management in decentralized applications, stacked vertically. The top layer is Governance, the middle is Transparency, and the bottom is Upgradeability.](dbe553cf16dd14073b89a8263a428664_img.jpg) + +| | | +|----|-------------------------------------------------------------------------------------------------| +| 3) | Governance
Ensuring that the application governance works appropriately | +| 2) | Transparency
Providing stakeholders with trust in the change management process | +| 1) | Upgradeability
Facilitating the evolution of an application that uses smart contracts | + +F.751.3(22) + +Diagram showing three layers of requirements for change management in decentralized applications, stacked vertically. The top layer is Governance, the middle is Transparency, and the bottom is Upgradeability. + +Figure 1 – The layers of requirements for change management in decentralized applications + +### 7.1 Facilitating the evolution of an application that uses smart contracts + +This requirement comprises correction of code error, the update of some business requirements, or the change of the state variables of the contracts. + +To fulfil this requirement, a change management framework: + +- is required to preserve the access to the data originally used by the current smart contracts to the future smart contracts, as a way to minimize data migration; +- in the context of a change, is required to make it feasible to alter data used by the current smart contracts in a way that is not possible without a change; + +- c) in the context of a change, is required to make it feasible to alter the data structure used by the current smart contracts in a way that is not possible without a change; +- d) is recommended to expose an immutable way of finding the current version of the smart contracts (sometimes called "proxy contract"); +- e) is recommended to enable the possibility of preserving some parts of the code from the current smart contracts as non-upgradeable; +- f) is recommended to preserve the quality of the smart contracts codes throughout time, even after various evolutions1; +- g) in the context of a change, is recommended to enable the application owners to make available a choice for application users to opt between staying managed by the current smart contracts, migrating to the future smart contracts or even finishing the relationship with the smart contracts (if the user can choose, it is known as "you-are-free-to-opt-out" principle). + +Note that according to the intrinsic characteristics of a DLT network, data stored in transactions for the network are not to be altered. It is only possible to alter the current state of the data stored in the variables of the smart contracts. Thus, the development of this item requirement is not able to solve the issue of the right to be forgotten present in privacy laws. + +### 7.2 Providing stakeholders with trust in the change management process + +Besides meeting the first requirement, a framework that meets this requirement grants for all stakeholders the trust that the process of change management is respected. The stakeholders include the ones that do not take part in the change process themselves, but that are interested in it. + +A change management framework that conforms to this requirement: + +- a) is recommended to establish that changes go through a life cycle, which typically includes proposal, approval, execution and conclusion or cancelling; +- b) is recommended to associate the change with off-chain information that spells out its motivation and the rationale adopted for the solution; +- c) is recommended to guarantee that every change is executed through an automatized change script; +- d) is recommended to make a change proposal (including change script) available for analysis of the application governance before the change approval; +- e) is required to prevent the execution of changes that do not go through the change process approval2; +- f) is required to provide transparency regarding which changes were proposed and what happened to these propositions; +- g) is recommended to work out a system to monitor the changes in progress; +- h) is recommended to enable the introduction of a time gap between the proposal and the approval or between the approval and the execution of the change, which would allow the users to take some time to decide what to do regarding the imminent application change as well as allow the development team to create an additional safety mechanism. + +--- + +1 This is more difficult to achieve in some cases, for example, when the upgrade mechanism is implemented as an independent layer of smart contracts. There are some solutions that include specific restrictions in the current and future smart contract code due to the way that the underlying DLT is implemented (for example, how data is stored, the difference of constructor and methods, etc) [b-Pal]. + +2 One possible way to achieve this is including the entire changing script on-chain before the change approval. Another approach is making the hash of the changing script on-chain before the change approval, in a common practice called commit and reveal [b-Pal]. + +Ideally, one should start discussing the change proposition in an off-chain way, to facilitate the debate and the evaluation of the impacts. Only when the members agree on what must be proposed should a change be proposed by using the process. To associate the change with the off-chain information (as described in item 'b'), it is possible, for instance, to register in the DLT the hash that documents the motivation, discussion and evaluation of the impact of the change and, possibly, the rationale adopted for the solution. A way to make the monitoring of the changes possible is (as described in item 'f') to emit an event for each modification of change state. + +### **7.3 Ensuring that the application governance works appropriately** + +This requirement aims to make sure that the correct set of stakeholders participate in the application governance and that they agree on the conformity of the smart contract. A development that meets the three requirements of this Recommendation must then have the concept of governance structure automatized, and it must be capable of deciding on changes to share the responsibility for the change. + +A change management framework that conforms to this requirement: + +- a) is recommended to include a mechanism to aid the group's decision-making process; +- b) is recommended to provide a mechanism for the creation of subgroups and delegation of votes to decide on specific changes; +- c) is recommended to support the classification of changes in different levels of formality regarding deployment, which will depend on how high the change impact will be and how urgent it needs to be implemented; +- d) is recommended to support the inclusion or elimination of members of the application governance; +- e) is recommended to enable the participation of the governance structure since the very first deployment of the smart contracts. + +## Bibliography + +- [b-ITU-T X.1400] Recommendation ITU-T X.1400 (2020), *Terms and definitions for distributed ledger technology*. +- [b-Pal] Paladino, S. (2020), *The State of Smart Contract Upgrades*, OpenZeppelin. +<, last visited on July 2021> +- [b-Wer] Werbach, K. (2018), *The Siren Song: Algorithmic Governance By Blockchain*, in *After the Digital Tornado: Networks, Algorithms, Humanity*. Available at SSRN. +<, last visited on July 2021> + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.751.31-202512-I_PDF-E/410562339ce067fdc6fa41940c118658_img.jpg b/marked/F/T-REC-F.751.31-202512-I_PDF-E/410562339ce067fdc6fa41940c118658_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..fabe6d820a9707dd0dd4f7816375b0045182dc28 --- /dev/null +++ b/marked/F/T-REC-F.751.31-202512-I_PDF-E/410562339ce067fdc6fa41940c118658_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a1e64e35421af14841e1b362d626cba3ad1f3e77007a0bd0e4500c5fd1499a1e +size 37482 diff --git a/marked/F/T-REC-F.751.31-202512-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.751.31-202512-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ac295b1b0c7f14b1c5a33ef6682b1b017c812477 --- /dev/null +++ b/marked/F/T-REC-F.751.31-202512-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37aed4fc114a39843438175326e0b7e1d09a9d6075723d82fea32040fd560ee4 +size 7192 diff --git a/marked/F/T-REC-F.751.31-202512-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/F/T-REC-F.751.31-202512-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..8ea2f0135721085f214196320241df095a6ae792 --- /dev/null +++ b/marked/F/T-REC-F.751.31-202512-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2ea0a62db238631c7facac3cf6b6436464e736670ec0cfe628c4bc09c65ee22b +size 70784 diff --git a/marked/F/T-REC-F.751.31-202512-I_PDF-E/f4fdd410cdb84df81274da55721e56fb_img.jpg b/marked/F/T-REC-F.751.31-202512-I_PDF-E/f4fdd410cdb84df81274da55721e56fb_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7669f04bf637e93ca5e8f3e3280a99780dc78d75 --- /dev/null +++ b/marked/F/T-REC-F.751.31-202512-I_PDF-E/f4fdd410cdb84df81274da55721e56fb_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aabf6311d3bc2bfe33699efb608126bdedbebf3ec56121a6575c97493c9545b2 +size 69932 diff --git a/marked/F/T-REC-F.751.31-202512-I_PDF-E/raw.md b/marked/F/T-REC-F.751.31-202512-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..dc90b7a5175bdf51250aeb3419ab799d68e5d90d --- /dev/null +++ b/marked/F/T-REC-F.751.31-202512-I_PDF-E/raw.md @@ -0,0 +1,339 @@ + + +Recommendation + +**ITU-T F.751.31 (12/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +**Reference framework of distributed ledger +technology-based cross-domain settlement for +electric vehicle public charging** + +![ITU logo](84a1d09fb489061482111515543b60dc_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +# ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.751.31 + +# Reference framework of distributed ledger technology-based cross-domain settlement for electric vehicle public charging + +## Summary + +Recommendation ITU-T F.751.31 specifies a reference framework of distributed ledger technology (DLT)-based cross-domain settlement to deliver efficient, a unified public charging settlement for electric vehicles among diverse public charging operators. The Recommendation includes stakeholders, framework and functional requirements of DLT-based cross-domain settlement. + +## History\* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.751.31 | 2025-12-14 | 21 | 11.1002/1000/16600 | + +## Keywords + +Cross-domain settlement, distributed ledger technology, electric vehicle, public charging. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2026 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|------------------------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Stakeholders of DLT-based cross-domain settlement ..... | 2 | +| 7 Framework and functional requirements of DLT-based cross-domain settlement ..... | 3 | +| 7.1 Framework of DLT-based cross-domain settlement ..... | 3 | +| 7.2 Functional requirements of DLT-based cross-domain settlement..... | 4 | +| Annex A – Workflow of DLT-based cross-domain settlement ..... | 7 | +| Bibliography..... | 8 | + + + +# Recommendation ITU-T F.751.31 + +## Reference framework of distributed ledger technology-based cross-domain settlement for electric vehicle public charging + +## 1 Scope + +This Recommendation specifies the reference framework of distributed ledger technology (DLT)-based cross-domain settlement for electric vehicle public charging among diverse charging operators. It includes: + +- Stakeholders of DLT-based cross-domain settlement. +- Framework and functional requirements of DLT-based cross-domain settlement. + +This Recommendation can be applicable to the design, implementation, deployment, and usage of the cross-domain settlement for electric vehicle public charging among diverse charging operators. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[[ITU-T F.751.0](#)] Recommendation ITU-T F.751.0 (2020), *Requirements for distributed ledger systems*. + +[[ITU-T F.751.2](#)] Recommendation ITU-T F.751.2 (2020), *Reference framework for distributed ledger technologies*. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 distributed ledger technology (DLT)** [b-ITU-T F.751.1]: Technology enabling large groups of nodes in distributed ledger networks to reach agreement and record information without the need for a central authority. + +**3.1.2 domain** [b-ITU-T Y.110]: A domain is the collection of segments which are owned and operated by a player and can include segments from more than one role. The extent of a domain is defined by a useful context, and one player can have more than one domain; however, a domain should not include more than one service provisioning platform. + +**3.1.3 electric vehicle (EV)** [b-IEC 61851-1]: Any vehicle propelled by an electric motor drawing current from a rechargeable energy storage system (RESS), intended primarily for use on public roads. + +**3.1.4 transaction** [b-ITU-T F.751.1]: An incident or an operation that leads to a change in the status of a ledger, such as adding a record or equivalent exchange based on currency. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following term: + +**3.2.1 cross-domain electric vehicle charging settlement:** A process that provides unified billing and financial transfer for electric vehicle public charging among diverse charging operator domains to ensure a transparent, traceable and automated charging transaction. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-----|-------------------------------| +| DLT | Distributed Ledger Technology | +| EV | Electric Vehicle | +| P2P | Peer to Peer | +| TEE | Trusted Execution Environment | + +## 5 Conventions + +In this Recommendation: + +- The keywords "**is required to**" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this Recommendation is to be claimed. +- The keywords "**is recommended**" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement needs not be present to claim conformance. + +## 6 Stakeholders of DLT-based cross-domain settlement + +The cross-domain electric vehicle (EV) charging settlement provides transaction clearing and transfer of funds for EV public charging, which enables electric vehicle users to charge their EV seamlessly by using diverse public charging operator's charging infrastructure. The associated charging fees are settled automatically and securely, eliminating the need for complex bilateral agreements or manual reconciliation. By leveraging distributed ledger technology (DLT), cross-domain settlement achieves greater transaction transparency and traceability, fostering trust among all stakeholders. Additionally, it enables decentralized fee settlement for EV public charging among diverse charging operators. + +Stakeholders involved in DLT-based cross-domain settlement are shown in Figure 1. + +![Diagram illustrating the stakeholders of DLT-based cross-domain settlement. The diagram shows a central 'DLT-based cross-domain settlement' area connected to various stakeholders: EV public charging operators (A, B, n), Financial service providers, Governance body, and EV users. Each stakeholder is connected to a 'DLT node', which is part of a 'Distributed ledger'. The connections are logical, represented by dashed lines.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +The diagram illustrates the stakeholders of DLT-based cross-domain settlement. At the top, three boxes represent 'EV public charging operator A', 'EV public charging operator B', and 'EV public charging operator n'. Each is connected by a dashed line to a 'DLT node' (represented by a cylinder icon with a document icon). These nodes are part of a 'Distributed ledger' (represented by a stack of document icons). The nodes are interconnected by dashed lines, forming a mesh. In the center, the text 'DLT-based cross-domain settlement' is written. Below the nodes, four more stakeholders are shown: 'Financial service providers' (a stack of boxes), 'Governance body' (a single box), and 'EV users' (a stack of boxes). Each is also connected to a 'DLT node', which is part of the distributed ledger. A legend at the bottom identifies the icons: a stack of documents for 'Distributed ledger', a cylinder for 'DLT node', a rectangle for 'Stakeholder', and a dashed line for 'Logical connection'. The reference number 'F.751.31(25)' is in the bottom right corner. + +Diagram illustrating the stakeholders of DLT-based cross-domain settlement. The diagram shows a central 'DLT-based cross-domain settlement' area connected to various stakeholders: EV public charging operators (A, B, n), Financial service providers, Governance body, and EV users. Each stakeholder is connected to a 'DLT node', which is part of a 'Distributed ledger'. The connections are logical, represented by dashed lines. + +**Figure 1 – Stakeholders of DLT-based cross-domain settlement** + +These stakeholders include: + +- a) **EV users:** These entities operate EVs and their terminals, enabling access to EV public charging and transaction settlement services. Upon registration and authentication, they use charging facilities provided by different operators, execute swift payments and monitor transaction records in real time. +- b) **EV public charging operators:** These entities are responsible for providing and operating the public charging infrastructure (including charging stations), managing electricity to provide the requested energy transfer services to EV users, and providing the billing and other value-added services to EV users. +- c) **Financial service providers:** These entities include third-party financial institutions involved in cross-domain charging settlement, which are responsible for authenticating user financial accounts, conducting credit evaluations, facilitating bonus points redemption and converting currencies. +- d) **Governance body:** This entity is responsible for monitoring the cross-domain transactions process and assigning data-access permissions across DLT nodes, to ensure secure, compliant data interoperability among stakeholders. + +## 7 Framework and functional requirements of DLT-based cross-domain settlement + +## 7.1 Framework of DLT-based cross-domain settlement + +The reference framework of DLT-based cross-domain settlement is mainly composed of three layers as shown in Figure 2. + +- a) **DLT basic layer:** This layer facilitates the secure and reliable integration of charging settlement data onto DLT nodes, including electricity consumption, charging prices, settlement outcomes, transaction evaluation results and account information. It provides functions including source verification, fingerprint generation and off-chain storage. +- b) **DLT core layer:** This layer comprises a specific DLT system with a trusted execution environment (TEE) support. It provides functions including smart contract, peer to peer (P2P) service, consensus algorithm and distributed ledger. + +- c) **DLT service layer:** This layer serves as a decentralized application on DLT nodes, providing cross-domain settlement services to stakeholders through smart contracts. It provides functions including cross-domain price management, cross-domain trusted settlement, cross-domain bonus points redemption and cross-domain information sharing. + +![Figure 2 – Reference framework of DLT-based cross-domain settlement. The diagram shows three stacked layers: DLT service layer (top), DLT core layer (middle), and DLT basic layer (bottom).](f4fdd410cdb84df81274da55721e56fb_img.jpg) + +The diagram illustrates the reference framework of DLT-based cross-domain settlement, organized into three hierarchical layers: + +- DLT service layer:** The top layer, containing four functional components: Cross-domain price management, Cross-domain trusted settlement, Cross-domain bonus points redemption, and Cross-domain information sharing. +- DLT core layer:** The middle layer, containing four foundational components: Smart contract, P2P service, Consensus algorithm, and Distributed ledger. +- DLT basic layer:** The bottom layer, containing three core components: Source verification, Fingerprint generation, and Off-chain storage. + +Figure 2 – Reference framework of DLT-based cross-domain settlement. The diagram shows three stacked layers: DLT service layer (top), DLT core layer (middle), and DLT basic layer (bottom). + +F.751.31(25) + +**Figure 2 – Reference framework of DLT-based cross-domain settlement** + +## 7.2 Functional requirements of DLT-based cross-domain settlement + +### 7.2.1 DLT basic layer + +The DLT basic layer maintains the integrity and security of cross-domain information. It includes: + +- Source verification:** It is required to validate the authenticity and origin of the incoming data, ensuring that it comes from trusted sources. +- Fingerprint generation:** It is required to create unique identifiers for on-chain information, including charging information, settlement information, account information, enabling efficient tracking and verification of data integrity. +- Off-chain storage:** It is required to store the verified raw data, including complete charging process data and basic management information of public charging facilities, in off-chain storage. + +### 7.2.2 DLT core layer + +The DLT core layer provides functionalities of the DLT system, ensuring the persistent storage and immutability of data throughout the cross-domain settlement process. It leverages the core capabilities provided by distributed ledgers, including smart contract, distributed ledger peer-to-peer service and consensus algorithm. + +- Consensus algorithm:** To ensure data integrity, the DLT network employs consensus algorithms to validate and confirm information related to cross-domain settlement. After the data has been recorded, the consensus algorithm is required to guarantee that all service nodes update and maintain a consistent copy of the data. Any attempt by an individual node to independently modify the data will be identified and prevented. +- Smart contract:** The DLT-based cross-domain settlement facilitates stakeholder to collaborate and streamline interactions by leveraging smart contracts, including: + - Automatic charging billing:** It is required to automatically calculate transaction fees and immutably record payment evidence on the DLT, ensuring transparency, auditability and fraud prevention. It will be achieved by using on-chain data, including real-time charging price, charging volume, facility details, user account information and applicable discounts. + +- ii) **Revenue allocation:** It is required to establish pre-agreed settlement rules among diverse public charging operators and financial service providers to automate the distribution of funds for diverse stakeholders. +- iii) **Transaction evaluation:** It is required to assess the creditworthiness of electric vehicle users and the service quality of charging operators based on historical behaviour. Charging operators with higher ratings are prioritized in user recommendations to strengthen service trust. +- iv) **Node permission management:** It is required to assign initial access rights to different stakeholders, including data query permissions and transaction verification permissions, and it is also required to update these rights dynamically in response to changes in trust scores. + +Distributed ledger is required to comply with clauses 7 and 8 of [ITU-T F.751.0], and P2P service is required to comply with clause 7 of [ITU-T F.751.2]. + +#### 7.2.3 DLT service layer + +The DLT service layer provides the operation and governance for the DLT-based cross-domain settlement. To ensure the efficiency and reliability of the entire cross-domain settlement process, it provides functions including cross-domain price management, cross-domain trusted settlement, cross-domain bonus points redemption and cross-domain information sharing. + +- a) **Cross-domain price management:** It is responsible for dynamically adjusting charging prices based on the power consumption of charging stations, operating costs and market demand. This service ensures fair and reasonable pricing by monitoring usage data of charging facilities, user payment records and publishing charging prices. + - i) It is required to support dynamic pricing capabilities for each public charging operator's charging infrastructure, where charging prices are automatically adjusted based on real-time charging demand, electricity costs, and equipment operating conditions. + - ii) It is required to support data security measures, ensuring that the transmission and storage of pricing information are encrypted using DLT, guaranteeing data integrity and immutability. + - iii) It is required to support transparency in charging price management, with price adjustment information published and shared across all relevant nodes to ensure users are informed of the current charging prices. +- b) **Cross-domain trusted settlement:** It is responsible for autonomously executing fund and bonus point settlement among diverse DLT nodes, reducing the risks and inefficiencies associated with manual intervention. + - i) It is required to record all transaction data during the settlement process on a distributed ledger, ensuring data immutability and providing a verifiable historical record for settlement. + - ii) It is required to perform settlement in real-time after the transaction is completed, ensuring timely and accurate financial clearing and profit distribution for all parties, and guaranteeing the transactional integrity of the transaction execution. + - iii) It is required to support the automatic execution of cross-domain settlement based on DLT-powered smart contracts, ensuring transparency and dispute-free transactions throughout the process. +- c) **Cross-domain bonus points redemption:** It facilitates users to use points as virtual tokens that can be exchanged across different nodes. Bonus point transactions are recorded on the decentralized ledger supported by DLT, ensuring that data cannot be tampered with and that transaction records remain transparent and traceable. + +- i) It is required to ensure that bonus points circulate freely between nodes, with all transaction data managed and verified by the financial service providers node. + - ii) It is required to base bonus point settlement on smart contracts, to enable automated point management and transaction processing, and to ensure that each transaction is carried out according to predefined rules. + - iii) It is recommended to ensure that bonus point redemption standards are transparent and publicly available, allowing users to clearly understand the redemption rules and required points. + - iv) It is recommended to encrypt and store the full lifecycle data of bonus points by using DLT, to ensure the security and immutability of the redemption process and enforce strict access control on user accounts. + - v) It is recommended to update user point redemption records in real-time and allows all redemption history to be verified through DLT, ensuring the transparency and traceability of point usage. +- d) **Cross-domain information sharing:** It securely and reliably exchanges the pivotal information from the cross-domain settlement process on DLT nodes. +- i) It is required to share charging information, including charging station location, availability, power consumption, charging records, charging price and critical fee information. + - ii) It is required to share charging facility information, including historical maintenance records, equipment ledger data, and the operation status. + - iii) It is recommended to share account information, including account identification, authentication results, user bonus points, payment records and settlement details. + - iv) It is recommended to share evaluation information, including EV user creditworthiness, operator service quality. + +## Annex A + +### Workflow of DLT-based cross-domain settlement + +(This annex forms an integral part of this Recommendation.) + +EV users can register and recharge their accounts through the client application provided by cross-domain settlement. After completing account registration, EV users can select charging facilities of different public charging operators to access charging services. After authenticating the EV user's charging request, the charging operator activates the corresponding charging facility to provide charging services. + +The workflow of DLT-based cross-domain settlement is shown in Figure A.1, including four main steps: + +![Figure A.1 – Workflow of DLT-based cross-domain settlement. The diagram shows a cycle between EV users, Public charging operators, and Financial service providers. Step 1: User registration (EV users to Financial service providers). Step 2: Charging service initialization (EV users to Public charging operators). Step 3: Charging settlement (Public charging operators to Financial service providers). Step 4: Account payment (Financial service providers to EV users). A central label 'DLT-based cross-domain settlement' is present.](410562339ce067fdc6fa41940c118658_img.jpg) + +``` +graph TD; EV[EV users] -- "Step 1. User registration" --> FSP[Financial service providers]; EV -- "Step 2. Charging service initialization" --> PCO[Public charging operators]; PCO -- "Step 3. Charging settlement" --> FSP; FSP -- "Step 4. Account payment" --> EV; subgraph DLT [DLT-based cross-domain settlement]; FSP; end; +``` + +Figure A.1 – Workflow of DLT-based cross-domain settlement. The diagram shows a cycle between EV users, Public charging operators, and Financial service providers. Step 1: User registration (EV users to Financial service providers). Step 2: Charging service initialization (EV users to Public charging operators). Step 3: Charging settlement (Public charging operators to Financial service providers). Step 4: Account payment (Financial service providers to EV users). A central label 'DLT-based cross-domain settlement' is present. + +F.751.31(25) + +**Figure A.1 – Workflow of DLT-based cross-domain settlement** + +- a) **Step 1. User registration:** A long-term user of public charging operator A completes account registration and points top-up upon first use of charging services provided by operator A. When the user subsequently uses charging facilities managed by operator B, there is no need to download the operator B application or register a new account. +- b) **Step 2. Charging service initialization:** After selecting a charging facility provided by operator B, the user connects the electric vehicle to the charging facility and performs information matching to initiate the charging service. Information including charging energy consumption and charging price is recorded and shared through the cross-domain settlement. +- c) **Step 3. Charging settlement:** Based on the charging price information, charging energy consumption information, and user account information provided by operator B, the financial service providers complete the cross-domain settlement for the charging service and generate the charging order. +- d) **Step 4. Account payment:** Based on the settlement information, the user completes payment for the charging payment with account points, thereby concluding the transaction. + +## Bibliography + +- [b-ITU-T F.751.1] Recommendation ITU-T F.751.1 (2020), *Assessment criteria for distributed ledger technology platforms.* +- [b-ITU-T Y.110] Recommendation ITU-T Y.110 (1998), *Global Information Infrastructure principles and framework architecture.* +- [b-IEC 61851-1] IEC 61851-1:2017, *Electric vehicle conductive charging system – Part 1: General requirements.* + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.751.4-202203-I_PDF-E/raw.md b/marked/F/T-REC-F.751.4-202203-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..854966dc86498f145574bfb09aee36a74e8793ef --- /dev/null +++ b/marked/F/T-REC-F.751.4-202203-I_PDF-E/raw.md @@ -0,0 +1,544 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.751.4** + +(03/2022) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +--- + +**General framework for distributed ledger +technology (DLT)-based invoices** + +Recommendation ITU-T F.751.4 + +ITU-T + +![ITU logo](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with a white grid pattern, overlaid by a stylized 'ITU' text in blue. + +ITU logo + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multidestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +## Recommendation ITU-T F.751.4 + +# General framework for distributed ledger technology (DLT)-based invoices + +## Summary + +A DLT-based invoice is an invoice that can be issued, transferred and received in a structured electronic format over digital ledgers which allow automatic, electronic transactions based on smart contracts. + +It presents as a novel invoice service category, emerging as a promising solution to tackle challenges by leveraging the capability of distributed ledger technology and the trust requirement of the stakeholders in the ecosystem. + +The usage of DLT-based invoices is driven mainly by seeking to optimize an end-to-end trustworthy business process across the jurisdictions in the major processes, e.g., issuance, routing, processing, reimbursement, auditing and so on. + +Recommendation ITU-T F.751.4 is not proposing a "regulatory" framework. However, for any tax implications, regulatory considerations must be addressed at a national level and are not the subject of the Recommendation. + +The electronic exchange of the invoice content between trading partners' accounts receivable and accounts payable business processes is to be recorded over the invoice digital ledger in a trustworthy way with local tax compliance. + +From a technology perspective, how the invoice will be transferred in a secure and interoperable way and how policies in different jurisdictions are enforced needs to be determined. Meanwhile, data privacy, security, trust and confidentiality have to be guaranteed, which is relevant to the following aspects: + +- secure messaging infrastructure to ensure that senders and receivers can trust the system and confidently exchange invoices; +- programmable government tax policies that can be securely enforced; +- invoice data validation schemes to ensure the integrity of the invoice content; +- immutability of the digital distributed ledgers to allow stakeholders to store and validate the invoice based on their corresponding privileges. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.751.4 | 2022-03-16 | 16 | 11.1002/1000/14966 | + +## Keywords + +DLT-based invoices, interoperability, security, smart contract. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2022 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-----------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Overview of DLT-based invoices..... | 2 | +| 6.1 Participants ..... | 2 | +| 6.2 Architecture ..... | 3 | +| 7 General requirements..... | 5 | +| 7.1 Identity requirement ..... | 5 | +| 7.2 Security requirement ..... | 6 | +| 7.3 Privacy requirement..... | 6 | +| 7.4 Interoperable requirement ..... | 6 | +| 7.5 Scalability requirement..... | 6 | +| 8 Use cases..... | 7 | +| 8.1 Invoicing process based on payment ..... | 7 | +| 8.2 Invoicing based on contract..... | 8 | +| Bibliography..... | 9 | + + + +## Recommendation ITU-T F.751.4 + +## General framework for distributed ledger technology (DLT)-based invoices + +# 1 Scope + +This Recommendation provides a general framework of DLT-based invoices on distributed ledger technology. The scope of this Recommendation also includes: + +- overview of DLT-based invoices; +- requirements of DLT-based invoices; +- structure of DLT-based invoices; +- use cases. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 distributed ledger** [b-ITU-T X.1400]: A type of ledger that is shared, replicated, and synchronized in a distributed and decentralized manner. + +**3.1.2 smart contract** [b-ITU-T X.1400]: Program written on the distributed ledger system that encodes the rules for specific types of distributed ledger system transactions in a way that can be validated, and triggered by specific conditions. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 distributed ledger system**: A system that implements a distributed ledger. + +**3.2.2 DLT-based invoices**: An invoice that can be issued, transferred and received in a structured electronic format over digital ledgers which allow automatic, electronic transactions based on smart contracts. + +**3.2.3 entity**: Anything that has a separately identifiable existence (e.g., organization, person, group or device). An entity uses distributed ledger technology to solve the problem of its business or information systems. + +**3.2.4 participating parties**: The participants of the transaction, e.g., sellers, buyers, payment networks etc. + +**3.2.5 transaction**: An incident or an operation that leads to a change in the status of a ledger, such as adding a record or an equivalent exchange based on currency. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|---------------------------------| +| CAE | Customer Associating Enterprise | +| DLT | Distributed Ledger Technology | +| SPV | Simplified Payment Verification | +| TAI | Trust Associated Index | +| TBI | Trust Base Index | +| UTXO | Unspent Transaction Output | +| VAT | Value Added Tax | + +# 5 Conventions + +This Recommendation uses the following conventions: + +- The keywords "**is required to**" indicate a requirement which must be strictly followed and from which no deviation is permitted, if conformance to this document is to be claimed. +- The keywords "**should**" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement needs not be present to claim conformance. + +# 6 Overview of DLT-based invoices + +## 6.1 Participants + +### 6.1.1 Payment platform + +A payment platform facilitates the transaction between customers and merchants. + +### 6.1.2 Customers + +Customers purchase items and may request an invoice for reimbursements. + +### 6.1.3 Merchant (sellers) + +Merchants initiate the invoice issuance together with the tax admins over the ledger based on the contracts or transactions. + +### 6.1.4 Destination tax admins + +The destination tax admin manages the inbound invoice validations together with the origin tax admin that issues the invoice. + +### 6.1.5 Origin tax admins + +Origin Tax Admin manages invoice issuance based on its local policy and the type of services that the sellers are supplying; the process could be processed by the smart contract issued by the origin tax admin. + +### 6.1.6 BER (billing service provider) + +BER is the billing service provider that collects the invoice billing request from the customers and submits the request to the tax admins. + +### 6.1.7 Acquirer (invoice acquire service provider) + +The acquirer acquires the invoice for interfacing with the Tax Admin for invoice validation and or reimbursement. + +#### 6.1.8 Wallet + +A wallet is as a customer's account for storing their invoices. + +### 6.1.9 Customer associating enterprise (CAE-Buyer) + +Customers submit the invoice in the wallet to its customer associating enterprise (CAE-Buyer) for reimbursement; the CAE-Buyer claims its tax with its tax admin, when claiming the VAT reverse and invoice validations. The tax admin is generally considered as the destination tax admin. + +### 6.1.10 Custom ledger + +A custom ledger is the ledger hosted in customs and interconnected with the ledger hosted in the tax admin. It is generally used for attestation in cross-border VAT reverse processes. + +#### 6.1.11 Payment domain + +The payment domain is a domain that the transactions are processed in; it includes the roles of merchants, payment platforms, customers and banks. + +### 6.1.12 Tax domain + +The tax domain is a domain that the invoices are issued, routed and reimbursed in, and includes the roles of tax admins, sellers, buyers and enterprises. + +## 6.2 Architecture + +### 6.2.1 Overview + +For the initial deployment of DLT-based invoices, a permissioned ledger is generally considered at the initial stage in different jurisdictions, thus the scope of this Recommendation is taking this assumption as a prerequisite. + +![Figure 1 – Overall architecture of DLT-based invoices. The diagram shows a hierarchical structure of layers. At the top is a single box labeled 'Cross-border routing layer'. Below it are three identical vertical stacks, each enclosed in a dashed blue box. Each stack contains four boxes stacked vertically: 'Regional routing layer' (top), 'Application layer', 'Attestation layer', and 'Core layer' (bottom). Lines connect the 'Cross-border routing layer' to the 'Regional routing layer' of each stack, and within each stack, lines connect the layers from top to bottom. A label 'F.751.4(22)' is in the bottom right corner.](65550960f5c5afd9d27706aaeddb0638_img.jpg) + +F.751.4(22) + +Figure 1 – Overall architecture of DLT-based invoices. The diagram shows a hierarchical structure of layers. At the top is a single box labeled 'Cross-border routing layer'. Below it are three identical vertical stacks, each enclosed in a dashed blue box. Each stack contains four boxes stacked vertically: 'Regional routing layer' (top), 'Application layer', 'Attestation layer', and 'Core layer' (bottom). Lines connect the 'Cross-border routing layer' to the 'Regional routing layer' of each stack, and within each stack, lines connect the layers from top to bottom. A label 'F.751.4(22)' is in the bottom right corner. + +**Figure 1 – Overall architecture of DLT-based invoices** + +The types of nodes anchored on the ledger typically include: + +- transaction accounting nodes; +- tax accounting nodes; +- custom nodes; +- audit nodes; +- routing nodes; +- SPV nodes. + +### 6.2.2 Core layer + +The core layer is mainly composed of the consensus nodes from the tax authorities; it is in charge of the following functionalities: + +- ledger consensus; +- security enforcement; +- identity management and verifications; +- smart contract management. + +### 6.2.3 Attestation layer + +In the attestation layer, the validators could be from relevant domains, e.g., payment domain, contract domain, tax domain etc. depending on different scenarios. The validators work together to redundantly validate and process the transactions, hence the data validation is not only depending on one domain but also linking to the relevant domains together with the core validators. + +In practical cases, it is not realistic to capture all UTXO dependencies in a transient way. Rather than the transitive UTXO dependence, the relationship of hash linking to the transactions could be accumulated to authenticate the data on chain in a different timing window instead. A given block can add back-pointers to a set of blocks, capturing the dependencies in an accumulative way. + +![Diagram illustrating the multilayer validation architecture of DLT-based invoices in the attestation layer. A Client on the left sends three transactions (TX) to a Payment validator (represented by a blue circle with three white cubes). The Payment validator is connected via dashed yellow lines to three Branch validators (represented by blue clouds with yellow cubes). Each Branch validator is connected to a Core validator (represented by a blue circle with yellow cubes). Each Core validator is also connected to a stack of four grey blocks. The diagram is labeled F.751.4(22) in the bottom right corner.](053f1077d592e6622cd21dc4bb4cb366_img.jpg) + +The diagram shows a 'Client' on the left sending three 'TX' (transaction) labels to a 'Payment validators' node, which is a blue circle containing three white cubes. From the 'Payment validators' node, three dashed yellow lines extend to three 'Branch validators' nodes, each represented by a blue cloud containing three yellow cubes. Each 'Branch validators' node is connected to a stack of four grey rectangular blocks. All three stacks of blocks are connected by dashed yellow lines to a single 'Core validators' node on the right, which is a blue circle containing three yellow cubes. The label 'F.751.4(22)' is in the bottom right corner. + +Diagram illustrating the multilayer validation architecture of DLT-based invoices in the attestation layer. A Client on the left sends three transactions (TX) to a Payment validator (represented by a blue circle with three white cubes). The Payment validator is connected via dashed yellow lines to three Branch validators (represented by blue clouds with yellow cubes). Each Branch validator is connected to a Core validator (represented by a blue circle with yellow cubes). Each Core validator is also connected to a stack of four grey blocks. The diagram is labeled F.751.4(22) in the bottom right corner. + +**Figure 2 – Multilayer validation architecture of DLT-based invoices in attestation layer** + +Apparently, the above method could further validate the transaction data in different dimensions or it could otherwise pick up the fake transaction with contradictory relevant background or foreground dimensions. + +Depending on the relationship of the dependencies, the direct trust level of transactions could be defined as Trust\_Base\_Index for one customer or merchant, e.g., the transactions TX1, TX4, TX6 and TXm compose Trust\_Base\_Index (TBI) over customer B; and the other trust factors that may have an indirect relationship with customer B could be defined as Trust\_Associated\_Index (TAI), e.g., TX1 and TX4 over Merchant A, TX5 and TX6 over Merchant C, TX5 and TX3 over customer D, and TX3 over Merchant E in a successive way. + +![Figure 3: Multilayer dependence compiling of DLT-based invoices. The diagram shows five vertical lines representing different entities: Merchant A, Customer B, Merchant C, Customer D, and Merchant E. Each line contains a series of circular nodes representing transactions. Merchant A has nodes TX1 (shaded), TX2 (white), TX4 (shaded), and TXn (white). Customer B has nodes TX6 (black), TX1 (shaded), TX4 (shaded), and TXn (white). Merchant C has nodes TX5 (black), TX6 (black), and TX9 (black). Customer D has nodes TX5 (black), TX3 (black), and TX3 (black). Merchant E has node TX3 (black). Dashed lines connect nodes between different entities: TX1 (Merchant A) to TX6 (Customer B), TX6 (Customer B) to TX5 (Merchant C), TX5 (Merchant C) to TX5 (Customer D), TX4 (Merchant A) to TX4 (Customer B), TX4 (Customer B) to TX6 (Merchant C), TX9 (Merchant C) to TX3 (Customer D), and TX3 (Customer D) to TX3 (Merchant E). The label F.751.4(22) is in the bottom right corner.](cfef993dcc8fb513de79eb1f93cf26ae_img.jpg) + +Figure 3: Multilayer dependence compiling of DLT-based invoices. The diagram shows five vertical lines representing different entities: Merchant A, Customer B, Merchant C, Customer D, and Merchant E. Each line contains a series of circular nodes representing transactions. Merchant A has nodes TX1 (shaded), TX2 (white), TX4 (shaded), and TXn (white). Customer B has nodes TX6 (black), TX1 (shaded), TX4 (shaded), and TXn (white). Merchant C has nodes TX5 (black), TX6 (black), and TX9 (black). Customer D has nodes TX5 (black), TX3 (black), and TX3 (black). Merchant E has node TX3 (black). Dashed lines connect nodes between different entities: TX1 (Merchant A) to TX6 (Customer B), TX6 (Customer B) to TX5 (Merchant C), TX5 (Merchant C) to TX5 (Customer D), TX4 (Merchant A) to TX4 (Customer B), TX4 (Customer B) to TX6 (Merchant C), TX9 (Merchant C) to TX3 (Customer D), and TX3 (Customer D) to TX3 (Merchant E). The label F.751.4(22) is in the bottom right corner. + +**Figure 3 – Multilayer dependence compiling of DLT-based invoices** + +Both TBI and TAI can be used to verify the invoice data in different dimensions, thus ensuring that the integrity of the invoice content is not only in one dimension, but also in a plane. + +### 6.2.4 Application layer + +The functionality of the application layer could include, but is not limited to, the following aspects: + +- invoice polling; +- reimbursements; +- VAT repayment. + +### 6.2.5 Regional routing layer + +To guarantee the interoperability, the regional routing of the invoice could be necessary depending on the transaction coverage jurisdiction, and the regional routing layer should be defined with the following functionalities: + +- numbering system to process the routing of the transaction; +- invoice validation; +- reimbursement facilities based on the invoices; +- VAT repayment across the region. + +The regional routing layer should be defined and enforced within a region, such as nationwide. + +### 6.2.6 Cross-border routing layer + +Aside from the regional routing layer, the cross-border routing layer should be defined to allow transactional validation across borders; the corresponding numbering and bridge should be defined to facilitate the capability. + +# 7 General requirements + +This clause identifies a set of general requirements that are required to be supported by the DLT-based invoices. + +## 7.1 Identity requirement + +The pre-authenticated node can be accepted to the permissioned ledger and recognized by an address in the ledger; the address maps to the registered profile of the nodes in the local shard ledger; and can be traced by header information of invoice blocks. + +Thus, the nodes are required to be authenticated in a local shard and uniquely identified in the ledger, and each invoice issued on the ledger is required to be traced to its origin issue shard ledger. + +## **7.2 Security requirement** + +The security requirements are classified as follows and mapped to the generic security principles, such as networking security, database security etc., and so no repeat definitions are needed here. + +### **7.2.1 Engineering security** + +It is required to guarantee engineering security, including the following aspects: + +- developing process security maintenance; +- deployment process security control. + +### **7.2.2 Operational security** + +It is required to guarantee operational security, including the following aspects: + +- operational security in tax admins; +- operational security in BERs; +- operational security in acquirers. + +### **7.2.3 Data accuracy** + +It is required to guarantee data accuracy, including the following aspects: + +- finality in invoice transaction, routing and settlement to avoid fraud; +- accuracy of oracle input/data. + +### **7.2.4 Security in customer management** + +It is required to guarantee security in customer management, including the following aspects: + +- ability to distinguish between un/authorized access; +- trust of custodial and safekeeping services; +- customer wallet and account management. + +## **7.3 Privacy requirement** + +It is required to guarantee the privacy that invoice transactional data can only be accessed between the participating parties and the auditing party. + +## **7.4 Interoperable requirement** + +Some level of consistency between at least similar DLTs is required to avoid unnecessary fragmentation, including: + +- interoperable routing protocols; +- interoperable data diagrams; +- interoperable policy mapping and configuration. + +## **7.5 Scalability requirement** + +The scalability requirement is to grow the total processing capacity with the number of validators in a permissioned ledger, including: + +- parallel intra-shard transaction processing; +- cross-shard commit protocol that atomically handles transactions affecting multiple shards; +- ledger life-cycle management by collectively signed state blocks; + +- low-latency "trust-but-verify" validation for low-value transactions. + +# 8 Use cases + +The invoicing process could be based on different scenarios, e.g.: + +- Use case 1: Invoicing based on payment; +- Use case 2: Invoicing based on contract. + +## 8.1 Invoicing process based on payment + +The invoicing process based on payment may include the following phases: + +- Phase A: transaction; +- Phase B: invoice issuance; +- Phase C: reimbursement; +- Phase D: tax clearance. + +![Figure 4 – Overall structure of DLT-based invoices. The diagram illustrates the flow of an invoice through four phases: A (Transaction), B (Invoice issuance), C (Reimbursement), and D (TAX clearance). It shows interactions between a Customer (Buyer), a Wallet, an Enterprise (Buyer), a Payment platform, a Seller, and two Tax Administrators (TAX Admin@Seller and TAX Admin@Buyer).](3ad00ce93ad9dea9ee0f47535e5355e6_img.jpg) + +The diagram illustrates the overall structure of DLT-based invoices across four phases: + +- Phase A. Transaction:** + - The Customer (Buyer) sends a payment request (A3) to the Payment platform. + - The Payment platform generates an order (A2) and sends it to the Seller. + - The Customer authorizes the payment (A4). + - The Seller generates a transaction (A6) and sends it to the Payment platform. + - The Payment platform sends a payment (A5) to the Seller. +- Phase B. Invoice issuance:** + - The Customer sends an invoice request (B1) to the Wallet. + - The Seller sends a digital signed request (B2) to the TAX Admin@Seller. + - The TAX Admin@Seller issues an invoice (B3) to the Seller. + - The Seller sends an invoice transfer (B4) to the Customer. + - The Customer authorizes the invoice to be stored in the wallet (B5). + - The Wallet stores the invoice (B6). +- Phase C. Reimbursement:** + - The Enterprise (Buyer) sends a reimbursement request (C1) to the Wallet. + - The Wallet sends an invoice reimbursed (C5) to the Enterprise (Buyer). + - The Enterprise (Buyer) sends a verification request (C2) to the TAX Admin@Buyer. + - The TAX Admin@Buyer sends a verification signed by SirNet (C4) to the Enterprise (Buyer). + - The Enterprise (Buyer) sends a verification (C3) to the TAX Admin@Seller. +- Phase D. TAX clearance:** + - The Enterprise (Buyer) sends a TAX clearance (D1) to the TAX Admin@Buyer. + +A dashed line labeled "TAX deductible" connects the Wallet and the TAX Admin@Seller. + +F.751.4(22) + +Figure 4 – Overall structure of DLT-based invoices. The diagram illustrates the flow of an invoice through four phases: A (Transaction), B (Invoice issuance), C (Reimbursement), and D (TAX clearance). It shows interactions between a Customer (Buyer), a Wallet, an Enterprise (Buyer), a Payment platform, a Seller, and two Tax Administrators (TAX Admin@Seller and TAX Admin@Buyer). + +**Figure 4 – Overall structure of DLT-based invoices** + +In the transaction phase, when the customer (buyer) places an order, the service provider (seller) generates the order over the payment platform, the payment platform then confirms the order after the customer authorization is received, and the receipt is generated when the payment has been processed. The receipt could be in the form of a payment block over a payment chain or a record in a centralized database. + +In the invoice issuance phase, the invoice is issued by the origin tax admin of the merchant based on the customer's request from the customer's wallet for instance, and the payment receipt is used as the UTXO for the invoice issuance. The participating nodes include the core consensus nodes anchored on the core layer of the ledger as well as the SPV nodes, such as the merchant node and personal wallet node. + +In the reimbursement phase, the customer associating enterprise (CAE) as an SPV node verifies the invoice when the customer initiates the reimbursement process and the invoice in the personal wallet is repaid as a UTXO. + +In the tax clearance process, the destination tax admin nodes and CAE SPV node join the process, the invoice is used as the UTXO to repay the VAT. + +## 8.2 Invoicing based on contract + +The DLT-based invoices can be issued over the tax ledger under the context that the associated contract has been signed; the contracts can be issued over a preregistered contract ledger endorsed by the concerned parties, as shown in Figure 5. + +![Figure 5: Overall structure of contract-based invoices. A flow diagram showing Node A and Node B connected to a 'Ledger for contract'. This ledger is connected via a dashed double-headed arrow to a 'Ledger for invoice', which is in turn connected via another dashed double-headed arrow to a 'Ledger for payment'. Solid lines also connect Node A and Node B directly to the 'Ledger for invoice' and 'Ledger for payment' respectively. A label F.751.4(22) is at the bottom right.](a33da0f14e456f92539ce3e9b7d81f9a_img.jpg) + +Figure 5: Overall structure of contract-based invoices. A flow diagram showing Node A and Node B connected to a 'Ledger for contract'. This ledger is connected via a dashed double-headed arrow to a 'Ledger for invoice', which is in turn connected via another dashed double-headed arrow to a 'Ledger for payment'. Solid lines also connect Node A and Node B directly to the 'Ledger for invoice' and 'Ledger for payment' respectively. A label F.751.4(22) is at the bottom right. + +Figure 5 – Overall structure of contract-based invoices + +The association of the relevant evident DLT-based objects can be programmable in assertions and can be displayed in the pages of the repository for the concerned owner in a way that is user friendly and configurable; the assertions can be compiled to run automatically if the predefined contexts are satisfied, as shown in Figure 6. + +| Assets repository | | | | | | | | | | | | | +|-------------------|--------------|------------------|-----------------------------|-----------------------|------------------|-----------------------------|-----------------------------|----------------|------|---------|----------------------------|-----------------------| +| Group name | Target | Target attribute | Concerned parties | Source | Source attribute | Delegate agent | Ledger selection | Ledger profile | Cost | Status | Data object address@Ledger | Execution | +| Project1 | XXX contract | contract |

Image: dropdown icon

| @local or cloud path | non-ledger |

Image: dropdown icon

|

Image: dropdown icon

| XX | XX | receipt | XXX | Manual trigger | +| Project1 | XXX invoice | Invoice |

Image: dropdown icon

| exist object@ledger | contract@ledger |

Image: dropdown icon

|

Image: dropdown icon

| XX | XX | Pending | Pending | Immediate automatic | +| Project1 | XXX payment | Payment |

Image: dropdown icon

| Pending object@ledger | invoice@ledger |

Image: dropdown icon

|

Image: dropdown icon

| XX | XX | Pending | Pending | Conditional automatic | + +F.751.4(22) + +Figure 6 – Assets repository over ledger objects + +The assertions are used to describe the basic building blocks of the asset repository. It defines the human readable logic, behaviour, cost, states and operations between digital ledger objects associating a predefined group for an asset repository of a person or an entity to enable automatic or manual triggered execution provided the pre-requisite conditions are met. + +The assertions link the ledger objects to one another by associating the existing or pending source objects to the target object in the repository without requiring an explicit reference, either within a single ledger object or between ledger objects. Ledger objects' linkages are a fundamental aspect of the repository. By associating ledger objects to the assets repository and cross-linking it to the ledger objects created by other people, the active participant repositories create useful interactions by linking the associated data objects in different ledgers. + +# Bibliography + +- [b-ITU-T F.751.0] Recommendation ITU-T F.751.0 (2020), *Requirements for distributed ledger systems*. +- [b-ITU-T F.751.2] Recommendation ITU-T F.751.2 (2020), *Reference framework for distributed ledger technologies*. +- [b-ITU-T F.751.3] Recommendation ITU-T F.751.3 (2022), *Requirements for change management in distributed ledger technology (DLT)-based decentralized applications*. +- [b-ITU-T X.1400] Recommendation ITU-T X.1400 (2020), *Terms and definitions for distributed ledger technology*. +- [b-ITU-T X.1404] Recommendation ITU-T X.1404 (2020), *Security assurance for distributed ledger technology*. + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.751.5-202212-I_PDF-E/raw.md b/marked/F/T-REC-F.751.5-202212-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..6abd5e982bee0aff71ac05c3704e4c48834928b5 --- /dev/null +++ b/marked/F/T-REC-F.751.5-202212-I_PDF-E/raw.md @@ -0,0 +1,323 @@ + + +# Recommendation **ITU-T F.751.5 (12/2022)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +## --- **Requirements for distributed ledger technology-based power grid data management** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessages service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.751.5 + +# Requirements for distributed ledger technology-based power grid data management + +## Summary + +Recommendation ITU-T F.751.5 defines requirements for distributed ledger technology (DLT)-based power grid data management, including framework of DLT-based power grid data management, requirements for infrastructure layer, requirements for service layer, requirements for application layer and requirements for data governance. This Recommendation can be used as a guideline for power grid data management with DLT technologies. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.751.5 | 2022-12-14 | 16 | 11.1002/1000/15199 | + +## Keywords + +Data management, distributed ledger technology, DLT, power grid. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2023 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|----------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Framework of DLT-based power grid data management..... | 2 | +| 7 Requirements for the infrastructure layer ..... | 3 | +| 8 Requirements for the service layer ..... | 3 | +| 9 Requirements for the application layer..... | 4 | +| 10 Requirements for data governance ..... | 5 | +| Bibliography..... | 6 | + + + +# Recommendation ITU-T F.751.5 + +# Requirements for distributed ledger technology-based power grid data management + +## 1 Scope + +This Recommendation defines requirements for DLT-based power grid data management, including: + +- requirements for infrastructure layer; +- requirements for service layer; +- requirements for application layer; and +- requirements for data governance. + +This Recommendation can be used as a guideline for power grid data management with DLT technologies. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 distributed ledger** [b-ITU-T X.1400]: A type of ledger that is shared, replicated, and synchronized in a distributed and decentralized manner. + +**3.1.2 DLT** [b-ITU-T X.1400]: A kind of distributed ledger with confirmed blocks organized in an append-only, sequential chain using cryptographic links. DLTs are designed to be tamper resistant and to create final, definitive and immutable ledger records. + +**3.1.3 consensus** [b-ITU-T F.751.0]: Agreement that a set of transactions is valid. + +**3.1.4 smart contract** [b-ITU-T F.751.0]: Program written on the distributed ledger system which encodes the rules for specific types of distributed ledger system transactions in a way that is recommended to be validated and triggered by specific conditions. + +**3.1.5 peer-to-peer** [b-ITU-T X.1400]: Relating to, using, or being a network of equal peers that share information and resources with each other directly without relying on a central entity. + +**3.1.6 on-chain** [b-ITU-T X.1400]: Located, performed or run inside a DLT system. + +**3.1.7 off-chain** [b-ITU-T X.1400]: Related to a DLT system, but located, performed or run outside that DLT system. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following term: + +**3.2.1 power grid:** The whole consisting of substations of various voltages and transmission and distribution lines in the power system. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +DLT Distributed Ledger Technology + +DPoS Deposit-based Proof of Stake + +P2P Peer-to-Peer networking + +PBFT Practical Byzantine Fault Tolerance + +PoS Proof of Stake + +PoW Proof of Work + +## 5 Conventions + +The following conventions are used in this Recommendation: + +- The keywords "**is required to**" indicate a requirement which must be strictly followed and from which no deviation is permitted, if conformance to this Recommendation is to be claimed; +- The keywords "**is recommended**" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance; +- The keywords "**is not recommended**" indicate a requirement which is not recommended but which is not specifically prohibited. Thus, conformance with this Recommendation still be claimed even if this requirement is present. + +## 6 Framework of DLT-based power grid data management + +DLT-based power grid data management improve the real-time interaction between power supply and demand response, and make the power grid dispatching process more transparent, more accurate, more economical. The DLT-based power grid data is stored in DLT, and the transaction data can be traced and tampered-proof in the whole process, thus ensuring the authenticity and uniqueness of the power grid data. + +DLT-based data management has the feature of tamper-proof, traceability, collective maintenance, and high transparency. A reference framework for DLT-based power grid data management is shown in Figure 1. This framework is composed of four layers, including the infrastructure layer, the service layer, the application layer and data governance. The requirements are described in the following clauses according to this framework. + +![Figure 1 – Framework of DLT-based power grid data management. The diagram shows three layers: Application layer, Service layer, and Infrastructure layer, with a corresponding Data governance column on the right.](ebff22fb5dd6f50a90e44dca0f82f285_img.jpg) + +| Application layer | Data governance | +|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +|
Compliance
Security
Permission grading
Scalability
Traceability
|
Data planning
Data execution
Data monitoring
Data disposing
| +| Service layer | | +|
Data reading
Data protection
Data sharing
On-chain and off-chain collaboration
| | +| Infrastructure layer | | +|
Hardware support
DLT core mechanism
System management
| | + +Figure 1 – Framework of DLT-based power grid data management. The diagram shows three layers: Application layer, Service layer, and Infrastructure layer, with a corresponding Data governance column on the right. + +F.751.5(22) + +**Figure 1 – Framework of DLT-based power grid data management** + +## 7 Requirements for the infrastructure layer + +The infrastructure layer provides the operating environment and basic components needed by the DLT-based power grid data management system, including hardware support, DLT core mechanism and system management. + +The following are the requirements for the infrastructure layer: + +IL-01: The hardware support of the DLT-based power grid data management is required to have the ability to prevent physical attacks. + +IL-02: The core mechanism is required to include the smart contract, consensus mechanism and peer-to-peer (P2P) network. The consensus mechanism includes but is not limited to practical Byzantine fault tolerance (PBFT) algorithm, proof of work (PoW) algorithm, proof of stake (PoS) algorithm, and deposit-based proof of stake (DPoS) algorithm. + +IL-03: System management is required to provide the function of account management, node management and access control. System management is required to support external interaction management functions. + +## 8 Requirements for the service layer + +The service layer provides services of data reading, data protection, data sharing, and on-chain and off-chain data collaboration. + +The following are the requirements for the service layer: + +BDS-01: Data reading is required to provide data access permissions. + +BDS-02: Data protection is required to provide security measures, including physical security measures and software security measures. + +BDS-03: Data sharing is required to ensure the security of the data link and to adopt an encrypted transmission protocol. + +BDS-04: On-chain and off-chain data collaboration is required to ensure relevance and consistency, and to increase the interoperability and scalability between chains. + +## 9 Requirements for the application layer + +The application layer provides customizable, visualized and interactive data applications for power grid data management users. + +The following are the requirements for the application layer: + +AL-01: The application layer of power grid data management is required to meet the compliance requirements, including: + +- the responsibilities and authorities of the DLT-based power grid data management system are required to be specified clearly in the design document; +- the DLT-based power grid data management system adopts encryption protection and security management measures for users' personal information; +- the management principles of independence, systematicity, full participation, mandatory, clear management status and responsibilities. + +AL-02: The application layer of power grid data management is required to meet the security requirements, including: + +- the confidentiality and security protection, the selection of a private key algorithm, and the security guarantee of data transmission at all levels of the system; +- to be equipped with an emergency response mechanism and practical measures; +- to ensure that the identities of all parties joining the node are safe and reliable, the assets are real and legal, and the node authority is controllable; +- to ensure the data security of data generation, transmission, storage and invocation, the key is required to be separated from the data, thus ensuring its confidentiality. + +AL-03: The application layer of power grid data management is required to meet permission grading requirements, including: + +- establish and improve the related party's account management system, such as account management, related party management, authority management, scope of authorization, etc; +- reduce unnecessary information collection, adopt the principle of least privilege and multi-role authorization scheme; +- corresponding permissions are granted according to the relevant party level, without overstepping the authority, and the division is clear. Different levels of relevant parties obtain corresponding data access permissions. + +AL-04: The application layer of power grid data management is required to meet scalability requirements, including: + +- the scalability of the internal functions of the DLT-based power grid data management system, and fully consider the future growth of the DLT-based power grid data management system; +- the construction of the system is required to make a scientific prediction of the future development trend, and the architecture component design is extensible; +- the modularization and componentization of the DLT-based power grid data management system is required to be fully considered and planned in the design document, and well-designed code allows more functions to be added to appropriate locations. + +AL-05: The application layer of power grid data management is required to meet traceability requirements, including: + +- unique identification and storage of asset-related data generated during the entire lifecycle; + +- the consideration of traceable data including data generation time, data type, data source, data query party information, authorization information, ownership change, traceability history record, data change and access records, etc.; +- the traceability of system, user, authorization information and other data is required to be divided according to the sensitivity and importance of the data, and the corresponding traceability plans are required to be adopted according to the nature of the data; +- the system displays different types of data in turn according to the sensitivity and importance of the data to meet different traceability requirements. + +## **10 Requirements for data governance** + +Data governance is the control of the entire stages of data management for the power grid, including planning stage, the execution stage and monitoring stage and the disposing stage. + +The following are the requirements for data governance: + +DG-01: The planning stage is required to ensure data usage, data specification and resource budgets. In the data planning stage, it is advisable to clarify data requirements, explain data usage, standardize data format and estimate resource budget. + +DG-02: The execution stage is required to ensure data generation, data collection, data processing and data application. In the implementation stage, the data production standards, data collection standards, data processing standards and data application standards are required to be strictly implemented. + +DG-03: The monitoring stage is required to ensure data status monitoring, data logic monitoring, data indicator monitoring, performance evaluation, etc. In the monitoring stage, monitoring configuration, monitoring alarms, monitoring processing, etc. is required to be set. + +DG-04: The disposing stage is required to ensure settlement of charges, audit records, summaries and review results, and final deletion, transfer or reuse of data collected and processed in the data management lifecycle. + +## Bibliography + +- [b-ITU-T F.751.0] Recommendation ITU-T F.751.0 (2020), *Requirements for distributed ledger systems.* +- [b-ITU-T X.1400] Recommendation ITU-T X.1400 (2020), *Terms and definitions for distributed ledger technology.* + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.751.8-202307-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.751.8-202307-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..25f8fbfaef317942ed3102c74b40a7d42f80608b --- /dev/null +++ b/marked/F/T-REC-F.751.8-202307-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:83cab7dcabfe939b223d97242d8523e461968d21f7c5432693c30a64f2d95521 +size 7050 diff --git a/marked/F/T-REC-F.751.8-202307-I_PDF-E/raw.md b/marked/F/T-REC-F.751.8-202307-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..bf03173d440d88599a5add6d1754f72b2de832bf --- /dev/null +++ b/marked/F/T-REC-F.751.8-202307-I_PDF-E/raw.md @@ -0,0 +1,322 @@ + + +Recommendation + +**ITU-T F.751.8 (07/2023)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +**Technical framework for distributed ledger technology (DLT) to cope with regulation** + +![ITU logo](84a1d09fb489061482111515543b60dc_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features the letters "ITU" in a bold, blue, sans-serif font, superimposed on a stylized globe icon with intersecting lines. + +ITU logo + +# ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.751.8 + +# Technical framework for distributed ledger technology (DLT) to cope with regulation + +## Summary + +Recommendation ITU-T F.751.8 defines the technical framework for distributed ledger technology (DLT) to cope with regulation, including regulatory challenges and technical capacities. The design of the technical framework of DLT in this Recommendation is closely related to DLT properties including decentralization, immutability and openness. This Recommendation can be used as guidance for the DLT system when facing regulation for DLT service providers and DLT system developers. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.751.8 | 2023-07-10 | 16 | 11.1002/1000/15174 | + +## Keywords + +Blockchain, distributed ledger technology, regulation. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2023 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|----------------------------------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Overview..... | 2 | +| 7 Regulatory challenges..... | 2 | +| 7.1 Leakage of confidential information ..... | 2 | +| 7.2 Immutability of records and absence of functionality to erase transactions/contents..... | 3 | +| 7.3 Security threats ..... | 3 | +| 7.4 Loss of tokens, digital currency and other assets ..... | 3 | +| 7.5 Financial and other crimes..... | 3 | +| 8 Technical capacities to cope with regulation..... | 3 | +| 8.1 Technical framework..... | 3 | +| 8.2 Application level capacities..... | 3 | +| 8.3 Privacy/Confidentiality capacities ..... | 3 | +| 8.4 Data erasure capacities ..... | 4 | +| 8.5 Data security capacities ..... | 5 | +| 8.6 Base level capacities..... | 5 | +| Bibliography..... | 6 | + + + +# Recommendation ITU-T F.751.8 + +## Technical framework for distributed ledger technology (DLT) to cope with regulation + +## 1 Scope + +This Recommendation defines a technical framework for distributed ledger technology (DLT) to address regulation. The scope of this Recommendation includes: + +- The regulatory challenges related to DLT; +- The technical capacities for DLT to cope with regulation. + +The target users of this Recommendation are DLT service providers and DLT system developers. DLT regulators are not target users. The objective of this Recommendation is not to provide regulatory solutions for DLT regulators, but rather to propose technical solutions to challenges related to DLT regulations. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[ITU-T F.751.0] Recommendation ITU-T F.751.0 (2020), *Requirements for Distributed Ledger Systems*. + +[ITU-T X.1401] Recommendation ITU-T X.1401 (2019), *Security threats of distributed ledger technology*. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 consensus** [b-ITU-T X.1400]: Agreement that a set of transactions is valid. + +**3.1.2 consensus mechanism** [b-ITU-T X.1400]: Rules and procedures by which consensus is reached. + +**3.1.3 distributed ledger** [b-ITU-T X.1400]: A type of ledger that is shared, replicated, and synchronized in a distributed and decentralized manner. + +**3.1.4 privacy** [b-ITU-T J.160]: A way to ensure that information is not disclosed to anyone other than the intended parties. Information is usually encrypted to provide confidentiality. Also known as "confidentiality". + +**3.1.5 threat** [b-ISO/IEC 27000]: Potential cause of an unwanted incident, which may result in harm to a system or organization. + +### 3.2 Terms defined in this Recommendation + +None. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|-------------------------------------------------| +| AML | Anti-Money-laundering | +| DLT | Distributed Ledger Technology | +| FATF | Financial Action Task Force on money laundering | +| IoT | Internet of Things | +| KYC | Know Your Customer | +| PKI | Public Key Infrastructure | +| TEE | Trusted Execution Environment | + +## 5 Conventions + +This Recommendation uses the following conventions: + +- The keyword "**is required to**" and "shall" indicate a requirement that must be strictly followed and from which no deviation is permitted, if conformance to this Recommendation is to be claimed. +- The keywords "**is recommended**" indicate a requirement that is recommended but which is not absolutely required. Thus, this requirement needs not be present to claim conformance. +- The keywords "**can optionally**" and "**may**" indicate an optional requirement that is permissible, without implying any sense of being recommended. These terms are not intended to imply that the vendor's implementation must provide the option, and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +## 6 Overview + +Distributed ledger technology (DLT) enables large groups of nodes in the distributed ledger network to reach agreements and record information without a central authority. The fast development of DLT brings challenges for regulation including privacy/confidentiality protection, data protection regulation, network attack handling and cryptocurrency regulation, among others. DLT has been applied to various domains such as cryptocurrency, supply chain management and IoT. Thus, it is necessary to summarize regulation requirements based on the analysis of DLT functions and properties and define the technical framework to fill the gap between openness and regulation. Precursor work was done in [b-DLT 4.1] and evolved in technical specifications found hereinafter. + +## 7 Regulatory challenges + +### 7.1 Leakage of confidential information + +Confidential information might be leaked directly on-chain, by combining external data with on-chain or off-chain data. Transactions and public keys might be directly available on-chain, particularly with permissionless DLT systems. External data might allow the identification of public keys with persons or companies and thereby identify transactions. + +In permissionless distributed ledgers, there is no access control and all data directly stored on-chain is accessible to everybody. Storing (unencrypted) confidential data on a permissionless distributed ledger is considered leakage of confidential data. + +### **7.2 Immutability of records and absence of functionality to erase transactions/contents** + +Depending on the DLT application and type of data, regulations require the erasure of personal data. This requirement is contained in laws and regulation, which can create conflicts with the immutability of DLT systems. + +NOTE – The "right to be forgotten" in Art. 17 of the EU General Data Protection Regulation ([b-GDPR]) is an example of the requirement of data erasure. + +Data protection regulation and other laws might also require correcting transactions. This is only possible through forks that put the reliability of distributed ledger systems at risk. + +### **7.3 Security threats** + +The security threats to DLT components refer to clauses 5.1 (threats to protocols), 5.2 (threats to networks) and 5.3 (threats to data) of [ITU-T X.1401]. + +### **7.4 Loss of tokens, digital currency and other assets** + +The loss of tokens, digital currency and other assets often result from threats to data. However, these losses might also be caused by wilful deceit regarding fraudulent coins, fraudulent distributed ledger systems or the entity behind account addresses. Phishing attacks might also trick people into making transactions they would not otherwise make. + +### **7.5 Financial and other crimes** + +Distributed ledger systems can be used for financial crimes such as money laundering or to facilitate other crimes such as ransomware attacks, among others. Transactions might contain illegal contents that obey the rules of the protocol but are in conflict with regulations and other laws. + +## **8 Technical capacities to cope with regulation** + +### **8.1 Technical framework** + +The design of the technical framework of DLT to cope with regulation is closely related to DLT properties including decentralization, autonomy, immutability, openness, transparency and anonymity. DLT technical capacities to cope with regulation are defined from the following aspects: application level, privacy/confidentiality, data erasure, data security and base level. In permissioned DLT, accountability is critical and DLT service providers are recommended to define an underlying orchestration legal entity to reduce the uncertainty of the regulatory actions. + +### **8.2 Application level capacities** + +- DLT systems that are used to transfer digital assets are recommended to include a module that supports FATF-rules and/or specific national regulations. +- It is recommended in permissioned DLT to support know your customer (KYC) and anti-money-laundering (AML) functionalities by using techniques such as public key infrastructure (PKI) based signatures. +- DLT service providers and system developers can optionally use techniques such as limiting the rights of transactions and block accounts to intervene in transactions created by malicious and abnormal nodes. +- DLT service providers and system developers can optionally provide a link to officially recognized PKI. This can optionally be used to identify transactions with a legal entity. + +### **8.3 Privacy/Confidentiality capacities** + +- A DLT system is recommended not to publish confidential information or personal data including account identity information, personal data/personal identifiable information, + +transaction data, digital assets information and other confidential information unless there is a justification for publishing it. + +- It is not recommended to store data containing this type of information on a permissionless DLT system. If this type of information needs to be verified by a DLT system, it is recommended to only store a commitment or other proof of the data on the DLT system. +- If it cannot be avoided to store the data on a DLT system, it is recommended to use cryptographic methods such as zero knowledge proofs, account confusion or methods based on hardware isolation (i.e., TEE) to protect confidential data (business data or personal data) from disclosure. +- It is recommended to properly secure confidential data stored off-chain by some appropriate cryptographic methods. +- DLT systems in which confidential information is stored are recommended to offer appropriate access control to ensure that only authorized access is possible. The access is recommended to be limited to those who are authorized and/or to situations where access is authorized. +- Off-chain storage is recommended to use the appropriate access control methods. +- Encryption, zero knowledge proofs and cryptographic hash-functions can optionally be used to secure information on-chain. However, immutable on-chain storage faces the challenge that compromised access methods cannot be blocked. It is recommended to perform a risk analysis and/or data protection impact analysis regarding on-chain storage of confidential/personal data. + +### 8.4 Data erasure capacities + +A high level of immutability is a key property of distributed ledger systems. Removing immutability completely would therefore often remove the reason why a distributed ledger system is chosen in the first place. + +When immutability is not desired for all attributes of a transaction, it is recommended to store data off-chain and only validate it, for example, by storing a proof to validate the off-chain data. For this purpose, technologies such as zero knowledge proofs, a commitment of the data or a hash-value can optionally be stored on-chain. Appropriate care is required to be taken that no undesired information can be derived from the data stored on-chain. + +In some use cases, verification of off-chain data might not be enough, and data persistence is needed. Still, the persistence of data might only be required for a specific period of time or while some conditions are met. Examples for such use cases are transaction data that needs to be preserved only as long as no following transaction exists, and records that need to be preserved for a specific time period (e.g., 10 years). In these cases, a distributed ledger can optionally be built that automatically removes this data when the following transaction has been completed or the time period has passed. Techniques such as pruning [b-Nakamoto] and chameleon hashes [b-Camenisch] can optionally be used. + +When such technologies are used + +- It shall be clearly defined what parts of the ledger shall remain immutable; +- It shall be clearly defined under which conditions other parts can be changed; +- The protocol of the distributed ledger shall define how this data is to be erased; +- The protocol of the distributed ledger shall ensure that data which is required not to be erased or modified is still protected against modification; +- It shall be ensured that archive copies of the ledger shall only be held according to applicable regulation. + +The objective of data erasure is to respond to regulation requests by local laws, recommendations or other requirements. The erasure functionality is recommended to be limited to some part of the data so that trust in the immutability of the rest of the data is maintained. + +### **8.5 Data security capacities** + +- DLT service providers and system developers are recommended to ensure that the design of DLT systems using cryptography and other technologies meets the security and performance requirements of DLT services, including reliability, completeness and immutability particularly by using appropriate cryptographic algorithms and consensus mechanisms. +- DLT service providers and system developers are recommended to ensure the consistency of data on-chain. + +### **8.6 Base level capacities** + +- DLT service providers and system developers are recommended to employ appropriate care to ensure the integrity, confidentiality, enforceability, availability and usage of DLT networks. +- DLT system developers are recommended to counter network threats with an appropriate security framework. +- DLT service providers and system developers can optionally use methods such as expanding computing power or stake resources to prevent consensus attacks. +- DLT service providers and system developers are recommended to support the security of smart contracts running on DLT systems by formal verification or other vulnerabilities detection methods. +- DLT system developers can optionally design fault-tolerance in DLT systems against malicious nodes. The DLT system is recommended to recover when a set of normal nodes are malfunctioning or become malicious nodes. + +## Bibliography + +- [b-ITU-T J.160] Recommendation ITU-T J.160 (2005), *Architectural framework for the delivery of time-critical services over cable television networks using cable modems*. +- [b-ITU-T X.1400] Recommendation ITU-T X.1400 (2020), *Terms and definitions for distributed ledger technology*. +- [b-Camenisch] Camenisch J., Derler D., Kernn S., Pöhls H. C., Samelin K., and Slamanig D. (2017), *Chameleon-hashes with ephemeral trapdoors and Applications to Invisible Sanitizable Signatures*, Public-Key Cryptography , Berlin, Springer, pp. 152–182. [https://doi.org/10.1007/978-3-662-54388-7\\_6](https://doi.org/10.1007/978-3-662-54388-7_6). +- [b-DIN SPEC 4997] DIN SPEC 4997 (2020), *Privacy by blockchain design: A standardised model for processing personal data using blockchain technology*. +- [b-DLT 4.1] ITU-T HSTP.DLT-RF (2019), *Distributed ledger technologies: Regulatory framework*. +- [b-GDPR] European Union General Data Protection Regulation law, Article 17, *Right to erasure ('right to be forgotten')*, . +- [b-ISO/IEC 27000] ISO/IEC 27000:2016, *Information technology – Security techniques – Information security management systems – Overview and vocabulary*. +- [b-Nakamoto] Nakamoto S. (2008), *Bitcoin: A Peer-to-Peer Electronic Cash System*. . + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.760.2-202404-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/F/T-REC-F.760.2-202404-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d4d8969f805414689ab79cfd08c333d454d482b3 --- /dev/null +++ b/marked/F/T-REC-F.760.2-202404-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:636c3839edff955a5b6208530b05fa1c983ab153d90f00216517a7a99478b988 +size 7074 diff --git a/marked/F/T-REC-F.760.2-202404-I_PDF-E/08441fa90c5fd11994626f662ac13f19_img.jpg b/marked/F/T-REC-F.760.2-202404-I_PDF-E/08441fa90c5fd11994626f662ac13f19_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..62850d18ee5e17c0cba0851d8d1c9e1a91be0d98 --- /dev/null +++ b/marked/F/T-REC-F.760.2-202404-I_PDF-E/08441fa90c5fd11994626f662ac13f19_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ee5e92b614596cdce8210f30d4490171c70a05ec14e6480c2bad319ab811e263 +size 46757 diff --git a/marked/F/T-REC-F.760.2-202404-I_PDF-E/3f1987804d7d78bc3b3bc560e974280a_img.jpg b/marked/F/T-REC-F.760.2-202404-I_PDF-E/3f1987804d7d78bc3b3bc560e974280a_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..8e5c4116368e349a54963ec228f045b6fddcb302 --- /dev/null +++ b/marked/F/T-REC-F.760.2-202404-I_PDF-E/3f1987804d7d78bc3b3bc560e974280a_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:acfedf5e093c7349428604fc6da4532234440cc27a40775758dbda25cb39011d +size 43607 diff --git a/marked/F/T-REC-F.760.2-202404-I_PDF-E/8fbdfc3d17fb1dae7b2d8f5a287fa9fc_img.jpg b/marked/F/T-REC-F.760.2-202404-I_PDF-E/8fbdfc3d17fb1dae7b2d8f5a287fa9fc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..5308b98ebba6a15c93b612f2eb96ec3b6c72e777 --- /dev/null +++ b/marked/F/T-REC-F.760.2-202404-I_PDF-E/8fbdfc3d17fb1dae7b2d8f5a287fa9fc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4c0b434fb738fb6cfc0cb84025f26396a394ee3e5d103826aad68bb6129fda56 +size 99950 diff --git a/marked/F/T-REC-F.760.2-202404-I_PDF-E/b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg b/marked/F/T-REC-F.760.2-202404-I_PDF-E/b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..105e8e14d0b5738377ba0ed23c934b1725f5e26a --- /dev/null +++ b/marked/F/T-REC-F.760.2-202404-I_PDF-E/b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1388b513b5a21d4ffb5defab2aff8f1c85878c19b96f7a9085aed64903eaa76f +size 71353 diff --git a/marked/F/T-REC-F.760.2-202404-I_PDF-E/ff7ce44f3fdd51bae7b231f34df07c6a_img.jpg b/marked/F/T-REC-F.760.2-202404-I_PDF-E/ff7ce44f3fdd51bae7b231f34df07c6a_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9049b7c30bdf5ac74683abeab1ed4a4e3731e0bd --- /dev/null +++ b/marked/F/T-REC-F.760.2-202404-I_PDF-E/ff7ce44f3fdd51bae7b231f34df07c6a_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2f1d9306dcdee179a596d91883337576e187928eb5506c3a418e48148c6a061f +size 34427 diff --git a/marked/F/T-REC-F.761-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.761-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f462aa76a1bd4abc43e71c76fb2d758330d8e893 --- /dev/null +++ b/marked/F/T-REC-F.761-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b69fbf51a775dd9a6d6f1a54889692b863303e635a263d2ff12b4e9196beb36b +size 7452 diff --git a/marked/F/T-REC-F.761-198811-I_PDF-E/raw.md b/marked/F/T-REC-F.761-198811-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..c06b947df3005ab8a6f932044aac12d7fe1796d9 --- /dev/null +++ b/marked/F/T-REC-F.761-198811-I_PDF-E/raw.md @@ -0,0 +1,119 @@ + + +![ITU logo](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +The logo of the International Telecommunication Union (ITU) features the letters 'ITU' in a bold, sans-serif font, superimposed on a stylized globe with intersecting lines. + +ITU logo + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.761** + +**OPERATIONS AND QUALITY OF SERVICE +AUDIOVISUAL SERVICES** + +--- + +**SERVICE-ORIENTED REQUIREMENTS +FOR TELEWRITING APPLICATIONS** + +**ITU-T Recommendation F.761** + +(Extract from the *Blue Book*) + +--- + +# NOTES + +1 ITU-T Recommendation F.761 (new numbering of Rec. F.730) was published in Fascicle II.5 of the *Blue Book*. This file is an extract from the *Blue Book*. While the presentation and layout of the text might be slightly different from the *Blue Book* version, the contents of the file are identical to the *Blue Book* version and copyright conditions remain unchanged (see below). + +2 In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +# **SERVICE-ORIENTED REQUIREMENTS FOR TELEWRITING APPLICATIONS** + +The CCITT, + +*considering* + +- (a) that telewriting could offer graphic-oriented communication on real time or store-and-forward basis; +- (b) that telewriting could serve as an optional facility to the general public telephone service, which could be of interest to several categories of users (e.g. deaf people, architects, advertising agencies, etc.); +- (c) that telewriting could support applications like educational communication services ("electronic blackboard"); +- (d) that telewriting could be a facility within a teleconference service; +- (e) that telewriting could add a graphic-oriented facility to telematic services, + +*concludes* + +that telewriting can be applied as either a communication service or a communication technique within other services, + +*recommends* + +that the service-oriented aspects of telewriting applications be in accordance with this Recommendation. + +## **1 General** + +### **1.1 *Scope*** + +This Recommendation specifies the service-oriented requirements for the application of telewriting in combination with telephony. + +Use of telewriting as an independent telecommunication service or in combination with other services than telephony is for further study. + +The technical characteristics of telewriting are specified in Recommendation T.150. + +### **1.2 *Definition*** + +Telewriting enables the transmission of graphic information to be displayed at the receiving side in accordance with the writing movements at the sending side. This display is normally effected on a real time basis; a delay in the transmission may be included. + +### **1.3 *Applications*** + +#### **1.3.1 Telewriting can support:** + +- a) explanatory information exchange; +- b) teleconferencing; +- c) distant teaching or lecturing; +- d) telecommunication between speech- and/or ear-handicapped people. + +#### **1.3.2 Applications in combination with services other than the telephone service and/or non-real time applications are for further study.** + +## 2 General characteristics of telewriting + +2.1 The main characteristics of telewriting in this context are real time display and interactive communication during one session, which requires availability of a terminal at each side. Consequently, both sides of the connection will be able to contribute to the same image. + +2.2 The presentation functionalities along with their attributes are described in Recommendation T.150. They allow mainly for: + +- the generation, transfer and representation of curves of arbitrary shapes; the effect of the movement of the writing instrument at the sending side is retained during reproduction; +- marking single positions in a telewriting image by means of a cursor; +- erasure of all or part of a telewriting image. + +2.3 The image input is provided by appropriate writing tools, such as a writing tablet and a writing pencil. + +2.4 At the originators' terminal, the input information is made visible either by a display unit or on the writing tablet itself. + +2.5 The reproduction at the recipients' terminal will take place on a screen, on paper or on any other means, normally resulting in a real time copy. The writing speed at the transmitting and receiving end should generally be the same. + +2.6 In principle, any type of network can be used as a carrier for telewriting signals. + +## 3 Telewriting in combination with telephony + +3.1 The general characteristics are as mentioned in § 2.1. + +3.2 The subset of presentation functionalities available are the default capabilities of the basic terminal as defined in Recommendation T.150. + +3.3 The following terminal operating modes are defined: + +- a) speech plus telewriting: speech signals and telewriting signals can be sent simultaneously; +- b) telewriting only: in this mode sending of speech signals is blocked, reception of speech signals is still possible. + +3.4 The public switched telephone network is used for carrying the telewriting information. In practice, a 300 bps sub-channel derived from the available speech channel bandwidth is used. + +*Note* – Use of the ISDN for carrying telewriting information is for further study. + +3.5 The quality of service depends on the characteristics of the telephone network and of the telewriting equipment. + +In case of simultaneous transmission of speech and telewriting signals, mutual interference should be relatively small. Faults caused by speech interference on the telewriting band are not tolerable. The influence of the telewriting signals on the quality of the speech conversation must be limited. \ No newline at end of file diff --git a/marked/F/T-REC-F.780.1-202304-I_PDF-E/raw.md b/marked/F/T-REC-F.780.1-202304-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..0feeae42c6c1a86d8dd51b61b377f9a7f2604a22 --- /dev/null +++ b/marked/F/T-REC-F.780.1-202304-I_PDF-E/raw.md @@ -0,0 +1,1251 @@ + + +# Recommendation **ITU-T F.780.1 (V3) (04/2023)** + +SERIES F: Non-telephone telecommunication services +Multimedia services + +# --- **Framework for telemedicine systems using ultra-high definition imaging** + +![ITU logo](390120de4fe440c42fea8154fcaad334_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +### **TELEGRAPH SERVICE** + +| | | +|-----------------------------------------------------------------|-------------| +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessages service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | + +### **MOBILE SERVICE** + +| | | +|--------------------------------------------------------|-------------| +| Mobile services and multdestination satellite services | F.110-F.159 | +|--------------------------------------------------------|-------------| + +### **TELEMATIC SERVICES** + +| | | +|-------------------------------------------|-------------| +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | + +### **MESSAGE HANDLING SERVICES** + +F.400-F.499 + +### **DIRECTORY SERVICES** + +F.500-F.549 + +### **DOCUMENT COMMUNICATION** + +| | | +|--------------------------------------|-------------| +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | + +### **DATA TRANSMISSION SERVICES** + +F.600-F.699 + +### **MULTIMEDIA SERVICES** + +**F.700-F.799** + +## **ISDN SERVICES** + +F.800-F.849 + +## **UNIVERSAL PERSONAL TELECOMMUNICATION** + +F.850-F.899 + +## **ACCESSIBILITY AND HUMAN FACTORS** + +F.900-F.999 + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.780.1 + +## Framework for telemedicine systems using ultra-high definition imaging + +## Summary + +Recommendation ITU-T F.780.1 describes requirements for using ultra-high definition (UHD) imaging, such as 4K and 8K video, for telemedicine. The purpose of these requirements is to use UHD systems for medical practices that use endoscopes and/or microscopes. This Recommendation also describes a list of requirements for using a UHD-based "endoscopic video camera" as a medical device. + +In addition, Annex A describes the requirements for the use of this technology as a medical device. + +This revision adds the clause for profiles of UHD imaging for medical services, as well as new definitions and abbreviations. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|--------------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.780.1 | 2018-10-14 | 16 | 11.1002/1000/13660 | +| 2.0 | ITU-T F.780.1 (V2) | 2022-03-16 | 16 | 11.1002/1000/14945 | +| 3.0 | ITU-T F.780.1 (V3) | 2023-04-29 | 16 | 11.1002/1000/15546 | + +## Keywords + +Endoscope, microscope, plastic optical fibre, ultra-high definition. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2023 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|-------------------------------------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 2 | +| 3.1 Terms defined elsewhere ..... | 2 | +| 3.2 Terms defined in this Recommendation ..... | 2 | +| 4 Abbreviations and acronyms ..... | 3 | +| 5 Conventions ..... | 3 | +| 6 Background..... | 4 | +| 7 Requirements for UHD endoscopic system..... | 5 | +| 7.1 System requirements on systems in local usage..... | 5 | +| 7.2 Wide-area network usage..... | 9 | +| 8 Environmental conditions ..... | 10 | +| 8.1 Operating room..... | 10 | +| 8.2 Safety considerations..... | 11 | +| 9 Profiles ..... | 11 | +| 9.1 Endoscopy ..... | 11 | +| 9.2 Microscopy ..... | 12 | +| 9.3 UHD medical imaging ..... | 12 | +| 9.4 Teleconferencing ..... | 13 | +| Annex A – Test standard of medical device "endoscopic video camera" in Japan ..... | 14 | +| Appendix I – Applications of UHD for telemedicine ..... | 18 | +| I.1 Teleconference..... | 18 | +| I.2 UHD medical imaging ..... | 19 | +| I.3 Endoscope with UHD resolution..... | 20 | +| I.4 Microscope with UHD resolution ..... | 20 | +| Appendix II – Example of endoscopic surgery using UHD..... | 23 | +| II.1 8K UHD endoscopic system configuration ..... | 23 | +| II.2 Mass production model of the 8K UHD endoscope..... | 24 | +| II.3 8K UHD endoscopic image performance ..... | 25 | +| II.4 Clinical case study..... | 27 | +| Appendix III – Real-time transmission of uncompressed UHD images ..... | 32 | +| III.1 Method of measuring the time delay of transmission of uncompressed UHD images ..... | 32 | +| III.2 Time delay in local connection between the 8K UHD camera and the 8K UHD LCD ..... | 32 | +| III.3 Time delay in local connection between the 8K UHD camera and the 8K UHD LCD ..... | 32 | +| III.4 Time delay in QG encoding / decoding transmission ..... | 32 | + +| | Page | +|------------------------------------------------------------------|-------------| +| III.5 Time delay in IP network between Japan and Singapore ..... | 33 | +| III.6 Discussion and conclusion..... | 34 | +| Bibliography ..... | 35 | + +# Recommendation ITU-T F.780.1 + +## Framework for telemedicine systems using ultra-high definition imaging + +## 1 Scope + +This Recommendation describes use cases and requirements for using ultra-high definition (UHD) imaging for telemedicine. The purpose of these requirements is to use UHD systems as part of an endoscope and/or microscope for medical practices. This Recommendation also describes a list of requirements for using a UHD-based "endoscopic video camera" as a medical device. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +- [ITU-T H.265] Recommendation ITU-T H.265 (2021), *High efficiency video coding*. +- [ITU-R BT.709-6] Recommendation ITU-R BT.709-6 (2015), *Parameter values for the HDTV standards for production and international programme exchange*. +- [ITU-R BT.1120-9] Recommendation ITU-R BT.1120-9 (2017), *Digital interfaces for studio signals with 1 920 × 1 080 image formats*. +- [ITU-R BT.2020-2] Recommendation ITU-R BT.2020-2 (2015), *Parameter values for ultra-high definition television systems for production and international programme exchange*. +- [ITU-R BT.2100-2] Recommendation ITU-R BT.2100-2 (2018), *Image parameter values for high dynamic range television for use in production and international programme exchange*. +- [ARIB STD-B58] ARIB STD-B58 (2014), *Interface for UHDTV Production Systems*. +- [CISPR 11] CISPR 11:2015+AMD1:2016+AMD2:2019 CSV Consolidated version, *Industrial, scientific and medical equipment – Radio-frequency disturbance characteristics – Limits and methods of measurement*. +- [IEC 60601-1] IEC 60601-1:2005+AMD1:2012, CSV Consolidated version, *Medical electrical equipment – Part 1: General requirements for basic safety and essential performance*. +- [IEC 60601-1-2] IEC 60601-1-2:2014+AMD1:2020 CSV Consolidated version, *Medical electrical equipment – Part 1-2: General requirements for basic safety and essential performance – Collateral Standard: Electromagnetic disturbances – Requirements and tests*. +- [IEC 61000-3-2] IEC 61000-3-2:2014, Electromagnetic compatibility (EMC) – Part 3-2: Limits – Limits for harmonic current emissions (equipment input current $\leq 16$ A per phase). + +- [IEC 61000-3-3] IEC 61000-3-3:2013+AMD1:2017+AMD2:2021 CSV Consolidated version, *Electromagnetic compatibility (EMC) – Part 3-3: Limits – Limitation of voltage changes, voltage fluctuations and flicker in public low-voltage supply systems, for equipment with rated current $\leq 16$ A per phase and not subject to conditional connection.* +- [IEC 61000-4-2] IEC 61000-4-2:2008, *Electromagnetic compatibility (EMC) – Part 4-2: Testing and measurement techniques – Electrostatic discharge immunity test.* +- [IEC 61000-4-3] IEC 61000-4-3:2006, *Electromagnetic compatibility (EMC) – Part 4-3: Testing and measurement techniques – Radiated, radio-frequency, electromagnetic field immunity test.* +- [IEC 61000-4-4] IEC 61000-4-4:2012, *Electromagnetic compatibility (EMC) – Part 4-4: Testing and measurement techniques – Electrical fast transient/burst immunity test.* +- [IEC 61000-4-5] IEC 61000-4-5:2014+AMD1:2017 CSV Consolidated version, *Electromagnetic compatibility (EMC) – Part 4-5: Testing and measurement techniques – Surge immunity test.* +- [IEC 61000-4-6] IEC 61000-4-6:2003/AMD2:2006, *Amendment 2 – Electromagnetic compatibility (EMC) – Part 4-6: Testing and measurement techniques – Immunity to conducted disturbances, induced by radio-frequency fields.* +- [IEC 61000-4-8] IEC 61000-4-8:2009, *Electromagnetic compatibility (EMC) – Part 4-8: Testing and measurement techniques – Power frequency magnetic field immunity test.* +- [IEC 61000-4-11] IEC 61000-4-11:2004/AMD1:2017, *Amendment 1 – Electromagnetic compatibility (EMC) – Part 4-11: Testing and measurement techniques – Voltage dips, short interruptions and voltage variations immunity tests.* +- [IETF RFC 3550] IETF RFC 3550 (2003), *RTP: A Transport Protocol for Real-Time Applications.* +- [IETF RFC 4175] IETF RFC 4175 (2005), *RTP Payload Format for Uncompressed Video.* +- [SMPTE 424M] SMPTE ST 424:2012 – SMPTE Standard – *3 Gb/s Signal/Data Serial Interface.* +- [SMPTE 425-1] SMPTE ST 425-1:2017 – SMPTE Standard – *Source Image Format and Ancillary Data Mapping for the 3 Gb/s Serial Interface.* +- [SMPTE 2022-6] SMPTE ST 2022-6:2012 – SMPTE Standard – *Transport of High Bit Rate Media Signals over IP Networks (HBRMT).* +- [SMPTE 2082-10] SMPTE ST 2082-10:2018 – SMPTE Standard – *2160-line and 1080-line Source Image and Ancillary Data Mapping for 12G-SDI.* +- [SMPTE 2110-20] SMPTE ST 2110-20:2017 – SMPTE Standard – *Professional Media Over Managed IP Networks: Uncompressed Active Video.* + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +None. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 4K UHD:** A video format in which the horizontal screen resolution is 3 840 and the vertical screen resolution is 2 160 pixels (2160p). + +**3.2.2 8K UHD:** A video format in which the horizontal screen resolution is 7 680 and the vertical screen resolution is 4 320 pixels (4320p). + +**3.2.3 dual green Bayer:** A colour filter array for arranging RGB colour filters on a square grid of image sensors. Its particular arrangement of colour filters is found in most single-chip digital image sensors used in digital cameras to create a colour image. The filter pattern is 50% green, 25% red (R) and 25% blue (B), hence it is also called RGBG, GRGB, RGGB. + +**3.2.4 microsurgery:** Surgery requiring an operating microscope. + +**3.2.5 ultra-high definition:** A video format of digital display and camera in which the horizontal screen resolution is on the order of over 4 000 pixels. + +**3.2.6 ultra-high definition imaging:** Video imaging using system parameters of an ultra-high definition television (UHDTV) as defined by [ITU-R BT.2020-2]. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-------|-----------------------------------------| +| B | Blue colour | +| CCU | Camera Control Unit | +| CMOS | Complementary Metal Oxide Semiconductor | +| DG | Dual Green | +| Fps | Frames per second | +| G | Green colour | +| GI | Graded-Index multimode | +| HD | High-Definition | +| HDMI | High-Definition Multimedia Interface | +| IP | Internet Protocol | +| IPS | In-Plane Switching | +| IR | Infra-Red | +| LCD | Liquid Crystal Display | +| LED | Light Emission Display | +| OLED | Organic LED | +| POF | Plastic Optical Fibres | +| R | Red colour | +| SDI | Serial Digital Interface | +| TN | Twisted Nematic | +| UHD | Ultra-High Definition | +| UHDTV | Ultra-high definition television | +| VA | Vertical Alignment | + +## 5 Conventions + +The following conventions are used in this Recommendation: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted, if conformance to this Recommendation is to be claimed. +- The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be present to claim conformance. +- The keywords "can optionally" indicate an optional requirement which is permissible, without implying any sense of being recommended. This term is not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with this Recommendation. + +Requirements are identified using the following conventions: + +- Requirement number xx in clause n.m is of the form R-xx; +- Recommended requirement number yy in clause n.m is of the form RR-yy; +- Optional requirement number zz in clause n.m is of the form OR-zz. + +## 6 Background + +Ultra-high definition (UHD) video technologies, including 8K technology, are advancing rapidly. The present burgeoning of 8K UHD imaging technology is ushering in an era of video with a 16-fold higher resolution ( $7\,680 \times 4\,320$ pixels, about 33 million pixels) than the current high-definition (HD) ( $1\,920 \times 1\,080$ pixels, about 2 million pixels) technology and enabling the development of the next-generation 8K broadcasting system. + +In this context, there are growing expectations for many important contributions of 8K UHD technology to innovative medical imaging in advanced image-guided diagnosis and treatment. Midway between HD and 8K UHD, 4K UHD imaging technology ( $3\,840 \times 2\,160$ pixels, about 8 million pixels) is also gradually progressing for medical use. + +Endoscopic surgery, where surgeons operate by watching a lesion on a video screen with a camera in the body, cannot be performed without utilizing video imaging technology. However, the use of low-resolution images creates issues such as poor depth perception and lack of detail when zoomed in. These problems can be solved by implementing UHD imaging technology such as 8K technology. Experimental and clinical studies have revealed the mechanical and technical feasibility of the 8K UHD endoscope, providing a positive outlook on its prospective use in clinical practice. Appendix II gives a summary of the outcome of a feasibility study on the current 8K UHD endoscopy. + +The operating microscope, like the endoscope, is indispensable in today's medical care. Microsurgery is an accurate operation performed by watching an enlarged view of a small lesion using a binocular microscope. The operating microscope differs from the endoscope in the following ways: + +- No video monitor is used in the operation, and the view of the operating field is shared only by the surgeon and an assistant surgeon. (Although operating microscopes generally have a built-in video camera function, it is not suitable for operations because the quality of the image drops significantly compared to that of the microscope due to poor resolution). +- It is optically advantageous because the microscope has fewer restrictions on the lens sizes than the endoscope. + +Essentially, the only drawback of the operating microscope is that the real-time HD video image of the operating field cannot be shared with other members of the operation staff besides the two surgeons. This problem, however, can be solved by displaying the operating field on an 8K monitor through an 8K camera connected to the operating microscope. + +Furthermore, an even wider variety of applications, such as teleconferencing and UHD medical imaging may be expected if the above system is utilized in remote environments. Appendix I + +describes some of these applications of UHD for telemedicine. As indicated above, UHD imaging technology has much to offer telemedicine. + +## 7 Requirements for UHD endoscopic system + +This clause describes the requirements for the framework of a telemedicine system using UHD imaging for the applications described in the previous clause. In particular, it describes the requirements of a UHD endoscopic system, especially requirements for a UHD rigid endoscopic system, an example of which is given in Appendix II. For actual medical uses, such a system is required to be approved as a medical device by a national regulatory authority, as described in Annex A. + +### 7.1 System requirements on systems in local usage + +This clause describes requirements for a UHD endoscopic system employed in a closed local use, such as in an operating room. It is assumed that there is no external network connected to the system. Figure 1 shows the general architecture of the system in this setting. + +![Figure 1: Typical configuration of a UHD endoscope system in a closed local setting. The diagram shows a linear flow of components: an 8K display, an interface converter, a camera control unit (CCU), and an 8K endoscope. The 8K display is on the left, connected to the interface converter. The interface converter is connected to the CCU. The CCU is connected to the 8K endoscope. The 8K endoscope is on the right. A red dot on the 8K display is labeled 'Display format'. A blue line labeled 'Video interface specification' connects the interface converter and the CCU. A red dot on the 8K endoscope is labeled 'Endoscope specification'. The text 'F.780.1(18)_F01' is in the bottom right corner.](7e670a2b556b53ea9002dfff3a420e08_img.jpg) + +``` +graph LR; A[8K display] --- B[Interface converter]; B --- C[Camera control unit (CCU)]; C --- D[8K endoscope]; E[Display format] -.-> A; F[Video interface specification] -.-> B; F -.-> C; G[Endoscope specification] -.-> D; +``` + +Figure 1: Typical configuration of a UHD endoscope system in a closed local setting. The diagram shows a linear flow of components: an 8K display, an interface converter, a camera control unit (CCU), and an 8K endoscope. The 8K display is on the left, connected to the interface converter. The interface converter is connected to the CCU. The CCU is connected to the 8K endoscope. The 8K endoscope is on the right. A red dot on the 8K display is labeled 'Display format'. A blue line labeled 'Video interface specification' connects the interface converter and the CCU. A red dot on the 8K endoscope is labeled 'Endoscope specification'. The text 'F.780.1(18)\_F01' is in the bottom right corner. + +Figure 1 – Typical configuration of a UHD endoscope system in a closed local setting + +A UHD endoscope system in this setting consists of the following components: + +- UHD endoscope lens; +- UHD endoscope camera head including a UHD image sensor; +- Camera control unit (CCU); +- Video signal converter; +- UHD display. + +#### 7.1.1 Physical requirements + +##### 7.1.1.1 Camera size + +Generally, a rigid endoscope for surgery is assumed to be held by the surgeon in one or both hands while adjusting its position and direction to optimize the endoscopic view of the surgical field during surgery. An endoscopic camera body is recommended to be small to avoid interference with a patient's body and other surgical instruments. In addition, the weight of the camera body is recommended to be light because surgeons must generally hold it for longer than a few hours. + +RR-01: A UHD endoscope system is recommended to have a camera head with the following characteristics: + +- A diameter less than 120 mm; +- A weight less than 500 g. + +NOTE – These characteristics are for hand-held operations. + +##### 7.1.1.2 Image sensor size + +A general surgical endoscopic camera is connected to a rigid endoscope including some combination of optical rod lenses. The F number of this optical system is roughly over 10, so in order to realize a proper image of 8K UHD resolution, the size of an 8K UHD image sensor is recommended to be as large as possible. + +On the other hand, because the size and weight of an endoscopic camera body are limited for surgical use, a proper sized image sensor should be selected. + +RR-02: A UHD endoscope system is recommended to support a complementary metal oxide semiconductor (CMOS) sensor with the following characteristics: + +- A size less than or equal to 'super-35 mm' (e.g., $24.6 \times 13.9$ mm, about 1.7 inches); +- A pixel pitch less than $3.2 \mu\text{m}$ . + +##### 7.1.1.3 Video interfaces + +###### 7.1.1.3.1 Output from camera + +A medical endoscopic camera is generally used in combination with a CCU, which outputs image signals to an endoscopic display. During surgery, a blackout of the display due to video cables being pulled out from devices in this system must be avoided, so serial digital interface (SDI) cables with strong connections to devices are generally used in operating rooms. Transformation of the 8K UHD image requires synchronized multi-SDI cables, of which the shape and length are the same. + +In Appendix II, the 8K UHD endoscopic camera has a dual-green image sensor with a Bayer pattern colour filter array and requires a total of 24 Gbit/s of transmission from the camera to the CCU. For the transmission of the 8K UHD video signal, HD-SDI ( $1.5 \text{ Gbps} \times 16 \text{ ch}$ ), 3G-SDI ( $3 \text{ Gbps} \times 8 \text{ ch}$ ), 12G-SDI ( $12 \text{ Gbps} \times 2 \text{ ch}$ ) or U-SDI (optical, $24 \text{ Gbps} \times 1 \text{ ch}$ ) are generally selected. In operating rooms, a smaller number of cables is desirable, but fewer cables require high-quality SDI cables. When using a U-SDI optical cable, it is easy to connect but the transmission may not be steady if the cable is bent, subjected to other external forces, or gathers dust. + +RR-03: A UHD endoscope system is recommended to support at least one of the following video interfaces for the source signal from the camera unit: + +*For full 8K signals:* + +- High-definition multimedia interface (HDMI) $2.1 \times 1 \text{ ch}$ ; +- 3G-SDI (as defined in [SMPTE 424M] and [SMPTE 425-1]) $\times 16 \text{ ch}$ ; +- 12G-SDI (as defined in [SMPTE 2082-10]) $\times 4 \text{ ch}$ ; +- U-SDI [ARIB STD-B58] $\times 1 \text{ ch}$ . + +*For 8K dual green (DG):* + +- 3G-SDI (as defined in [SMPTE 424M] and [SMPTE 425-1]) $\times 8 \text{ ch}$ ; +- HD-SDI $\times 16 \text{ ch}$ ; + +###### 7.1.1.3.2 Input to display + +The input interface for the 8K UHD display should be matched to the output interface of the 8K UHD camera (CCU). In the operating room, endoscopic displays are mounted on, e.g., carts, stands or arms suspended from the ceiling, and their height, position and direction should be adjustable during surgery. Therefore, SDI interfaces are desirable. + +The high-definition multimedia interface (HDMI), which is mainly for consumer use must be locked while in use in the operating room so that it is not accidentally pulled out from the endoscopic display. + +RR-04: A UHD endoscope system is recommended to support at least one of the following video interfaces for the input signal to the video display: + +*For a full 8K signal* + +- 3G-SDI (as defined in [SMPTE 424M] and [SMPTE 425-1]) $\times$ 16 ch +- 12G-SDI (as defined in [SMPTE 2082-10]) $\times$ 4 ch +- U-SDI [ARIB STD-B58] $\times$ 1 ch +- HDMI 2.0 $\times$ 4 ch +- HDMI 2.1 $\times$ 1 ch. + +*For 8K dual green:* + +- HD-SDI (i.e., 1.5G-SDI as defined in [ITU-R BT.1120-9], [b-SMPTE 292M]) $\times$ 16 ch +- 3G-SDI (as defined in [SMPTE 424M] and [SMPTE 425-1]) $\times$ 8 ch + +##### 7.1.1.4 Video display panel type + +Displays in the operating room are observed from various directions by medical staff, so in-plane switching (IPS) displays are often introduced with a wide viewing angle and less colour-shift. + +However, for the 8K UHD display, vertical alignment (VA) displays are also used to obtain high contrast from the front side, high presence and real 8K UHD resolution with a large panel size. + +RR-05: A UHD endoscope system is recommended to support at least one of the following display panel types: + +- For LC: IPS, twisted nematic (TN), VA. + +NOTE – Other display formats, such as the UHD projector and UHD-organic light emission display (OLED), are for future study. + +#### 7.1.2 Signal requirements + +##### 7.1.2.1 Resolution of capture format + +For advanced endoscopic observation with high resolution and a wide viewing angle, 8K UHD resolution ( $7\,680 \times 4\,320$ pixels) is recommended. During endoscopic surgery, surgeons can control the digital zooming of an 8K UHD image up to 4 times to observe the surgical field in detail by not closing the tip of the rigid endoscope to the surgical field. + +RR-06: A UHD endoscope system is recommended to support the following resolution: + +- $7\,680 \times 4\,320$ pixels + +NOTE – Other resolutions, such as square resolutions are for future study. + +##### 7.1.2.2 Colour filter + +To minimise endoscopic camera size so that it can be held in one hand for a long period of time during surgery, a single plate type colour image sensor is recommended. Here, both the image sensor unit and the circuit boards for signal processing and signal conversion should be small with as much power saving as possible. + +Human eyes are especially sensitive to the colour green, so the Bayer pattern colour filter array with two green, one red (R) and one blue (B) is recommended. + +RR-07: A UHD endoscope system is recommended to support: + +- Bayer pattern colour filter array (i.e., green1, green2, red, blue). + +##### **7.1.2.3 Frame frequency and scan mode** + +###### **7.1.2.3.1 Frame frequency** + +Endoscopic surgery requires close observation of the endoscopic display so that the surgeon can control the endoscope itself and the various surgical instruments. Therefore, a frame frequency of less than 30 frames per second (fps) is not recommended, whereas more than 50 fps is recommended to support safer surgical procedures. + +RR-08: A UHD endoscope system is recommended to support at least one of the following frame frequencies (Hz): + +- 120, 120/1.001, 100, 60, 60/1.001, and 50. + +NOTE – The frame frequencies 30, 30/1.001, 25, 24, and 24/1.001 as defined in [ITU-R BT.2020-2] are not suitable for medical purposes. + +###### **7.1.2.3.2 Scan mode** + +RR-09: A UHD endoscope system is recommended to support progressive scan mode, as defined in [ITU-R BT.2020-2]. + +##### **7.1.2.4 Display format** + +###### **7.1.2.4.1 Resolution** + +In operating rooms, HD medical displays are conventionally used. For example, to record endoscopic images for the entire length of a surgery, it is common to use down-conversion from 8K UHD to 4K UHD/HD resolution for downsizing video data. Therefore, both 8K UHD and also 4K UHD/HD resolutions are recommended. + +RR-10: A UHD endoscope system is recommended to support the following resolutions: + +- 7 680 × 4 320 pixels; +- 3 840 × 2 180 pixels; +- 1 920 × 1 080 pixels. + +###### **7.1.2.4.2 Colour signal** + +In endoscopic surgery, it is important to be able to recognize subtle colour differences across tissues and organs on the display with high-bit gradation. It is also useful to compress the colour data of the UHD video data without being noticeable to surgeons. + +RR-11: A UHD endoscope system is recommended to support both the following: + +- RGB (4:4:4); +- YCbCr (4:2:2). + +###### **7.1.2.4.3 Colour space** + +In endoscopic surgery, it is important to recognize subtle colour differences across tissues and organs on a display with wide colour space. + +RR-12: A UHD endoscope system is recommended to support one of the following: + +- [ITU-R BT.2020-2]; +- [ITU-R BT.2100-2]. + +NOTE 1 – For some legacy systems, [ITU-R BT.709-6] is a valid option. + +NOTE 2 – It is preferable for the display used for real-time image displays of medical procedures to have a low delay so as to have as little effect as possible on the transmission delay, while maintaining high-contrast resolution and performing quality image calibration. + +### 7.2 Wide-area network usage + +Advanced endoscopic images are not for surgeons in operating rooms but for medical staff outside the operating room (including outside the hospital) to share for conferences, education, telemedicine, and so on. + +This clause describes the requirements for a UHD endoscopic system employed with an external Internet protocol (IP) network (remotely) connected to the system. + +Figure 2 shows the general architecture of the system in this setting. + +![Figure 2: Typical remote UHD endoscope system using an IP network. The diagram shows the flow of video from an 8K endoscope through a Camera control unit (CCU), an 8K IP encoder, and a Media converter into an IP network. On the receiving side, the video passes through another Media converter, an 8K IP decoder, and an Interface converter to an 8K display. Various specifications are indicated: Video interface specification, Encoded video format, Endoscope specification, Physical layer network interface, and Video codec interface specification.](75f0cb39f1cd165dfe4a6aa6c4d9388d_img.jpg) + +The diagram illustrates the architecture of a remote UHD endoscope system. On the right, an **8K endoscope** (labeled with **Endoscope specification**) is connected to a **Camera control unit (CCU)**. The CCU is connected to an **8K IP encoder** (labeled with **Video interface specification** and **Encoded video format**). The encoder is connected to a **Media converter** (labeled with **Physical layer network interface**). This Media converter is connected to a central **IP network** cloud. On the other side of the IP network, another **Media converter** (labeled with **Video codec interface specification**) is connected to the network. This Media converter is connected to an **8K IP decoder** (labeled with **Video interface specification**). The decoder is connected to an **Interface converter**, which is finally connected to an **8K display** (labeled with **Display format**). The label **F.780.1(18)\_F02** is in the bottom right corner. + +Figure 2: Typical remote UHD endoscope system using an IP network. The diagram shows the flow of video from an 8K endoscope through a Camera control unit (CCU), an 8K IP encoder, and a Media converter into an IP network. On the receiving side, the video passes through another Media converter, an 8K IP decoder, and an Interface converter to an 8K display. Various specifications are indicated: Video interface specification, Encoded video format, Endoscope specification, Physical layer network interface, and Video codec interface specification. + +**Figure 2 – Typical remote UHD endoscope system using an IP network** + +A UHD endoscopic system in this setting consists of the following ten components: + +- 1) UHD endoscope lens; +- 2) UHD endoscope camera head including UHD image sensor; +- 3) Camera control unit; +- 4) Video encoder; +- 5) Media converter (encoder); +- 6) IP network; +- 7) Media converter (decoder); +- 8) Video decoder; +- 9) Video signal converter; +- 10) UHD display. + +#### 7.2.1 Video codec + +Endoscopic surgery requires real-time and low-delay transmission times of endoscopic images into the display. Therefore, it is recommended to introduce uncompressed transmission at least in the operating room. If real-time transmission is not necessary, for example, while recording video images that are only for monitoring outside the operating room, compressed video transmission is also useful. + +RR-13: A UHD endoscope system with an IP connection is recommended to support either "uncompressed" or "compressed" video formats. + +For compressed video, it is recommended that at least one of the following coding formats is used: + +- [ITU-T H.265]; +- SMPTE RDD 35 (TICO); +- SMPTE RDD 34 (LLVC). + +#### 7.2.2 Video over IP interface + +To transmit HD endoscopic video outside the operating room, it is recommended to use the established IP networking infrastructure using an IP encoder / decoder in the hospital or outside the hospital. + +RR-14: A UHD endoscope system is recommended to support the transport of SDI signals over IP, according to either of the following: + +- [SMPTE 2022-6] or [SMPTE 2110-20] for uncompressed video; +- [IETF RFC 3550] for compressed video. + +#### 7.2.3 Physical layer network interface + +RR-15: A UHD endoscope system should specify the network interface of the physical layer such that the system can be physically connected to a network. + +#### 7.2.4 End-to-end latency + +In endoscopic surgery, the delay time between the camera and the display should be as short as possible for safe and comfortable surgical procedures. Realistically, there must be some type of signal conversion between the camera and the display and a certain delay time for some frames. Therefore, it is recommended that the total delay time is kept within the allowance range of the endoscopic surgical procedures. + +RR-16: A UHD endoscope system is recommended to keep the total delay of video transmission for real-time medical procedures, such as operations, at no more than 85 ms, or no more than 5.1 frames with 60/1.001 or 10.2 frames with 120/1.001. + +NOTE – For remote interactive service, it is preferable for a UHD endoscope system to be able to keep the total delay of video transmission for real-time medical procedures, such as operations, at no more than 150 ms between the camera input and display output [b-Bate]. + +#### 7.2.5 Encapsulation + +RR-17: A UHD endoscope system is recommended to support the encapsulation of its signal data according to at least [IETF RFC 3550] or [IETF RFC 4175] for uncompressed data. + +#### 7.2.6 Forward error correction + +RR-18: A UHD endoscope system is recommended to support forward error correction (FEC), e.g., according to the following: + +- [SMPTE ST 2022-5]. + +## 8 Environmental conditions + +This clause describes the requirements on the environment surrounding a UHD endoscopic system. + +### 8.1 Operating room + +An endoscope is a surgical device used in an operating room and must be sterilized for use on patients. As shown in Appendix II, an example of an 8K UHD endoscopic camera is covered by a transparent + +sterile drape. An endoscope for surgery is often not fixed in the surgical field and is instead held in a surgeon's hand (and rarely mounted on an arm holder). Therefore, the cable between the endoscopic camera and the CCU should be flexible and durable for easy handling. For transmission from the camera to the CCU, optical fibre cables should be used. These cables must be examined regularly for failures from bending, outer forces, etc., because the fibres are usually glass and therefore fragile. Failures can cause blackouts of the endoscopic image on the display. For example, it is recommended to adopt durable protective tubing or plastic optical fibre (POF) cable [b-Koike]. + +### **8.2 Safety considerations** + +An example of a test standard for "endoscopic video camera" medical devices in Japan is shown in Annex A. The abovementioned requirements include a variety of conditions in the 8K UHD endoscopic system (as shown in Appendix II) approved for class 1 medical devices in Japan. + +For example, the temperature of the camera body surface should be no more than 41 degrees celsius if it is to be held in a surgeon's hand for an extended period of time during surgery (regulated by [b-JIS T 0601-1]). + +In addition, the leakage of current flow to a patient from the endoscopic camera must be no more than 10 µA for a Type cardiac floating (CF) device used in heart surgery (regulated by [b-JIS T 0601-1]). Regarding leakage current from CCU to earth, it must be no more than 10 µA for Type body (B) devices that are not used in heart surgery (also regulated by [b-JIS T 0601-1]). + +## **9 Profiles** + +In this clause, different profiles of UHD medical devices are described. These are not exhaustive, and there may be more profiles as the use cases of UHD medical devices develop. Such additions are for future study. + +### **9.1 Endoscopy** + +Endoscopes are divided into three broad categories, depending on their shape: + +- 1) rigid endoscopes, +- 2) flexible endoscopes, and +- 3) capsule endoscopes. + +#### **9.1.1 Types of endoscopes** + +##### **9.1.1.1 Rigid endoscopes** + +Rigid endoscopes capture images with a base camera that uses a relay lens system, with little restriction on the size of the camera, so the applicability of UHD resolution is very high. + +##### **9.1.1.2 Flexible endoscopes** + +Flexible endoscopes capture images with a camera mounted on the distal end or a camera guided by an optical fibre to be inserted into the body. + +##### **9.1.1.3 Capsule endoscopes** + +#### **9.1.2 Application areas of endoscopy** + +Endoscopy is used for the following application areas: + +- transnasal endoscopy, +- laryngeal endoscopy, +- bronchoscopy, +- upper gastrointestinal endoscopy, + +- duodenal endoscopy, +- enteroscopy, colonoscopy, +- thoracoscopic laparoscopy, +- fetoscopy, cystoscopy, +- cholangioscopy, +- arthroscopy, +- spinal (neuro) endoscopy, +- vascular endoscopy, +- epidural endoscopy, +- capsule endoscopy. + +The list is not exhaustive. + +### 9.2 Microscopy + +Microsurgery, surgery requiring an operating microscope, is an accurate operation performed by watching an enlarged view of a small lesion using a binocular microscope. The operating microscope is indispensable in today's medical care as is the endoscope. The current widely used components of a surgical microscope are an illumination device, an objective lens, a zoom lens, a focus lens, and an eyepiece (for surgeon and the assistant), which is connected to a camera using one assistant port. The image from the camera is displayed on the display/monitor. + +#### 9.2.1 Application areas of UHD microscopy + +UHD microscopy is used in the following areas: + +- surgical microscopes, +- pathological microscopes, +- biological microscopes. + +The list is not exhaustive. + +#### 9.2.2 UHD microsurgery + +Using UHD resolution makes it possible to capture surgical field images at a higher resolution compared to images obtained by looking through the eyepiece with the naked eye. This type of surgery, performed while watching the display/monitor, is called "heads-up surgery", and it is likely that there will be a shift to this style of microscopic surgery. + +##### 9.2.2.1 Use of 3D + +As microscopic surgery is performed while obtaining precise depth information of the surgical field using stereoscopic vision with images obtained by both eyes, it is best to use a 3D display/monitor system in a heads-up surgery. Current surgical microscopes are equipped with two ports for assistants, so it is possible to connect two UHD cameras. UHD resolution and 3D heads-up surgery can be achieved by acquiring the image for the right eye and the image for the left eye and outputting them to a 3D display/monitor with UHD resolution. + +### 9.3 UHD medical imaging + +UHD medical imaging refers to real-time or recorded images of a medical procedure using a UHD resolution camera. Each field of examination and treatment has its own specific name, and the main types are: + +- endoscopic imaging, + +- microscopic imaging, +- surgical procedure imaging (e.g., recordings of a laparotomy or thoracotomy or the like taken from above, usually near a shadow-less light). + +### **9.4 Teleconferencing** + +UHD teleconferencing is a practice of sharing medical images, such as those related to medical examinations, tests, diagnoses, treatment, or rehabilitation, taken with a UHD resolution camera with other locations within a premise (e.g., within the same hospital, clinic, university, research institution) or remote locations (i.e., locations not within the same building or place), see Figure I.3, and used for presentations, training, lectures, and tests in real time. + +#### **9.4.1 In-premise teleconferencing** + +Teleconferencing images are shared within a premise. + +#### **9.4.2 Distance teleconferencing** + +Teleconferencing images are shared with a location distantly situated. + +## Annex A + +### Test standard of medical device "endoscopic video camera" in Japan + +(This annex forms an integral part of this Recommendation.) + +**Table A.1 – Test standard of [IEC 60601-1-2], EMI: [CISPR 11] Group 1 Class B (230 V / 50 Hz)** + +| No. | Standard | Test item | Test port | Content | Note | Test voltage | Time [H] | +|-----|---------------------------|--------------------------------|-------------------------------|--------------------------------------------------------------------------------------------|--------------------------------|----------------------------------------|----------| +| 1 | [CISPR 11]
Gr1-Class B | Radiation EMI (10 m) | Body | 30-230 MHz; 30 dB | | AC 230V/
50 Hz | 1 | +| | | | | 230-1000 MHz; 37 dB | | | | +| 2 | [CISPR 11]
Gr1-Class B | Conduction EMI including click | AC | 0.15-0.5 MHz;
QP66-56 dB/AV56-46 dB | | AC 230V/
50 Hz | 0.5 | +| | | | | 0.5-5 MHz; QP56 dB/AV46 dB | | | | +| | | | | 5-30 MHz;
QP60 dB/AV50 dB
AC 230 V/50 Hz | | | | +| 3 | [IEC 61000-4-2] | Electrostatic radiation | Body | Contact: $\pm(2,4,6)$ kV,
Air: $\pm(2,4,8)$ kV (Horizontal / Vertical) | | AC 230V/
50 Hz | 3 | +| 4 | [IEC 61000-4-3] | Radiation immunity | Body | 80 MHz-2.5 GHz, 3 V/m,
4 phase (Horizontal / Vertical) not
for a life support system | 80% AM
1 kHz | AC 230V/
50 Hz | 4 | +| 5 | [IEC 61000-4-4] | Fast transient/
burst | AC | $\pm 2.0$ kV (L,N,PE,L+N+PE) | | AC 240V/
50 Hz
AC 100V/
50 Hz | 1 | +| | | | Camera cable
(camera side) | $\pm 1.0$ kV | | AC 230V/
50 Hz | 0.25 | +| | | | Camera cable
(CCU side) | $\pm 1.0$ kV | | AC 230V/
50 Hz | 0.25 | +| 6 | [IEC 61000-4-5] | Lightning surge | AC | Common $\pm(0.5, 1, 2)$ kV/
normal $\pm(0.5, 1)$ kV per 5 times | $0^\circ, 90^\circ, 270^\circ$ | AC 240V/
50 Hz
AC 100V/
50 Hz | 6 | +| 7 | [IEC 61000-4-6] | Conduction EMI immunity | AC | 0.15-80 MHz, 3 Vrms | 80% AM
1 kHz | AC 230V/
50 Hz | 0.75 | +| | | | Camera cable
(camera side) | 0.15-80 MHz, 3 Vrms | | AC 230V/
50 Hz | 0.5 | +| | | | Camera cable
(CCU side) | 0.15-80 MHz, 3 Vrms | | AC 230V/
50 Hz | 0.5 | +| 8 | [IEC 61000-4-8] | Magnetic field | Body | 50/60 Hz, 3 A/m | X/Y/Z | AC 230V/
50 Hz,
60 Hz | 1 | +| 9 | [IEC 61000-4-11] | Dip | AC | More than 95% drop, 10 ms ( $0^\circ, 180^\circ$ )
60% drop, 100 ms
30% drop, 500 ms | | AC 240V/
50 Hz
AC 100V/
50 Hz | 2 | + +**Table A.1 – Test standard of [IEC 60601-1-2], EMI: [CISPR 11] Group 1 Class B +(230 V / 50 Hz)** + +| No. | Standard | Test item | Test port | Content | Note | Test voltage | Time [H] | +|-----|-----------------|-----------------------------|-----------|---------------------------------------|--------------|----------------|----------| +| | | Instantaneous power failure | AC | More than 95% drop, 5000 ms | | | | +| 10 | [IEC 61000-3-2] | Power harmonics | AC | For devices driven by more than 220 V | only Class B | AC 230V/ 50 Hz | 0.5 | +| 11 | [IEC 61000-3-3] | Flicker | AC | For devices driven by more than 220 V | only Class B | AC 230V/ 50 Hz | 1 | + +**Table A.2 – Test standard of [IEC 60601-1-2], EMI: [CISPR 11] Group 1 Class B +(120 V / 60 Hz)** + +| No. | Standard | Test item | Test port | Content | Note | Test voltage | Time [H] | +|-----|------------------------|--------------------------------|----------------------------|--------------------------------------------------------------------------------------|--------------------------------|----------------|----------| +| 1 | [CISPR 11] Gr1-Class B | Radiation EMI (10 m) | Body | 30-230 MHz; 30 dB | | AC 120V/ 60 Hz | 1 | +| | | | | 230-1000 MHz; 37 dB | | | | +| 2 | [CISPR 11] Gr1-Class B | Conduction EMI including click | AC | 0.15-0.5 MHz; QP66-56 dB/AV56-46 dB | | AC 120V/ 60 Hz | 0.5 | +| | | | | 0.5-5 MHz; QP56 dB/AV46 dB | | | | +| | | | | 5-30 MHz; QP60 dB/AV50 dB | | | | +| 3 | [IEC 61000-4-2] | Electrostatic radiation | Body | Contact: $\pm(2, 4, 6)$ kV, Air: $\pm(2, 4, 8)$ kV (Horizontal / Vertical) | | AC 120V/ 60 Hz | 3 | +| 4 | [IEC 61000-4-3] | Radiation immunity | Body | 80 MHz-2.5 GHz, 3 V/m, 4 phase (Horizontal / Vertical) not for a life support system | 80% AM 1 kHz *1 | AC 120V/ 60 Hz | 4 | +| 5 | [IEC 61000-4-4] | Fast transient/ burst | AC | $\pm 2.0$ kV (L,N,PE,L+N+PE) | | AC 120V/ 60 Hz | 0.5 | +| | | | Camera cable (camera side) | $\pm 1.0$ kV | *3 | AC 120V/ 60 Hz | 0.25 | +| | | | Camera cable (CCU side) | $\pm 1.0$ kV | *3 | AC 120V/ 60 Hz | 0.25 | +| 6 | [IEC 61000-4-5] | Lightning surge | AC | Common $\pm(0.5, 1, 2)$ kV/ normal $\pm(0.5, 1)$ kV per 5 times | $0^\circ, 90^\circ, 270^\circ$ | AC 120V/ 60 Hz | 3 | +| 7 | [IEC 61000-4-6] | Conduction EMI Immunity | AC | 0.15-80 MHz, 3 Vrms | 80% AM 1 kHz *1 | AC 120V/ 60 Hz | 0.75 | +| | | | Camera cable (camera side) | 0.15-80 MHz, 3 Vrms | *1*2 | AC 120V/60 Hz | 0.5 | +| | | | Camera cable (CCU side) | 0.15-80 MHz, 3 Vrms | *1*2 | AC 120V/ 60 Hz | 0.5 | +| 8 | [IEC 61000-4-8] | Magnetic field | Body | 60 Hz, 3 A/m | X/Y/Z | AC 120V/ 60 Hz | 1 | + +**Table A.2 – Test standard of [IEC 60601-1-2], EMI: [CISPR 11] Group 1 Class B +(120 V / 60 Hz)** + +| No. | Standard | Test item | Test port | Content | Note | Test voltage | Time [H] | +|-----|------------------|-----------------------------|-----------|---------------------------------------------------------------------------------|------|-------------------|----------| +| 9 | [IEC 61000-4-11] | Dip | AC | More than 95% drop, 10 ms
(0°, 180°)
60% drop, 100 ms
30% drop, 500 ms | | AC 120V/
60 Hz | 1 | +| | | Instantaneous power failure | AC | More than 95% drop, 5 000 ms | | | | + +**Table A.3 – Test standard of [IEC 60601-1-2], EMI: [CISPR 11] Group 1 Class B +(100 V / 50 Hz)** + +| No. | Standard | Test item | Test port | Content | Note | Test voltage | Time [H] | +|-----|---------------------------|--------------------------------|-----------|----------------------------------------|--------------------|-------------------|----------| +| 1 | [CISPR 11]
Gr1-Class B | Radiation EMI (10 m) | Body | 30-230 MHz; 30 dB | | AC 100V/
50 Hz | 1 | +| | | | | 230-1000 MHz; 37 dB | | | | +| 2 | [CISPR 11]
Gr1-Class B | Conduction EMI including click | AC | 0.15-0.5 MHz;
QP66-56 dB/AV56-46 dB | | AC 100V/
50 Hz | 0.5 | +| | | | | 0.5-5 MHz;
QP56 dB/AV46 dB | | | | +| | | | | 5-30 MHz;
QP60 dB/AV50 dB | | | | +| 7 | [IEC 61000-4-6] | Conduction EMI Immunity | AC | 0.15-80 MHz, 3 Vrms | 80% AM
1 kHz *1 | AC 100V/
50 Hz | 0.75 | + +**Table A.4 – Test standard of [IEC 60601-1-2], EMI: [CISPR 11] Group 1 Class B (DC)** + +| No. | Standard | Test item | Test port | Content | Note | Time [H] | +|-----|---------------------------|----------------------------|-------------------------------|-----------------------------------------------------------------------------------------------|-----------------|----------| +| 1 | [CISPR 11]
Gr1-Class B | Radiation EMI
(10 m) | Body | 30-230 MHz; 30 dB | | 1 | +| | | | | 230-1000 MHz; 37 dB | | | +| 3 | [IEC 61000-4-2] | Electrostatic
radiation | Body | Contact: $\pm(2,4,6)$ kV, Air:
$\pm(2,4,8)$ kV (Horizontal /
Vertical) | | 3 | +| 4 | [IEC 61000-4-3] | Radiation
immunity | Body | 80 MHz-2.5 GHz, 3 V/m,
4 phase (Horizontal /
Vertical) not for a life
support system | 80% AM
1 kHz | 4 | +| 5 | [IEC 61000-4-4] | Fast transient /
burst | Camera cable
(camera side) | $\pm 1.0$ kV | | 1 | +| 7 | [IEC 61000-4-6] | Conduction
EMI immunity | Camera cable
(camera side) | 0.15-80 MHz, 3 Vrms | | 0.75 | +| 8 | [IEC 61000-4-8] | Magnetic field | Body | 50/60 Hz, 3 A/m | X/Y/Z | 1 | + +**Table A.5 – Product security test standard of [IEC 60601-1] +(AC 100-240 V, 50/60 Hz)** + +| No. | Test item | +|-----|------------------------------| +| 1 | Electricity test | +| 2 | Structural measurement | +| 3 | Ball pressure test | +| 4 | Modelling stress test | +| 5 | Noise measurement | +| 6 | Document check | +| 7 | Risk management confirmation | +| 8 | Usability check | + +## Appendix I + +### Applications of UHD for telemedicine + +(This appendix does not form an integral part of this Recommendation.) + +Real-time or recorded images taken with a UHD resolution camera are sent to, or shared with, other locations within a premise (e.g., within the same hospital, clinic, university, or research institution) or remote locations (i.e., locations not within the same building or place) and used for medical procedures such as medical examinations, tests, diagnoses, treatment, and rehabilitation. The images do not necessarily have to be real-time images. Figure I.1 shows an 8K endoscope with a real-time recorder. + +![Diagram of an 8K endoscope system with real-time recorder.](4b87467ad9642943235f48f7d4b59449_img.jpg) + +The diagram illustrates an 8K endoscope system within an operating room, enclosed by a dashed rectangular border. On the left, three rectangular boxes are stacked vertically: 'Debayer', 'Recorder', and 'Camera control unit (CCU)'. Red lines connect these boxes to a large '8K UHD monitor' on the right, which displays an endoscopic image of internal tissue. Below the monitor, an '8K UHD endoscope' is shown. Red lines connect the endoscope to the 'CCU' and the 'Recorder'. The text 'Operating room' is located to the right of the endoscope. A small label 'F.780.1(18)\_FI.1' is at the bottom right of the diagram. + +Diagram of an 8K endoscope system with real-time recorder. + +**Figure I.1 – 8K endoscope system with real-time recorder** + +### I.1 Teleconference + +Medical images, such as those related to medical examinations, tests, diagnoses, treatment, or rehabilitation, taken with a UHD resolution camera are sent to, or shared with, other locations within a premise (e.g., within the same hospital, clinic, university, research institution), see Figure I.2 or remote locations (i.e., locations not within the same building or place), see Figure I.3 and used for presentations, training, lectures, and tests in real time. + +![Diagram of an 8K endoscope system with in-house teleconference. The operating room contains an 8K endoscope connected to a camera control unit (CCU), which is connected to a distributor and a debayer. The debayer is connected to an 8K monitor displaying the endoscopic image. A second debayer is connected to a conference room monitor, also displaying the 8K image. The diagram is labeled F.780.1(18)_Fl.2.](7133ccf78043568ca62ecbcd43628a4a_img.jpg) + +The diagram illustrates an 8K endoscope system with in-house teleconference. In the operating room, an 8K endoscope is connected to a camera control unit (CCU). The CCU is connected to a distributor, which is connected to a debayer. The debayer is connected to an 8K monitor displaying the endoscopic image. A second debayer is connected to a conference room monitor, also displaying the 8K image. The diagram is labeled F.780.1(18)\_Fl.2. + +Diagram of an 8K endoscope system with in-house teleconference. The operating room contains an 8K endoscope connected to a camera control unit (CCU), which is connected to a distributor and a debayer. The debayer is connected to an 8K monitor displaying the endoscopic image. A second debayer is connected to a conference room monitor, also displaying the 8K image. The diagram is labeled F.780.1(18)\_Fl.2. + +**Figure I.2 – 8K endoscope system with teleconference (in-house)** + +![Diagram of an 8K endoscope system with teleconference at a remote site. The operating room contains an 8K endoscope connected to a camera control unit (CCU), which is connected to a distributor and a debayer. The debayer is connected to an 8K monitor displaying the endoscopic image. A second debayer is connected to a conference room monitor, also displaying the 8K image. The conference room monitor is connected to an other hospital. The diagram is labeled F.780.1(18)_Fl.3.](552265bdbcf6d43d341fd018a9076269_img.jpg) + +The diagram illustrates an 8K endoscope system with teleconference at a remote site. In the operating room, an 8K endoscope is connected to a camera control unit (CCU). The CCU is connected to a distributor, which is connected to a debayer. The debayer is connected to an 8K monitor displaying the endoscopic image. A second debayer is connected to a conference room monitor, also displaying the 8K image. The conference room monitor is connected to an other hospital. The diagram is labeled F.780.1(18)\_Fl.3. + +Diagram of an 8K endoscope system with teleconference at a remote site. The operating room contains an 8K endoscope connected to a camera control unit (CCU), which is connected to a distributor and a debayer. The debayer is connected to an 8K monitor displaying the endoscopic image. A second debayer is connected to a conference room monitor, also displaying the 8K image. The conference room monitor is connected to an other hospital. The diagram is labeled F.780.1(18)\_Fl.3. + +**Figure I.3 – 8K endoscope system with teleconference (remote site)** + +### I.2 UHD medical imaging + +UHD medical imaging refers to real-time or recorded images of a medical procedure using a UHD resolution camera. + +Each field of examination and treatment has its own specific name, and the main types are endoscopic imaging (e.g., transnasal endoscopy, laryngeal endoscopy, bronchoscopy, upper gastrointestinal endoscopy, duodenal endoscopy, enteroscopy, colonoscopy, thoracoscopic laparoscopy, fetoscopy, cystoscopy, cholangioscopy, arthroscopy, spinal (neuro) endoscopy, vascular endoscopy, epidural endoscopy, and capsule endoscopy), microscopic imaging (e.g., surgical microscopes, pathological microscopes, and biological microscopes), and surgical procedure imaging (recordings of a laparotomy or thoracotomy or the like taken from above, usually near a shadow-less light). + +### **I.3 Endoscope with UHD resolution** + +Endoscopes are divided into three broad categories, depending on their shape: + +- 1) rigid endoscopes, +- 2) flexible endoscopes, and +- 3) capsule endoscopes. + +Rigid endoscopes capture images with a base camera that uses a relay lens system, with little restriction on the size of the camera, so the applicability of the UHD resolution is very high. + +Flexible endoscopes capture images with a camera mounted on the distal end or a camera guided by an optical fibre to be inserted into the body. + +With the former, UHD conversion is difficult due to limitations to the size of the image sensor. With the latter, it is difficult to incorporate the number of optical fibres needed for UHD resolution while maintaining flexibility. + +As with flexible endoscopes, UHD conversion is difficult due to limitations in the size of the image sensor that can be mounted on capsule endoscopes. + +### **I.4 Microscope with UHD resolution** + +The standard components of a surgical microscope is an illumination device, an objective lens, a zoom lens, a focus lens, and an eyepiece (for surgeon and assistant), which is connected to a camera using one assistant port. The image from the camera is displayed on the display/monitor, but currently, surgeons mostly perform surgery while looking through the eyepiece and not while watching the display/monitor. + +Setting the camera and the display/monitor to UHD resolution makes it possible to capture surgical field images at a higher resolution compared to the images obtained by looking through the eyepiece with the naked eye. This type of surgery, performed while watching the display/monitor, is called "heads-up surgery", and it is likely that there will be a shift to this style of microscopic surgery. + +As microscopic surgery is performed while obtaining precise depth information of the surgical field using stereoscopic vision with images obtained by both eyes, it is best to use a 3D display/monitor system in a heads-up surgery. Current surgical microscopes are equipped with two ports for assistants, so it is possible to connect two UHD cameras. UHD resolution and 3D heads-up surgery can be achieved by acquiring the image for the right eye and the image for the left eye and outputting them to a 3D display/monitor with UHD resolution. + +The operating microscope is indispensable in today's medical care as is the endoscope. Microsurgery is an accurate operation performed by watching an enlarged view of a small lesion using a binocular microscope. Figure I.4 shows an example microscopic eye surgery set-up. The operating microscope is different from the endoscope in the following ways: + +- No video monitor is used in the operation, and the view of the operating field is shared only by the surgeon and an assistant surgeon. (Although operating microscopes, in general, have a built-in video camera function, it is not suitable for operation because the quality of the image largely drops as compared to that of the microscope due to its poor resolution.); and +- Optically advantageous because the microscope has less restriction on lens sizes than the endoscope. + +![Figure I.4: Set-up for 8K microscopic eye surgery. The image shows a surgical procedure in progress. A large 85-inch 8K LCD monitor on the left displays a high-resolution, magnified view of the eye's internal structures. A surgical microscope is positioned over the patient's eye, with an 8K camera attached to it to capture the surgical field. Two surgeons in blue sterile gowns are visible, focused on the procedure. Red callout boxes with arrows point to each component: '85-inch 8K LCD monitor', '8K camera', and 'Surgical microscope'.](397d7c5f170e27a78f44fd8a5b0a234b_img.jpg) + +Figure I.4: Set-up for 8K microscopic eye surgery. The image shows a surgical procedure in progress. A large 85-inch 8K LCD monitor on the left displays a high-resolution, magnified view of the eye's internal structures. A surgical microscope is positioned over the patient's eye, with an 8K camera attached to it to capture the surgical field. Two surgeons in blue sterile gowns are visible, focused on the procedure. Red callout boxes with arrows point to each component: '85-inch 8K LCD monitor', '8K camera', and 'Surgical microscope'. + +F.780.1(18)\_FI.4 + +**Figure I.4 – Set-up for 8K microscopic eye surgery** + +A potential drawback of the operating microscope is the inability to share the real-time HD video image of the operating field with other members of the operation staff except for the two operating surgeons. Connecting the 8K camera to an operating microscope and displaying the view of the operating field on an 8K monitor can solve this problem. + +In an ophthalmic surgery using an 8K endoscope performed at Miyake eye hospital in Tokyo, Japan on 5 December 2014, an operating microscope (which was suspended from the ceiling) was used and was connected to an 8K camera via a lens adaptor. An 85-inch 8K monitor was located near the surgeon and patient so that all the surgeons, assistants, and operation staff present could closely watch the UHD images at the same time. + +A total of eleven cases of cataract, glaucoma, corpus vitreum and retina surgeries were performed, and each surgery was done within the planned timeline. Since removal of the cortex lentis and the insertion of an intraocular lens in cataract surgery, fibre pillar zone incision in goniotomy, express surgery, and trabeculotomy (operation that cuts open and removes part of the column zone with an electrosurgical knife) were all surgeries on the surface side of the eye (anterior ocular segment), the quality of view obtained and shared was equally bright and had UHD, as that which was viewed by the surgeon through the operating microscope. See Figure I.5 for example eye surgeries using an 8k microscope. + +![Two side-by-side images showing eye surgery. The top image shows glaucoma surgery with surgical instruments visible. The bottom image shows retinal surgery with a laser probe being used on the retina.](cce1f4715971176232d2f8cf7b80acf7_img.jpg) + +a) Glaucoma surgery by 8K microscope + +b) Retinal surgery by 8K microscope + +Two side-by-side images showing eye surgery. The top image shows glaucoma surgery with surgical instruments visible. The bottom image shows retinal surgery with a laser probe being used on the retina. + +F.780.1(18)\_FI.5 + +**Figure I.5 – Example of eye surgery using an 8K microscope** + +## Appendix II + +### Example of endoscopic surgery using UHD + +(This appendix does not form an integral part of this Recommendation.) + +### II.1 8K UHD endoscopic system configuration + +An 8K UHD endoscopic system comprises an 8K UHD endoscope, a light source with a 300 W xenon lamp, CCU, 8K UHD recorder, an interface converter, and an 8K UHD liquid crystal display (LCD) as shown in Figure II.1. The 8K UHD endoscope consists of an 8K UHD camera head, lens adapter, and rigid endoscope. The CCU has a signal processing board with various functions, such as calibration of the colour balance and gain in real-time, digital zooming ( $\times 1.0 \sim \times 4.0$ ), gamma curve compensation, and resolution enhancement of the 8K UHD endoscopic images. + +![Figure II.1: System configuration example of an 8K UHD endoscope. The diagram shows the flow of video signals from an 8K endoscope (KairoScope-E) through a Camera control unit (KairoScope-CCU) to an 8K SSD recorder (HR-7518) via 3G-SDI (x8). The recorder is connected to an Interface converter (FC-8216) via 3G-SDI (x8), which then connects to an 8K LCD (LV70002) via HDMI (x4). The CCU also connects to a 2K recorder (UR-4MD) via HD-SDI, which in turn connects to a 2K LCD (EJ-MLA26N2) via HD-SDI. A Light source (CL-300X) is connected to the endoscope.](6ca05954842b17f14dfd52f26b9d43d2_img.jpg) + +``` +graph TD; E[8K endoscope KairoScope-E] --> CCU[Camera control unit KairoScope-CCU]; LS[Light source CL-300X] --> E; CCU -- 3G-SDI x8 --> R[8K SSD recorder HR-7518]; R -- 3G-SDI x8 --> IC[Interface converter FC-8216]; IC -- HDMI x4 --> D1[8K LCD LV70002]; CCU -- HD-SDI --> R2[2K recorder UR-4MD]; R2 -- HD-SDI --> D2[2K LCD EJ-MLA26N2]; +``` + +Figure II.1: System configuration example of an 8K UHD endoscope. The diagram shows the flow of video signals from an 8K endoscope (KairoScope-E) through a Camera control unit (KairoScope-CCU) to an 8K SSD recorder (HR-7518) via 3G-SDI (x8). The recorder is connected to an Interface converter (FC-8216) via 3G-SDI (x8), which then connects to an 8K LCD (LV70002) via HDMI (x4). The CCU also connects to a 2K recorder (UR-4MD) via HD-SDI, which in turn connects to a 2K LCD (EJ-MLA26N2) via HD-SDI. A Light source (CL-300X) is connected to the endoscope. + +Figure II.1 – System configuration example of an 8K UHD endoscope + +The 8K UHD recorder is used to keep and reproduce 8K UHD endoscopic images of a clinical laparoscopic cholecystectomy. Therefore, the 8K UHD recorder should have a long recording time and very high-speed access. An 8TB storage capacity solid state drive (SSD) installed in the 8K recorder makes it possible to store the 8K images for as long as 160 minutes by using signal compression technologies. + +The 8K UHD LCD can clearly present 8K UHD endoscopic images of a clinical laparoscopic cholecystectomy with a resolution of $7\,680 \times 4\,320$ pixels and colour depth of 10 bits. This large display is preferable to surgeons and operating-room staff because the most favourable viewing distance from the 8K UHD display has been estimated to be around 0.75 times the display/screen height in the operation room. A 2K recorder and 2K LCD are used in a standard and usual surgical setting. The total delay time of the 8K UHD endoscopic system is about 75 ms – sufficiently shorter than the human response time of 150 ms. + +Figure II.2 shows a typical 8K UHD endoscopic system. All signal processing equipment is installed in one pack, and the total size of the 8K UHD endoscopic system is the same as that of a conventional 2K endoscopic system. A surgeon can operate the 8K UHD endoscope freely with a single hand while simultaneously viewing the 8K UHD display image. + +![A photograph of a typical 8K UHD endoscopic system. On the left, a white medical cart holds various electronic components, including a monitor on top and a camera unit below. A hand is visible holding a long, thin endoscope probe. To the right, a large flat-screen monitor displays a dark image. Below the monitor is a silver-colored processing unit on a black stand. The floor has a red line near the base of the cart and stand.](b99bfa6a26ede115d00d81059ee4ce6b_img.jpg) + +A photograph of a typical 8K UHD endoscopic system. On the left, a white medical cart holds various electronic components, including a monitor on top and a camera unit below. A hand is visible holding a long, thin endoscope probe. To the right, a large flat-screen monitor displays a dark image. Below the monitor is a silver-colored processing unit on a black stand. The floor has a red line near the base of the cart and stand. + +F.780.1(18)\_FII.2 + +**Figure II.2 – A typical 8K UHD endoscopic system** + +### **II.2 Mass production model of the 8K UHD endoscope** + +The first prototype model, P1-model, of an 8K UHD camera head was developed in 2014 based on a broadcasting camera [b-Yamashita]. The 8K UHD camera head had a single-chip 2.5-inch CMOS image sensor with an 8K UHD resolution and dual-green colour (G) filters (red $\times$ 1, green $\times$ 2, blue $\times$ 1). The dimensions of the 8K UHD camera head were 125 mm (W) $\times$ 125 mm (H) $\times$ 185 mm (L), and its weight was 2.2 kg. A handle was attached to the outer case of the camera head to be easily held with a single hand. A standard type of a 300-W xenon lamp for conventional endoscopes was used as the light source. A lens adapter was mounted between the rigid endoscope and the 8K camera head to enlarge the endoscopic field of view and adjust the focus point of the endoscopic image. + +The second prototype model, the P2-model, of an 8K UHD camera head was made in 2016, and its dimensions were 80 mm (W) $\times$ 80 mm (H) $\times$ 215 mm (L), and its weight was 450 g. The mass production model of an 8K UHD camera head had succeeded in getting more compact in size with the dimensions of 75 mm (W) $\times$ 75 mm (H) $\times$ 170 mm (L) and a reduced weight of 370 g in 2017 (Figure II.3). + +The production model of the 8K UHD camera head has a super-35-mm CMOS imaging sensor that is smaller than the 2.5-inch CMOS image sensor and offers up to 120 frames per second (fps) at 8K UHD resolution of 7 680 $\times$ 4 320 pixels with a wide dynamic range ( $>$ 68 dB), very high sensitivity ( $>$ 1.6 V/lux-s), and low power consumption ( $<$ 3 W). A rigid endoscope is employed for medical applications using an 8K UHD camera head, because the rigid endoscope can be 8K friendly as it has a series of built-in relay lenses of high quality and a specific camera that can be mounted on the eyepiece. A new lens adapter has been uniquely developed as both ranges of the view of the 8K UHD camera and endoscope are adjusted absolutely when they are connected and operated with a single and manual focus. The outer diameter of the rigid endoscope is 10 mm, and inner relay lenses of 6 mm are incorporated in the endoscope. These relay lenses are specially selected for 8K UHD + +imaging, and the total f-number of the rigid endoscope is over 16. A fogless lens is installed on the tip of the 8K UHD rigid endoscope to defog an objective lens when inserted into an abdominal cavity. The fogless lens functions at all times due to the heat generated from a part of the xenon light. Using the control buttons on the 8K UHD camera head, a surgeon can operate the digital zooming ( $\times 1.0 \sim \times 4.0$ ) to change the endoscopic image from a normal image to a magnified image. + +In Japan, this production model of the 8K UHD endoscope was approved as a new medical device by the pharmaceuticals and medical devices agency (PMDA) on 15 September 2017. + +![Diagram of the 8K UHD endoscope showing its components and specifications.](045c7af3c493369642487e4d52e15065_img.jpg) + +The diagram shows a mass production model of the 8K UHD endoscope. It is composed of two main parts: the 8K camera head on the left and the 8K rigid endoscope on the right. + +**8K camera head specifications:** + +- Super-35 mm CMOS +- 33 million pixels, Dual-green filters +- 370 g, 75×75×170 mm + + The camera head has "Control buttons" and a "Lens adapter single and manual focus". A red arrow points to the connection point for "Xenon light". + +**8K rigid endoscope specifications:** + +- Ø10 mm, 0° and 30° angled +- Inner lenses: Ø6 mm +- F-number: 16 + + The tip of the endoscope is labeled as a "Fogless lens" with a circular inset image showing the lens tip. + +Diagram of the 8K UHD endoscope showing its components and specifications. + +F.780.1(18)\_FII.3 + +**Figure II.3 – An example of a mass production model of the 8K UHD endoscope** + +To downsize and minimize the weight of the 8K UHD camera head, a specially developed air cooling system was installed in the camera head. The camera case temperature was kept at less than 41°C within the safety standard temperature by the air cooling system. Therefore, the 8K camera could be designed to be a compact-size figure without a handle to be held in the surgeon's hand. + +### **II.3 8K UHD endoscopic image performance** + +This clause compares the performance between an 8K UHD endoscopic system and a conventional 2K endoscopic system. It presents the evaluation results of an 8K UHD endoscopic system regarding the resolution, zooming, and sensitivity. + +#### **II.3.1 8K UHD endoscopic image resolution** + +Figure II.4 shows examples of a surgical thread in the abdominal cavity of a pig for comparison between an 8K UHD endoscope and a conventional 2K endoscope. The images of a surgical 3-0 thread (200 ~ 249 µm) used in a standard laparoscopic surgery are shown in Figure II.4, of which the left-hand side is a 2K image, and the right-hand side is an 8K UHD image. In the 2K image, the outline and texture of the thread look ambiguous and jaggy. On the contrary, in the 8K UHD image, the outline and texture of the thread look clear and smooth. Furthermore, the capillaries on the small intestine can be observed in the 8K UHD image. In addition, a surgical 12-0 thread (1 ~ 9 µm), the thinnest thread used in microscopic surgery, was observed, and the thread was clear in the 8K UHD endoscopic image (Figure II.5). + +![Figure II.4: A side-by-side comparison of two endoscopic images. The left image, labeled '2K', shows a close-up of a surgical site with a 3-0 thread suture. The right image, labeled '8K', shows the same surgical site with a 3-0 thread suture, but with significantly higher resolution and clarity. A small text label 'F.780.1(18)_FII.4' is visible in the bottom right corner of the 8K image.](1ce750ac0d53b8e532abdeabf7740611_img.jpg) + +Figure II.4: A side-by-side comparison of two endoscopic images. The left image, labeled '2K', shows a close-up of a surgical site with a 3-0 thread suture. The right image, labeled '8K', shows the same surgical site with a 3-0 thread suture, but with significantly higher resolution and clarity. A small text label 'F.780.1(18)\_FII.4' is visible in the bottom right corner of the 8K image. + +**Figure II.4 – Surgical 3-0 thread images of the 2K endoscope (left) and 8K UHD endoscope (right)** + +![Figure II.5: An endoscopic image labeled '8K' showing a surgical site. A red rectangle highlights a specific area, which is magnified in a larger inset on the right. The inset shows a close-up of a thin suture thread, labeled '12-0 thread' with a white arrow pointing to it. A small text label 'F.780.1(18)_FII.5' is visible in the bottom right corner of the main image.](1d2cfd9625ccddd870b9c4e0dafccc8e_img.jpg) + +Figure II.5: An endoscopic image labeled '8K' showing a surgical site. A red rectangle highlights a specific area, which is magnified in a larger inset on the right. The inset shows a close-up of a thin suture thread, labeled '12-0 thread' with a white arrow pointing to it. A small text label 'F.780.1(18)\_FII.5' is visible in the bottom right corner of the main image. + +**Figure II.5 – Surgical 12-0 thread is also observed with an 8K UHD endoscope** + +#### **II.3.2 Zooming and sensitivity** + +Figure II.6 shows an example of a digital image zooming into the surgical field of an abdominal cavity. The 4-times magnified image of the surgical field is still clear in detail, and the resolution is good enough for the surgery. Figure II.7 also shows the improvement of image sensitivity in the surgical field compared between (a) the P1-model of the 8K endoscope with "double" xenon light sources and (b) the P2-model of the 8K UHD endoscope with a "single" xenon light source. The image brightness of the new 8K UHD endoscope is superior to that of the P1-model of the 8K endoscope despite the illumination being halved. This improvement is due to the 4-times sensitivity, gamma curve optimization, and noise reduction of the image sensor of the P2-model of the 8K UHD camera. + +![Figure II.6: Digital image zooming into the surgical field of an abdominal cavity. The image shows a split-screen view of an endoscopic surgical procedure. The top-left portion shows a wider view of the surgical field with a blue box highlighting a specific area. A blue arrow labeled 'Zoom' points from this box to a larger, magnified view of the same area on the right. Text in the top right corner indicates 'Digital zoom × 1.0 ~ 4.0'. The magnified view on the right shows a close-up of the surgical instruments and tissue, with a '×4.0' label in the bottom right corner. The file name 'F.780.1(18)_FII.6' is visible at the bottom right of the image frame.](84cebc242b1af66723a02cdf67fc98c9_img.jpg) + +Figure II.6: Digital image zooming into the surgical field of an abdominal cavity. The image shows a split-screen view of an endoscopic surgical procedure. The top-left portion shows a wider view of the surgical field with a blue box highlighting a specific area. A blue arrow labeled 'Zoom' points from this box to a larger, magnified view of the same area on the right. Text in the top right corner indicates 'Digital zoom × 1.0 ~ 4.0'. The magnified view on the right shows a close-up of the surgical instruments and tissue, with a '×4.0' label in the bottom right corner. The file name 'F.780.1(18)\_FII.6' is visible at the bottom right of the image frame. + +**Figure II.6 – Digital image zooming into the surgical field of an abdominal cavity** + +![Figure II.7: Comparison of surgical field images obtained by (a) the first model 8K UHD endoscope and (b) the second model 8K endoscope. The image shows two side-by-side endoscopic views of a surgical procedure. The left image (a) shows a wider view of the surgical field with a red arrow pointing to the right image (b), which is a magnified view of the same area. The file name 'F.780.1(18)_FII.7' is visible at the bottom right of the image frame.](82c2ebb7e3c8b55a525d6b91748243f4_img.jpg) + +Figure II.7: Comparison of surgical field images obtained by (a) the first model 8K UHD endoscope and (b) the second model 8K endoscope. The image shows two side-by-side endoscopic views of a surgical procedure. The left image (a) shows a wider view of the surgical field with a red arrow pointing to the right image (b), which is a magnified view of the same area. The file name 'F.780.1(18)\_FII.7' is visible at the bottom right of the image frame. + +**Figure II.7 – Comparison of surgical field images obtained by (a) the first model 8K UHD endoscope and (b) the second model 8K endoscope** + +### **II.4 Clinical case study** + +#### **II.4.1 Four cases of cholecystectomy using the P2-model of 8K UHD endoscopic system** + +A study was conducted where the P2-model of the 8K UHD endoscopic system was evaluated during four cases of cholecystectomy performed in Kyorin university hospital (Tokyo, Japan) as described in the following: + +- The first patient was a female in her sixties with right-sided hypochondriac pain; +- The second patient was a female in her seventies with epigastric pain; +- The third patient was a female in her seventies with right-sided stomach pain; +- The fourth patient was a male in his fifties with epigastric and back pain and was diagnosed with choledocholithiasis. + +The four cases of cholecystectomy, similar to the two cases of cholecystectomy in 2014 [b-Yamashita], were performed by the same surgeons and in the same operating room at Kyorin University Hospital. An 8K LCD was placed adjacent to the operating table so that the surgeon could stand near, and in front of, the 8K LCD as much as possible because the 8K image is viewed with the best quality when it is in front from a distance of 0.75 times the height of the display as shown in Figure II.8. + +![Figure II.8: Surgeon's view of 8K UHD LCD in the operation room. The image shows a surgical team in blue scrubs and caps performing a procedure. In the foreground, a large 85-inch 8K UHD LCD monitor displays a high-resolution endoscopic view of the surgical site. To the right, a smaller conventional 30-inch 2K LCD monitor shows a lower-resolution view. The surgeons are focused on the larger monitor.](7adda64806b35031fadb1fd5a227a9f8_img.jpg) + +Figure II.8: Surgeon's view of 8K UHD LCD in the operation room. The image shows a surgical team in blue scrubs and caps performing a procedure. In the foreground, a large 85-inch 8K UHD LCD monitor displays a high-resolution endoscopic view of the surgical site. To the right, a smaller conventional 30-inch 2K LCD monitor shows a lower-resolution view. The surgeons are focused on the larger monitor. + +F.780.1(18)\_FII.8 + +**Figure II.8 – Surgeon's view of 8K UHD LCD in the operation room** + +Figure II.9 shows the P2-model of the 8K UHD endoscope used during a cholecystectomy. One laparoscopist held the 8K UHD endoscope covered with a sterilized drape and monitored the surgical field in the patient's abdominal cavity. + +![Figure II.9: 8K UHD endoscope in a cholecystectomy. The image shows a close-up of a surgeon's hands in blue gloves holding a long, thin 8K UHD endoscope. The endoscope is covered with a clear, sterilized plastic drape. The tip of the endoscope is inserted into a patient's abdomen, which is draped with blue sterile cloth.](483c43eac5aa582dfcb6e0aa8cee163f_img.jpg) + +Figure II.9: 8K UHD endoscope in a cholecystectomy. The image shows a close-up of a surgeon's hands in blue gloves holding a long, thin 8K UHD endoscope. The endoscope is covered with a clear, sterilized plastic drape. The tip of the endoscope is inserted into a patient's abdomen, which is draped with blue sterile cloth. + +F.780.1(18)\_FII.9 + +**Figure II.9 – 8K UHD endoscope in a cholecystectomy** + +#### **II.4.2 Clinical evaluation of a cholecystectomy** + +In the cholecystectomy of the two patients operated on with the P1-model of the 8K UHD endoscope, the surgeons and other surgical staff reported the following positive evaluations of the 8K UHD endoscope: + +- With sufficient illumination, laparoscopic images with quite high resolution were viewed; +- 8K laparoscopic images were excellent in reproducing the appearances of solidity and reality; +- Vessels on the multilayer membrane around the bile duct and the gallbladder were clearly recognized; +- Viewing of 8K laparoscopic images caused less eye strain. + +However, the following limitations were also reported: + +- The heavy weight of the 8K camera head made it difficult to hold it in position even by two laparoscopists; +- The large size of the 8K camera head interfered with the surgical field; + +- Without sufficient illumination, darkness and low quality of colour reproduction of the images made laparoscopic surgery impossible. + +The P2-model of the 8K UHD endoscope was successfully used to complete the four cases of cholecystectomy. + +The P2-model of the 8K endoscope was able to solve the issues raised above. The P2-model of the 8K UHD endoscope is light enough for one laparoscopist to hold it in position, and the compact 8K camera head did not interfere with the surgical field. Furthermore, the laparoscopic view was sufficiently bright with a single 300-W xenon light source. + +The operating surgeons and other members of the surgical staff reported new evaluations of the 8K UHD endoscope and provided the following advice: + +- The fogless lens was a very useful system because the conventional endoscope was taken out of the abdominal cavity many times and the operation was stopped when the tip of the endoscope became cloudy; +- The 8K laparoscopic images had a more natural 3D effect than that of the 3D endoscope with 2K resolution; +- The wide colour gamut of the 8K images covered an infrared area and made it possible to recognize the different colours of the texture in detail; +- The 8K endoscopic system is applicable to surgical fields involving fine sutures in addition to laparoscopic surgery; +- The microscopic images of lymph vessels and membrane structure were clearly visible. New knowledge of anatomy could be obtained; +- In mediastinoscopic surgery, the degree of freedom for forceps increased by moving the field of view backward, so there was the benefit of reducing one forceps; +- The 8K endoscopic system led to a reduction in the time required for taking instruments in and out since the switching of the wide image to a magnified image was done freely; +- The blurring of the HD image was eliminated in combination with the robot arm. + +#### **II.4.3 Improvement of 8K UHD camera cable using durable and noiseless plastic optical fibre** + +The above-mentioned cable between the 8K UHD camera head and the camera control unit (CCU) includes glass optical fibres for transmitting the 8K UHD video signal. It is reasonable and useful to use glass optical fibres for transmitting a large amount of data over a long distance. On the other hand, they perform poorly with respect to bending motion and external force, such as when they are trampled or run over by a surgical table or surgical instrument cart in the operating room. When part of the glass fibre is broken, it becomes impossible to transmit the video signal. + +By replacing glass fibres with POF ([b-Koike]), the 8K UHD camera was resilient against external force. Graded-index multimode (GI) plastic optical fibres (POF) possesses the material characteristics of being highly flexible, difficult to break, easy to handle due to its large diameter, and not requiring high-precision alignment. GI POF achieves high-speed transmission (40 Gbit/s@1.55 μm) exceeding silica-based multimode fibre (Figure II.10). + +In addition, a special connector with 12 GI-POF cables with a ball lens on each cable end was developed (Figure II.11). As each GI-POF cable transmits up to 10 Gbps, the total transmission speed increases to 120 Gbps maximum. The unique point of the connector is the method of attaching a ball lens on each fibre end. Technologies that were used to produce ballpoint pens were applied and the connector was made at a low cost with precise alignment of the ball and the fibre. + +A prototype was developed where GI-POF for the 8K UHD camera cable is connected to the CCU and a ballpoint-pen interconnection, and to output an 8K UHD real-time image on the 8K UHD LCD (Figure II.12). + +![Figure II.10: Graded index-plastic optical fibre (GI-POF) interface for 8K UHD video transmission. The diagram shows a camera control unit for 8K endoscope connected to an 8K endoscope camera via a GI-POF connector with a ballpoint pen interconnect. The GI-POF is connected to an optical module for GI-POF via a connector with butt-coupling. The optical module for GI-POF is connected to a connector with butt-coupling, which is connected to an optical module for GI-POF. The diagram is labeled F.780.1(18)_FII.10.](339be989b91d5b1e73e5ecdc8401ca75_img.jpg) + +Camera control unit for 8K endoscope + +8K endoscope camera + +GI-POF connector with Ballpoint pen interconnect + +GI-POF + +Connector with butt-coupling + +Optical module for GI-POF + +GI-POF connector with Ballpoint pen interconnect (Plug and receptacle) + +Connector with butt-coupling + +Optical module for GI-POF + +F.780.1(18)\_FII.10 + +Figure II.10: Graded index-plastic optical fibre (GI-POF) interface for 8K UHD video transmission. The diagram shows a camera control unit for 8K endoscope connected to an 8K endoscope camera via a GI-POF connector with a ballpoint pen interconnect. The GI-POF is connected to an optical module for GI-POF via a connector with butt-coupling. The optical module for GI-POF is connected to a connector with butt-coupling, which is connected to an optical module for GI-POF. The diagram is labeled F.780.1(18)\_FII.10. + +**Figure II.10 – Graded index-plastic optical fibre (GI-POF) interface for 8K UHD video transmission** + +![Figure II.11: Ballpoint-pen interconnection. The figure includes a photograph of a liquid type ballpoint-pen, a diagram of a ball lens connected to a GI-POF, and two graphs comparing conventional and ball lens connections. The conventional connection requires precise alignment, while the ball lens connection has a gap tolerance and rough axial alignment. The graphs show evaluated loss (dB) vs. axial misalignment (μm) and gap tolerance (μm). A red box at the bottom states: 'Ballpoint-pen interconnect => A low-cost easy connection using the flexibility of the ingredient is achieved.' The figure is labeled F.780.1(18)_FII.11.](c53adc66eab40abd5f8b9107d950ad04_img.jpg) + +Conventional connection (physical contact + precise alignment) + +Ball lens connection (gap tolerance + rough axial alignment) + +Fibre core + +Optical fibre + +Spherical collimator lens + +Liquid type ballpoint-pen + +Ball lens + +GI-POF + +Evaluated loss [dB] + +with lens (Y axis) + +without lens (Y axis) + +Axial misalignment [μm] + +Gap tolerance [μm] + +with lens (Z axis) + +without lens (Z axis) + +**Ballpoint-pen interconnect => A low-cost easy connection using the flexibility of the ingredient is achieved.** + +F.780.1(18)\_FII.11 + +Figure II.11: Ballpoint-pen interconnection. The figure includes a photograph of a liquid type ballpoint-pen, a diagram of a ball lens connected to a GI-POF, and two graphs comparing conventional and ball lens connections. The conventional connection requires precise alignment, while the ball lens connection has a gap tolerance and rough axial alignment. The graphs show evaluated loss (dB) vs. axial misalignment (μm) and gap tolerance (μm). A red box at the bottom states: 'Ballpoint-pen interconnect => A low-cost easy connection using the flexibility of the ingredient is achieved.' The figure is labeled F.780.1(18)\_FII.11. + +**Figure II.11 – Ballpoint-pen interconnection** + +![A photograph showing a prototype of an 8K UHD endoscope system. A white rectangular unit (CCU) sits on top of a red rectangular unit. A thick, braided cable (GI-POF cable) connects them. The cable is plugged into a transparent ballpoint-pen-like interconnection device. This device is connected to a large monitor displaying a high-resolution image of a red, textured object (possibly a strawberry). The monitor shows '8K' in the top left corner and 'ZOOM x1.0' and 'A.GAIN -3' in the bottom right corner. The setup is on a wooden table with a small '8K' sticker.](19fd552435a80d0ffda518b710d16908_img.jpg) + +A photograph showing a prototype of an 8K UHD endoscope system. A white rectangular unit (CCU) sits on top of a red rectangular unit. A thick, braided cable (GI-POF cable) connects them. The cable is plugged into a transparent ballpoint-pen-like interconnection device. This device is connected to a large monitor displaying a high-resolution image of a red, textured object (possibly a strawberry). The monitor shows '8K' in the top left corner and 'ZOOM x1.0' and 'A.GAIN -3' in the bottom right corner. The setup is on a wooden table with a small '8K' sticker. + +F.780.1(18)\_FII.12 + +**Figure II.12 – Prototype of the 8K UHD endoscope cable including a GI-POF cable, which connects the 8K UHD camera head and the CCU with a ballpoint-pen interconnection** + +## Appendix III + +### Real-time transmission of uncompressed UHD images + +(This appendix does not form an integral part of this Recommendation.) + +### III.1 Method of measuring the time delay of transmission of uncompressed UHD images + +- 1) A tester for measuring time delay is connected via HDMI to the LCD. The LCD is then measured through the interpretation of the test pattern shown in an image captured by the installed image sensor. +- 2) The image is taken with an 8K endoscope camera. +- 3) The test pattern of the lag tester is shown on an 8K display after the transmission by a real-time video transmission system. +- 4) The amount of delay of the overall system is measured by reading the test pattern with the image sensor of the lag tester. +- 5) The time delay of the real-time image transmission system is calculated by subtracting the delay of the LCD for measurement (point 1 listed above) from the overall delay. + +### III.2 Time delay in local connection between the 8K UHD camera and the 8K UHD LCD + +The delay time of the liquid crystal display (LCD) is calculated by subtracting the time delay of the system from the overall delay. The top and bottom of the display patterns are, respectively, the fastest and slowest parts shown. The results for the measurements at the top, middle and bottom were: 0.6 ms, 8.5 ms, and 15.4 ms, respectively. + +### III.3 Time delay in local connection between the 8K UHD camera and the 8K UHD LCD + +When the delay was measured in a situation similar to that of an operating room environment, where an 8K endoscope camera was directly connected to an 8K display, the delay was 82.1 ms at the top of the display, 82.4 ms at the middle and 84.4 ms at the bottom, for an average of 82.3 ms. With an 8K display alone, the delay was 30.8 ms at the top of the display, 34.2 at the middle and 38.2 ms at the bottom, a delay of roughly two frames. See Figure III.1. + +![Diagram showing the signal flow from an 8K UHD camera to an 8K UHD LCD. The camera connects via Optical fibre to a CCU. The CCU connects via 3G-SDI x 8 to an FC8216 unit. The FC8216 unit connects via HDMI 2.0 x 4 to the 8K UHD LCD. Labels include '8K-DG -> HDMI' and 'F.780.1(18)_FIII.1'.](a8050baf48cc2c8e25d5ea2d1a67ef39_img.jpg) + +``` + +graph LR + Camera[8K UHD camera] -- Optical fibre --> CCU[CCU] + CCU -- 3G-SDI x 8 --> FC8216[FC8216] + FC8216 -- HDMI 2.0 x 4 --> LCD[8K UHD LCD] + subgraph Conversion + FC8216 + end + style Conversion fill:none,stroke:none + note right of FC8216: 8K-DG -> HDMI + +``` + +8K-DG -> HDMI +F.780.1(18)\_FIII.1 + +Diagram showing the signal flow from an 8K UHD camera to an 8K UHD LCD. The camera connects via Optical fibre to a CCU. The CCU connects via 3G-SDI x 8 to an FC8216 unit. The FC8216 unit connects via HDMI 2.0 x 4 to the 8K UHD LCD. Labels include '8K-DG -> HDMI' and 'F.780.1(18)\_FIII.1'. + +**Figure III.1 – Local connection between 8K UHD camera and 8K UHD LCD** + +### III.4 Time delay in QG encoding / decoding transmission + +Next, delay time was measured in a case where an 8K endoscope camera image was displayed on an 8K display after being encoded for IP transmission from a 3G-SDI×8ch dual (dual green) signal, transmitted through four optical fibres in a local environment, and then decoded into a 3G-SDI×8ch (dual green) signal. In the encoding / decoding process, four 8K / 4K / HD uncompressed transmission + +devices were used synchronously with an RS-232 connection. The time needed for encoding / decoding with one device alone was about 19 ms. + +The delay time of the system was 101 ms at the top of the display, 101.6 ms at the middle and 101.4 ms at the bottom, for an average of 101.3 ms. See Figure III.2. + +![Figure III.2: Connection between 8K UHD camera and 8K UHD LCD through the encoding / decoding transmission. The diagram shows an 8K UHD camera connected to a CCU via an optical fibre. The CCU outputs 3G-SDI x 8 to four QG (Quantum) units. These units are connected to an RS-232C unit and a GENLock unit. The QG units output Optical fibre x 4 to another set of four QG units. These units are connected to another RS-232C unit and a GENLock unit. The output of these QG units is 3G-SDI x 8 to an FC8216 unit. The FC8216 unit outputs HDMI2.0 x 4 to an 8K UHD LCD. The diagram is labeled F.780.1(18)_FIII.2.](47a7beddcb8a1b7abdca746967e32bb4_img.jpg) + +Figure III.2: Connection between 8K UHD camera and 8K UHD LCD through the encoding / decoding transmission. The diagram shows an 8K UHD camera connected to a CCU via an optical fibre. The CCU outputs 3G-SDI x 8 to four QG (Quantum) units. These units are connected to an RS-232C unit and a GENLock unit. The QG units output Optical fibre x 4 to another set of four QG units. These units are connected to another RS-232C unit and a GENLock unit. The output of these QG units is 3G-SDI x 8 to an FC8216 unit. The FC8216 unit outputs HDMI2.0 x 4 to an 8K UHD LCD. The diagram is labeled F.780.1(18)\_FIII.2. + +**Figure III.2 – Connection between 8K UHD camera and 8K UHD LCD through the encoding / decoding transmission** + +### III.5 Time delay in IP network between Japan and Singapore + +Finally, delay was measured where an 8K endoscope camera image was converted to IP using a QG70 displayed on an 8K image after being sent from Japan to Singapore, and then sent back to Japan. + +The delay time of the system was 176.5 ms at the top of the display, 176.9 ms at the middle and 177.8 ms at the bottom, for an average of 177.1 ms. + +In addition, the round-trip transmission time between Japan and Singapore alone was 75.8 ms; when this delay was added to the delay time for QG encoding / decoding transmission calculated in the preceding paragraph (average of 101.3 ms), the overall delay time roughly matched the result calculated here. See Figure III.3 and Figure III.4. + +![Figure III.3: Connection between 8K UHD camera and 8K UHD LCD by way of a round trip between Japan (KAIT) and Singapore. The diagram shows an 8K UHD camera connected to a CCU via an optical fibre. The CCU outputs 3G-SDI x 8 to four QG (Quantum) units. These units are connected to an RS-232C unit and a GENLock unit. The QG units output Optical fibre x 4 to a cloud representing a round trip between KAIT and Singapore. The output of the cloud is Optical fibre x 4 to another set of four QG units. These units are connected to another RS-232C unit and a GENLock unit. The output of these QG units is 3G-SDI x 8 to an FC8216 unit. The FC8216 unit outputs HDMI2.0 x 4 to an 8K UHD LCD. The diagram is labeled F.780.1(18)_FIII.3.](f1091147d93cee4dfa88498610e395a7_img.jpg) + +Figure III.3: Connection between 8K UHD camera and 8K UHD LCD by way of a round trip between Japan (KAIT) and Singapore. The diagram shows an 8K UHD camera connected to a CCU via an optical fibre. The CCU outputs 3G-SDI x 8 to four QG (Quantum) units. These units are connected to an RS-232C unit and a GENLock unit. The QG units output Optical fibre x 4 to a cloud representing a round trip between KAIT and Singapore. The output of the cloud is Optical fibre x 4 to another set of four QG units. These units are connected to another RS-232C unit and a GENLock unit. The output of these QG units is 3G-SDI x 8 to an FC8216 unit. The FC8216 unit outputs HDMI2.0 x 4 to an 8K UHD LCD. The diagram is labeled F.780.1(18)\_FIII.3. + +**Figure III.3 – Connection between 8K UHD camera and 8K UHD LCD by way of a round trip between Japan (KAIT) and Singapore** + +![IP network diagram between Japan (KAIT) and Singapore. The diagram shows a path from UDX and Koganei through various routers (kote, note, kote (mx2010), Kote (asr99)) to Singapore. Data rates are shown in M (Megabits) and K (Kilobits).](5eb69662cc4fa7d0d49b4eb22951c204_img.jpg) + +The diagram illustrates an IP network path between Japan and Singapore. The path starts at the top left with UDX (0K) and Koganei (1334.2M) connected to a router labeled 'kote'. From 'kote', the path continues through another 'kote (mx2010)' router, then to 'Kote (asr99)', and finally reaches Singapore. The path is marked with various data rates in M (Megabits) and K (Kilobits). The 'note' router in the center is connected to NFV\_sapp (0K), KAIT (rou) (0.1K), and KAIT (0.2K). The Singapore node is highlighted with a red box. The diagram also includes a reference code F.780.1(23)\_FIII.4 in the bottom right corner. + +IP network diagram between Japan (KAIT) and Singapore. The diagram shows a path from UDX and Koganei through various routers (kote, note, kote (mx2010), Kote (asr99)) to Singapore. Data rates are shown in M (Megabits) and K (Kilobits). + +**Figure III.4 – IP network between Japan (marked as KAIT) and Singapore** + +### III.6 Discussion and conclusion + +In recent measurements, it was found that the delay is large even with a direct connection between an 8K endoscope camera and an 8K display because an 8K image by itself is particularly large. Because the existing systems are square-division systems that connect four 4K-resolution screens, it appears that much time is spent in buffering the data. It may be possible to reduce the delay time by using a two-sample interleave (2SI) system. In addition, the 19 ms of overhead for the QG70 used to convert to/from the IP transmitted signal is extremely small; this can be considered a strong point of the uncompressed transmission. + +Lastly, it is important to note that further efficiency in the low-latency exchange of uncompressed UHD images over a wide-area network can be achieved by all-optical, end-to-end path transmission through all the IP networking equipment in the network. In fact, uncompressed 8K UHD video generated at a rate of 72 Gbit/s has been successfully exchanged between Tokyo and Ibaraki, in Japan, over 173-km fibre distance via optically switched "dynamic optical path network" to transmit a remote interactive musical session, where the round-trip time of the uncompressed 60-fps full 8K image exchange was less than approximately 30 ms, including the processing time from camera capture, transmission, video display. Details are reported in [b-Kurosuj]. + +## Bibliography + +- [b-Bate] Bate, L., Cook, C. (2001), *The feasibility of force control over the Internet*, Proc. 2001 Australian Conference on Robotics and Automation, Sydney, 14-15 November. Pages 146-151. +<[https://www.academia.edu/70246320/The\\_Feasibility\\_of\\_Force\\_Control\\_Over\\_the\\_Internet](https://www.academia.edu/70246320/The_Feasibility_of_Force_Control_Over_the_Internet)> +- [b-JIS T 0601-1] Japanese Standards Association (2023), JIS T 0601-1, *Medical electrical equipment – Part 1: General requirements for basic safety and essential performance*. +<[https://global.ihs.com/doc\\_detail.cfm?document\\_name=JIS%20T%200601%2D1&item\\_s\\_key=00341940](https://global.ihs.com/doc_detail.cfm?document_name=JIS%20T%200601%2D1&item_s_key=00341940)> +- [b-Koike] Koike, Y. (2014), *Fundamentals of Plastic Optical Fibers*, Wiley-VCH. +<> +- [b-Kurosu] Kurosu, T., Suda, S., Kakibayashi, H., Solis-Trapala, K., Kurumida, J., Ishii, K., Tanizawa, K., and Namiki, S. (2015), *What is the true value of dynamic optical path switching?* Opto-Electronics and Communications Conference. +<[https://www.researchgate.net/publication/308813317\\_What\\_is\\_the\\_true\\_value\\_of\\_dynamic\\_optical\\_path\\_switching](https://www.researchgate.net/publication/308813317_What_is_the_true_value_of_dynamic_optical_path_switching)> +- [b-SMPTE 292M] SMPTE ST 292-1:2012 – SMPTE Standard – *1.5 Gb/s Signal/Data Interface*. +<> +- [b-Yamashita] Yamashita, H., Aoki, H., Tanioka, K., Mori, T., and Chiba, T. (2016), *Ultra-high definition (8K UHD) endoscope: our first clinical success*, SpringerPlus, 5:1445. +<> + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.780.3-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.780.3-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..abda52cbb1047d6649f89e40c898a1c33bed7267 --- /dev/null +++ b/marked/F/T-REC-F.780.3-202212-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6f3b7e07231c0f3e116c99c493bdda98ecd52313b6735faa6cea23815cf3162b +size 7190 diff --git a/marked/F/T-REC-F.780.3-202212-I_PDF-E/a24e89a6fe9bb70c83f8bf5202baba95_img.jpg b/marked/F/T-REC-F.780.3-202212-I_PDF-E/a24e89a6fe9bb70c83f8bf5202baba95_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..bae139b595cbc0579c0c5c71626878b1a4927b17 --- /dev/null +++ b/marked/F/T-REC-F.780.3-202212-I_PDF-E/a24e89a6fe9bb70c83f8bf5202baba95_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d7fb97274c8c8dcd660746ca42ff882eea26995fc82cd1e7b0efa835de6ef5db +size 26230 diff --git a/marked/F/T-REC-F.780.3-202212-I_PDF-E/b8661c6c54f72ecc7ff6cb05e47b2891_img.jpg b/marked/F/T-REC-F.780.3-202212-I_PDF-E/b8661c6c54f72ecc7ff6cb05e47b2891_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1dac545e9210bcc2583187a36f41c3e964a24466 --- /dev/null +++ b/marked/F/T-REC-F.780.3-202212-I_PDF-E/b8661c6c54f72ecc7ff6cb05e47b2891_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:744f30a5e53ea8c3793de5c0ef56c8aa04c681decad16de588e2307c1a9c77e7 +size 24071 diff --git a/marked/F/T-REC-F.780.3-202212-I_PDF-E/e9314c83043183351ed74908e9bf2f90_img.jpg b/marked/F/T-REC-F.780.3-202212-I_PDF-E/e9314c83043183351ed74908e9bf2f90_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..055830e07508a3eb94a3e3a89e82e3f20dc3c615 --- /dev/null +++ b/marked/F/T-REC-F.780.3-202212-I_PDF-E/e9314c83043183351ed74908e9bf2f90_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ee6af11cf8710b359ff4925b259e68054e07add3da919cd3c3535ecc4782a412 +size 40605 diff --git a/marked/F/T-REC-F.780.4-202309-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.780.4-202309-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..25f8fbfaef317942ed3102c74b40a7d42f80608b --- /dev/null +++ b/marked/F/T-REC-F.780.4-202309-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:83cab7dcabfe939b223d97242d8523e461968d21f7c5432693c30a64f2d95521 +size 7050 diff --git a/marked/F/T-REC-F.780.4-202309-I_PDF-E/cfda9df1319e04207eb28bcefd1dab7b_img.jpg b/marked/F/T-REC-F.780.4-202309-I_PDF-E/cfda9df1319e04207eb28bcefd1dab7b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..38bc6dccc49786ac4dc34168f9ccdeb34eb8d538 --- /dev/null +++ b/marked/F/T-REC-F.780.4-202309-I_PDF-E/cfda9df1319e04207eb28bcefd1dab7b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7ab2f33480f3979e4abe6be55b2fbd592b1bad3066798e189b8da5ef010d7940 +size 143144 diff --git a/marked/F/T-REC-F.780.5-202401-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/F/T-REC-F.780.5-202401-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c1629394c02f697f60ae954ad36bf274fa8082cb --- /dev/null +++ b/marked/F/T-REC-F.780.5-202401-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3763acd7a0ed56aa30f3f90468c1fb0cb54573eb9045bca3eddf8976b4f0f41d +size 7219 diff --git a/marked/F/T-REC-F.780.5-202401-I_PDF-E/366a77fdefb0097b3289b4a011911390_img.jpg b/marked/F/T-REC-F.780.5-202401-I_PDF-E/366a77fdefb0097b3289b4a011911390_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7fd3dbb08b1d48c4820ca5b3459daa1a8d06438d --- /dev/null +++ b/marked/F/T-REC-F.780.5-202401-I_PDF-E/366a77fdefb0097b3289b4a011911390_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ab94f2d95bc3e942907ad0db75c1a5a21d7a4ca621b0771aafff471761cc06b9 +size 38054 diff --git a/marked/F/T-REC-F.780.5-202401-I_PDF-E/5b8a756d9a71c35f17db8bcb90b438a3_img.jpg b/marked/F/T-REC-F.780.5-202401-I_PDF-E/5b8a756d9a71c35f17db8bcb90b438a3_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..68dfa61a7bfc5bd73cf8af31f827ad4549149457 --- /dev/null +++ b/marked/F/T-REC-F.780.5-202401-I_PDF-E/5b8a756d9a71c35f17db8bcb90b438a3_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d4d40cb48f0fc545f4b033a16875373aa2291d584563842195755c214aef80ac +size 66393 diff --git a/marked/F/T-REC-F.780.5-202401-I_PDF-E/5cab96b2d23174c25919840ecd50aa48_img.jpg b/marked/F/T-REC-F.780.5-202401-I_PDF-E/5cab96b2d23174c25919840ecd50aa48_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9d8333160393bc711177e1ae2004e6bc5e7b2ca8 --- /dev/null +++ b/marked/F/T-REC-F.780.5-202401-I_PDF-E/5cab96b2d23174c25919840ecd50aa48_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cf3fd2a85e2e518dfd02caa90a955bb4be7b98073668beb57b1c80b2cbe8b6b2 +size 128851 diff --git a/marked/F/T-REC-F.780.5-202401-I_PDF-E/7133ccf78043568ca62ecbcd43628a4a_img.jpg b/marked/F/T-REC-F.780.5-202401-I_PDF-E/7133ccf78043568ca62ecbcd43628a4a_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3e5f7425cc182d86fa397e7603c74dca3b69bdf9 --- /dev/null +++ b/marked/F/T-REC-F.780.5-202401-I_PDF-E/7133ccf78043568ca62ecbcd43628a4a_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:784e821a76f65ad7e57f03271d2873ec5f59b0ff62fd20c60d716340d539af1c +size 48693 diff --git a/marked/F/T-REC-F.780.5-202401-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg b/marked/F/T-REC-F.780.5-202401-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1a66fcb24a8ae52e80a0ca6a6453c7ef8a95fb8d --- /dev/null +++ b/marked/F/T-REC-F.780.5-202401-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2cd6ee05e7b34400e131d4f79d2dc6ece3d497566e72c37dda91465ee40046ed +size 40610 diff --git a/marked/F/T-REC-F.780.5-202401-I_PDF-E/8307f6b04df072c9332f9987e034272c_img.jpg b/marked/F/T-REC-F.780.5-202401-I_PDF-E/8307f6b04df072c9332f9987e034272c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d595b6f80c3417aa351b8eaa696f816f8b06aaae --- /dev/null +++ b/marked/F/T-REC-F.780.5-202401-I_PDF-E/8307f6b04df072c9332f9987e034272c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:569350a8038f87c3c5263817fff216e00c3181c178f12b2d28d786d138125ee6 +size 37216 diff --git a/marked/F/T-REC-F.780.5-202401-I_PDF-E/898fb89a50d9ec1dfb4e425c816976a7_img.jpg b/marked/F/T-REC-F.780.5-202401-I_PDF-E/898fb89a50d9ec1dfb4e425c816976a7_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..465ff7f1a909467580d80902faa0856966a44198 --- /dev/null +++ b/marked/F/T-REC-F.780.5-202401-I_PDF-E/898fb89a50d9ec1dfb4e425c816976a7_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e6143df9677b863f4f60e13d8555056a055b78a9e6e7151d5aea4513580c9c5a +size 22503 diff --git a/marked/F/T-REC-F.780.5-202401-I_PDF-E/9c6461e1e94afae4dec455e69a2ce152_img.jpg b/marked/F/T-REC-F.780.5-202401-I_PDF-E/9c6461e1e94afae4dec455e69a2ce152_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f5d27a8b76ee7e9c8a732343b6cea581e7ede0d1 --- /dev/null +++ b/marked/F/T-REC-F.780.5-202401-I_PDF-E/9c6461e1e94afae4dec455e69a2ce152_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:923de9268cacec81738f8ea5724c4def26b9fa344adb4c0de0ac173c51b5b52e +size 44805 diff --git a/marked/F/T-REC-F.780.5-202401-I_PDF-E/c5452f95f3b28f1bfe29e84fbc2e1267_img.jpg b/marked/F/T-REC-F.780.5-202401-I_PDF-E/c5452f95f3b28f1bfe29e84fbc2e1267_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..6d9df24a0a2987d233d12d45e88f0cd249ec6d5e --- /dev/null +++ b/marked/F/T-REC-F.780.5-202401-I_PDF-E/c5452f95f3b28f1bfe29e84fbc2e1267_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:89c0b4f5b74bb980272a3259e774ae26df78fdc7fa547de424089a70e133266d +size 64851 diff --git a/marked/F/T-REC-F.780.5-202401-I_PDF-E/c5655e700cc3e9aac7e9f4f07f30264d_img.jpg b/marked/F/T-REC-F.780.5-202401-I_PDF-E/c5655e700cc3e9aac7e9f4f07f30264d_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1a1d41dd4d39663e8ef42575ce10ae107181ec76 --- /dev/null +++ b/marked/F/T-REC-F.780.5-202401-I_PDF-E/c5655e700cc3e9aac7e9f4f07f30264d_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bd8aa9c72fb6518bf2d55e0aecf077f8528fd0ddaf0ee69b703dd5ab37ad051d +size 69729 diff --git a/marked/F/T-REC-F.780.5-202401-I_PDF-E/c914f51f4427bc672dd0526cfc90ebe9_img.jpg b/marked/F/T-REC-F.780.5-202401-I_PDF-E/c914f51f4427bc672dd0526cfc90ebe9_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..131a3983af3ae1b03a55b80043d5c6d873a72dfd --- /dev/null +++ b/marked/F/T-REC-F.780.5-202401-I_PDF-E/c914f51f4427bc672dd0526cfc90ebe9_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6af818d1dfad3e321d516057fa2cb0b8cea459354cbf795f05abd63826074367 +size 49571 diff --git a/marked/F/T-REC-F.780.5-202401-I_PDF-E/d53cd0fd1cf896a9353fd63de1505ba2_img.jpg b/marked/F/T-REC-F.780.5-202401-I_PDF-E/d53cd0fd1cf896a9353fd63de1505ba2_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..831e3732a38854f02df81bbb25df88c67ce3db64 --- /dev/null +++ b/marked/F/T-REC-F.780.5-202401-I_PDF-E/d53cd0fd1cf896a9353fd63de1505ba2_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9618fe576b11d33cebde8840dc25740770277d9b89882102a74d8c243bdb56e1 +size 53617 diff --git a/marked/F/T-REC-F.780.5-202401-I_PDF-E/e69b9188aa2c14ec6b21c83f711fef65_img.jpg b/marked/F/T-REC-F.780.5-202401-I_PDF-E/e69b9188aa2c14ec6b21c83f711fef65_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7982051b2792c3c4d5494cd5634de55b8f23fefc --- /dev/null +++ b/marked/F/T-REC-F.780.5-202401-I_PDF-E/e69b9188aa2c14ec6b21c83f711fef65_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b0dd4352c444a458c870a82464f3869c6afe054b06c507d89b8f531ff02f43d6 +size 53037 diff --git a/marked/F/T-REC-F.780.5-202401-I_PDF-E/raw.md b/marked/F/T-REC-F.780.5-202401-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..3e03ee38f716faa7e75280463a04f015841bc323 --- /dev/null +++ b/marked/F/T-REC-F.780.5-202401-I_PDF-E/raw.md @@ -0,0 +1,692 @@ + + +# Recommendation **ITU-T F.780.5 (01/2024)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +## **Requirements, reference framework and use cases for telemonitoring systems in rapid deployment hospitals** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with white lines representing latitude and longitude, and the letters 'ITU' in a bold, blue, sans-serif font overlaid on the globe. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.780.5 + +# Requirements, reference framework and use cases for telemonitoring systems in rapid deployment hospitals + +## Summary + +Recommendation ITU-T F.780.5 describes the application scenarios, functional requirements, and reference architecture of telemonitoring systems in rapid deployment hospitals (RDHs) and applies them to planning and designing in RDHs. The appendix to this Recommendation includes use cases of the proposed reference system. + +## History\* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.780.5 | 2024-01-13 | 16 | 11.1002/1000/15763 | + +## Keywords + +Rapid deployment hospital, reference framework, requirement, telemonitoring, use cases. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2024 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|------|----------------------------------------------------------------------------------|----| +| 1 | Scope..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 2 | +| 3.1 | Terms defined elsewhere ..... | 2 | +| 3.2 | Terms defined in this Recommendation ..... | 3 | +| 4 | Abbreviations and acronyms ..... | 3 | +| 5 | Conventions ..... | 4 | +| 6 | Background..... | 5 | +| 7 | Requirements for telemonitoring systems in rapid deployment hospitals ..... | 5 | +| 7.1 | System functional requirements ..... | 5 | +| 7.2 | Requirements for application platforms ..... | 6 | +| 7.3 | Requirements for telemonitoring terminals ..... | 7 | +| 7.4 | Requirements for telemonitoring networks ..... | 8 | +| 7.5 | Requirements for telemonitoring cloud platform ..... | 9 | +| 7.6 | Compatibility requirements ..... | 9 | +| 7.7 | Scalability requirements ..... | 9 | +| 7.8 | Requirements for special needs ..... | 9 | +| 8 | Reference framework for telemonitoring system in rapid deployment hospital ..... | 9 | +| 8.1 | Reference architecture of the overall system designing ..... | 9 | +| 8.2 | Reference architecture of the network layer ..... | 10 | +| 8.3 | Reference architecture of private cellular network for the RDH..... | 11 | +| 8.4 | Reference architecture of non-3GPP network optimization..... | 12 | +| 8.5 | Reference architecture of the cloud-edge collaboration ..... | 13 | +| | Appendix I – Use cases ..... | 14 | +| I.1 | Introduction ..... | 14 | +| I.2 | Local rapid networking of the RDH ..... | 14 | +| I.3 | Establishment of telecommunication in non-regular scenarios ..... | 15 | +| I.4 | Remote IoT data collection ..... | 16 | +| I.5 | Remote live video monitoring ..... | 17 | +| I.6 | Non-real-time data transferring ..... | 17 | +| I.7 | Tele-radiotherapy application scenario in RDH ..... | 17 | +| I.8 | Private IMT-2020 network for the RDH ..... | 18 | +| I.9 | FHIR-based interoperability for telemonitoring..... | 19 | +| I.10 | Digital twin technology for telemonitoring ..... | 19 | +| I.11 | Remote nursing services ..... | 19 | +| | Bibliography..... | 20 | + + + +# Recommendation ITU-T F.780.5 + +# Requirements, reference framework and use cases for telemonitoring systems in rapid deployment hospitals + +# 1 Scope + +This Recommendation describes the use cases, reference framework and technical requirements for telemonitoring systems in a rapid deployment hospitals (RDHs). The scope of this Recommendation includes: + +- Requirements for telemonitoring systems in rapid deployment hospitals. +- Reference framework for telemonitoring systems in rapid deployment hospitals. +- Use cases for telemonitoring systems in rapid deployment hospitals. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +- [ITU-T F.760.1] Recommendation ITU-T F.760.1 (2022), *Requirements and reference framework for emergency rescue systems*. +- [ITU-T F.780.2] Recommendation ITU-T F.780.2 (V2) (2023), *Accessibility of telehealth services*. +- [ITU-T H.860] Recommendation ITU-T H.860 (2014), *Multimedia e-health data exchange services: Data schema and supporting services*. +- [ITU-T Y.3500] Recommendation ITU-T Y.3500 (2014), *Information technology – Cloud computing – Overview and vocabulary*. +- [ITU-T Y.3502] Recommendation ITU-T Y.3502 (2014), *Information technology – Cloud computing – Reference architecture*. +- [ITU-T Y.3601] Recommendation ITU-T Y.3601 (2018), *Big data – Framework and requirements for data exchange*. +- [ITU-T Y.4101] Recommendation ITU-T Y.4101/Y.2067 (2017), *Common requirements and capabilities of a gateway for Internet of things applications*. +- [ITU-T Y.4208] Recommendation ITU-T Y.4208 (2020), *Internet of things requirements for support of edge computing*. +- [Bluetooth CS5.4] Bluetooth CS5.4 (2023), *Bluetooth Core Specification, v5.4*. +- [ETSI GS MEC 003] ETSI GS MEC 003 V3.1.1 (2022), *Multi-access Edge Computing (MEC): Framework and Reference Architecture*. +- [ETSI TS 123 501] ETSI TS 123 501 V17.10.0 (2023), *5G; System Architecture for the 5G System (5GS) (3GPP TS 23.501 version 17.10.0 Release 17)*. + +- [IEEE 802.11] IEEE 802.11 (2020), *IEEE Standard for Information Technology – Telecommunications and Information Exchange between Systems Local and Metropolitan Area Networks – Specific Requirements Part 11: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) Specifications*. +- [IEEE 802.15.4] IEEE 802.15.4 (2020), *IEEE Standard for Low-Rate Wireless Networks*. +- [IETF RFC 6120] IETF RFC 6120 (2011), *Extensible Messaging and Presence Protocol (XMPP): Core*. +- [IETF RFC 7252] IETF RFC 7252 (2014), *The Constrained Application Protocol (CoAP)*. +- [IETF RFC 8259] IETF RFC 8259 (2017), *The JavaScript Object Notation (JSON) Data Interchange Format*. +- [ISO 18308] ISO 18308:2022, *Health informatics – Requirements for an electronic health record architecture*. +- [ISO 19605-1] ISO 19605-1:2018, *Organization and digitization of information about buildings and civil engineering works, including building information modelling (BIM) – Information management using building information modelling – Part 1: Concepts and principles*. +- [ISO 23247-1] ISO 23247-1:2021, *Automation systems and integration – Digital twin framework for manufacturing – Part 1: Overview and general principles*. +- [ISO/IEC 19464] ISO/IEC 19464:2014, *Information technology, Advanced Message Queuing Protocol (AMQP) v1.0 specification*. +- [ISO/IEC 20922] ISO/IEC 20922:2016, *Information technology, Message Queuing Telemetry Transport (MQTT) v3.1.1*. + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 service level agreement (SLA)** [ITU-T Y.3500]: Documented agreement between the service provider and customer that identifies services and service targets. + +NOTE 1 – A service level agreement can also be established between the service provider and a supplier, an internal group or a customer acting as a supplier. + +NOTE 2 – A service level agreement can be included in a contract or another type of documented agreement. + +**3.1.2 Internet of things (IoT)** [ITU-T Y.4000]: A global infrastructure for the information society, enabling advanced services by interconnecting (physical and virtual) things based on existing and evolving interoperable information and communication technologies. + +NOTE 1 – Through exploiting identification, data capture, processing and communication capabilities, the IoT makes full use of things to offer services to all kinds of applications, ensuring that security and privacy requirements are fulfilled. + +NOTE 2 – From a broader perspective, the IoT can be perceived as a vision with technological and societal implications. + +**3.1.3 electronic health record** [ISO 18308]: It is a longitudinal electronic record of patient health information generated by one or more encounters in any care delivery settings. This information includes patient's demographics, progress notes, problems description, medication receipts, vital signs, past medical treatment history, immunizations, laboratory data and radiology reports. + +**3.1.4 personal health record** [b-ISO/TR 14292]: It is an electronic, universally available, lifelong health information resource for individuals who need to make health decisions. Individuals own and manage the information in the PHR, which comes from healthcare providers and the individual. + +**3.1.5 fast healthcare interoperability resources** [b-HL7 FHIR]: It is a set of rules and specifications for exchanging electronic healthcare data, such as EHR and PHR. It is designed to be flexible and adaptive and can be applied in a wide range of settings and different healthcare information systems. + +**3.1.6 user plane function** [ETSI TS 123 501]: The UPF is a user plane network element in IMT-2020 system architecture and is responsible for all user plane functions, including traffic routing and forwarding, policy enforcement. Depending on different application scenarios, UPFs with different function sets are derived, including roles such as PSA UPF and ULCL UPF. + +**3.1.7 building information modelling (BIM)** [ISO 19605-1]: The process of generating and managing building data through the whole life cycle of a built asset, from its initial design, and construction, to maintaining and de-commissioning. Implementing BIM facilitates project management and cross-disciplinary collaboration by sharing project information between engineers, owners, architects and contractors. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 rapid deployment hospital (RDH)**: The most important characteristic of RDH is rapid deployment to accept patients as soon as possible, including deployment of containers and facilities and deployment of computed tomography (CT) and information technology (IT) systems to support medical services. + +**3.2.2 telemonitoring**: A continuous or non-continuous monitoring process that allows a healthcare professional to remotely interpret the data necessary for a patient's medical follow-up and, if necessary, make decisions regarding the patient's state of health. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|----------|------------------------------------------------------| +| 5G | 5th Generation of Mobile Networks | +| 5GC | 5G Core Network | +| AMQP | Advanced Message Queuing Protocol | +| BIM | Building Information Modelling | +| CoAP | Constrained Application Protocol | +| COVID-19 | Corona Virus Disease 2019 | +| CT | Computed Tomography | +| DICOM | Digital Imaging and Communications in Medicine | +| DNN | Data Network Name | +| ECG | Electrocardiogram | +| EHR | Electronic Health Record | +| EMR | Electronic Medical Record | +| FHIR | Fast Healthcare Interoperability Resources | +| HTTPS | Hypertext Transfer Protocol over Secure Socket Layer | + +| | | +|---------|---------------------------------------------------| +| IaaS | Infrastructure as a Service | +| IEEE | Institute of Electrical and Electronics Engineers | +| IMT | International Mobile Telecommunications | +| IoT | Internet of Things | +| IT | Information Technology | +| LR-WPAN | Low-Rate Wireless Personal Area Network | +| MEC | Multi-access Edge Computing | +| MEP | MEC Platform | +| MQTT | Message Queuing Telemetry Transport | +| NTN | Non-terrestrial Networks | +| PaaS | Platform as a Service | +| PACS | Picture Archiving and Communication System | +| PCF | Policy Control Function | +| PHR | Personal Health Record | +| PLMN | Public Land Mobile Network | +| PSA | PDU Session Anchor | +| QoS | Quality of Service | +| RDH | Rapid Deployment Hospital | +| REST | Representational State Transfer | +| SaaS | Software as a Service | +| SIM | Subscriber Identity Module | +| SLA | Service Level Agreement | +| TAC | Type Allocation Code | +| UAV | Unmanned Aerial Vehicle | +| ULCL | Uplink Classifier | +| URI | Uniform Resource Identifier | +| VPN | Virtual Private Network | +| WLAN | Wireless Local Area Network | +| WPAN | Wireless Personal Area Network | +| XML | Extensible Markup Language | +| XMPP | Extensible Messaging and Presence Protocol | + +# 5 Conventions + +In this Recommendation: + +- The keywords "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. + +- The keywords "is recommended" and "should" indicate a requirement which is recommended but which is not absolutely required. Thus this requirement needs not be present to claim conformance. +- The keywords "can optionally" and "may" indicate an optional requirement which is permissible, without implying any sense of being recommended. These terms are not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +# 6 Background + +Both natural and complex disasters may produce a massive number of casualties that outstrip the ability of the local healthcare system to provide the necessary care. As a consequence, temporary hospitals, generally called field hospitals, mobile hospitals, or cabin hospitals, have been used to meet emergency requirements and have achieved notable success. + +As defined by WHO in 2003, a field hospital is a mobile, self-contained, self-sufficient healthcare facility capable of rapid deployment and expansion or contraction to meet immediate emergency requirements for a specified period. A field hospital can be set up in an existing structure or a temporary structure, tent, or similar. Field hospitals are deployed in a wide range of scenarios, including natural disasters, epidemic outbreaks, armed conflicts, and refugee crises. + +A mobile hospital is a medical centre or a small hospital with a full range of medical equipment that can be moved and settled in a new place and situation swiftly in order to provide medical services to patients or wounded persons in critical conditions such as wars or natural disasters. In particular, a mobile hospital is bigger than first-aid units and smaller than a permanent hospital. Also, the mobile hospital can be connected to another field hospitals to increase ward section capacity anytime and anywhere. + +Cabin hospitals are intended for large-scale medical isolation and are instituted either by establishing rapidly constructed modular/portable buildings or through the acquisition of indoor spaces within existing venues and even in temporarily renovated gyms and dorms in colleges and universities with enclosed cubicles to assist social distancing. + +Whether field hospital, mobile hospital, or cabin hospital, their most important characteristic is rapid deployment in order to accept patients as soon as possible, including not only deployment of containers and facilities, but also deployment of computed tomography (CT) and information technology (IT) systems to support medical services. It is recommended to unify the category name of these hospitals as rapid deployment hospitals (RDHs). + +Telemonitoring refers to the transmission of symptom scores, physiological data, including heart rate, blood pressure, oxygen saturation, and weight, directly to care providers either via automated electronic means or by web-based or phone-based data entry. Integration of monitoring technologies in smartphones and wireless devices is a key trend in patient care. + +Most RDHs face a shortage of doctors and nurses, and on the other hand, medical workers should avoid face-to-face contact with patients in epidemic scenarios. Telemonitoring technology in RDHs has thereby become a research hotspot. + +# 7 Requirements for telemonitoring systems in rapid deployment hospitals + +## 7.1 System functional requirements + +This clause will provide a detailed description of the requirements of a telemonitoring system: + +- a) It should support the modular deployment and prefabricated construction of hospitals in open areas such as suburbs and the wild, that can be relocated and reused in different places, and + +the transformation of existing infrastructure based on ships, large gymnasiums, buses, and campervans; + +- b) It should support remote data exchange such as e-health data exchange [ITU-T H.860], big data exchange [ITU-T Y.3601] and video interaction to provide interoperable exchange among the RDH and a remote central hospital; +- c) It should support a contactless medical environment to reduce or avoid cross-infection among patients and between patients and medical staff; +- d) It should support at least one kind of wide-area network (WAN) to provide basic telemonitoring communication services, such as public cellular network (e.g., 2G, IMT-2000, IMT-advanced, IMT-2020), microwave network, or satellite network; the dedicated network should be supported if the public cellular network were available, providing private communication resources for guaranteed service level agreement (SLA) performance under massive medical terminal connection scenarios; +- e) It should support multiple network connections for numerous category sensors and medical devices, such as a cellular network and non-3GPP network. The non-3GPP network includes wired network, wireless local area network (WLAN) [IEEE 802.11], wireless personal area network (WPAN) [Bluetooth CS5.4], and low-rate wireless personal area network (LR-WPAN) [IEEE 802.15.4]; +- f) It should support cloud-edge management to provide unified management on resource scheduling of the cloud platform, applications management, and services management, etc.; +- g) It should support various applications under different telemonitoring scenarios, such as tele guidance, teleconsultation, telemedicine, and tele-radiotherapy; +- h) It can optionally support building information modelling (BIM) [ISO 19605-1] technology during the RDH's deployment process, which enables three-dimensional visualization of building information and relative data to help reliable deployment for the RDH. + +## 7.2 Requirements for application platforms + +The requirements of telemonitoring application platforms are listed as follows: + +- a) It should support real-time data acquisition and monitoring of the use of hospital beds, drugs, medical equipment, and a series of equipment-related factors, such as heat dissipation, vibration, noise, safety, cross pollution, operation and maintenance of medical equipment; +- b) It should support real-time data acquisition and monitoring of the patients' multiple physiological parameter data, such as electrocardiogram (ECG), heart rate, blood pressure, blood sugar, blood oxygen saturation, and body temperature; +- c) It should support real-time data acquisition and monitoring of various environmental indicators, such as pressure, oxygen concentration, temperature, humidity, wind speed, noise, water quality and the concentration of solid particles in RDHs; +- d) It should support the local storage of collected data by law-compliant and standards-based electronic health records, such as electronic health record (EHR) [ISO 18308] and personal health record (PHR) [b-ISO/TR 14292]. It should be noted that collecting PHR data must be authorized by individuals within the legal framework; +- e) It should support different types of collected data (e.g., text, voice, image, and video) to be encoded in a common health data schema, such as extensible markup language (XML) [b-W3C XML 1.0], for data exchange [ITU-T H.860]; +- f) It should support the transmission of collected data between the RDH and remote central hospitals using fast healthcare interoperability resources (FHIR) [b-HL7 FHIR] representational state transfer (REST) APIs over a secure and encrypted channel (e.g., web service call over HTTPS); + +- g) It should support various telemonitoring applications, such as telemonitoring guidance, nursing, operation management, training and teaching, and multi-point telemonitoring; +- h) It can optionally support a digital twin [ISO 23247-1] implementation for telemonitoring, which replicates the physical model and continuously adapts to operational changes based on real-time data from various sensors and devices. Using its digital twin model will enable remote control and monitoring of the actual equipment; +- i) It can optionally support monitoring the generous and unpredictable demand for medical services to reduce the pressure on the medical staff; +- j) It can optionally support intelligent robots, which can serve patients as human caregivers (e.g., delivering basic living supplies and providing clinical assistance) and perform predefined tasks autonomously in different scenarios (e.g., logistics, cleaning, and monitoring of patients) through remote manipulation and intelligent algorithms; +- k) It can optionally support an indoor positioning system to realise the position monitoring of personnel and equipment in the cabins; +- l) It can optionally support tele-radiotherapy management function to realise hierarchical diagnosis and treatment between the RDH and remote central hospitals. The management function includes formulating, verifying and implementing radiotherapy plans for patients in the RDH; +- m) It can optionally support body posture detection and analysis functions based on smart wearable technology, multi-sensor fusion, or deep learning algorithms for bedridden patients; +- n) It can optionally support the three-dimensional visualization, statistical analysis, and efficient storage of the telemonitoring data via integrating BIM and IoT technologies. BIM-IoT integration uses BIM models' geometric and parametric properties and the real-time streaming of environmental data (e.g., temperature, humidity) gathered by multiple sensors to create a basic example of digital twin for visualising telemonitoring data in real time; +- o) It can optionally support monitoring the dietary structure (e.g., the variety of food, the taste of meals) of inpatients to improve nutritional status and dietary satisfaction in the RDH. + +## 7.3 Requirements for telemonitoring terminals + +A telemonitoring terminal is used for information collection and transmission. Two types of telemonitoring terminals are considered: + +- cellular terminals; and +- access gateway. + +Cellular terminals can access the cellular networks directly, and access gateways can support those traditional medical devices that access to the wireless network by converting a non-3GPP network to a cellular network. To better support data collection via numerous Internet of things (IoT) devices, the function of the access gateway is recommended to be enhanced with additional requirements. + +### 7.3.1 General requirements + +The cellular terminals should meet the general requirements [ITU-T F.760.1]. + +### 7.3.2 Access gateway requirements + +The access gateway should meet the general and access gateway requirements [ITU-T F.760.1]. + +### 7.3.3 Enhanced access gateway requirements + +Some additional requirements are recommended for the access gateway to support IoT gateway functions: + +- a) It is recommended to support common requirements and capabilities of a gateway for IoT applications, such as communication, scalability, addressing, and quality of service (QoS) [ITU-T Y.4101]; +- b) It is recommended to support computation, storage and connectivity functions and provide capabilities to support performance balance and collaboration among IoT devices [ITU-T Y.4208]; +- c) It is recommended to support at least one kind of IoT messaging protocol such as message queuing telemetry transport (MQTT) [ISO/IEC 20922], advanced message queuing protocol (AMQP) [ISO/IEC 19464], extensible messaging and presence protocol (XMPP) [IETF RFC 6120], constrained application protocol (CoAP) [IETF RFC 7252] for data acquisition from many connected devices or sensors in the RDH; +- d) It is recommended to support connecting to multiple cloud platforms (e.g., public cloud, private cloud [ITU-T Y.3500], edge cloud) with IoT messaging protocol; +- e) It is recommended to support converting the collected raw data to a semi-structured format such as JavaScript object notation (JSON) [IETF RFC 8259], or XML and adding global device ID to the semi-structured format. + +## 7.4 Requirements for telemonitoring networks + +This clause will provide a detailed description of the requirements of telemonitoring networks: + +- a) It should support continuous cellular network coverage in the urban and rural scenarios and should provide stable, fast and low-latency network connections to meet the needs of a significant amount of data creation, access and transmission between the RDH and remote central hospitals; +- b) It should support the private cellular network deployed in the RDH, such as IMT-2020 LAN [ETSI TS 123 501], which should be isolated from the public cellular networks. Moreover, the cellular terminals should be able to access public and private networks simultaneously; +- c) It should support discontinuous network coverage such as WLAN, WPAN, LR-WPAN, for vital signs data collection and medical equipment data acquisition in all cabins; +- d) It should support stable and reliable telecommunication between the local RDH and remote central hospital in non-regular scenarios, including uninhabited areas, deserts, virgin forests, etc. Here, multiple types of backhaul modules (e.g., emergency communication vehicles, microwave links, drone communications, satellite communications) should be provided for network connection establishment; +- e) It should support providing dedicated physical or virtualized network resources for video interaction in the diagnosis and treatment cabin; +- f) It should support the modularization of base stations (e.g., macro base station, small cell base station) and other network equipment (e.g., router and switch), which could be relocated and reused in different places; +- g) It should support interactive remote video monitoring (e.g., video conferencing, medical image transmission) for medical instrument operations, drug administration, and tele-radiotherapy guidance between the RDH and remote central hospital, which includes the performance requirements [b-3GPP TR 22.826] listed in Table 7-1. + +**Table 7-1 – Performance requirements of interactive remote video monitoring** + +| Interactive remote video monitoring | Communication service availability: target value in % | Communication service reliability: Mean time between failure | End-to-end latency: maximum | Bit rate | +|----------------------------------------------|--------------------------------------------------------------|---------------------------------------------------------------------|------------------------------------|-----------------| +| Compressed 4K video stream | 99.99 | > 1 month | < 100 ms | 25 Mbit/s | +| High quality audio stream | 99.99 | > 1 month | < 100 ms | 128 kbit/s | +| CT/MRI scan data, max 300Mbyte of DICOM data | 99.99 | > 1 month | seconds | > 240 Mbit/s | + +## **7.5 Requirements for telemonitoring cloud platform** + +- a) It should support cloud computing management [ITU-T Y.3500], [ITU-T Y.3502], which provides multiple cloud service categories such as infrastructure as a service (IaaS), platform as a service (PaaS), and software as a service (SaaS) and cloud deployment models (e.g., public cloud, private cloud, edge cloud); +- b) It should support multi-access edge computing (MEC) [ETSI GS MEC 003] functions if the IMT-2020 network is available; +- c) It should support multiple cloud management functions over telemonitoring applications, user, operation and maintenance; +- d) It should support telemonitoring-specific cloud capabilities, including data processing (data acquisition, preprocessing, data buffering, data interpretation, data storage), video functions, object storage functions (database and file storage), etc.; +- e) It should support the same IoT messaging protocol as the enhanced access gateway in the RDH. + +## **7.6 Compatibility requirements** + +- a) It should support telemonitoring applications to run seamlessly on different cloud platforms; +- b) It should support telemonitoring cloud capabilities to work on different bearer networks; +- c) It should support multiple emergency bearer networks to switch seamlessly in non-regular scenarios. + +## **7.7 Scalability requirements** + +It should support the scalability of the capacities, performance and functions of networks and cloud platforms, which provides better coverage and connectivity in a high-density crowd environment. + +## **7.8 Requirements for special needs** + +It should support the accessible provision of telehealth services for people with different types of impairments [ITU-T F.780.2]. + +# **8 Reference framework for telemonitoring system in rapid deployment hospital** + +## **8.1 Reference architecture of the overall system designing** + +The reference architecture of the telemonitoring system, shown in Figure 8-1, is required to include four layers, namely, terminal layer, network layer, cloud layer, and application layer. + +- a) The terminal layer is used for information collection and transmission as the medical tools and applications carrier. It can be classified into integrated terminals and data acquisition terminals. The integrated terminals are taken as an access gateway, connecting multiple medical terminals and transferring those data to a cellular network signal. The data acquisition terminals can correctly and continuously monitor and collect information on the cabin environment, medical equipment, and patients' physical signs. +- b) The network layer plays a crucial role in telemonitoring data transmission and provides guaranteed SLA, is highly reliable and stable, and provides an interconnection link between the terminal and cloud layers. The communication link can be carried as cellular, microwave, satellite, etc. +- c) The cloud layer cooperates with the network layer for the telemonitoring data and provides the virtualization infrastructure, including computing, storage, and network resources. Moreover, it can provide specific telemonitoring services (e.g., data processing, video, object storage, and collaboration with another cloud layer) to support specific applications. It can also provide telemonitoring-specific 5GC network ability (positioning, QoS management, etc.) if the IMT-2020 network is available. +- d) Various applications should be provided in the application layer to support telemonitoring scenarios, such as telemonitoring guidance, nursing, operation management, training and teaching, tele-radiotherapy planning system, and big data analysis systems. + +![Reference architecture of the telemonitoring system diagram showing four layers: Application, Cloud, Network, and Terminal.](5cab96b2d23174c25919840ecd50aa48_img.jpg) + +The diagram illustrates the reference architecture of the telemonitoring system, organized into four layers: + +- Application layer:** Contains six dashed boxes representing various applications: Telemonitoring guidance, Telemonitoring nursing, Telemonitoring operation management, Telemonitoring training and teaching, Tele-radiotherapy planning system, and Big data analysis system. +- Cloud layer:** Contains four dashed boxes representing cloud services: Collaboration capability, Data processing, Video, and Object storage. +- Network layer:** Contains three dashed boxes representing network types: Cellular network, Microwave network, and Satellite network. +- Terminal layer:** Contains two main categories: + - Integrated terminal:** Includes the Access gateway. + - Dynamic acquisition terminal:** Includes Video terminal, Imaging equipment, Robot, Wearable device, Sensor, Tablet terminal, and Radiotherapy equipment. + +Reference architecture of the telemonitoring system diagram showing four layers: Application, Cloud, Network, and Terminal. + +F.780.5(24) + +**Figure 8-1 – Reference architecture of the telemonitoring system** + +## 8.2 Reference architecture of the network layer + +Figure 8-2 shows the reference network architecture of the network layer. Under typical telemonitoring scenarios for the RDH, a wide-connected link between the local RDH and the remote central hospital is established through the bearer network and cellular core network, and the monitoring data will be transmitted to the telemonitoring application platform on the remote central hospital. + +The RDH is usually covered by a non-3GPP network and a cellular network composed of macro base stations and small cell base stations. Medical terminals of different network types in the RDH can directly connect to the cellular network or indirectly connect to the cellular network through an access + +gateway. All cellular data in the RDH shall be transmitted to the cellular core network through the bearer network. + +The bearer network can be classified into a backbone network and an emergency network. The backbone network is used as the bearer network in telemonitoring systems with continuous cellular network coverage. In contrast, emergency networks such as satellites and microwaves are used as the bearer network in unconventional telemonitoring scenarios, and the backhaul modules can be emergency communication vehicles, drones, etc. The cellular core network is responsible for the necessary switching functions for telemonitoring data. + +The control and data link shall be established between the RDH and the remote central hospital to provide specific and detailed telemonitoring services. The control link is used for the RDH and remote hospital scheduling, and the transmitted information includes the network type, service type, configuration information, etc. After the control link is created, a corresponding data link is established between the local RDH and remote hospitals for transmitting telemonitoring data. + +![Figure 8-2: Reference network architecture of the network layer. The diagram shows a flow from left to right: Two User Equipment (UE) units connected to a 'Non-3GPP network' box, which is part of a larger 'RDH' (Remote Device Hub) box. The 'Non-3GPP network' box is connected to a 'Cellular network' box. The 'Cellular network' box is connected to a 'Bear network' box, which is enclosed in a dashed blue rectangle and contains two sub-boxes: 'Emergency network' and 'Backbone network'. The 'Bear network' box is connected to a 'Core network' box. The 'Core network' box is connected to a 'Telemonitoring application platform' box, which is part of a larger 'Remote central hospital' box. The labels 'RDH', 'Bear network', 'Core network', and 'Remote central hospital' are placed below their respective boxes. A small label 'F.780.5(24)' is at the bottom right.](8307f6b04df072c9332f9987e034272c_img.jpg) + +``` +graph LR; UE1[UE] --- Non3GPP[Non-3GPP network]; UE2[UE] --- Non3GPP; subgraph RDH [RDH]; Non3GPP; end; Non3GPP --- CN[Cellular network]; CN --- BN[Bear network]; subgraph BN [Bear network]; EN[Emergency network]; BNN[Backbone network]; end; BN --- CoreN[Core network]; CoreN --- TAP[Telemonitoring application platform]; subgraph RCH [Remote central hospital]; TAP; end; +``` + +Figure 8-2: Reference network architecture of the network layer. The diagram shows a flow from left to right: Two User Equipment (UE) units connected to a 'Non-3GPP network' box, which is part of a larger 'RDH' (Remote Device Hub) box. The 'Non-3GPP network' box is connected to a 'Cellular network' box. The 'Cellular network' box is connected to a 'Bear network' box, which is enclosed in a dashed blue rectangle and contains two sub-boxes: 'Emergency network' and 'Backbone network'. The 'Bear network' box is connected to a 'Core network' box. The 'Core network' box is connected to a 'Telemonitoring application platform' box, which is part of a larger 'Remote central hospital' box. The labels 'RDH', 'Bear network', 'Core network', and 'Remote central hospital' are placed below their respective boxes. A small label 'F.780.5(24)' is at the bottom right. + +**Figure 8-2 – Reference network architecture of the network layer** + +## 8.3 Reference architecture of private cellular network for the RDH + +Traditionally, a remote access virtual private network (VPN) extends a private network across a public network over an IPsec-encrypted tunnel. However, several VPN disadvantages (scalability, performance, and security) could make deploying a remote access VPN challenging in the RDH. + +Under a typical telemonitoring scenario in the RDH, the acquired data needs both local storage and processing and remote transmission, so the terminals need access to the private and public IMT-2020 networks in the RDH. As shown in Figure 8-3, the modular radio access network (RAN), uplink classifier (ULCL) user plane network (UPF), PSA2 UPF and MEC platform (MEP) [ETSI GS MEC 003] equipment are jointly deployed to establish a 3GPP-based private IMT-2020 network. During the initial configuration process, a traffic steering policy with a destination IP address and a type allocation code (TAC) list ought to be configured on the policy control function (PCF) [ETSI TS 123 501]. Smartphone users should subscribe to a private IMT-2020 business with an RDH-dedicated data network name (DNN). + +When smartphones are outside of the RDH's network coverage, they can only access the public IMT-2020 network. After smartphones enter the RDH's network coverage, the traffic steering policy configured on the PCF will be triggered and distributed to the ULCL UPF through the SMF [ETSI TS 123 501]. The traffic accessing the MEP will be diverted to the private IMT-2020 network through the ULCL UPF and PSA2 UPF, which are often deployed together. The traffic accessing the telemonitoring cloud will be diverted to the public IMT-2020 network through the ULCL UPF and PSA1 UPF. Here, smartphones shall be able to access the public and private networks simultaneously without changing the mobile number and SIM card. + +![Figure 8-3: Reference architecture of the private cellular network. The diagram shows two main network components connected by a horizontal line. On the left, labeled '3GPP-based private IMT-2020 network in the RDH', there is a vertical stack of four boxes: 'UE' at the top, '(R) AN' below it, 'ULCL UPF' below that, and 'MEP' at the bottom. 'UE' is connected to '(R) AN', which is connected to 'ULCL UPF'. 'ULCL UPF' is connected to 'PSA2 UPF', which is connected to 'MEP'. On the right, labeled 'Public IMT-2020 network', there is a vertical stack of three boxes: 'PSA1 UPF' at the top, 'Telemonitoring cloud' to its right, and 'IMT-2020 core network' at the bottom. 'PSA1 UPF' is connected to 'Telemonitoring cloud' and 'IMT-2020 core network'. The 'ULCL UPF' box in the private network is connected to the 'PSA1 UPF' box in the public network. A small label 'F.780.5(24)' is in the bottom right corner of the diagram area.](76b0cd79baaedd942af4dc42f2e764b8_img.jpg) + +Figure 8-3: Reference architecture of the private cellular network. The diagram shows two main network components connected by a horizontal line. On the left, labeled '3GPP-based private IMT-2020 network in the RDH', there is a vertical stack of four boxes: 'UE' at the top, '(R) AN' below it, 'ULCL UPF' below that, and 'MEP' at the bottom. 'UE' is connected to '(R) AN', which is connected to 'ULCL UPF'. 'ULCL UPF' is connected to 'PSA2 UPF', which is connected to 'MEP'. On the right, labeled 'Public IMT-2020 network', there is a vertical stack of three boxes: 'PSA1 UPF' at the top, 'Telemonitoring cloud' to its right, and 'IMT-2020 core network' at the bottom. 'PSA1 UPF' is connected to 'Telemonitoring cloud' and 'IMT-2020 core network'. The 'ULCL UPF' box in the private network is connected to the 'PSA1 UPF' box in the public network. A small label 'F.780.5(24)' is in the bottom right corner of the diagram area. + +**Figure 8-3 – Reference architecture of the private cellular network** + +## 8.4 Reference architecture of non-3GPP network optimization + +In major telemonitoring scenarios for the RDH, medical terminals without cellular capabilities connect to the aforementioned access gateway through non-3GPP networks, commonly used as an access network, such as WLAN, WPAN, or wired network. Different non-3GPP networks have diverse network performance requirements. Only services with minimal throughput but high latency requirements can use WPAN, such as location-based services. WLAN applies to services like the big data analysis system and tele-radiotherapy planning system that don't need a reliable network. Optical fibre networks are implemented for services requiring high-quality networks, such as telemonitoring guidance and nursing. + +As shown in Figure 8-4, the access gateway could convert a typical non-3GPP network into a cellular network. Transceiver, acquisition, and policy configuration modules are three functional modules installed on the access gateway that allow network equipment (e.g., MEP, 5GC) at different network layers to identify specific non-3GPP network information. + +The acquisition module collects the following non-3GPP network information. + +- 1) Medical terminal type's name and identifier; +- 2) Medical terminal manufacturer's name and identifier; +- 3) Non-3GPP access network type's name and identifier; +- 4) Non-3GPP network access point's identifier; +- 5) Non-3GPP network capabilities (e.g., uplink rate, downlink rate, latency, jitter, reliability); +- 6) Telemonitoring application data carried by non-3GPP networks. + +The transceiver module transmits the gathered data to the network equipment periodically or when any non-3GPP network information changes. By integrating the incoming data and the cellular network's current status, the network equipment produces an association and affiliated priority between a specific non-3GPP network and network slice and then sends the following data to the transceiver module. + +- 1) Specific non-3GPP network information sourced from the access gateway and corresponding network slice; +- 2) The priority of the above correspondence. + +After the transceiver module receives the above data, the policy configuration module will implement or update the accepted policies to meet differentiated network performance requirements. Telemonitoring data from medical terminals will be delivered to the hospital's MEP via the access gateway on the specific network slice. + +![Figure 8-4: Reference architecture of non-3GPP network optimization. The diagram shows three types of terminals (WLAN, WPAN, and Wired) connected to a central 'Access gateway' box. These terminals are grouped under the label 'Non-3GPP network'. The 'Access gateway' box contains three modules: 'Transceiver module', 'Acquisition module', and 'Policy configuration module'. The 'Acquisition module' is connected to 'Network equipment' on the right, which is labeled 'Cellular network'. Dashed ovals indicate the interfaces between the terminals and the gateway, and between the gateway and the network equipment. A reference label F.780.5(24) is in the bottom right.](9c6461e1e94afae4dec455e69a2ce152_img.jpg) + +Figure 8-4: Reference architecture of non-3GPP network optimization. The diagram shows three types of terminals (WLAN, WPAN, and Wired) connected to a central 'Access gateway' box. These terminals are grouped under the label 'Non-3GPP network'. The 'Access gateway' box contains three modules: 'Transceiver module', 'Acquisition module', and 'Policy configuration module'. The 'Acquisition module' is connected to 'Network equipment' on the right, which is labeled 'Cellular network'. Dashed ovals indicate the interfaces between the terminals and the gateway, and between the gateway and the network equipment. A reference label F.780.5(24) is in the bottom right. + +**Figure 8-4 – Reference architecture of non-3GPP network optimization** + +## 8.5 Reference architecture of the cloud-edge collaboration + +Figure 8-5 shows reference architecture of the cloud-edge collaboration. Under a typical telemonitoring scenario in the RDH, edge computing is deployed in the RDH, and cloud computing is deployed in a remote central hospital. The cloud layer includes the IaaS layer and PaaS layer [ITU-T Y.3500], and cloud-edge collaboration can be classified into resource collaboration for the IaaS layer and service collaboration for the PaaS layer. + +- a) Resource collaboration includes the underlying hardware abstraction from the single-node perspective and the resource scheduling from the global perspective. The edge cloud platform and central cloud platform resources can be used efficiently. +- b) Service collaboration provides key capabilities and flexible docking mechanisms for building edge applications under a telemonitoring scenario. The central cloud provides basic management services (e.g., distribution subscription, access, discovery, usage, operation, maintenance) and telemonitoring-specific services (e.g., data processing, video, object storage) for edge services through a cloud-native architecture. It is convenient for the edge cloud to use the central cloud service capabilities. + +![Figure 8-5: Reference architecture of the cloud-edge collaboration. The diagram shows two columns of boxes. The left column is labeled 'Remote central hospital' and contains 'Cloud PaaS' and 'Cloud IaaS' boxes. The right column is labeled 'RDH' and contains 'Edge PaaS' and 'Edge IaaS' boxes. A double-headed arrow labeled 'Service collaboration' connects 'Cloud PaaS' and 'Edge PaaS'. Another double-headed arrow labeled 'Resource collaboration' connects 'Cloud IaaS' and 'Edge IaaS'. A reference label F.780.5(24) is in the bottom right.](898fb89a50d9ec1dfb4e425c816976a7_img.jpg) + +Figure 8-5: Reference architecture of the cloud-edge collaboration. The diagram shows two columns of boxes. The left column is labeled 'Remote central hospital' and contains 'Cloud PaaS' and 'Cloud IaaS' boxes. The right column is labeled 'RDH' and contains 'Edge PaaS' and 'Edge IaaS' boxes. A double-headed arrow labeled 'Service collaboration' connects 'Cloud PaaS' and 'Edge PaaS'. Another double-headed arrow labeled 'Resource collaboration' connects 'Cloud IaaS' and 'Edge IaaS'. A reference label F.780.5(24) is in the bottom right. + +**Figure 8-5 – Reference architecture of the cloud-edge collaboration** + +# Appendix I + +## Use cases + +(This appendix does not form an integral part of this Recommendation.) + +This clause describes various scenarios for the telemonitoring system in rapid deployment hospitals. + +## I.1 Introduction + +Use cases generated as part of this Recommendation are categorized as follow: + +- Use cases covering the local rapid networking and telecommunication of RDHs. Those use cases are captured in the "Local rapid networking of the RDH" and "Establishment of telecommunication in non-regular scenarios" and "Private IMT-2020 network for the RDH" sections of this document. +- Use cases covering the delivery of data acquisition and monitoring from RDHs to monitoring centres. Those use cases are captured in the "Remote IoT data collection", "Remote live video monitoring", "Non-real-time data transferring", "Tele-radiotherapy application scenario in RDH", "FHIR-based interoperability for telemonitoring", "Digital twin technology for telemonitoring" and "Remote nursing services" sections of this document. + +## I.2 Local rapid networking of the RDH + +The RDH usually consists of individual modular cabins, such as medical, communication, and business cabins. The communication and business cabins are typically deployed together. The communication cabin can also be replaced by an emergency communication vehicle or a large fixed-wing unmanned aerial vehicle (UAV) loaded with a IMT-2020 macro base station. + +The medical cabin is responsible for the locally deployed network equipment (e.g., small base station, IMT-2020 access gateway) and medical services such as diagnosis and treatment, ward, inspection, surgery, and drug storage. Cellular terminals access the cellular networks directly, and traditional non-3GPP medical devices (e.g., medical bracelets, video terminal, smart box) access the cellular network through the IMT-2020 access gateway. + +As depicted in Figure I.1, the communication and business cabin is responsible for the locally deployed network equipment (e.g., macro base station, UPF, MEC) and the backhaul link (e.g., satellite base station, microwave base station) with the public land mobile network (PLMN). In a modular deployment, automatic network configuration and cellular network connection can only be completed by power. + +As depicted in Figure I.2, the emergency communication vehicle is responsible for the locally deployed macro base station and the backhaul link (e.g., satellite base station, microwave base station) with the PLMN. + +![Figure I.1: Scenario 1: Local rapid networking with a macro base station. The diagram shows three main components: Medical cabin, Communication and business cabin, and Central hospital. The Medical cabin contains a 5G access gateway connected to various devices (wearables, camera, medical monitor, smartphone). The gateway is connected to a Macro base station in the Communication and business cabin. The Macro base station is connected to a UPF (User Plane Function) and a MEC (Multi-access Edge Computing) unit. The UPF is connected to a PLMN 5GC (5G Core Network) via a Backhaul link. The PLMN 5GC is connected to the Central hospital, which contains a Central cloud.](5b8a756d9a71c35f17db8bcb90b438a3_img.jpg) + +F.780.5(24) + +Figure I.1: Scenario 1: Local rapid networking with a macro base station. The diagram shows three main components: Medical cabin, Communication and business cabin, and Central hospital. The Medical cabin contains a 5G access gateway connected to various devices (wearables, camera, medical monitor, smartphone). The gateway is connected to a Macro base station in the Communication and business cabin. The Macro base station is connected to a UPF (User Plane Function) and a MEC (Multi-access Edge Computing) unit. The UPF is connected to a PLMN 5GC (5G Core Network) via a Backhaul link. The PLMN 5GC is connected to the Central hospital, which contains a Central cloud. + +**Figure I.1 – Scenario 1: Local rapid networking with a macro base station** + +![Figure I.2: Scenario 2: Local rapid networking with an emergency communication vehicle. The diagram shows three main components: Medical cabin, Business cabin, and Remote central hospital. The Medical cabin contains a 5G access gateway connected to various devices (wearables, camera, medical monitor, smartphone). The gateway is connected to an Emergency communication vehicle in the Business cabin. The vehicle contains a Macro base station connected to a UPF (User Plane Function) and a MEC (Multi-access Edge Computing) unit. The UPF is connected to a PLMN 5GC (5G Core Network) via a Backhaul link. The PLMN 5GC is connected to the Remote central hospital, which contains a Central cloud.](c5655e700cc3e9aac7e9f4f07f30264d_img.jpg) + +F.780.5(24) + +Figure I.2: Scenario 2: Local rapid networking with an emergency communication vehicle. The diagram shows three main components: Medical cabin, Business cabin, and Remote central hospital. The Medical cabin contains a 5G access gateway connected to various devices (wearables, camera, medical monitor, smartphone). The gateway is connected to an Emergency communication vehicle in the Business cabin. The vehicle contains a Macro base station connected to a UPF (User Plane Function) and a MEC (Multi-access Edge Computing) unit. The UPF is connected to a PLMN 5GC (5G Core Network) via a Backhaul link. The PLMN 5GC is connected to the Remote central hospital, which contains a Central cloud. + +**Figure I.2 – Scenario 2: Local rapid networking with an emergency communication vehicle** + +## I.3 Establishment of telecommunication in non-regular scenarios + +When deploying RDHs in an environment lacking a backbone network (e.g., deserts, uninhabited areas, and far seas), satellite or microwave communication can be used as the backhaul link according to the transmission distance. In this scenario, satellite or microwave base stations should be integrated into the communication cabin. + +As depicted in Figure I.3, satellite communication has a wide transmission range. It is suitable for environments far from the backbone network. + +As depicted in Figure I.4, the transmission range of microwave communication is about tens of kilometres, so several microwave relay stations should be deployed in a scenario far from the backbone network. + +It is worth noticing that the non-terrestrial networks (NTN) family [b-3GPP TR 38.811], [b-3GPP TR 38.821], still being studied in 3GPP, is expected to provide scalable and efficient network solutions globally. The NTN use an airborne or space-borne vehicle to embark in a transmission equipment relay node or base station, including satellite communication networks, high-altitude platform systems (HAPS), and air-to-ground networks. + +![Figure I.3: Scenario 1: Satellite communication as the backhaul link. The diagram shows three main components: 'Communication and business cabin', 'Backhaul link', and 'Central hospital'. In the cabin, a 'Base station' is connected to a 'UPF' and a 'MEC'. The 'Base station' is connected to a 'Satellite station' via a microwave link. The 'Satellite station' is connected to a 'Satellite' in the 'Backhaul link'. The 'Satellite' is connected to another 'Satellite station' on the ground, which is connected to a 'PLMN 5GC'. The 'PLMN 5GC' is connected to a 'Central cloud' in the 'Central hospital'.](e69b9188aa2c14ec6b21c83f711fef65_img.jpg) + +F.780.5(24) + +Figure I.3: Scenario 1: Satellite communication as the backhaul link. The diagram shows three main components: 'Communication and business cabin', 'Backhaul link', and 'Central hospital'. In the cabin, a 'Base station' is connected to a 'UPF' and a 'MEC'. The 'Base station' is connected to a 'Satellite station' via a microwave link. The 'Satellite station' is connected to a 'Satellite' in the 'Backhaul link'. The 'Satellite' is connected to another 'Satellite station' on the ground, which is connected to a 'PLMN 5GC'. The 'PLMN 5GC' is connected to a 'Central cloud' in the 'Central hospital'. + +**Figure I.3 – Scenario 1: Satellite communication as the backhaul link** + +![Figure I.4: Scenario 2: Microwave communication as the backhaul link. The diagram shows three main components: 'Communication and business cabin', 'Backhaul link', and 'Central hospital'. In the cabin, a 'Base station' is connected to a 'UPF' and a 'MEC'. The 'Base station' is connected to a 'Microwave station' via a microwave link. The 'Microwave station' is connected to a 'Microwave relay station' in the 'Backhaul link'. The 'Microwave relay station' is connected to another 'Microwave station' on the ground, which is connected to a 'PLMN 5GC'. The 'PLMN 5GC' is connected to a 'Central cloud' in the 'Central hospital'.](d53cd0fd1cf896a9353fd63de1505ba2_img.jpg) + +F.780.5(24) + +Figure I.4: Scenario 2: Microwave communication as the backhaul link. The diagram shows three main components: 'Communication and business cabin', 'Backhaul link', and 'Central hospital'. In the cabin, a 'Base station' is connected to a 'UPF' and a 'MEC'. The 'Base station' is connected to a 'Microwave station' via a microwave link. The 'Microwave station' is connected to a 'Microwave relay station' in the 'Backhaul link'. The 'Microwave relay station' is connected to another 'Microwave station' on the ground, which is connected to a 'PLMN 5GC'. The 'PLMN 5GC' is connected to a 'Central cloud' in the 'Central hospital'. + +**Figure I.4 – Scenario 2: Microwave communication as the backhaul link** + +## I.4 Remote IoT data collection + +Message queuing telemetry transport (MQTT) protocol is based on the publish/subscribe model for communication and data exchange. As an MQTT subscriber, the remote manager will subscribe to a data collection topic to the MQTT broker. The broker forwards the message under the topic to the IoT gateway. + +As an MQTT client, the enhanced access gateway collects multiple data (i.e., patient-related, equipment-related, environmental, and drug-related data) every 10 seconds and converts those raw data to a JSON-based format. Those JSON-based data is remotely transmitted to the MQTT broker deployed in the central cloud. The central cloud will continue to process the collected data under the batch or stream processing model and store calculation results in the database (e.g., HBase, Redis, Elasticsearch). Based on those calculation results, the remote manager can monitor the vital signs of all patients, environmental factors such as temperature and humidity in all cabins, the operating status of the equipment, and the use of medicines and beds. + +Figure I.5 shows remote IoT data gathering in the RDH. + +![Diagram illustrating Remote IoT data gathering in the RDH. The diagram shows two main components: RDH (Remote Data Hub) and Central hospital. In the RDH, an 'Enhanced access gateway' is connected to various sensors (represented by icons of a watch, clock, and thermometer). The gateway sends 'Modbus polling' requests to the sensors and receives 'Data reply' from them. The gateway is also connected to an 'MQTT client'. The MQTT client sends data to a cloud (represented by a blue cloud icon) via 'Publish'. The cloud then sends data to the 'MQTT broker' in the Central hospital via 'Subscribe'. The MQTT broker sends data to an 'MQTT subscriber' (represented by a person sitting at a desk) via 'Publish'. The MQTT subscriber sends data back to the MQTT broker via 'Subscribe'. The MQTT broker then sends data to the cloud via 'Publish'. The cloud sends data to the MQTT client via 'Subscribe'. The MQTT client sends data to the Enhanced access gateway via 'Data reply'. The Enhanced access gateway then sends data to the sensors via 'Modbus polling'. A label 'F.780.5(24)' is present in the bottom right corner of the diagram.](c914f51f4427bc672dd0526cfc90ebe9_img.jpg) + +Diagram illustrating Remote IoT data gathering in the RDH. The diagram shows two main components: RDH (Remote Data Hub) and Central hospital. In the RDH, an 'Enhanced access gateway' is connected to various sensors (represented by icons of a watch, clock, and thermometer). The gateway sends 'Modbus polling' requests to the sensors and receives 'Data reply' from them. The gateway is also connected to an 'MQTT client'. The MQTT client sends data to a cloud (represented by a blue cloud icon) via 'Publish'. The cloud then sends data to the 'MQTT broker' in the Central hospital via 'Subscribe'. The MQTT broker sends data to an 'MQTT subscriber' (represented by a person sitting at a desk) via 'Publish'. The MQTT subscriber sends data back to the MQTT broker via 'Subscribe'. The MQTT broker then sends data to the cloud via 'Publish'. The cloud sends data to the MQTT client via 'Subscribe'. The MQTT client sends data to the Enhanced access gateway via 'Data reply'. The Enhanced access gateway then sends data to the sensors via 'Modbus polling'. A label 'F.780.5(24)' is present in the bottom right corner of the diagram. + +**Figure I.5 – Remote IoT data gathering in the RDH** + +## I.5 Remote live video monitoring + +Using remote video monitoring, the remote managers can monitor the patient's condition and medical staff's working status, assign medical resources rationally, and maintain RDH stability and orderly 24/7. Based on HD video monitoring, specialists in remote central hospitals can guide doctors to treat patients with intractable diseases. + +## I.6 Non-real-time data transferring + +Electronic medical records (EMRs) contain patients' medical and treatment history, and picture archiving and communication system (PACS) handles images from various medical imaging instruments. Data of EMR and PACS may be required to be securely uploaded to the telemonitoring centre for archiving, inspection, and further analysis. + +## I.7 Tele-radiotherapy application scenario in RDH + +The typical tele-radiotherapy application scenario in China appears in the National Cancer Centre or Cancer Hospital, Chinese Academy of Medical Sciences (CAMS). The radiotherapy centre in the cancer hospital includes a clinical trial management system (CTMS), radiotherapy quality assurance platform, radiotherapy planning system and cloud storage centre, etc. + +Radiotherapy planning systems, at the heart of radiation therapy (RT), provide a set of computerized tools that allow the radiation oncologist, medical physicist, and treatment planner to create and visualize radiotherapy treatments, given the imaging data available. + +As depicted in Figure I.6, DICOM-formatted CT images generated in RDHs are uploaded to the cloud storage centre through an IMT-2020 network. Based on those CT images, experts in the radiotherapy centre will carry out multidisciplinary consultation, target delineation, radiotherapy quality assurance, and plan evaluation. The prepared radiotherapy treatment plans will be sent to the edge cloud (e.g., MEC) deployed in the RDHs, and then be implemented by doctors. + +As depicted in Figure I.7, remote video conferencing based on high-end camera and audio systems could be conducted by a remote expert in the radiotherapy centre. + +![Figure I.6: Tele-radiotherapy application scenario in RDH. The diagram shows a flow from a Radiotherapy room in RDHs or branches (containing a CT scan machine, DICOM CT image, and Edge cloud) to a Radiotherapy centre in Cancer Hospital, CAMS (containing a Cloud storage centre and Radiotherapy planning system). The flow is mediated by a Cellular network (4G, 5G). The steps are: 1. Upload CT image from the CT scan to the Cloud storage centre; 2. Expert consultation from the Cloud storage centre to the Radiotherapy planning system; 3. Send radiotherapy treatment plannings from the Radiotherapy planning system to the Edge cloud in the RDH, which then leads to Local radiotherapy.](c5452f95f3b28f1bfe29e84fbc2e1267_img.jpg) + +F.780.5(24) + +Figure I.6: Tele-radiotherapy application scenario in RDH. The diagram shows a flow from a Radiotherapy room in RDHs or branches (containing a CT scan machine, DICOM CT image, and Edge cloud) to a Radiotherapy centre in Cancer Hospital, CAMS (containing a Cloud storage centre and Radiotherapy planning system). The flow is mediated by a Cellular network (4G, 5G). The steps are: 1. Upload CT image from the CT scan to the Cloud storage centre; 2. Expert consultation from the Cloud storage centre to the Radiotherapy planning system; 3. Send radiotherapy treatment plannings from the Radiotherapy planning system to the Edge cloud in the RDH, which then leads to Local radiotherapy. + +**Figure I.6 – Tele-radiotherapy application scenario in RDH** + +![Figure I.7: Remote video conferencing in RDH. The diagram shows a Radiotherapy centre in Cancer Hospital, CAMS (with a doctor at a computer) and a Radiotherapy room in RDHs or branches (with a doctor and a patient on a treatment table). They are connected via a Cellular network (4G, 5G) for remote video conferencing.](366a77fdefb0097b3289b4a011911390_img.jpg) + +F.780.5(24) + +Figure I.7: Remote video conferencing in RDH. The diagram shows a Radiotherapy centre in Cancer Hospital, CAMS (with a doctor at a computer) and a Radiotherapy room in RDHs or branches (with a doctor and a patient on a treatment table). They are connected via a Cellular network (4G, 5G) for remote video conferencing. + +**Figure I.7 – Remote video conferencing in RDH** + +## I.8 Private IMT-2020 network for the RDH + +In a typical telemonitoring situation, as shown in Figure I.8, the lightweight and modular RAN, MEP, and ULCL UPF are co-deployed in the RDH. To reduce network latency and ensure data security, the MEP offers local service processing capabilities, and the ULCL UPF sends service data to the MEP server for processing. + +Smartphone users (such as patients and medical staff) who subscribe to a private network business can only access the Internet over the public IMT-2020 network outside the RDH. They can securely access the private and public IMT-2020 networks when they enter the RDH's private IMT-2020 network coverage. + +![Diagram of Private IMT-2020 network coverage of the RDH. The diagram is split into two main areas by a vertical line labeled 'N9'. The left area, 'Private IMT-2020 coverage area for the RDH', contains a 'MEP' (Medical Equipment) connected to a 'ULCL UPF' (User Plane Function). A smartphone is shown with a green dashed line (labeled '2') connecting it to the ULCL UPF. A red dashed line (labeled '1') connects the ULCL UPF to the right area. The right area, 'Public IMT-2020 coverage area', contains a 'PSA UPF' (PDN Gateway UPF) connected to a 'Telemonitoring cloud platform' (represented by a cloud with a hospital icon). The PSA UPF is connected to the ULCL UPF via the N9 interface.](7133ccf78043568ca62ecbcd43628a4a_img.jpg) + +Diagram of Private IMT-2020 network coverage of the RDH. The diagram is split into two main areas by a vertical line labeled 'N9'. The left area, 'Private IMT-2020 coverage area for the RDH', contains a 'MEP' (Medical Equipment) connected to a 'ULCL UPF' (User Plane Function). A smartphone is shown with a green dashed line (labeled '2') connecting it to the ULCL UPF. A red dashed line (labeled '1') connects the ULCL UPF to the right area. The right area, 'Public IMT-2020 coverage area', contains a 'PSA UPF' (PDN Gateway UPF) connected to a 'Telemonitoring cloud platform' (represented by a cloud with a hospital icon). The PSA UPF is connected to the ULCL UPF via the N9 interface. + +**Figure I.8 – Private IMT-2020 network coverage of the RDH** + +## I.9 FHIR-based interoperability for telemonitoring + +Fast healthcare interoperability resources (FHIR) enables mapping healthcare concepts into "resources", such as the patient and observation. Although Turtle, XML and JSON formats can be used to encode HL7 FHIR resources, JSON is usually chosen because it is popular and has a relatively small overall data size. The communication between local and remote systems is done through most-commonly-used HTTP requests. An FHIR client in the remote system posts resources to an FHIR server in the RDH, and then the server returns a result with the URI address of the created resources. The FHIR standard enables the RDH to quickly and securely exchange telemonitoring data with remote systems and makes it simple to integrate new monitoring terminals and external applications. + +## I.10 Digital twin technology for telemonitoring + +Under the sudden COVID-19 epidemic, remote monitoring and control of various terminals (e.g., medical sensors) have become necessary. Digital twin technology can collect data through multiple sensors to realistically represent a virtual patient. By providing comprehensive data about each patient, a digital twin could assist doctors in monitoring and treating infected patients continuously and remotely. + +## I.11 Remote nursing services + +Due to the unexpected COVID-19 epidemic, there was a sharp increase in the number of patients admitted to the RDH as well as a shortage of medical staff and distant care services. + +Remote nursing could control a camera remotely, mark the shared patient case data cooperatively, and perform nursing warnings and medical order processing. It could provide rich expert services to the RDH through video conferences without sufficient medical staff. + +# Bibliography + +- [b-ITU-T Y.4000] Recommendation ITU-T Y.4000/Y.2060 (2012), *Overview of the Internet of things*. +- [b-3GPP TR 22.826] 3GPP TR 22.826 V17.2.0 (2021), *Study on communication services for critical medical applications*. +- [b-3GPP TR 38.811] 3GPP TR 38.811 V15.4.0 (2020), *Study on New Radio (NR) to support non-terrestrial networks (Release 15)*. +- [b-3GPP TR 38.821] 3GPP TR 38.821 V16.2.0 (2023), *Solutions for NR to support Non-Terrestrial Networks (NTN) (Release 16)*. +- [b-HL7 FHIR] HL7 FHIR (2022), *Fast Healthcare Interoperability Resources (FHIR) 4.3.0*. +- [b-ISO/TR 14292] ISO/TR 14292:2012, *Health informatics – Personal health records – Definition, scope and context*. +- [b-W3C XML 1.0] W3C XML1.0 (2008), *Extensible Markup Language (XML) 1.0*. + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.780.6-202512-I_PDF-E/16c69c0dacd3c57ae91acd114e5f5bd2_img.jpg b/marked/F/T-REC-F.780.6-202512-I_PDF-E/16c69c0dacd3c57ae91acd114e5f5bd2_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..595b4adc61359502d9148605ef90a555d92c158d --- /dev/null +++ b/marked/F/T-REC-F.780.6-202512-I_PDF-E/16c69c0dacd3c57ae91acd114e5f5bd2_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3b605d2f16b2b51f5c1ca804cfb30327982a4331b834559fb8e2fc969a3ce9d2 +size 67215 diff --git a/marked/F/T-REC-F.780.6-202512-I_PDF-E/303fadfb9def251d1575d6221199b158_img.jpg b/marked/F/T-REC-F.780.6-202512-I_PDF-E/303fadfb9def251d1575d6221199b158_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..cf4fe29d2cae3e2832d163d2a8890ad0f0599e32 --- /dev/null +++ b/marked/F/T-REC-F.780.6-202512-I_PDF-E/303fadfb9def251d1575d6221199b158_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9b9cee9f78d20b4b549b3ff440ec55f1efbde4382cd0d684ac50f2e558683287 +size 34688 diff --git a/marked/F/T-REC-F.780.6-202512-I_PDF-E/32f80b2d70752613b6577797811f3a95_img.jpg b/marked/F/T-REC-F.780.6-202512-I_PDF-E/32f80b2d70752613b6577797811f3a95_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f7e4969de0b15abaa9509df13b575770b9eb4648 --- /dev/null +++ b/marked/F/T-REC-F.780.6-202512-I_PDF-E/32f80b2d70752613b6577797811f3a95_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7f20e9806757bc4b4d372d4de12b9cc54b6f624d66227e4725441c2e87bf224d +size 92154 diff --git a/marked/F/T-REC-F.780.6-202512-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg b/marked/F/T-REC-F.780.6-202512-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c1629394c02f697f60ae954ad36bf274fa8082cb --- /dev/null +++ b/marked/F/T-REC-F.780.6-202512-I_PDF-E/390120de4fe440c42fea8154fcaad334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3763acd7a0ed56aa30f3f90468c1fb0cb54573eb9045bca3eddf8976b4f0f41d +size 7219 diff --git a/marked/F/T-REC-F.780.6-202512-I_PDF-E/397d7c5f170e27a78f44fd8a5b0a234b_img.jpg b/marked/F/T-REC-F.780.6-202512-I_PDF-E/397d7c5f170e27a78f44fd8a5b0a234b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..b0fc6bfffdb931ebe1cb80d6550b2c9089be9f00 --- /dev/null +++ b/marked/F/T-REC-F.780.6-202512-I_PDF-E/397d7c5f170e27a78f44fd8a5b0a234b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bedaeecd9f6bc9e9438a49854b7c6eb7bea7ea61c75234a68905082054fd3855 +size 16096 diff --git a/marked/F/T-REC-F.780.6-202512-I_PDF-E/483c43eac5aa582dfcb6e0aa8cee163f_img.jpg b/marked/F/T-REC-F.780.6-202512-I_PDF-E/483c43eac5aa582dfcb6e0aa8cee163f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..101417d32af4f98dc2248d6dd253399b6e124871 --- /dev/null +++ b/marked/F/T-REC-F.780.6-202512-I_PDF-E/483c43eac5aa582dfcb6e0aa8cee163f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7db50c370c032254959151ecda2a22c39e22dea4a5ca46033ddd602a3caf2f39 +size 15019 diff --git a/marked/F/T-REC-F.780.6-202512-I_PDF-E/62a4c9055642dbb00663e633332f04d3_img.jpg b/marked/F/T-REC-F.780.6-202512-I_PDF-E/62a4c9055642dbb00663e633332f04d3_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a08d3028c1bb17ff13f6520afc3fe912d2bbac45 --- /dev/null +++ b/marked/F/T-REC-F.780.6-202512-I_PDF-E/62a4c9055642dbb00663e633332f04d3_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8acbbfbef89d98c6a788ccb5f19c2d0bfa856c47c3da45a4dbb5ead14a760d66 +size 32957 diff --git a/marked/F/T-REC-F.780.6-202512-I_PDF-E/69b7bd65e85cdef6fdd7fb0a8194257c_img.jpg b/marked/F/T-REC-F.780.6-202512-I_PDF-E/69b7bd65e85cdef6fdd7fb0a8194257c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..19d046bcb7a40962254d3ef4b7c33dfeb02b7d6a --- /dev/null +++ b/marked/F/T-REC-F.780.6-202512-I_PDF-E/69b7bd65e85cdef6fdd7fb0a8194257c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3045e5d0b8ee62f3ab4d0a4df5f1fe19acaf23cf48ca10a9f0abd48e27585ed4 +size 30860 diff --git a/marked/F/T-REC-F.780.6-202512-I_PDF-E/b8661c6c54f72ecc7ff6cb05e47b2891_img.jpg b/marked/F/T-REC-F.780.6-202512-I_PDF-E/b8661c6c54f72ecc7ff6cb05e47b2891_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..46e82bcf35130d329cdf6b8bc71c89dc831d6cea --- /dev/null +++ b/marked/F/T-REC-F.780.6-202512-I_PDF-E/b8661c6c54f72ecc7ff6cb05e47b2891_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3a601848e3b2bbc01c85fc44e1e5f743fe8a68657ff441234005dc96b5058cfd +size 57200 diff --git a/marked/F/T-REC-F.780.6-202512-I_PDF-E/c74ee8a60918baf4b06f002a72f86111_img.jpg b/marked/F/T-REC-F.780.6-202512-I_PDF-E/c74ee8a60918baf4b06f002a72f86111_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..35147d0271bfbba4c7341e847f894fd4312d61ea --- /dev/null +++ b/marked/F/T-REC-F.780.6-202512-I_PDF-E/c74ee8a60918baf4b06f002a72f86111_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8b7c2942c3bf78421ed51aa3ba3c4a33bcde4931705d6b7bfd8fc94e1478066d +size 9710 diff --git a/marked/F/T-REC-F.780.6-202512-I_PDF-E/cce1f4715971176232d2f8cf7b80acf7_img.jpg b/marked/F/T-REC-F.780.6-202512-I_PDF-E/cce1f4715971176232d2f8cf7b80acf7_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..08b64cd1082784aabd4a363095d30117bd561a0f --- /dev/null +++ b/marked/F/T-REC-F.780.6-202512-I_PDF-E/cce1f4715971176232d2f8cf7b80acf7_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ef3d3ba2db9fc60bfeefdddd14bc373dfc38f50412c0c16b00a7d49ed30ccb3d +size 25181 diff --git a/marked/F/T-REC-F.780.6-202512-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg b/marked/F/T-REC-F.780.6-202512-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..cfff06c93726f9d16a8a576fde114aeab031db2b --- /dev/null +++ b/marked/F/T-REC-F.780.6-202512-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c92a278fbe85e5ea2e8c8c8b8d887e4ae6d96a51c72e25c39f03320594f51cf3 +size 49293 diff --git a/marked/F/T-REC-F.780.6-202512-I_PDF-E/d9843ac5dfc4bbe9e4d21b8898723b5e_img.jpg b/marked/F/T-REC-F.780.6-202512-I_PDF-E/d9843ac5dfc4bbe9e4d21b8898723b5e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..5a9ea783e02e4345ca6fcc9c11f0687889d52083 --- /dev/null +++ b/marked/F/T-REC-F.780.6-202512-I_PDF-E/d9843ac5dfc4bbe9e4d21b8898723b5e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dda8d143b09f4229ac4c6d588bc54c8bd20323a3129035614340c95db643df3d +size 80980 diff --git a/marked/F/T-REC-F.780.6-202512-I_PDF-E/e3c1d325b828c44049a7ac37e5b34603_img.jpg b/marked/F/T-REC-F.780.6-202512-I_PDF-E/e3c1d325b828c44049a7ac37e5b34603_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1bf0a8c24f8cd888928cd5543a2e4e5761858ac6 --- /dev/null +++ b/marked/F/T-REC-F.780.6-202512-I_PDF-E/e3c1d325b828c44049a7ac37e5b34603_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d0a7c71fb57d79cd7dd8d543e285f549482f4883188b0f92475bcc9a76ef129d +size 14573 diff --git a/marked/F/T-REC-F.780.6-202512-I_PDF-E/ee84ba98f797de76074797ce486078de_img.jpg b/marked/F/T-REC-F.780.6-202512-I_PDF-E/ee84ba98f797de76074797ce486078de_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..78f09f16e95db23eb23946f5cf6188ec9ac5481e --- /dev/null +++ b/marked/F/T-REC-F.780.6-202512-I_PDF-E/ee84ba98f797de76074797ce486078de_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e308dcbb839499d48308a96059b05b807a05e10001b3f86490d1adcaba9f27cf +size 22622 diff --git a/marked/F/T-REC-F.780.6-202512-I_PDF-E/f9c64800d9bace9b4315646d1057be3c_img.jpg b/marked/F/T-REC-F.780.6-202512-I_PDF-E/f9c64800d9bace9b4315646d1057be3c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..015595f9bb3a94cb05819611d955192ece31c556 --- /dev/null +++ b/marked/F/T-REC-F.780.6-202512-I_PDF-E/f9c64800d9bace9b4315646d1057be3c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f5d66601bcd4be43f31deb0eaf962d79d9612c3ddf4b53fd3b1f0694c6d7c78d +size 32240 diff --git a/marked/F/T-REC-F.780.7-202508-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/F/T-REC-F.780.7-202508-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..25f8fbfaef317942ed3102c74b40a7d42f80608b --- /dev/null +++ b/marked/F/T-REC-F.780.7-202508-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:83cab7dcabfe939b223d97242d8523e461968d21f7c5432693c30a64f2d95521 +size 7050 diff --git a/marked/F/T-REC-F.780.7-202508-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg b/marked/F/T-REC-F.780.7-202508-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a89ba94f6a0f674b23ba0129c4ba8a2a2c3359d2 --- /dev/null +++ b/marked/F/T-REC-F.780.7-202508-I_PDF-E/a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a342349fa82de6e3e114b343a0cd9004f2fe68b001d12081dbf1a73c09a78ec8 +size 66195 diff --git a/marked/F/T-REC-F.780.7-202508-I_PDF-E/raw.md b/marked/F/T-REC-F.780.7-202508-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..8c52b6c2c49e73cd95a54b4009abdc7f0c7755c4 --- /dev/null +++ b/marked/F/T-REC-F.780.7-202508-I_PDF-E/raw.md @@ -0,0 +1,682 @@ + + +# Recommendation **ITU-T F.780.7 (08/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +## --- **Technical requirements of medical equipment management platforms for digital hospitals** + +![ITU logo](0538daaa5583c23e17db3a12f2281a55_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue circular emblem with a stylized globe and the letters 'ITU' in white. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.780.7 + +# Technical requirements of medical equipment management platforms for digital hospitals + +## Summary + +Digital hospitals refer to health-care institutions that integrate information and communication technologies into health care, medical service and management. As a critical component of digital hospital management, medical equipment (ME) management plays an essential role in improving cost-effectiveness and operational efficiency. + +Recommendation ITU-T F.780.7 provides the technical requirements of medical equipment management platform for digital hospitals. It describes a three-layer architecture encompassing a resource layer, service layer and application layer to support ME inventory management, lifecycle management, and, in particular, maintenance management. It also outlines the requirements for data and business services in the service layer, as well as infrastructure requirements in the resource layer. + +## History\* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.780.7 | 2025-08-13 | 21 | 11.1002/1000/16433 | + +## Keywords + +Cloud computing, data management, digital hospital, medical equipment management. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2025 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|---------------------------------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 3 | +| 6 Reference architecture for medical equipment management platform in digital hospital..... | 3 | +| 7 Technical requirements for a ME management platform ..... | 4 | +| 7.1 Application-layer requirements ..... | 4 | +| 7.2 Service layer requirements ..... | 9 | +| 7.3 Resources layer requirements..... | 12 | +| Bibliography..... | 14 | + + + +# Recommendation ITU-T F.780.7 + +## Technical requirements of medical equipment management platforms for digital hospitals + +## 1 Scope + +This Recommendation specifies the technical requirements for building a medical equipment (ME) management platform within a digital hospital, which involves the integration of information and communication technology. The platform is based on a three-layer reference architecture encompassing resource, service and application layers. + +This Recommendation focuses on only the medical equipment which uses information and communication technology. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +- [ITU-T F.748.38] Recommendation ITU-T F.748.38 (2024), *Technical specification for artificial intelligence cloud platform: General architecture*. +- [ITU-T X.1361] Recommendation ITU-T X.1361 (2018), *Security framework for the Internet of things based on the gateway model*. +- [ITU-T Y.3510] Recommendation ITU-T Y.3510 (2016), *Cloud computing infrastructure requirements*. +- [ITU-T Y.3600] Recommendation ITU-T Y.3600 (2015), *Big data – Cloud computing based requirements and capabilities*. +- [ITU-T Y.4101] Recommendation ITU-T Y.4101/Y.2067 (2017), *Common requirements and capabilities of a gateway for Internet of things applications*. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 calibration** [b-WHO-1]: Some medical equipment, particularly those with therapeutic energy output..., needs to be calibrated periodically. This means that energy levels are to be measured and if there is a discrepancy from the indicated levels, adjustments must be made until the device functions within specifications. Devices that take measurements ... also require periodic *calibration* to ensure accuracy compared to known standards. + +NOTE 1 – Examples of devices with therapeutic energy output include defibrillators, electrosurgical units, and physical therapy stimulators. + +NOTE 2 – Examples of devices that perform measurements include electrocardiographs, laboratory equipment, patient scales, and pulmonary function analysers. + +**3.1.2 clinical engineer** [b-WHO-1]: A professional who supports and advances patient care by applying engineering and managerial skills to health-care technology ... While a *clinical engineer* is a specialized biomedical engineer, the terms are often used interchangeably. + +**3.1.3 corrective maintenance** [b-ITU-T E.800]: The maintenance carried out after fault recognition and intended to restore an item to a state in which it can perform a required function. + +**3.1.4 data interoperability** [b-ITU-T Y.4563]: The ability of two or more systems or components to exchange data and to use the data that has been exchanged. + +**3.1.5 inspection and preventive maintenance** [b-WHO-1]: ... all the scheduled activity necessary to ensure a piece of medical equipment is functioning correctly and is well maintained. [It] therefore includes inspection and preventive maintenance. + +**3.1.6 gateway** [ITU-T Y.4101]: A unit in the Internet of things which interconnects the devices with the communication networks. It performs the necessary translation between the protocols used in the communication networks and those used by devices. + +**3.1.7 medical equipment** [b-WHO-2]: Medical devices requiring calibration, maintenance, repair, user training, and decommissioning – activities usually managed by clinical engineers. Medical equipment is used for the specific purposes of diagnosis and treatment of disease or rehabilitation following disease or injury; it can be used either alone or in combination with any accessory, consumable, or other piece of medical equipment. Medical equipment excludes implantable, disposable or single-use medical devices. + +### **3.2 Terms defined in this Recommendation** + +This Recommendation defines the following terms: + +**3.2.1 digital hospital:** A hospital integrating information and communication technology into its health care, medical service and management. + +**3.2.2 medical equipment inventory:** A detailed itemized list of medical equipment held by a health organization or institution. + +**3.2.3 medical equipment inventory management:** The process of maintaining and updating a detailed itemized list of medical equipment held by a health organization or institution to reflect the current status of each asset. + +**3.2.4 medical equipment lifecycle in hospital:** The process including procurement, installation, commissioning, operation, maintenance and decommissioning and disposal of medical equipment. + +**3.2.5 medical equipment metadata:** Structured information that characterizes the essential attributes, lifecycle activities and associated documentation of medical equipment. + +## **4 Abbreviations and acronyms** + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|------------------------------------------| +| CA | Certificate Authority | +| CASP | Consumables, Accessories and Spare Parts | +| CM | Corrective Maintenance | +| CPU | Central Processing Unit | +| CSV | Comma-Separated Values | +| IoT | Internet of Things | +| IPM | Inspection and Preventive Maintenance | + +| | | +|------|------------------------------| +| I/O | Input/Output | +| JSON | JavaScript Object Notation | +| ME | Medical Equipment | +| PDF | Portable Document Format | +| RBAC | Role-Based Access Control | +| SMS | Short Message Service | +| SQL | Structured Query Language | +| UDI | Unique Device Identification | + +## 5 Conventions + +In this Recommendation: + +The keyword "is required" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. + +The keywords "is recommended" and "should" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement need not be followed to claim conformance. + +The keywords "can optionally" and "may" indicate an optional requirement which is permissible, without implying any sense of being recommended. These terms are not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +## 6 Reference architecture for medical equipment management platform in digital hospital + +Digital hospitals integrate information and communication technologies into health care, medical services and management processes. This integration supports resource coordination and process streamlining, thereby improving service quality and reducing costs compared with traditional hospitals that rely on manual operations or fragmented information systems. As a critical component of digital hospital management, a ME management platform plays an essential role in improving cost-effectiveness and operational efficiency. + +A reference architecture for a ME management platform consists of a resource layer, a service layer and an application layer (see Figure 6-1). Each white box in the reference architecture represents a functional module of the ME management platform. + +![Reference architecture for ME management platform diagram showing three layers: Application, Service, and Resource.](a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg) + +The diagram illustrates the reference architecture for the ME management platform, organized into three horizontal layers: + +- Application layer:** Contains three modules: ME inventory management, ME lifecycle management, and ME maintenance management. +- Service layer:** Divided into two columns: + - Data service:** Includes Data collection, Data transmission, Data storage, Data analysis, Data visualization, and Data management. + - Business service:** Includes User identification, Access control, Gateway management, and Logging service. +- Resource layer:** Contains three modules: Computing resources, Networking resources, and Storage resources. + +F.780.7(25) + +Reference architecture for ME management platform diagram showing three layers: Application, Service, and Resource. + +**Figure 6-1 – Reference architecture for ME management platform** + +The application layer is conceived to manage hospitals' unaddressed management difficulties regarding medical equipment by covering essential application scenarios including ME inventory management, lifecycle management and maintenance management. + +The service layer consists of data services and business services. Data services include data collection, data transmission, data storage, data analysis, data visualization and data management. Business services include user identification, access control, gateway management and logging services. + +The resource layer provides computing, storage and networking resources for the ME management platform. + +## 7 Technical requirements for a ME management platform + +### 7.1 Application-layer requirements + +#### 7.1.1 ME inventory management module + +A ME inventory is a detailed, itemized list of medical equipment held by a health organization or institution [b-WHO-3]. To be worthwhile, an inventory is continually maintained and updated to reflect the current status of each asset. The goal of ME inventory is to have an accurate, up-to-date record of all assets held by the organization, reflecting the status at any given moment in time. + +ME inventory management module involves not only recording basic information such as the quantity and location of ME but also organizing and structuring this information for systematic management. In modern ME management systems, this Recommendation divides ME inventory management into two major functions: + +- ME metadata management; +- consumables, accessories and spare parts (CASP) management. + +This is aimed at achieving detailed management of ME and related items and ensuring efficient utilization, compliance and continuity of clinical services. + +##### 7.1.1.1 ME metadata management + +The following recommendations apply to the ME metadata management: + +MEMDM-01: The ME metadata management is required to assign a unique identification number to each ME. + +NOTE 1 – This number act as a central identifier, linking all related data collected during the ME's lifecycle management for optimal organization and traceability. An example of identification number system is the unique device identification (UDI) system. + +MEMDM-02: The ME metadata management is recommended to classify the ME inventory metadata into three categories: + +- a) Core metadata: this is fundamental information necessary for the identification and management of ME including: + - i) Equipment identification number; + - ii) Type; + - iii) Function; + - iv) Manufacturer; + - v) Model; + - vi) Serial number; + - vii) Physical location within the hospital; + - viii) Operating status ("in service" or "out of service", the reason for being out of service); + - ix) Power requirements; + - x) Date inventory performed; + - xi) Maintenance service provider; + - xii) Purchase supplier. +- b) Lifecycle metadata: this includes data generated across the ME lifecycle in hospital. +- c) File attachment metadata: this provides supplementary information related to the ME, including: + - i) Standard operating procedures; + - ii) Maintenance procedures; + - iii) Software and firmware documentation; + - iv) Risk classification documentation. + +NOTE 2 – Risk classification of the devices is used to determine the frequency of inspections. + +MEMDM-03: The ME metadata management is recommended to create ME inventory metadata including at least core metadata. + +MEMDM-04: The ME metadata management can optionally create lifecycle metadata and file attachment metadata. + +MEMDM-05: The ME metadata management is recommended to allow authorized personnel to create, update, archive and delete metadata throughout the ME's lifecycle. + +MEMDM-06: The ME metadata management is recommended to support query and indexing functionalities to facilitate the efficient retrieval of metadata. + +##### 7.1.1.2 Consumable, accessories and spare parts management + +The following recommendations apply to CASP management: + +CASPM-01: The CASP management is recommended to maintain a separate inventory for ME consumables, accessories and spare parts. + +NOTE – CASP inventory includes: item identification number, type, description, quantity in stock, reorder threshold and supplier details. + +CASPM-02: The CASP management is recommended to include a notification module that triggers alerts when the quantity of a specific type of consumable, accessory or spare part falls below a predefined reorder threshold. + +CASPM-03: The CASP management is recommended to support one or more of the following notification methods: + +- Email; +- SMS; +- Push notifications via mobile applications. + +#### **7.1.2 ME lifecycle management module** + +Rather than referring to the product lifecycle of an item of ME, the term 'ME lifecycle' refers in a hospital to the process from procurement to decommissioning and disposal. ME lifecycle management in hospital ensures efficient utilization, cost control and compliance with regulatory standards and safety protocols throughout the ME lifecycle in hospital. + +This clause specifies the functional requirements of the ME management platform in the following phases of ME lifecycle in hospital: + +- Procurement; +- Installation, commissioning operation; +- Maintenance and repair; +- Decommissioning and disposal; +- Calibration (optional). + +NOTE – Calibration is applicable only to MEs that deliver therapeutic energy or perform measurements, which require periodic calibration to ensure outputs or readings remain within specified tolerances. While calibration is an essential aspect of ME management, it will be addressed in future studies. + +Maintenance management, being more complex, is covered separately in clause 7.1.3. + +##### **7.1.2.1 Procurement management** + +The following recommendations apply to the procurement management of ME lifecycle management: + +MELPM-01: It is recommended that the procurement management support the submission of ME procurement requests. + +MELPM-02: It is recommended that the procurement management support a role-based multilevel approval workflow for procurement requests to ensure compliance and transparency in the procurement process. + +MELPM-03: The procurement management is recommended to support procurement order tracking. + +##### **7.1.2.2 Installation and commissioning management** + +The following recommendation applies to the installation and commissioning management of ME lifecycle management: + +MELICM-01: The installation and commissioning management is recommended to support the documentation of installation and commissioning activities. + +NOTE – The documented information includes: initial performance and safety inspection reports, records of user operation training, installation date, commissioning outcomes and assigned personnel. + +##### **7.1.2.3 Decommissioning and disposal management** + +The following recommendations apply to the decommissioning and disposal management of ME lifecycle management: + +MELDM-01: The decommissioning and disposal management is recommended to support the submission of ME disposal requests. + +MELDM-02: The decommissioning and disposal management is recommended to support a role-based multilevel approval workflow for disposal requests to ensure compliance with disposal process. + +MELDM-03: The decommissioning and disposal management is recommended to support the querying and export of ME disposal records to facilitate asset verification by financial and audit departments. + +#### **7.1.3 ME maintenance management module** + +ME maintenance management is one of the core modules of the ME management platform. It aims to ensure the normal operation, prolonged lifespan, reduced failure rate and improved efficiency of ME. + +ME maintenance management can be divided into two major categories: + +- Inspection and preventive maintenance (IPM) [b-WHO-1] and +- Corrective maintenance (CM) [b-WHO-1]. + +These two categories represent the core activities in ME maintenance. + +In addition to the core activities (IPM and CM), there are three auxiliary maintenance activities that support the overall maintenance process: + +- Maintenance recording: records all maintenance activities, providing essential data for the other modules. +- Task notifications: ensure timely task execution, enhancing maintenance efficiency. +- Maintenance reporting: generates reports based on maintenance data to support decision-making. + +##### **7.1.3.1 IPM management** + +IPM includes all scheduled activities that ensure equipment functionality and prevent breakdowns or failures. The following recommendations apply to IPM management: + +MEIPM-01: The IPM management is recommended to allow administrators to set a periodic IPM schedule for each item of ME. + +MEIPM-02: The IPM management is recommended to automatically create an IPM work order according to the predefined schedules. + +MEIPM-03: The IPM management is recommended to support the lifecycle management of IPM work orders, including creation, assignment, execution and completion, to facilitate efficient task execution and tracking. + +##### **7.1.3.2 CM management** + +CM is a process used to restore the physical integrity, safety and/or performance of a device after a failure. The reduction of equipment downtime and minimization of overall expenses are at the centre of ME maintenance management. The following requirement and recommendations apply to CM management: + +MECM-01: The CM management is required to support failure reporting for ME. + +MECM-02: The CM management is recommended to automatically create a CM work order based on the submitted failure report. + +MECM-03: The CM management is recommended to support the assignment of work orders to designated clinical engineers. + +NOTE – If third-party service providers are involved, the assigned clinical engineers should be responsible for coordination. + +MECM-04: The CM management is recommended to support the lifecycle management of CM work orders, including creation, assignment, execution and completion, to facilitate efficient task execution and tracking. + +##### 7.1.3.3 Maintenance record management + +The following recommendations apply to the maintenance record management: + +MEMRM-01: The maintenance record management is recommended to ensure the integrity and accuracy of maintenance records once after submission, preventing unauthorized modifications. + +MEMRM-02: The maintenance record management is recommended to automatically generate a maintenance record upon the completion of a maintenance work order. + +MEMRM-03: The maintenance record management is recommended to support manual entry of maintenance records for situations where automatic generation is not feasible. + +MEMRM-04: The maintenance record management is recommended to log the user ID and timestamp for all manually entered records to ensure accountability. + +MEMRM-05: The maintenance record management is recommended to support the archiving and querying of maintenance records. + +##### 7.1.3.4 Maintenance task notification + +The following recommendation applies to the maintenance task notification: + +MEMTN-01: The maintenance task notification module is recommended to support maintenance task notification capabilities. + +NOTE 1 – Notifications are directed to the clinical engineer responsible for the maintenance task. + +NOTE 2 – Maintenance task notifications include assignment, completion, modification updates and reminders before the due date. + +##### 7.1.3.5 Maintenance reporting + +The following recommendations apply to the maintenance reporting management: + +MEMP-01: The maintenance reporting management is recommended to support the generation of reports based on user-defined criteria such as ME type, time range and maintenance task type. + +MEMP-02: The maintenance reporting management is recommended to analyse maintenance data to identify trends and patterns, such as fault frequency, downtime, and cost efficiency. + +MEMP-03: The maintenance reporting management is recommended to support graphical visualization of data, including charts, graphs and dashboards. + +MEMP-04: The maintenance reporting management is recommended to support report export. + +NOTE – Examples of formats for maintenance report export include comma-separated values (CSV) [b-IETF RFC 4180], portable document format (PDF) [b-ISO 32000-1] or JavaScript Object Notation (JSON) [b-ECMA]. + +### **7.2 Service layer requirements** + +#### **7.2.1 Data services** + +##### **7.2.1.1 Data collection module** + +Data collection module recommendations include: + +MESDC-01: The data collection module is recommended to support at least one of the following data collection modes: + +- Direct acquisition from ME; +- Data exchange with hospital information systems; +- Manual data entry. + +MESDC-02: The data collection module is recommended to support the direct acquisition of ME metadata through network interfaces. + +MESDC-03: The data collection module is recommended to support protocol adaptation mechanisms, such as protocol bridging or message translation, through embedded drivers or middleware to ensure interoperability with different communication protocols used by ME. + +MESDC-04: The data collection module is recommended to support data interoperability. + +NOTE – Three types of data interoperability include semantic interoperability, syntactical interoperability and object abstraction representation interoperability. + +MESDC-05: The data collection module is recommended to support collecting multiple types of data, such as structured data, unstructured data and semi-structured data. + +MESDC-06: The data collection module is recommended to support collecting data in different formats, such as text, image, video and audio. + +MESDC-07: The data collection module is recommended to support the configuration of data collection modes, including periodic and event-driven data collection. + +MESDC-08: The data collection module is recommended to support data aggregation, extraction, transformation and de-noising [ITU-T Y.3600]. + +##### **7.2.1.2 Data transmission module** + +Data transmission module requirements and recommendations include: + +MESDT-01: The data transmission module is required to support the transmission of ME metadata across the resource layer, service layer and application layer to enable bidirectional data flow within the system architecture. + +MESDT-02: The data transmission module is recommended to support rule-based data routing and filtering mechanisms to ensure that only relevant data is exchanged between layers in the ME management platform. + +MESDT-03: The data transmission module is recommended to support queuing, buffering and retransmission mechanisms to maintain data continuity under unstable network conditions or processing delays. + +##### **7.2.1.3 Data storage module** + +Data storage module recommendations include: + +MESDS-01: The data storage module is recommended to support the storage of multiple data formats, such as text, image, video and audio. + +MESDS-02: The data storage module is recommended to support the elastic storage capacity to accommodate different data categories and volumes of ME metadata. + +NOTE 1 – For categories of ME metadata, refer to clause 7.1.1. + +MESDS-03: The data storage module is recommended to support at least one type of storage, including block storage, object storage and file storage. + +MESDS-04: The data storage module is recommended to support at least one type of database to accommodate different categories of ME metadata. + +NOTE 2 – For categories of ME metadata, refer to clause 7.1.1. + +MESDS-05: The data storage module is recommended to support a storage hierarchy policy and data isolation mechanism to enable the separation of critical and non-critical ME metadata when needed. + +NOTE 3 – The classification of specific metadata fields into "critical" or "non-critical" is defined by the implementing hospital based on safety, regulatory compliance and essential management decisions. + +MESDS-06: The data storage module is recommended to support storage backup strategies to ensure data availability and recoverability. + +##### **7.2.1.4 Data analysis module** + +The platform can optionally support a data analysis module. If the module is implemented, the requirements and recommendations include: + +MESDA-01: The data analysis module is required to support at least one type of common data-processing algorithm. + +NOTE – Common data-processing algorithms include classification, clustering, regression and ranking. + +MESDA-02: The data analysis module is required to support batch processing capabilities for performing periodic analysis of accumulated ME metadata. + +MESDA-03: The data analysis module is recommended to support stream processing for enabling near-real-time analysis of incoming ME metadata. + +MESDA-04: The data analysis module is recommended to integrate artificial intelligence technologies to enhance data analysis capabilities. + +MESDA-05: The data analysis module is recommended to support distributed computing. + +MESDA-06: The data analysis module is recommended to support a scalable data-processing architecture to accommodate increasing volumes of ME metadata. + +MESDA-07: The data analysis module can optionally support priority-based scheduling for computing resource allocation according to task importance and urgency. + +MESDA-08: The data analysis module is recommended to support fault recovery capabilities to ensure continuous data analysis processes. + +##### **7.2.1.5 Data visualization module** + +The platform can optionally support a data visualization module. If the module is implemented, the requirements and recommendations include: + +MESDV-01: The data visualization module is required to support data visualization styles for different categories of ME metadata. + +NOTE 1 – Data visualization styles include chart-based, table-based, geographical, dynamic and interactive visualization. + +NOTE 2 – For categories of ME metadata, refer to clause 7.1.1. + +MESDV-02: The data visualization module is recommended to support customization of visual elements such as colours, labels and layouts. + +MESDV-03: The data visualization module is recommended to support application programming interfaces to enable developers to extend data visualization functionalities, such as custom chart type development and interactive feature customization. + +MESDV-04: The data visualization module is recommended to support an accessible user interface. + +##### **7.2.1.6 Data management module** + +Data management module requirements and recommendations include: + +MESDM-01: The data management module is required to support ME metadata lifecycle management, including metadata creation, storage, archiving and deletion. + +MESDM-02: The data management module is recommended to support data modelling. + +MESDM-03: The data management module is recommended to support monitoring capabilities. + +#### **7.2.2 Business services** + +##### **7.2.2.1 General requirements** + +General recommendations for business services include: + +MEBG-01: The business services are recommended to support the life cycle management of business services, including deployment, start/stop, upgrade, rollback and other operations. + +MEBG-02: The business services are recommended to support the scalability of business services. + +MEBG-03: The business services are recommended to provide an integrated task and workflow management engine to coordinate approval flows, task assignment and exception handling as specified by application-layer scenarios. + +##### **7.2.2.2 User identification module** + +User identification module recommendations include: + +MEBUI-01: The user identification module is recommended to support unique user identity verification through multifactor identification in comply with the hospital's cybersecurity policies. + +NOTE – Three types of multifactor identification are commonly used: knowledge-based (e.g., passwords), possession-based (e.g., SMS authentication codes, hardware tokens and CA certificates.), and inheritance-based (e.g., biometric identification). + +##### **7.2.2.3 Access control module** + +Access control module recommendations include: + +MEBAC-01: The access control module is recommended to support a role-based access control (RBAC) mechanism to define access scopes for users, MEs and subsystems. + +MEBAC-02: The access control module is recommended to support a master-sub-account system management with configurable access control policies for sub-accounts based on roles and responsibilities. + +MEBAC-03: The access control module is recommended to support fine-grained permission definitions at the level of services, APIs or data fields. + +##### **7.2.2.4 Gateway management module** + +Gateway management module recommendations include: + +MEBGM-01: The gateway management module is recommended to support integration with gateway components that provide basic gateway functions, such as protocol translation, traffic monitoring and traffic load balancing, and request routing. + +MEBGM-02: The gateway management module is recommended to support gateway security mechanisms [ITU-T X.1361]. + +MEBGM-03: The gateway management module is recommended to support integration with IoT gateway components and associated capabilities [ITU-T Y.4101]. + +##### **7.2.2.5 Logging service module** + +Logging service module requirements and recommendations include: + +MEBLS-01: The logging service module is required to support logging of user actions, service executions, data exchanges and system events. + +MEBLS-02: The logging service module is recommended to support near real-time log data collection and indexing. + +MEBLS-03: The logging service module is recommended to support log storage policies, such as those defining retention periods, archiving rules and access permissions, according to the specific operational requirements of the hospital. + +MEBLS-04: The logging service module is recommended to support near real-time log retrieval with multidimensional query methods, such as keywords, fuzzy, context, timestamp and SQL aggregation. + +### **7.3 Resources layer requirements** + +The resources layer of the platform includes computing, network, storage and other hardware and software resources. + +#### **7.3.1 Computing resources module** + +Computing resources module recommendations include: + +MECR-01: The computing resources module is recommended to be deployed over a hospital-managed private cloud infrastructure based on virtualization technology in [ITU-T Y.3510]. + +MECR-02: The computing resources module is recommended to support dynamic resource adjustment for virtual machines. + +MECR-03: The computing resources module is recommended to support distributed serving in [ITU-T F.748.38] to enable the coordinated scheduling and allocation of computing resources across multiple nodes deployed in different hospital locations. + +#### **7.3.2 Networking resources module** + +Networking resources module recommendations include: + +MENR-01: The networking resources module is recommended to support multiple networks, and three types of networking resources are considered: + +- Hospital extranet, which provides wired and wireless access and connects to the Internet through a secure exit; +- Hospital intranet, which provides wired and wireless access, connects to the hospital intranet cloud data centre, inter-campus networks and regulatory agencies, and connects to the hospital extranet and hospital IoT through a secure border; +- Hospital IoT, provides wireless IoT access, connects IoT platforms and IoT gateways, and connects to the hospital intranet through a secure border. + +MENR-02: The networking resources module is recommended to support the general requirements of networking resources defined in clause 8.1 of [ITU-T Y.3510]. + +MENR-03: The networking resources module is recommended to support dynamic flow adaptation based on the traffic generated by cloud services. + +MENR-04: The networking resources module can optionally support edge computing [b-ITU-T Y.3540] for performance optimization and security considerations of networking resources. + +#### **7.3.3 Storage resources module** + +Storage resources module recommendations include: + +MESR-01: The storage resources module is recommended to provide performance monitoring and statistics in [ITU-T Y.3510]. + +NOTE – Networking performance statistics include disk I/O speed, disk space usage, CPU utilization, memory utilization and job completion status. + +MESR-02: The storage resources module is recommended to support integration with data centres or storage services distributed across multiple hospital locations. + +## Bibliography + +- [b-ITU-T E.800] Recommendation ITU-T E.800 (2008), *Definitions of terms related to quality of service.* +- [b-ITU-T Y.3540] Recommendation ITU-T Y.3540 (2023), *Edge computing – Overview and high-level requirements.* +- [b-ITU-T Y.4563] Recommendation ITU-T Y.4563 (2021), *Requirements and functional model to support data interoperability in Internet of things environments.* +- [b-ECMA] ECMA 404 (2017), *The JSON data interchange format.* + +- [b-IETF RFC 4180] IETF RFC 4180 (2005), *Common Format and MIME Type for Comma-Separated Values (CSV) Files.* + +- [b-ISO 32000-1] ISO 32000-1:2008, *Document management – Portable document format – Part 1: PDF 1.7.* +- [b-WHO-1] World Health Organization (2011), *Medical equipment maintenance programme overview .* + +- [b-WHO-2] World Health Organization (2011), *Procurement process resource guide.* + +- [b-WHO-3] World Health Organization (2011), *Introduction to medical equipment inventory management.* + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.780.8-202508-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg b/marked/F/T-REC-F.780.8-202508-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f33c053bdb172a78c522a4178dfb659414106e2e --- /dev/null +++ b/marked/F/T-REC-F.780.8-202508-I_PDF-E/4801720824e4b5e2361a5564f91cfb70_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0a398c7b70fee4a4ecbdf625fdab3caa2b956126952322ebe2a7d4a9b2973bff +size 61451 diff --git a/marked/F/T-REC-F.780.8-202508-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/F/T-REC-F.780.8-202508-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..25f8fbfaef317942ed3102c74b40a7d42f80608b --- /dev/null +++ b/marked/F/T-REC-F.780.8-202508-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:83cab7dcabfe939b223d97242d8523e461968d21f7c5432693c30a64f2d95521 +size 7050 diff --git a/marked/F/T-REC-F.780.8-202508-I_PDF-E/raw.md b/marked/F/T-REC-F.780.8-202508-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..e0562d5d7efca851b07ef647c4a681fc8ba7d6b6 --- /dev/null +++ b/marked/F/T-REC-F.780.8-202508-I_PDF-E/raw.md @@ -0,0 +1,498 @@ + + +Recommendation + +**ITU-T F.780.8 (08/2025)** + +SERIES F: Non-telephone telecommunication services + +Multimedia services + +--- + +**Technical requirements of proactive health +data collection terminal equipment** + +![ITU logo](84a1d09fb489061482111515543b60dc_img.jpg) + +The logo of the International Telecommunication Union (ITU) is located in the bottom right corner. It features a blue globe with white lines representing latitude and longitude, and the letters 'ITU' in a bold, blue, sans-serif font overlaid on the globe. + +ITU logo + +# ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.780.8 + +# Technical requirements of proactive health data collection terminal equipment + +## Summary + +Recommendation ITU-T F.780.8 defines the concept of proactive health and distinguishes it from passive health. Proactive health data collection terminal equipment (PHTE) is designed to collect and transmit various types of proactive health data in order to monitor an individual's lifestyle and thereby maintain and enhance overall health status. + +This Recommendation provides the technical requirements of PHTE. It establishes a comprehensive framework covering functionality, performance, interoperability, safety, electromagnetic compatibility, environmental adaptability and durability. These requirements serve as a reference for the design, production and deployment of PHTE. The Recommendation provides definitions for key terms and concepts, including proactive health, proactive health data, proactive health system and proactive health data collection terminal equipment. Technical requirements for other components of the proactive health system architecture are beyond the scope of this Recommendation. + +## History\* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.780.8 | 2025-08-13 | 21 | 11.1002/1000/16434 | + +## Keywords + +Data collection, proactive health, proactive health data collection terminal equipment. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2025 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|---------------------------------------------------------------------------------------|------| +| 1 Scope..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 2 | +| 3.1 Terms defined elsewhere ..... | 2 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 3 | +| 6 Overview..... | 3 | +| 6.1 Proactive health ..... | 3 | +| 6.2 Proactive health data collection terminal equipment..... | 3 | +| 6.3 Proactive health data..... | 4 | +| 7 Technical requirements of proactive health data collection terminal equipment ..... | 4 | +| 7.1 Functions ..... | 4 | +| 7.2 Performance..... | 5 | +| 7.3 Interoperability ..... | 6 | +| 7.4 Safety ..... | 6 | +| 7.5 Electromagnetic compatibility..... | 6 | +| 7.6 Environmental adaptation and durability ..... | 6 | +| Appendix I – Architecture of proactive health system ..... | 7 | +| Appendix II – Representative data elements of proactive health data ..... | 9 | +| Bibliography..... | 11 | + + + +# Recommendation ITU-T F.780.8 + +## Technical requirements of proactive health data collection terminal equipment + +# 1 Scope + +This Recommendation specifies technical requirements of proactive health data collection terminal equipment regarding the following aspects: + +- Functions; +- Performance; +- Interoperability; +- Safety; +- Electromagnetic compatibility; +- Environmental adaptation and durability. + +# 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +- [[ITU-T F.743.21](#)] Recommendation ITU-T F.743.21 (2020), *Framework for data asset management*. +- [[ITU-T H.810](#)] Recommendation ITU-T H.810 (V5) (2019), *Interoperability design guidelines for personal connected health systems: Introduction*. +- [[ITU-T H.811](#)] Recommendation ITU-T H.811 (2017), *Interoperability design guidelines for personal connected health systems: Personal Health Devices interface*. +- [[ITU-T K.42](#)] Recommendation ITU-T K.42 (2024), *General principle for the definition of emission and immunity requirements for telecommunication/ICT equipment*. +- [[ITU-T L.1015](#)] Recommendation ITU-T L.1015 (2019), *Criteria for evaluation of the environmental impact of mobile phones*. +- [IEC 60529] IEC 60529:2019, *Degrees of protection provided by enclosures (IP code)*. +- [IEC 62368-1] IEC 62368-1: 2023, *Audio/video, Information and communication technology equipment – Part 1: Safety Requirements*. +- [ISO/IEEE 11073-10101] ISO/IEEE 11073-10101: 2020, *Health informatics – Device interoperability – Part 10101: Point-of-care medical device communication – Nomenclature*. +- [ISO/IEEE 11073-10441] ISO/IEEE 11073-10441: 2015, *Health informatics – Personal health device communication – Part 10441: Device specialization – Cardiovascular fitness and activity monitor*. + +| | | +|------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------| +| [ISO/IEEE 11073-10442] | ISO/IEEE 11073-10442: 2015, Health informatics – Personal health device communication – Part 10442: Device specialization – Strength fitness equipment . | +| [ISO/IEEE 11073-20601] | ISO/IEEE 11073-20601: 2019, Health informatics – Personal health device communication – Part 20601: Application profile – Optimized Exchange Protocol . | + +# 3 Definitions + +## 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 assistive technology (AT)** [b-ITU-T F.791]: Piece of equipment, product system, hardware, software or service that is used to enable, maintain or improve functional capabilities of individuals with disabilities. + +**3.1.2 data quality** [ITU-T F.743.21]: The integrity, normalization, consistency, accuracy, uniqueness, and relevance of data, which is the premise and guarantee for data mining. + +**3.1.3 person with age-related disabilities** [b-ITU-T F.791]: A person with cognitive or physical disabilities caused by the aging process. Examples are impaired eyesight, deafness in varying degrees, reduced mobility or cognitive abilities. + +**3.1.4 person with specific needs** [b-ITU-T F.791]: Includes persons with disabilities..., persons who are not literate, those with learning disabilities, children, indigenous people, older persons with age-related disabilities and anyone who has a temporary disability. + +## 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 personal health device interface (PHD-IF)**: Interface that connects one or more personal health device (e.g., sensor/actuator) client components to one or more personal health device (sensor/actuator) service components using transport media such as universal serial bus (USB), Bluetooth low energy, Bluetooth basic rate/enhanced data rate (BR/EDR), ZigBee or near-field communications. + +NOTE 1 – Examples of sensor service components include glucose meters, weighing scales and heart rate monitors. + +NOTE 2 – Definition based on [b-ITU-T H.810]. + +**3.2.2 proactive health**: A health management approach that continuously monitors an individual's lifestyle and optimizes it as the primary means of health intervention to maintain and enhance overall health status. + +NOTE – Lifestyle refers to physical activity, sleep, diet, music listened to and meditation and other aspects. + +**3.2.3 proactive health data collection terminal equipment (PHTE)**: A collection of devices designed to achieve proactive health goals that are capable of collecting and transmitting individual proactive health data, which is used to generate lifestyle-based health intervention recommendations. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|-----|----------------------| +| AT | Assistive Technology | +| BLE | Bluetooth Low Energy | + +| | | +|--------|-----------------------------------------------------| +| BR/EDR | Basic Rate/Enhanced Data Rate | +| DIM | Domain Information Model | +| GATT | Generic Attribute profile | +| iOS | iPhone Operating System | +| NFC | Near Field Communication | +| PHD-IF | Personal Health Device Interface | +| PHTE | Proactive Health data collection Terminal Equipment | + +# 5 Conventions + +In this Recommendation: + +- The keyword "is required" indicates a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. +- The keywords "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus, this requirement needs not be present to claim conformance. +- The keyword "can optionally" indicates an optional requirement which is permissible, without implying any sense of being recommended. These terms are not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +# 6 Overview + +## 6.1 Proactive health + +Proactive health is a health management approach that continuously monitors an individual's lifestyle – such as physical activity, sleep, diet, music listened to and meditation – and optimizes it as the primary means of health intervention, aiming to maintain and enhance overall health status. + +The differences between proactive health and passive health are illustrated in Table 1: + +**Table 1 – Differences between proactive health and passive health** + +| Dimension | Proactive health | Passive health
(Traditional medicine) | +|--------------------|---------------------------------------------------------------|------------------------------------------------------| +| Intervention stage | Pre-disease stage (before disease onset or subclinical stage) | Post-disease stage (after diagnosis) | +| Target objective | Maintain health status | Eliminate disease symptoms, control complications | +| Methods | Lifestyle optimization | Primarily medication and surgery | +| Data nature | High-frequency, continuous | Low-frequency, discrete | +| Individual role | Active participant (lifestyle decision-maker) | Passive recipient (follower of medical instructions) | + +## 6.2 Proactive health data collection terminal equipment + +The term proactive health data collection terminal equipment (PHTE) refers to a set of devices capable of collecting and transmitting proactive health data. The collected data is used to generate + +lifestyle-based health intervention recommendations, enabling users to achieve their active health objectives. + +As shown in Appendix I, the PHTE serves as a core component of the proactive health system. Proactive health primarily relies on non-medical health interventions to maintain and enhance an individual's health status. The primary function of this type of equipment is to collect health-related data from individuals while supporting data transmission. Through a gateway, the collected proactive health data is uploaded to a health platform, where it undergoes processing and analysis. Based on the analysis, the platform generates personalized lifestyle-based health intervention recommendations and provides feedback to the user, thereby forming a closed-loop intervention process to achieve proactive health objectives. + +In terms of data transmission, the proactive health system architecture follows [ITU-T H.810]. + +## **6.3 Proactive health data** + +All data collected by the PHTE is referred to as proactive health data. + +Proactive health data encompasses the following four categories: + +- User information: refers to data that, either individually or in combination with other information, can be used to identify a specific individual. +- Human body-related metrics: refers to measurable indicators of an individual's physiological and biochemical state. +- Behavioural data: refers to information related to an individual's lifestyle including dietary habits, exercise routines, sleep patterns and emotional state. +- Environmental data: refers to information about the external conditions and surroundings that influence an individual's health and well-being. + +The recommended measurement contents for each data categories are presented in Appendix II. + +# **7 Technical requirements of proactive health data collection terminal equipment** + +## **7.1 Functions** + +### **7.1.1 Data collection** + +FAC-01: It is recommended that PHTE support the collection of four categories of proactive health data, user information, human body-related metrics, behavioural data and environmental data. + +NOTE – The recommended measurement contents for each data categories are presented in Appendix II. + +FAC-02: It is recommended that PHTE collect the corresponding categories of proactive health data in accordance with the device specialization in [ISO/IEEE 11073-10441] and [ISO/IEEE 11073-10442] for different device types. + +FAC-03: If the PHTE supports multimodal sensors, it is recommended that the device perform synchronized data collection across multiple channels and report the data with unified timestamp packaging. + +### **7.1.2 Data transmission** + +FAT-01: It is recommended that PHTE support multiple communication modes concurrently and be capable of automatically selecting the optimal link based on the deployment scenario. + +FAT-02: It is recommended that PHTE support a local caching mechanism to temporarily store data when the network is unavailable or the health platform is inaccessible, and to automatically retransmit the cached data once communication is restored. + +FAT-03: It is recommended that PHTE support identity authentication and data access authorization mechanisms to ensure that data transmission adheres to privacy protection principles and the principle of least privilege. + +### 7.1.3 Cloud computing + +FCC-01: It is recommended that PHTE support the establishment of communication with gateway devices via standardized communication protocols and enable the forwarding of data to cloud interfaces through the gateway. + +FCC-02: PHTE can optionally support receiving feedback instructions from the cloud via the gateway and be capable of triggering corresponding local actions, thereby enabling timely response to personalized lifestyle-based health intervention tasks. + +NOTE – Local actions include vibration alerts or adjustment of data collection frequency. + +### 7.1.4 Artificial intelligence + +FAI-01: It is recommended that PHTE support artificial intelligence modules deployed either locally or in the cloud, for purposes including: + +- Health status classification; +- Predictive analysis; +- Health intervention recommendations based on the individual's lifestyle. + +## 7.2 Performance + +### 7.2.1 Data quality + +PDQ-01: It is recommended that the proactive health data collected by PHTE provide the relevant data quality dimensions described in [ITU-T F.743.21] to guarantee the subsequent use of data in proactive health systems. + +PDQ-02: It is recommended that the PHTE include a comprehensive contextual metadata collection mechanism, to enable reasonable assessment of data quality and semantic interpretation and to ensure applicability across different use scenarios. + +### 7.2.2 Accessibility and usability + +PAC-01: It is recommended that the user interface of PHTE be designed to be accessible to persons with disabilities, persons with age-related disabilities and those with specific needs [b-ITU-T Y.4219]. + +PAC-02: It is recommended that PHTE be compatible with assistive technologies used by the user [b-ITU-T Y.4219]. + +### 7.2.3 Compatibility + +PC-01: It is recommended that PHTE support the deployment of drivers and the execution of associated applications on major operating system platforms. + +NOTE 1 – This clause is applicable to PHTE that configures parameters or transmits data through desktop or mobile applications. + +NOTE 2 – Examples of major operating system platforms include Windows, Linux, Android and iOS. + +PC-02: It is recommended that PHTE interoperate with at least one type of gateway device and adapt to the corresponding capabilities. + +NOTE 3 – Types of gateways include the Internet of things (IoT) gateway [b-ITU-T Y.4101] and edge computing gateway [b-ITU-T Y.4122]. + +## **7.3 Interoperability** + +According to the interface classification defined in [ITU-T H.811], the personal health device interface (PHD-IF) supports two technical pathways: the [ISO/IEEE 11073-20601]-based pathway (referred to as the "X73" interface pathway) and the Bluetooth low energy (BLE) pathway. + +### **7.3.1 General requirements** + +IG-01: It is required that PHTE support either the X73 pathway or the BLE pathway, depending on its communication architecture. + +IG-02: It is recommended that PHTE be interoperable. + +IG-03: It is recommended that PHTE support the nomenclature conventions specified in [ISO/IEEE 11073-10101]. + +IG-04: It is recommended that PHTE define its domain information model (DIM) and configuration details based on its functional characteristics, with reference to [ISO/IEEE 11073-10441] and [ISO/IEEE 11073-10442] device specialization standards. + +### **7.3.2 Requirements based on the X73 pathway** + +For PHTE implementing the X73 pathway, the following interoperability requirements apply: + +- IG-05: It is required that PHTE support physical transport technologies in [ITU-T H.811]. +- IG-06: It is required that the application-layer protocol remain consistent across different physical transport layers in order to ensure semantic interoperability during data exchange. +- IG-07: It is recommended that PHTE implement the three models, the DIM, the service model and the communication model, defined in [ISO/IEEE 11073-20601]. + +### **7.3.3 Requirements based on the BLE pathway** + +For PHTE implementing the BLE pathway, the following interoperability requirements apply: + +- IG-08: It is recommended that PHTE construct its generic attribute profile (GATT) service structure in [ITU-T H.811]. +- IG-09: It is recommended that the semantics of the PHTE transmitted data conform to the terminology in [ISO/IEEE 11073-10101] and remain mappable to the DIM. + +## **7.4 Safety** + +### **7.4.1 Electrical safety** + +SES-01: It is required that PHTE comply with the relevant provisions of [IEC 62368-1]. + +### **7.4.2 Material safety** + +SMS-01: It is recommended that the materials used in PHTE comply with the materials restrictions specified in [ITU-T L.1015]. + +## **7.5 Electromagnetic compatibility** + +EMC-01: It is recommended that PHTE comply with the relevant provisions of [ITU-T K.42]. + +## **7.6 Environmental adaptation and durability** + +ER-01: It is recommended that PHTE comply with international ingress protection standards [IEC 60529]. + +## Appendix I + +### Architecture of proactive health system + +(This appendix does not form an integral part of the Recommendation) + +To achieve proactive health objectives, the proactive health system consists of four key components: + +- Proactive health data collection terminal equipment: responsible for collecting and transmitting individual proactive health data. These terminals serve as the primary data source for the proactive health system and encompass multidimensional information, including user information, human body-related metrics, behavioural data and environmental data. +- Gateway: receives data from the PHTE and transmits it to the cloud platform via a communication network. The gateway can be a physical device or virtualized (i.e., integrated within the PHTE, directly interacting with the communication network). +- Health cloud platform: stores, processes and analyses proactive health data, providing a comprehensive assessment of an individual's health status and generating personalized lifestyle optimization recommendations. +- Four key stakeholders: include user, hospital, manufacturer/third party health service provider and public sector. + +Figure I.1 shows the ecosystem of proactive health data. After proactive health data reaches the health cloud platform, it can be subsequently utilized by four stakeholders: + +- User: retrieves proactive health data from the health cloud platform, receives health intervention recommendations and improves lifestyle choices to prevent diseases and enhance individual health. +- Hospital: accesses patients' proactive health data to support diagnosis and chronic disease management, thereby improving diagnostic efficiency, reducing misdiagnosis rates and enabling precision medicine. +- Manufacturer / third party health service provider: collects users' proactive health data to offer personalized health services. +- Public sector: analyses population health data to formulate public health policies, optimize resource allocation and enhance overall population health. + +![Diagram of the ecosystem of proactive health data showing data flow from active health data collection terminal equipment through a gateway to a health cloud platform, which then serves users, hospitals, manufacturers, and the public sector. Protocols like IEEE, ISO, X73, BLE, Bluetooth, Zigbee, USB, HL7, IHE, IETF, and W3C are indicated along the data paths.](4801720824e4b5e2361a5564f91cfb70_img.jpg) + +The diagram illustrates the ecosystem of proactive health data. On the left, 'Active health data collection terminal equipment' (represented by icons of a smartwatch, a person, and a smartphone) connects to a 'Gateway' (represented by a wireless signal icon). The connection is labeled with protocols: IEEE, ISO, X73 pathway, BLE pathway, Bluetooth, Zigbee, and USB. The Gateway connects to a 'Health cloud platform' (represented by a cloud and three computer monitors). This connection is labeled with protocols: HL7, IHE, IETF, and W3C. The Health cloud platform then connects to four entities: 'User' (represented by a person icon), 'Hospital' (represented by a building icon), 'Manufacturer' (represented by a factory icon), and 'Public sector' (represented by a classical building icon). These connections are labeled with protocols: HL7, IHE, IETF, and W3C. The Public sector label also includes the reference 'F.780.8(25)'. + +Diagram of the ecosystem of proactive health data showing data flow from active health data collection terminal equipment through a gateway to a health cloud platform, which then serves users, hospitals, manufacturers, and the public sector. Protocols like IEEE, ISO, X73, BLE, Bluetooth, Zigbee, USB, HL7, IHE, IETF, and W3C are indicated along the data paths. + +Figure I.1 – Ecosystem of proactive health data + +## Appendix II + +### Representative data elements of proactive health data + +(This appendix does not form an integral part of the Recommendation) + +Table II.1 shows categories and corresponding representative data elements of proactive health data. + +**Table II.1 – Categories and corresponding representative data elements** + +| Category | Representative data elements | +|-----------------------------------|---------------------------------------------| +| User information | User profile attributes: User ID | +| | User profile attributes: Gender | +| | User profile attributes: Age | +| Human body-related metrics | Body height | +| | Body weight | +| | Heart rate | +| | Respiration rate/volume | +| | Blood oxygen saturation (SpO 2 ) | +| | Skin/body temperature | +| | Electrocardiogram | +| | Electrodermal activity | +| | Skin conductance | +| | Bio-impedance | +| | Electromyography | +| Behavioural data | Location: Latitude | +| | Location: Longitude | +| | Location: Altitude | +| | Step count | +| | 3D Acceleration X/Y/Z | +| | 3D Angular acceleration (pitch/roll/yaw) | +| | Distance | +| | Speed | +| | Stride length | +| | Slope | +| | Energy expended | +| | Activity time | +| | Activity intensity | +| | Program identifier | +| | Sustained activity threshold | +| | Session/sub-session | +| | Calories intake | +| | Carbohydrate calories ingested | + +**Table II.1 – Categories and corresponding representative data elements** + +| Category | Representative data elements | +|---------------------------|-------------------------------------| +| Environmental data | Ambient temperature | +| | Humidity | +| | Light | +| | Noise level | +| | Air quality | +| | Environmental scenario tag: Outdoor | +| | Environmental scenario tag: Indoor | + +## Bibliography + +- [b-ITU-T F.791] Recommendation ITU-T F.791 (2018), *Accessibility terms and definitions*. +- [b-ITU-T Y.4101] Recommendation ITU-T Y.4101/Y.2067 (2017), *Common requirements and capabilities of a gateway for Internet of things applications*. +- [b-ITU-T Y.4122] Recommendation ITU-T Y.4122 (2021), *Requirements and capability framework of the edge-computing-enabled gateway in the Internet of things*. +- [b-ITU-T Y.4219] Recommendation ITU-T Y.4219 (2023), *Accessibility requirements for user interface of smart applications supporting IoT*. + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.781.2-202406-I_PDF-E/raw.md b/marked/F/T-REC-F.781.2-202406-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..12d631ee7793073457d030c9c38769462e28ef7d --- /dev/null +++ b/marked/F/T-REC-F.781.2-202406-I_PDF-E/raw.md @@ -0,0 +1,353 @@ + + +# Recommendation **ITU-T F.781.2 (06/2024)** + +SERIES F: Non-telephone telecommunication services +Multimedia services + +--- + +## **Quality assessment requirements for artificial intelligence/machine learning-based software as a medical device** + +![ITU logo](390120de4fe440c42fea8154fcaad334_img.jpg) + +The logo of the International Telecommunication Union (ITU), featuring the letters 'ITU' in blue inside a circle with a globe-like grid pattern. + +ITU logo + +## ITU-T F-SERIES RECOMMENDATIONS **Non-telephone telecommunication services** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| MULTIMEDIA SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.781.2 + +# Quality assessment requirements for artificial intelligence/machine learning-based software as a medical device + +## Summary + +With the advent of artificial intelligence/machine learning (AI/ML) and its strength in faster and more accurate disease detection and diagnosis, more timely and widespread adoption of decision-making assistant (DMA) software as a medical device (SaMD) could improve health for human beings. However, this does not mean the AI/ML-based DMA-SaMD for decision-making is ready for the clinic; AI/ML technology can only be used with complete confidence if it has been quality controlled through a rigorous evaluation in a standardized way. Performance and usability shall be assessed under a reliable and rigorous evaluation with a robust method to substantiate AI/ML-based DMA-SaMD quality. + +Recommendation ITU-T F.781.2 provides a requirement framework for the quality assessment with a perspective of lifecycle management for AI/ML-based DMA-SaMD. It describes the quality assessment principles and process in the life cycle of AI/ML-based DMA-SaMD, including requirement analysis, data collection, algorithm design, verification and validation, change control and other stages when using AI/ML technology to assist medical staff in making clinical decisions by providing suggestions on diagnostic and treatment activities. + +## History \* + +| Edition | Recommendation | Approval | Study Group | Unique ID | +|---------|----------------|------------|-------------|--------------------| +| 1.0 | ITU-T F.781.2 | 2024-06-13 | 16 | 11.1002/1000/15929 | + +## Keywords + +Artificial intelligence, machine learning, quality assessment, software as a medical device. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, and information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents/software copyrights, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the appropriate ITU-T databases available via the ITU-T website at . + +© ITU 2024 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|--------------------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 2 | +| 4 Abbreviations and acronyms ..... | 2 | +| 5 Conventions ..... | 2 | +| 6 Background..... | 2 | +| 7 Main principles ..... | 2 | +| 7.1 Algorithm characteristics..... | 3 | +| 7.2 Risk-oriented ..... | 3 | +| 7.3 Total life cycle management ..... | 3 | +| 8 Quality assessment life cycle of AI/ML-based DMA-SaMD..... | 3 | +| 8.1 Requirement analysis..... | 4 | +| 8.2 Data collection..... | 4 | +| 8.3 Algorithm design..... | 4 | +| 8.4 Verification and validation ..... | 5 | +| 8.5 Change control..... | 5 | +| Bibliography..... | 6 | + + + +# Recommendation ITU-T F.781.2 + +# Quality assessment requirements for artificial intelligence/machine learning-based software as a medical device + +## 1 Scope + +This Recommendation sets a basic framework for the quality assessment of artificial intelligence/machine learning (AI/ML) based decision-making assistant (DMA) software as a medical device (SaMD), which uses AI/ML technology to assist medical staff in making clinical decisions by providing suggestions on diagnostic and treatment activities. Intended uses include assisting triage, assisting evaluation, assisting detection and assisting diagnosis. The total life cycle for quality assessment includes requirements analysis, data collection, algorithm design, verification and validation, change control and other stages. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +[IEC 62304] IEC 62304:2006, *Medical device software – Software life cycle processes*. + +[ISO 13485] ISO 13485:2016, *Medical devices – Quality management systems – Requirements for regulatory purposes*. + +[ISO 14971] ISO 14971:2019, *Medical devices – Application of risk management to medical devices*. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 artificial intelligence** [b-ITU-T F.748.17]: An interdisciplinary field, usually regarded as a branch of computer science, dealing with models and systems for the performance of functions generally associated with human intelligence, such as reasoning and learning. + +**3.1.2 machine learning** [b-ITU-T Y.3172]: Processes that enable computational systems to understand data and gain knowledge from it without necessarily being explicitly programmed. + +NOTE 1 – This definition is adapted from [b-ETSI GR ENI 004]. + +NOTE 2 – Supervised machine learning and unsupervised machine learning are two examples of machine learning types. + +**3.1.3 medical device** [b-ITU-T K.145]: Instrument, apparatus, appliance, software, material or other article, whether used alone or in combination, together with any accessories, including the software intended by its manufacturer to be used specifically for diagnostic and/or therapeutic purposes and necessary for its proper application, intended by the manufacturer to be used for human beings for the purpose of: + +- diagnosis, prevention, monitoring, treatment or alleviation of disease, + +- diagnosis, monitoring, treatment, alleviation of or compensation for an injury or handicap, +- investigation, replacement or modification of the anatomy or of a physiological process, +- control of conception, + +and which does not achieve its principal intended action in or on the human body by pharmacological, immunological or metabolic means, but which may be assisted in its function by such means. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following term: + +**3.2.1 decision-making assistant software as a medical device:** Software designed for one or more medical purposes, independently of any medical hardware device, with the primary objective of assisting users, such as medical staff and patients, in making medical decisions. It provides suggestions and guidance pertaining to various diagnostic and treatment activities, encompassing functions such as triage assistance, assessment support, detection aid and diagnosis assistance. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|----------|--------------------------------------------------------| +| AI | Artificial Intelligence | +| DMA-SaMD | Decision-Making Assistant Software as a Medical Device | +| ML | Machine Learning | + +## 5 Conventions + +In this Recommendation: + +The keywords "shall" and "is required to" indicate a requirement which must be strictly followed and from which no deviation is permitted if conformance to this document is to be claimed. + +The keywords "can" and "is recommended" indicate a requirement which is recommended but which is not absolutely required. Thus this requirement needs not be present to claim conformance. + +The keywords "can optionally" and "may" indicate an optional requirement which is permissible, without implying any sense of being recommended. These terms are not intended to imply that the vendor's implementation must provide the option and the feature can be optionally enabled by the network operator/service provider. Rather, it means the vendor may optionally provide the feature and still claim conformance with the specification. + +## 6 Background + +With the advent of ML/AI and its strength in faster and more accurate disease detection and diagnosis, more timely and widespread adoption of AI/ML-based DMA-SaMD could improve health for human beings. Many reports conclude that an AI/ML algorithm can identify a particular condition from medical images as effectively as a radiologist; however, that alone does not mean the diagnosis is ready for the clinic. Reliable and rigorous evaluation with a robust method is required to substantiate AI/ML-based DMA-SaMD performance and usability. + +## 7 Main principles + +The quality assessment lifecycle management for AI/ML-based DMA-SaMD shall be based on the intended applications, application scenarios and core functionalities of software. It shall take algorithm characteristics as the core, take risk management as the orientation and meet the quality management system requirements. + +### 7.1 Algorithm characteristics + +Algorithm is the core of AI technology. Different types of AI algorithms, such as supervised learning and unsupervised learning, data-driven and model-driven, white-box and black-box algorithms, have different characteristics and different application scenarios. Therefore, the quality control requirements of AI/ML-based DMA-SaMD is recommended to focus on the generalization ability, bias control, robustness, interpretability and other issues of the algorithm based on the algorithm characteristics, combined with its intended use, application scenarios and core functions. At the same time, AI algorithms can optionally be used in combination. It is recommended to consider the overall quality control requirements of the algorithm combination based on the characteristics of these algorithms. + +### 7.2 Risk-oriented + +The risk levels and requirements for risk control measures of AI/ML-based DMA-SaMD depend on its intended uses, application scenarios and core functions; see [IEC 62304] for details. + +The software risk level may be expressed by the software safety class – which is divided into three levels, A, B and C – and can be comprehensively judged based on the intended use, application scenario and core function of the software. The specific determination method is detailed in [IEC 62304]. + +Software risk management activities shall be carried out according to [ISO 14971] requirements. Risk identification and risk control measures are required to reduce risks such as over-fitting and under-fitting, false negative and false positive to acceptable levels. + +### 7.3 Total life cycle management + +AI/ML-based DMA-SaMD is required to comply to the requirements of the quality management system (see [ISO 13485] for details) and refer to relevant standards and good engineering practices. The life cycle process of AI medical devices shall be considered and the product quality assurance work shall match the software safety level to ensure that risk management and traceability analysis run through the total life cycle. The following points apply: + +- Adequate, appropriate and effective verification and validation activities are required before marketing, ensuring that the generalization ability of the algorithm meets the needs of users, in order to identify foreseeable risks and reduce them to an acceptable level, and clarify product use restrictions and necessary warning information. +- Research on algorithm generalization ability is recommended to be continued after the listing. It may identify unforeseen risks in the early stage based on user complaints, adverse events and recalls, and enable effective risk control measures to reduce risks to an acceptable level. +- According to the software change requirements, appropriate verification and validation activities are required when implementing change activities after evaluation, to ensure that the generalization ability of the algorithm can continuously meet the needs of users. + +## 8 Quality assessment life cycle of AI/ML-based DMA-SaMD + +The life cycle of AI/ML-based DMA-SaMD shall be based on the software life cycle process, mainly including requirement analysis, data collection, algorithm design, verification and validation, change control and other stages. + +![Figure 1 – Quality assessment life cycle of AI/ML-based DMA-SaMD. The diagram shows a five-step flow: Requirement analysis, Data collection, Algorithm design, Verification and validation, and Change control. Each step is in a box with sub-points. Requirement analysis includes Data sources and Algorithm performance. Data collection includes Data collection, Data cleansing, Data annotation, and Data set construction. Algorithm design includes Algorithm selection, Algorithm training, and Algorithm performance evaluation. Verification and validation include Software verification and Software validation. Change control is the final step.](a5ee5c23b6dc52ec1d724b76d5a5f58f_img.jpg) + +``` + +graph LR + A[Requirement analysis +◆ Data sources +◆ Algorithm performance] --> B[Data collection +◆ Data collection +◆ Data cleansing +◆ Data annotation +◆ Data set construction] + B --> C[Algorithm design +◆ Algorithm selection +◆ Algorithm training +◆ Algorithm performance evaluation] + C --> D[Verification and validation +◆ Software verification +◆ Software validation] + D --> E[Change control] + +``` + +F.781.2(24) + +Figure 1 – Quality assessment life cycle of AI/ML-based DMA-SaMD. The diagram shows a five-step flow: Requirement analysis, Data collection, Algorithm design, Verification and validation, and Change control. Each step is in a box with sub-points. Requirement analysis includes Data sources and Algorithm performance. Data collection includes Data collection, Data cleansing, Data annotation, and Data set construction. Algorithm design includes Algorithm selection, Algorithm training, and Algorithm performance evaluation. Verification and validation include Software verification and Software validation. Change control is the final step. + +**Figure 1 – Quality assessment life cycle of AI/ML-based DMA-SaMD** + +### 8.1 Requirement analysis + +The requirement analysis is recommended to combine the intended use, application scenarios and core functions of the software, and comprehensively consider the requirements of regulations, standards, users, products, data, functions, performance, interfaces, user interfaces, cybersecurity and warning prompts to meet the clinical needs and compliance requirements, focusing on data sources, algorithm performance and other requirements. The following points apply: + +- Data source: the compliance, sufficiency and diversity of data sources, the scientific and rational distribution of data, and the sufficiency, effectiveness and accuracy of data quality controlling shall be considered. +- Algorithm performance: the applicability and requirements of false negative and false positive, repeatability and reproducibility, robustness, real-time and other performance indicators shall be comprehensively considered in combination with medical practice and product positioning. + +### 8.2 Data collection + +Data collection shall consider the quality control requirements of data acquisition, sorting, annotation, data set construction and other activities to ensure data quality and algorithm design quality. The following points apply: + +- Data acquisition: considering the quality control requirements for acquisition equipment, acquisition process and data desensitization. +- Data sorting: considering the quality control requirements for data cleaning and data pre-processing. +- Data annotation: considering the requirements of annotation resource management, annotation process quality control and annotation quality evaluation. +- Data set construction: considering the construction requirements of training set, tuning set and testing set. Specify the division method, division basis and data distribution proportion of training set, tuning set and testing set. Also, consider the risks introduced by data amplification. + +### 8.3 Algorithm design + +Algorithm design shall consider the quality control requirements of algorithm selection, algorithm training, algorithm performance evaluation and other activities. For the black-box algorithm, the algorithm design shall analyse the factors affecting the algorithm performance, and it is recommended to establish the association with the existing medical knowledge to improve the interpretability of the algorithm. The following points apply: + +- The algorithm selection shall consider the basic information such as algorithm type, algorithm structure, input/output data type, flow chart, algorithm programming framework and operating environment. +- Algorithm training: training and tuning shall be conducted based on training set and tuning set, and the requirements for evaluation indicators, training methods, training objectives, tuning methods and training data volume evaluation indicator curve shall be specified. The training data volume evaluation index curve should be able to verify the adequacy and effectiveness of the algorithm training. +- Algorithm performance evaluation: the algorithm design results shall be evaluated based on the testing set to verify that the algorithm performance meets the algorithm design requirements, such as for false negative and false positive, repeatability and reproducibility, robustness, real-time and other performance indicators. + +### **8.4 Verification and validation** + +This phase consists of two steps: + +- Software verification + +Software verification is required to confirm that the output of a certain stage of software development and software change meets the input requirements by providing objective evidence, including software verification testing (unit testing, integration testing, system testing), design review and other activities. + +- Software validation + +Software validation is required to confirm that the software meets user needs and expected purposes by providing objective evidence, including software validation testing (user testing), clinical evaluation, design review and other series of activities. + +### **8.5 Change control** + +In case of algorithm change and software change, verification and validation activities appropriate to the type, content and extent of algorithm change and software change shall be carried out according to the requirements of the quality management system. Risk management and traceability analysis shall be carried out throughout the change process. + +A rollback mechanism shall be considered for algorithm change and software change to ensure the continuity of medical business, especially for software with high risk. + +## Bibliography + +- [b-ITU-T F.748.17] Recommendation ITU-T F.748.17 (2022), *Technical specification for artificial intelligence cloud platform – Artificial intelligence model development*. +- [b-ITU-T K.145] Recommendation ITU-T K.145 (2019), *Assessment and management of compliance with radio frequency electromagnetic field exposure limits for workers at radiocommunication sites and facilities*. +- [b-ITU-T Y.3172] Recommendation ITU-T Y.3172 (2019), *Architectural framework for machine learning in future networks including IMT-2020*. +- [b-ETSI GR ENI 004] ETSI GR ENI 004 (2021), *Experiential Networked Intelligence (ENI); Terminology for Main Concepts in ENI*. +- [b-FDA] United States Food and Drug Administration (2019), *Proposed Regulatory Framework for Modifications to Artificial Intelligence/Machine Learning (AI/ML) Based Software as Medical Device*, FDA discussion paper. + +- [b-IEC/TR 80001-2-2] IEC/TR 80001-2-2:2012, *Application of risk management for IT-networks incorporating medical devices – Part 2-2: Guidance for the disclosure and communication of medical device security needs, risks and controls*. +- [b-IEC/TR 80001-2-8] IEC/TR 80001-2-8:2016, *Application of risk management for IT-networks incorporating medical devices – Part 2-8: Application guidance – Guidance on standards for establishing the security capabilities identified in IEC 80001-2-2*. +- [b-IEC/TR 80001-2-9] IEC/TR 80001-2-9:2017, *Application of risk management for IT-networks incorporating medical devices – Part 2-9: Application guidance – Guidance for use of security assurance cases to demonstrate confidence in IEC/TR 80001-2-2 security capabilities*. +- [b-IMDRF/AIMD WG/N67] International Medical Device Regulators Forum (IMDRF) (2021), *Machine Learning-enabled Medical Devices – A subset of Artificial Intelligence-enabled Medical Devices: Key Terms and Definitions*. +- [b-IMDRF/SaMD WG/N10] International Medical Device Regulators Forum (IMDRF) (2013), *Software as a Medical Device: Key Definitions*. +- [b-IMDRF/SaMD WG/N12] International Medical Device Regulators Forum (IMDRF) (2014), *Software as a Medical Device: Possible Framework for Risk Categorization and Corresponding Considerations*. +- [b-IMDRF/SaMD WG/N23] International Medical Device Regulators Forum (IMDRF) (2015), *Software as a Medical Device: Application of Quality Management System*. +- [b-IMDRF/SaMD WG/N41] International Medical Device Regulators Forum (IMDRF) (2017), *Software as a Medical Device: Clinical Evaluation*. + +- [b-IMDRF/CYBER WG/N60] International Medical Device Regulators Forum (IMDRF) (2020), *Principles and Practices for Medical Device Cybersecurity*. +- [b-NMPA-1] National Medical Products Administration (NMPA) (2022), *Guidelines for the premarket review of artificial intelligence medical devices*, NMPA guidance [in Chinese]. + +- [b-NMPA-2] National Medical Products Administration (NMPA) (2022), *Guidelines for the premarket review of Medical Device Software* [in Chinese]. + +- [b-NMPA-3] National Medical Products Administration (NMPA) (2022), *Guidelines for the premarket review of Medical Device Cybersecurity* [in Chinese]. + + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.790-200701-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.790-200701-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..0431ebd0a0489f1035bb5c760e82ed38e2d8bfd0 --- /dev/null +++ b/marked/F/T-REC-F.790-200701-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8c2df115963fdd4621278e7f9cc5eca9240a8469058b4785a87cad40c18e740f +size 8162 diff --git a/marked/F/T-REC-F.790-200701-I_PDF-E/e6df2733626a85205c1db682e6259c46_img.jpg b/marked/F/T-REC-F.790-200701-I_PDF-E/e6df2733626a85205c1db682e6259c46_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..cc5c6554385bac273cd8a739311a8391c3ae6a4f --- /dev/null +++ b/marked/F/T-REC-F.790-200701-I_PDF-E/e6df2733626a85205c1db682e6259c46_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:becbd9bffff5ae38f44c522da22f9ef83b0cb8d5a8d246891563913971d838c3 +size 61524 diff --git a/marked/F/T-REC-F.791-201808-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/F/T-REC-F.791-201808-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..5295e5f3b359f1cb7eb589492906ba42372ce460 --- /dev/null +++ b/marked/F/T-REC-F.791-201808-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b606969d95fd7f6efdad13ce9473bd814fc626737a78b4ace581548907b98c87 +size 7056 diff --git a/marked/F/T-REC-F.791-201808-I_PDF-E/raw.md b/marked/F/T-REC-F.791-201808-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..4550dde8b570e06e7b0b59a993599c0a9378e594 --- /dev/null +++ b/marked/F/T-REC-F.791-201808-I_PDF-E/raw.md @@ -0,0 +1,767 @@ + + +International Telecommunication Union + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.791** + +(08/2018) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Multimedia services + +# --- Accessibility terms and definitions + +Recommendation ITU-T F.791 + +ITU-T + +![ITU logo](1d7527f4316cfe2d342b08d1653d1592_img.jpg) + +The logo of the International Telecommunication Union (ITU) features a stylized globe with a red lightning bolt striking across it. The letters 'ITU' are prominently displayed in blue and white on the globe. + +ITU logo + +International +Telecommunication +Union + +# ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.791 + +# Accessibility terms and definitions + +## Summary + +Recommendation ITU-T F.791 defines terms describing accessibility, disability and technical vocabulary to be used for improving the drafting of standards, and to facilitate accurately the needs and the mainstreaming of accessibility in standards that will include persons with disabilities (PWDs), older persons with age-related disabilities and persons with specific needs. WTDC Res. 58, WTDC AD and Recommendation ITU-T F.790 provide additional background information + +With the passage of the United Nations *Convention on the Rights of Persons with Disabilities* [b-UNCRPD] in 2006, and its ratification by numerous countries, many new terms and definitions were created, some of which at the behest of PWDs themselves, to eradicate terms that were demeaning, insulting and inaccurate. + +Standards writers are able to mainstream accessibility features into standards, as well as to write specific standards for accessibility. To design products and services successfully, there needs to be a common language. Recommendation ITU-T F.791 gives definitions to make it easier for industries to implement these accessibility features and accessibility standards by establishing a common language and vocabulary. It is also important that governments, government agencies, non-government organizations (NGOs), the UN and its respective agencies be "normalized" in mainstream everyday language. + +Article 9 of [b-UNCRPD] makes clear the need to include PWDs, older persons with age-related disabilities and persons with specific needs by mainstreaming them into all aspects of modern life. This can only be done by including PWDs in the design of modern technology and information and communication technologies (ICTs) using universal design as defined in [b-UNCRPD] and using a common language and vocabulary. + +This edition updates some of the definitions based on feedback received from the user community. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.791 | 2015-11-29 | 16 | 11.1002/1000/12624 | +| 2.0 | ITU-T F.791 | 2018-08-29 | 16 | 11.1002/1000/13661 | + +## Keywords + +Closed captioning, open captioning, persons with disabilities, persons with specific needs, UNCRPD, universal design. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2019 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +###### Page + +| | | | +|------|-------------------------------------------------------------------------------|----| +| 1 | Scope..... | 1 | +| 2 | References..... | 1 | +| 3 | Definitions ..... | 1 | +| 4 | Abbreviations and acronyms ..... | 4 | +| 5 | Conventions ..... | 5 | +| 6 | Terminology ..... | 5 | +| 6.1 | Access service; accessibility service ..... | 5 | +| 6.2 | Accessibility ..... | 5 | +| 6.3 | Accessibility content ..... | 5 | +| 6.4 | Accessibility feature ..... | 6 | +| 6.5 | Assistive listening device ..... | 6 | +| 6.6 | Assistive listening system..... | 6 | +| 6.7 | Assistive technology..... | 6 | +| 6.8 | Audio subtitles; spoken subtitles ..... | 6 | +| 6.9 | Audio captions; audio captioning..... | 6 | +| 6.10 | Audio description; video description; visual description; described video.... | 7 | +| 6.11 | Captions; captioning..... | 7 | +| 6.12 | Clean audio ..... | 7 | +| 6.13 | Closed/open accessibility services ..... | 7 | +| 6.14 | Design for all [deprecated] ..... | 8 | +| 6.15 | Disability ..... | 8 | +| 6.16 | Human factors; ergonomics..... | 8 | +| 6.17 | Impairment ..... | 8 | +| 6.18 | Inclusive design (deprecated) ..... | 9 | +| 6.19 | Interface accessibility ..... | 9 | +| 6.20 | Keyboard emulator ..... | 9 | +| 6.21 | Lip reading; lip-reading interpretation ..... | 9 | +| 6.22 | Lip speaker; oral interpreter ..... | 10 | +| 6.23 | Mainstreaming..... | 10 | +| 6.24 | Person with age-related disabilities ..... | 10 | +| 6.25 | Person with disabilities..... | 10 | +| 6.26 | Person with specific needs..... | 10 | +| 6.27 | Pixellation..... | 10 | +| 6.28 | Platform accessibility feature ..... | 11 | +| 6.29 | Profile setting..... | 11 | +| 6.30 | Real time..... | 11 | +| 6.31 | Relay service ..... | 11 | +| 6.32 | Remote participation ..... | 12 | + +| | Page | +|-----------------------------------------------------------|------| +| 6.33 Respeaking ..... | 12 | +| 6.34 Screen magnification software ..... | 12 | +| 6.35 Screen reader software ..... | 12 | +| 6.36 Sign language; signed language; visual signing ..... | 12 | +| 6.37 Sign language interpretation ..... | 13 | +| 6.37 bis Sign language presentation ..... | 13 | +| 6.38 Special needs ..... | 13 | +| 6.39 Specific needs ..... | 13 | +| 6.40 Subtitles ..... | 14 | +| 6.41 Supplementary audio service ..... | 14 | +| 6.42 Universal design ..... | 14 | +| 6.43 Barrier ..... | 14 | +| 6.44 Facilitator ..... | 15 | +| 6.45 User experience ..... | 15 | +| 6.46 Speech to text interpretation (STTI) ..... | 15 | +| Appendix I – Deprecated terminology ..... | 16 | +| Bibliography ..... | 17 | + +# Accessibility terms and definitions + +# 1 Scope + +This Recommendation defines terms to describe accessibility and terms that are needed by standards writers when drafting international standards. This Recommendation takes account of terms already used in existing ITU Recommendations and Resolutions, as well as those in the UN *Convention on the Rights of Persons with Disabilities* [b-UNCRPD]. These terms and definitions aim to reduce confusion, on the part of not only standards writers and implementers, but also the general public. This Recommendation also deprecates terminology that is no longer used, offensive or demeaning to persons with disabilities (PWDs) and others. + +The terminology in this Recommendation is for use in international work when English is used to refer to telecommunication/information and communication technology (ICT) accessibility matters. This Recommendation also applies to everyday life and all usages, including website content and other literature, as well as ICT, telecommunications and broadcasting standardization. It should also be mainstreamed into future policy, regulatory and academic documents as to be consistent with global compatibility and understanding. In the future, work that is yet to be created, written or approved may have new terms that can be later added as appropriate by consensus in a revision. + +NOTE – This drafting of this document complies with [b-ITU-T F.790]. + +# 2 References + +None. + +# 3 Definitions + +This Recommendation uses the following terms: + +**3.1 access service; accessibility service:** Provision of features intended to make primary audiovisual content accessible to users with specific needs, preferences or in specific environmental contexts. + +**3.2 accessibility:** The degree to which a product, device, service or environment (virtual or real) is available to as many people as possible. + +**3.3 accessibility content:** Content delivered by an audiovisual media solution, e.g., captions, subtitles, audio description, audio subtitles, and differentiated from the solution's interface accessibility, that is accessible to persons with disabilities (PWDs), as well as persons with specific needs. + +NOTE – Access services are a primary means of delivering accessibility content. + +**3.4 accessibility feature:** An additional content component that is intended to assist people hindered in their ability to perceive an aspect of the main content. + +**3.5 assistive listening device (ALD):** Devices that enable persons who are hard of hearing to hear sounds and speech on an improved basis. + +**3.6 assistive listening system (ALS):** Assistive technology (AT) systems utilizing electromagnetic radiation, commonly radio or light waves, or a combination of the two, enabling transmission of an acoustic signal from a sound source (e.g., a loudspeaker or a person talking) directly to the hearing aid or cochlear implant processor of a person who is hard of hearing. + +**3.7 assistive technology (AT):** Piece of equipment, product system, hardware, software or service that is used to enable, maintain or improve functional capabilities of individuals with disabilities. + +**3.8 audio subtitles; spoken subtitles:** Subtitle text rendered into speech by a human voice artist or a synthetic voice from text-to-speech software. + +**3.9 audio captions; audio captioning:** Captions that are read aloud and reflected as speech. Audio captioning may also be called "audio subtitles" or "spoken subtitles" in the case of foreign language dialogue. Captions can also be used to designate the audio content of an audiovisual work or sequence in any language along with action. Captions are read aloud by a human or a specific apparatus that converts the text into speech. + +**3.10 audio description; video description; visual description; described video:** An additional audio track to aid persons with visual impairments who cannot follow the visual content. + +**3.11 captions; captioning:** A real-time transcription of spoken words, sound effects, relevant musical cues and other relevant audio information in live or pre-recorded events. Captions can be open, not adjustable by the user, or closed where they can be turned on and off by the users at will. See clause 3.13 for further explanation of open and closed accessible services. + +**3.12 clean audio:** An enhanced audio signal by means of signal processing, with improved intelligibility of the dialogue with respect to ambient noise, background noise, music, etc. This can also apply to the quality of the audio used for audio captioning (see clause 3.9), audio description (see clause 3.10), and subtitles (see clause 3.40). + +**3.13 closed/open accessibility service:** An accessibility service – audio description, audio subtitling, captioning and sign language – that can have the option of be selected by the end user. If this is the case, it is closed. If cannot be selected or turned of, by the user it is an open service, i.e., open caption. + +**3.14 design for all [deprecated]:** The design of mainstream products or services that are accessible to and usable by all persons, especially including persons with disabilities (PWDs), and persons who were born with specific needs. This term is replaced by "universal design" (see clause 3.42 and Appendix I). + +**3.15 disability:** An evolving concept, which refers to the interaction between persons with impairments, and attitudinal and environmental barriers that hinder their full and effective participation in society on an equal basis with others. + +**3.16 human factors; ergonomics:** Factors relating to usability and proper interaction between persons and products and devices; services, systems or environments, both real and virtual. + +**3.17 impairment:** Any loss or abnormality of psychological, physiological, or anatomical structure or function. + +**3.18 inclusive design [deprecated].** + +This term is replaced by "universal design" (see clause 3.42 and Appendix I). + +**3.19 interface accessibility:** Accessibility of the set of provisions that allow a user to operate and control audiovisual media solutions. + +**3.20 keyboard emulator:** Hardware or software input device that emulates the key press outputs of an alphanumeric keyboard. + +**3.21 lip reading; lip-reading interpretation:** A form of communication or interpretation used by persons that are deaf or hard of hearing, regardless of whether they use sign language. + +**3.22 lip speaker; oral interpreter:** A trained interpreter for persons who are deaf and hard of hearing, who speaks silently the dialogue in the audiovisual content or in any other event in real time, + +so that the speech is clearly discernible for persons with hearing disabilities who can lip-read the words from the interpreter's mouth without the use of sign language. + +**3.23 mainstreaming:** Inclusion of persons with disabilities (PWDs) in everyday life without segregation from the environment, education, technology, i.e., access to telephones, the Internet, the worldwide web and all information and communication technologies (ICTs). + +**3.24 person with age-related disabilities:** A person with cognitive or physical disabilities caused by the aging process. Examples are impaired eyesight, deafness in varying degrees, reduced mobility or cognitive abilities. + +**3.25 person with disabilities (PWD):** The correct way to refer to a person with a disability [b-UNCRPD]. + +**3.26 person with specific needs:** Includes persons with disabilities (PWDs), persons who are not literate, those with learning disabilities, children, indigenous people, older persons with age-related disabilities and anyone who has a temporary disability. + +**3.27 pixellation:** Phenomenon caused by displaying a bitmap or a section of a bitmap at such a large size that individual pixels become visible, making the image blurred and more difficult to decipher. + +**3.28 platform accessibility feature:** Accessibility functionality provided as standard on a particular hardware or software platform. + +**3.29 profile setting:** The ability for users to store and retrieve multiple profiles containing sets of user interface preference settings without having to reset them each time, including accessibility settings. + +**3.30 real time:** Data or services (e.g., broadcasting) that are transmitted with virtually no delay. + +**3.31 relay service:** A telephone service that enables a person who is deaf or hard of hearing, or whose speech is not clearly understood, or who prefers to use sign language, to place and receive telephone calls in real time. + +**3.32 remote participation:** [b-ITU-T A-Sup. 4]: Participation in a meeting from a separate geographical location, using communication technologies. + +**3.33 respeaking:** A technique to produce captions where a person ("the respeaker") listens to the speech and respeaks it, such that the vocal input of the respeaker is processed by speech recognition software that transcribes it and produces captions. + +**3.34 screen magnification software:** A software application used by a person with impaired vision to magnify a portion of the text or graphics displayed on a video screen sufficiently to enable reading and comprehension. + +**3.35 screen reader software:** Software application used by a person who is blind or cannot easily read print to identify and interpret what is shown on a video display and read aloud using speech synthesis. + +**3.36 sign language; signed language; visual signing:** A natural language that, instead of relying on acoustically conveyed sound patterns, uses signs made by moving the hands combined with facial expressions and postures of the body to convey meaning. + +NOTE – Sign language varies from country to country, including many dialects, in a similar manner to spoken languages. + +**3.37 sign language interpretation:** Synchronized showing of an interpreter who uses sign language to convey the main audio content and dialogue to people who use sign language. + +**3.37bis sign language presentation:** The process of presenting, in a unidirectional manner, a topic to an audience using sign language. + +NOTE – In certain cases, a synthetic construct (e.g., an animated avatar) can be used in place of an interpreter. + +**3.38 special needs:** In the context of accessibility, term used only for countries, on a form or a sign to identify any accessibility accommodation that a person with disabilities (PWD) or a person with specific needs may require to help them be able to participate. + +NOTE – This term is not used to refer to persons (see clauses 3.39, 6.38 and Appendix I). For reference to countries, see [b-ITU WTDC Res.58] and [b-ITU WTDC AP]. + +**3.39 specific needs:** This replaces the use of the term 'special needs'. This term refers to a wide range of categories including women, children, youth, indigenous people, older persons with age-related disabilities, persons with illiteracy, as well as persons with disabilities (PWDs). See [b-ITU PP Res.175], [b- ITU WTDC Res.58], [b- ITU WTDC AP] and clause 6.39. + +**3.40 subtitles:** On-screen text translation of language(s) of the dialogue in any audiovisual content. + +**3.41 supplementary audio service:** An additional audio soundtrack that provides additional features or function over and above that provided by the main audio stream. + +**3.42 universal design:** The design of products, environments, programmes and services to be usable by all people, to the greatest extent possible, without the need for adaptation or specialized design. Universal design shall not exclude assistive devices for particular groups or persons with disabilities where this is needed. + +NOTE – See [b-UNCRPD]. This term succeeds "design for all" (see clause 3.14) and "inclusive design" (see clause 3.18). See also Appendix I. + +**3.43 barrier:** Attitudinal or environmental factor that, in relation to an impairment, limits functioning and participation in society on an equal basis with others. + +**3.44 facilitator:** Attitudinal or environmental factor, such as a person, environment or tool, that improves functioning and reduces disability through its absence or presence. + +**3.45 user experience:** Person's perceptions and responses resulting from the use or anticipated use of a product, system or service, including navigation of physical and virtual environment. + +NOTE 1 – User experience includes all the user's emotions, beliefs, preferences, perceptions, physical and psychological responses, behaviours and accomplishments that occur before, during and after use. + +NOTE 2 – User experience is a consequence of brand image, presentation, functionality, system performance, interactive behaviour and assistive capabilities of the interactive system, the user's internal and physical state resulting from prior experiences, attitudes, skills and personality, as well as the context of use. + +NOTE 3 – Usability, when interpreted from the perspective of the user's personal goals, can include the kind of perceptual and emotional aspects typically associated with user experience. Usability criteria can be used to assess aspects of user experience. + +NOTE 4 – Adapted from [b-ISO 9241-210]. + +**3.46 speech to text interpretation (STTI):** A simultaneous form of text interpretation conveying spoken content. + +# 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|------|--------------------------------------------| +| ALD | Assistive Listening Device | +| ALS | Assistive Listening System | +| AT | Assistive Technology | +| CART | Communication Access Real-Time Translation | +| EPG | Electronic Programme Guide | + +| | | +|------|------------------------------------------| +| HDTV | High Definition Television | +| ICT | Information and Communication Technology | +| IPTV | Internet Protocol Television | +| PC | Personal Computer | +| PWD | Person with Disabilities | +| SA | Supplementary Audio | + +# 5 Conventions + +The terms and definitions in this Recommendation are matched in numbering across clauses 3 and 6, for easier reference to the reader. The essential definitions are provided in clause 3, while more nuanced considerations on the use of different terms associated with the definitions are provided in clause 6. + +# 6 Terminology + +## 6.1 Access service; accessibility service + +See clause 3.1 for a concise definition. + +Examples related to impairments include users who are deaf, hard of hearing, blind or partially sighted. Examples related to preferences include people who turn down the sound on a TV when others are trying to sleep. Examples related to environmental contexts of use include being unable to hear a phone in a loud environment or see a screen in strong light. Examples of common access services are captioning, subtitling, audio description and sign language interpretation. A service such as captioning, audio description or visual signing improves the accessibility of the audiovisual content for which it was made. Metadata shall be available in an electronic programme guide (EPG) to allow a user to be able to determine the access services available. + +## 6.2 Accessibility + +See clause 3.2 for a concise definition. + +Accessibility can be viewed as the ability to access and possibly benefit from some system or entity. Accessibility is often used to focus on PWDs or to identify persons with specific needs who may or may not be PWDs and their right of access to entities directly or through use of AT or access services. Accessibility is the property of being accessible and usable upon demand by an authorized entity. Accessibility is strongly related to universal design (see article 2 of [b-UNCRPD]), when the approach involves "direct access". Universal design is about making (and designing) things, from concept to realization, to be accessible to as many people as possible, whether or not they have disabilities or specific needs. Using the principles of universal design at the beginning of the design process makes products, services and environments, both virtual and real, accessible to a greater number of people and prevents expensive reformulation later when accessibility is possibly required by regulation. An alternative method of providing accessibility is to provide "indirect access" by having the entity support the use of a person's AT to achieve access. + +## 6.3 Accessibility content + +See clause 3.3 for a concise definition. + +In order for PWDs and those with specific needs to follow audiovisual content, certain technology must be implemented. Examples are as follow. For persons who are deaf or profoundly hard of hearing, open and closed captioning (including audio captioning) is necessary to understand dialogue and background noises. An example is sign language, which can also be open or closed. For persons + +who are visually impaired, audio description is necessary to describe the non-spoken components of audiovisual content. + +## **6.4 Accessibility feature** + +See clause 3.4 for a concise definition. + +Examples of accessible features are: captions for persons who are deaf or hard of hearing, subtitles in various languages, as well as sign language interpretation for video and audio description for the visually impaired. + +## **6.5 Assistive listening device** + +See clause 3.5 for a concise definition. + +ALDs work with hearing aids or cochlear implant processors with or without radio assistance and radio communication systems comprising narrow band transmitter(s). These can be hand-held, placed on a table or worn physically around the neck. ALDs can have a wired or an inductive connection to a hearing aid. + +## **6.6 Assistive listening system** + +See clause 3.6 for a concise definition. + +Assistive listening systems (ALSs) are wireless sound systems designed specifically to help people with hearing loss to access a sound source via their hearing device so that they can hear more clearly. The function of an ALS is to eliminate the degradation of intelligibility, unwanted noise or acoustical distortion between the sound source and the person's hearing device. ALSs include a hearing loop (an audio inductive loop), infrared systems or radio frequency-based systems that are found in mobile phones. They can be found installed at shop counters and in large venues e.g., theatres, museums and conference centres. + +## **6.7 Assistive technology** + +See clause 3.7 for a concise definition. + +AT is an interface to an ICT device that allows access to technology. It also includes the process used in selecting, locating and using ICTs. The use of AT and AT devices promotes greater independence by enabling people to perform tasks that they were formerly unable to accomplish or had great difficulty accomplishing by providing enhancements to or changed methods of interacting with the technology. AT provides indirect access and this is supported by universal design. An example is a screen reader that enables persons who are blind to hear printed text read aloud. + +## **6.8 Audio subtitles; spoken subtitles** + +See clause 3.8 for a concise definition. + +Subtitles become the audio content of an audiovisual work or sequence in a foreign language through specific techniques that convert the text into speech from subtitles on the screen in the target language by text recognition. Conversion may be done at the source or on the user's device using speech synthesis. + +## **6.9 Audio captions; audio captioning** + +See clause 3.9 for a concise definition. + +Audio captioning may be done at the source or on the user's device using speech synthesis. This is for the benefit of blind or visually impaired persons when the only meaningful content comes from captions or subtitles. + +## **6.10 Audio description; video description; visual description; described video** + +See clause 3.10 for a concise definition. + +Audio description provides additional audible narrative and is mixed with the dialogue. This describes the significant aspects of the visual content, settings, actions and ambiguous sounds of the audiovisual media that cannot be understood from the main dialogue soundtrack alone. Audio description should be synchronous with this list to describe any actions and identify who is doing what, where and when. Special attention should be paid to the audio mix to make sure that the volume of the main dialogue and the audio description are not in conflict. + +## **6.11 Captions; captioning** + +See clause 3.11 for a concise definition. + +Captions provide a real-time on-screen transcript of the dialogue and any sound effects when an Internet or television broadcast is live. Ideally, users have some control over the position and size of the presentation. Captions can also be pre-recorded and synchronized with the dialogue of the content. [b-ITU-T Y.1901]. There are different ways to identify speakers, e.g., with different colours. + +There is also real-time captioning [also referred to as communication access real-time translation (CART)], which is used for meetings to enable participation for PWDs. This service can be provided by means of either textual or graphical supplementary content. The captions and the dialogue are usually in the same language. The service is primarily to assist users having difficulty hearing the sound. They are also beneficial to those persons who do not understand the spoken language for other reasons e.g., the language is not that person's first language (including sign language). + +At the time of publication, captions are created in two basic ways, either spoken or typed. Most commonly, text is produced by stenographic methods using technology similar to court reporting or using a respeaking technique through speech recognition software. There are technical forms of creating captions in foreign languages that do not have a script or keyboard, or for which speech technology solutions are still unavailable. Captions might require the use of different keyboards and techniques. One example is in Japan, where two captioners work together simultaneously using a specialized keyboard to render captions in their language. + +## **6.12 Clean audio** + +See clause 3.12 for a concise definition. + +Clean audio enables everyone to hear the sound of broadcast content clearly, including dialogue and narrative, audio description, audio captions and spoken subtitles, which is imperative for persons with hearing difficulties (including those who are hard of hearing or with age-related hearing disabilities). + +## **6.13 Closed/open accessibility services** + +See clause 3.13 for a concise definition. + +There is a need for both the use of open and closed services. The user should have the ability to turn on and turn off a service, such as captioning or sign language, if it is not needed. Equally, when a person with disabilities is in situation away from home, they should be able to turn these services on or off at will. Open captions are also vital for announcing emergencies and other public announcements. It is necessary to have open services in giving the general public important live information as in news broadcasting. These open services are not in the control of the user for safety reasons, or as in the case of a live broadcast. There is also the possibility that technically, for whatever reason, it is not possible to provide closed services. + +## 6.14 Design for all [deprecated] + +See clause 3.14 and Appendix I on deprecated terminology. This term is replaced by "universal design" (see clause 3.42) as is "inclusive design" (see clause 3.18). + +The term "design for all" was used before the creation of the term "universal design", but is not recognized or used in article 2 of [b-UNCRPD]. Unfortunately, its implied implementation is not practical from the industrial point of view; it is preferred to use UN language, but "design for all" does not carry a social stigma if it is used in addition to help explain the concept of universal design, which has a slightly more flexible meaning. + +## 6.15 Disability + +See clause 3.15 for a concise definition. + +PWDs include those who have short- or long-term physical, sensory, cognitive or psychosocial impairments, which in interaction with various barriers may hinder their full and effective participation in society on an equal basis with others. See also clause 6.25 for more information on PWDs. + +NOTE – For more detailed information about the definition of "disability", please refer to appropriate documents such as [b-WHO ICF]. + +## 6.16 Human factors; ergonomics + +See clause 3.16 for a concise definition. + +This category can deal with accessibility for PWD, but not exclusively. Examples of human factor accessibility in practical applications could be: + +- curb cuts, which can be used by wheelchairs, motorized disability scooters and perambulators for babies; +- the tactile marker in the keypad "5" key in telephone sets conforming to [b-ITU-T E.161], that guide blind users in placing voice calls; +- captioning in audiovisual programmes, which helps an audience to understand its content in a noisy environment. + +## 6.17 Impairment + +See clause 3.17 for a concise definition. Qualifications of "impairment" include the following: + +- a) **age-related impairment:** Sensory or cognitive degeneration that increases with age, generally covering items such as the deterioration of sight and hearing, memory impairment or memory loss and motor impairment. +- b) **cognitive impairment:** Deterioration that affects an individual's ability to think, concentrate, formulate ideas, reason and remember. +- c) **dexterity impairment:** Reduction in function of the arms and hands that makes activities related to moving, turning or pressing objects difficult or impossible. If other motor features of the body do not function well, dexterity impairment may impair speech communication. +- d) **functional impairment:** A person's loss of functional ability of an organ or physical or mental capability to perform its specified function. The existence of a medical condition may not necessarily restrict functional capacity and does not define disability. The [b-WHO ICI] definition of impairment is "any loss or abnormality of psychological or anatomical structure or function." +- e) **hearing loss impairment:** Please see "hearing impairment" (deprecated) in Appendix I. + +This term should not be used to cover both deaf and hard of hearing persons. Instead, "hearing disability" should be used as a general term. The terms to be used for those affected are "persons who are deaf" or "persons who are hard of hearing". + +Hearing loss can affect the whole auditory range and can be profound, making hearing speech and sound impossible through auditory means; this type of hearing loss describes a person who is deaf. Alternatively, hearing loss can vary in degree affecting only a part of the auditory spectrum and in this case AT can assist (e.g., a hearing aid placed in one or both ears); this type of loss describes a person who is hard of hearing. The use of the word "profound" means severe. Technology is usually needed in both cases, but of different types. Persons who are deaf need visual solutions and persons who are hard of hearing need enhanced audio solutions often together with visual text content, see [b-WFD-IFHHP]. + +- f) **visual impairment; vision impairment:** Vision loss (of a person) to such a degree as to qualify for additional support through a significant limitation of visual capability resulting from disease, trauma, age or congenital defect, all of which can include degenerative conditions that cannot be corrected by conventional means like glasses or refractive correction, medication or surgery. The loss may cover visual acuity, significant central or peripheral field defects, or reduced contrast sensitivity. + +## **6.18 Inclusive design (deprecated)** + +See clause 3.18 and Appendix I on deprecated terminology. This term is replaced by "universal design" (see clause 3.42) as is "design for all" (clause 3.14). + +The design of mainstream products or services that are accessible to and usable by as many people as reasonably possible without the need for special adaptation or specialized design. Inclusive design is a predecessor term to universal design, but is not recognized or used in article 2 of [b-UNCRPD]. The use of UN language is preferred, but "inclusive design" does not carry a social stigma if the term is used in addition to help explain the concept of universal design. + +## **6.19 Interface accessibility** + +See clause 3.19 for a concise definition. + +Interface accessibility is concerned with components and controls, such as those of the physical (hardware and remote) and virtual types, e.g., displays, menus, EPGs, windows and selection grids. An accessible interface does not automatically imply content accessibility. + +## **6.20 Keyboard emulator** + +See clause 3.20 for a concise definition. + +Keyboard emulators are used by individuals who are unable to physically enter text using a keyboard. + +## **6.21 Lip reading; lip-reading interpretation** + +See clause 3.21 for a concise definition. + +Many persons become profoundly or partially deaf later in life. Persons who do not have sign language skills often rely on recognizing the spoken words on the lips of others without sound and can interpret speech if trained or are naturally able to lip read. This technique works better on a one to one basis in real time as lip reading is often guesswork. Dialects and moustaches sometimes get in the way, but with high definition television (HDTV), it is more feasible to virtually see speech that persons with hearing disabilities cannot recognize by sound on the mouths of speakers. Lip reading can make it possible to understand speech without sound only when the speaker is facing the camera. This is, of course, not always the case and many persons have to rely on captioning as well. + +## **6.22 Lip speaker; oral interpreter** + +See clause 3.22 for a concise definition. + +A lip speaker or oral interpreter is trained in a technique to relay spoken words silently through lip movements, providing communication support to deaf persons. This technique provides useful support for deaf people who are confident lip readers. + +## **6.23 Mainstreaming** + +See clause 3.23 for a concise definition. + +Whenever possible, mainstreaming should be transparent or give equivalent access using ATs. ATs can be necessary to assist inclusion, by recognizing and providing for specific needs or requirements, especially upon request. ATs will provide equivalency of integration. Mainstreaming is necessary to achieve integration, see [b-UNCRPD]. This also applies to the writing of accessibility features into mainstream standardization. See "accessibility features" (clause 3.4). + +## **6.24 Person with age-related disabilities** + +See clause 3.24 for a concise definition. + +People grow older but do not necessarily see themselves as old or elderly. They often resist seeing the changes and the diminishing capabilities that age brings as making them a person with disabilities. The use of the term elderly is discouraged and not accurate, especially when age-related disabilities can begin in the early forties. The simple beginning of wearing glasses due to the inability to focus or see distant objects clearly is a classic example. Deterioration of hearing with age and the loss of the ability to separate sounds makes it difficult to understand speech in noisy environments. Older persons can also be referred to as persons with specific needs, especially if there is a strong resistance to being labelled as an older person or a PWD. + +## **6.25 Person with disabilities** + +See clause 3.25 for a concise definition. + +It is incorrect to refer to a PWD as a disabled or handicapped person. This definition was at the behest of the consulting panel of PWDs when [b-UNCRPD] was created. The basis of this choice was that they were persons first and had a disability second. Many organizations use the term "people with disabilities", but [b-UNCRPD] uses "persons with disabilities". + +## **6.26 Person with specific needs** + +See clause 3.26 for a concise definition. + +The term "persons with specific needs" can be used for persons with age-related disabilities and others, i.e., women, youth, children and indigenous people as well as persons with impaired literacy, both with and without disabilities. Curb cuts are an example of mainstreaming for persons with specific needs that accommodate many people with and without disabilities. They were designed for the accommodation of wheel chair users and motorized scooters for PWDs. However, they are also invaluable to mothers with baby buggies or prams (see [b- ITU WTDC AP]). "Person with specific needs" covers everyone, but should not entirely replace the term "person with disabilities". + +## **6.27 Pixellation** + +See clause 3.27 for a concise definition. + +When a person with low vision tries to read text, view graphics or photos on a personal computer (PC) screen, the person may use screen magnification software to enlarge content. However, if the software is not well designed, the text in an image can become too "pixellated" to be legible when the + +image is magnified. Pixellation (see clause 3.27) can occur, making the text or image indiscernible. This also applies when a person is without disabilities in general, if the image is poor. + +## **6.28 Platform accessibility feature** + +See clause 3.28 for a concise definition. + +Examples of platform accessibility features are screen magnification or "zoom" functionality provided within an operating system, as well as the sizes of font and icon for photos or items on a desktop. These are usually built into the operating system and can be activated without adding any other software. + +## **6.29 Profile setting** + +See clause 3.29 for a concise definition. + +Internet protocol television (IPTV) and EPG architecture are required to support the capability to store and retrieve multiple profiles containing sets of user interface preference settings and language settings so that it will be easy to switch to different personal preferences for different users without the user having to reset them each time. This should also include accessibility features in the settings for captioning, sign language and audio description. + +## **6.30 Real time** + +See clause 3.30 for a concise definition. + +Real time text, live television emissions, live voice telephone calls and text phone telephone calls are generally real time services. Instant messaging and e-mail are not delivered in real time as the transmission of messages must be activated by the user doing an action to send the data. + +## **6.31 Relay service** + +See clause 3.31 for a concise definition. + +A relay service is usually a human-operated service for media and mode (voice, text and video) translation during phone conversations. The provision of relay services that are well integrated into the phone system is an important capability of accessible phone services for persons who are deaf or hard of hearing, as well as those whose speech cannot be clearly understood. + +Relay services are usually provided as community-supported services, as their operation requires many more human resources than regular person-to-person calls, see [b-ITU-G3ict]. + +Existing types of relay services are: + +- video relay services, translating between sign language in video and speech in a voice phone; +- text relay services, translating between real-time text in the text part of a phone and speech in a voice phone, usually for persons with speech impairments, hearing disabilities, including those who are hard of hearing, deaf or deaf-blind; +- speech-to-speech relay services, supporting speech calls for people with speech impairments or cognitive disabilities; +- captioned speech relay services (captioned telephony), adding real-time text captions to a voice call, for people who are hard of hearing or deaf. + +In all these cases, other media (voice, video, text) may also be handled in the call, ideally connected in two- or three-party call mode between the parties in the call who have terminals that can handle these media. + +Integration of relay services in the phone system implies: + +- calls to a number for a person with disabilities should be able to invoke a relay service selected by the user if the user so decides; +- calls from a person with disabilities to another number should be able to invoke a relay service selected by the user if the user so decides; +- calls between two users who can and want to use the same modes and media in the call should be possible without invoking any relay service. + +Relay services should work with all commonly used handsets and terminals. Users need to be able to use the same terminal for calls directly in the modes they handle, as is used for calls through relay services and with emergency services. The relay and emergency services can only feasibly support a limited number of connection types or protocols. Thus, it is important to coordinate the specifications of access to relay services, emergency services and terminals used by people with disabilities so that maximum interoperability in all available media is achieved for all occurring call combinations. + +## **6.32 Remote participation** + +See clause 3.32 for a concise definition. + +The online process that enables a person to participate without actually being physically present at a meeting location by the use of online communication tools, including video, audio and text communication in chat boxes. Remote participation also allows for captioning to be included for those PWDs who cannot access sound. + +More details concerning remote participation can be found in [b-ITU-T FSTP.ACC-RemPart] and [b-ITU-T A-Sup. 4]. + +## **6.33 Respeaking** + +See clause 3.33 for a concise definition. + +This technique involves a longer delay in producing captions as there are three steps in producing the captions and the loss of synchronization is an issue to be considered. + +Unless there is a supplementary typing input device, respeaking is not accurate when it comes to the reproduction of foreign names or nouns, unless they are programmed into the software in advance. + +## **6.34 Screen magnification software** + +See clause 3.34 for a concise definition. + +Screen magnification is an accessibility feature included in some hardware. An example can be found in mobile phones. This accessibility feature may already be included in computer operating systems. If this is not available, this accessibility feature can often be downloaded. Screen magnification is useful for persons with varying degrees of sight problems and is also used by those with learning difficulties. + +## **6.35 Screen reader software** + +See clause 3.35 for a concise definition. + +Screen readers can come in an audible format, in which a person can listen to a word printed on the screen, or in a refreshable braille display, enabling a person who is blind to read in braille what was printed on the screen, + +## **6.36 Sign language; signed language; visual signing** + +See clause 3.36 for a concise definition. + +As is the case with any language, sign language has its own distinct grammar and syntax. There is a different sign language for every spoken language, and dialects are found within each country just + +like any spoken language. There does not exist an international sign language *per se*, but deaf individuals who use different signed languages adapt their signing to communicate with each other and make themselves understood. This has led some to misconstrue that there exists a single uniform signed language understood by all deaf persons. This is in fact an incorrect assumption. What does exist is a collection of very basic signs with a limited vocabulary that does not include specific terminology and is used to communicate between deaf persons who use different signed languages. This is referred to as International Sign. It is constantly changing and adapting to the specific needs of participants in specific situations. It is used primarily within the international deaf community to have a communication method that most can understand without having to resort to having so many different national sign languages present at a large international meeting for persons who are from many different countries. However, many deaf people do not use or know any form of International Sign and it is interpreted differently in every country. It can be used under certain circumstances when no other choice is available, but not in detailed technical discussions. There is no equivalent spoken language translation. Thus, a real true understandable International Sign is not standardized or consistent as explained above. + +## **6.37 Sign language interpretation** + +See clause 3.37 for a concise definition. + +Sign language interpretation is often done by human beings who are fluent in a particular national sign language. There must always be two persons for each language, just as there are two persons who interpret orally expressed languages. This can be done live with the interpreters present or remotely via video. There are avatars that have been created to simulate human sign language interpretation. However, the problem with avatars is that one of the functions for a sign language interpreter is to voice out what a person with disabilities is signing if that person does not speak. + +## **6.37bis Sign language presentation** + +See clause 3.37bis for a concise definition. + +No additional discussion is provided for this term. + +## **6.38 Special needs** + +See clause 3.38 for a concise definition. + +This term can be used only for countries or on a form to identify any special requirements that a person with disabilities or a person with specific needs requires to be able to participate. It should not be used to describe a person with disabilities. One cannot refer to a person having special needs. The term "specific needs" (see clause 3.39) is now used instead of "special needs" with regard to people, see [b-ITU WTDC AP] [b-WTDC Res.68]. + +The use of the term "special needs" to describe a person with disabilities countermands the desire of PWDs to be mainstreamed into technology, education, work and society. The use of this term to describe PWDs also indicates that they are deemed not capable and to include them would create huge difficulties. It gives industry the view that the word "special" means a market too small (and therefore not profitable) to be considered as a mainstream item for implementation of accessibility features in technology and services that if implemented would eliminate barriers to inclusiveness. + +See Appendix I on deprecated terminology. + +## **6.39 Specific needs** + +See clause 3.39 for a concise definition. + +This term was created to cover a wider range of needs and includes more than just PWDs. It replaces the term "special needs" (see clause 6.38), which now refers to countries and not people. "Specific + +needs" refers to a wide range of categories including women, children, youth, indigenous people, older persons with age-related disabilities, persons with illiteracy, as well as PWDs. It allows for the concept of mainstreaming, human factors and encompasses the needs of many, which supports the UNCRPD concept of universal design, see [b-UNCRPD], [b-ITU PP Res.175], [b-ITU WTDC Res.58], [b-ITU WTDC AP]. + +## **6.40 Subtitles** + +See clause 3.40 for a concise definition. + +The assumed audience for subtitling is for persons who can hear, as well as those who cannot hear, and those who do not understand the language of the dialogue. Subtitles can be on smartphones, tablets, computers or any screen-based content. They can be produced manually or by respeaking. It is technically possible to choose their position on the screen, their size and colours. They should be synchronized with the spoken content. Subtitles in the past have usually been open in content for the purpose of language translation or to clarify the speech of any language spoken that is unclear and was separate from closed captioning or subtitling that is superimposed later for the entire dialogue and captioned audio on the content for accessibility. + +Subtitle text is not always a real-time dialogue text transcription. Sometimes it is pre-embedded in the content and may not follow all the dialogue when translating the speech of one individual speaking a language other than the main language of the content. If captions or captioning is introduced later, which translates all the main dialogue and sound effects with text, care needs to be taken to place the two forms of text successfully on the screen without one interfering with the placement of the other. + +NOTE – This service can be provided by means of either textual or graphical supplementary content. + +Subtitles are the translation of spoken words only, while captions include spoken words plus any meaningful sound whose perception is important to understanding the content of the audiovisual program. See clause 3.11. + +## **6.41 Supplementary audio service** + +See clause 3.41 for a concise definition. + +Supplementary audio (SA) can be an audio description that describes the scenario and the unspoken action of the content for the benefit of persons with disabilities who cannot see the content. It is in addition to the spoken dialogue. It can also be audio captioning or spoken subtitles in which the text written on the screen is read out audibly for persons with reading or visual difficulties. + +The SA stream may be provided using one of two schemes (see [b-ETSI TS 101-154]) as follows. + +- Broadcast mixed: Pre-mixed by the broadcaster and offered as an alternative audio stream, as in open accessibility services. +- Receiver mixed: Mixed in the receiver under the control of signalling provided by the broadcaster plus some limited control of the user, as in closed accessibility services. + +## **6.42 Universal design** + +See clause 3.42 for a concise definition. + +No additional discussion is provided for this term. + +## **6.43 Barrier** + +See clause 3.43 for a concise definition. + +No additional discussion is provided for this term. + +## **6.44 Facilitator** + +See clause 3.44 for a concise definition. + +No additional discussion is provided for this term. + +## **6.45 User experience** + +See clause 3.45 for a concise definition. + +No additional discussion is provided for this term. + +## **6.46 Speech to text interpretation (STTI)** + +See clause 3.46 for a concise definition. + +This form of interpretation has been developed in North European countries. A trained professional interprets a spoken language into text in another language. Example: a conference with an English speaker is translated into Finnish in text form for the deaf or hard of hearing delegates. + +# Appendix I + +## Deprecated terminology + +(This appendix does not form an integral part of this Recommendation.) + +The following is terminology that should be avoided when referring to accessibility issues. + +**Elderly:** Not to be used, as the term is deemed to be offensive to "older persons". See "person with age-related disabilities (clause 3.24). The word "elderly" is a negative term that is counterproductive to assisting older persons in dealing with age degenerative problems. + +**Handicapped:** Not to be used unless it is in colloquial use for a specific language translation purpose. See "person with disabilities" (clause 3.25). + +**Hearing impairment:** The use of the word "impairment" is restricted and cannot be used for those with hearing disabilities because the solutions are completely the opposite for persons who are deaf and persons who are hard of hearing. If a person is deaf, they need visual communication or assistive technologies as in captioning or text that can be read and video for sign language. If a person is hard of hearing, they need improved audio communication or assistive technologies that enhance audio i.e., hearing aids and ALDs see [b-WFD-IFHHP]. See also clause 3.17. + +**Inclusive design, design for all or accessible design:** These terms were previously used in place of and to mean "universal design", which is the correct UN language taken from [b-UNCRPD]. These terms have many different definitions in the various documents from the range of organizations that used them. Migration to recognized UN terminology is recommended. See the disclaimer in [b-ITU-T H-Sup. 17]. + +**Persons with special needs:** The term "special needs" is not to be applied to a person because it has been found to be offensive to PWDs. It also gives the wrong impression to industry that the market of PWDs is small and thus marginalizes these persons with regards to standardization and implementation of accessibility features in standardization. The term "special needs" can be used in the context of a registration form in order to indicate anyone who needed special access of some kind, for instance, to attend a meeting or to access a building. It should not be used to define a person. "Special needs" can be used for countries, as in "countries with special needs" see [b-ITU WTDC AP], (see also clauses 3.23, 3.25, 3.38, 6.23, 6.38). + +# Bibliography + +- [b-ITU-T A-Sup. 4] ITU-T A-series Recommendations – Supplement 4 (2015), *Supplement on guidelines for remote participation*. +- [b-ITU-T E.161] Recommendation ITU-T E.161 (2001), *Arrangement of digits, letters and symbols on telephones and other devices that can be used for gaining access to a telephone network*. +- [b-ITU-T F.790] Recommendation ITU-T F.790 (2007), *Telecommunications accessibility guidelines for older persons and persons with disabilities*. +- [b-ITU-T H-Sup. 17] Recommendation ITU-T H-Series Supplement 17 (2014), *Guide for addressing accessibility in standards*. +- [b-ITU-T Y.1901] Recommendation ITU-T Y.1901 (2009), *Requirements for the support of IPTV services*. +- [b-ITU-T FSTP.ACC-RemPart] ITU-T Technical Paper FSTP.ACC-RemPart (2015), *Guidelines for supporting remote participation in meetings for all*. +- [b-ITU-G3ict] ITU (Internet). *G3ict e-accessibility policy toolkit for persons with disabilities*. Available [viewed 2018-11-09] at: [http://www.e-accessibilitytoolkit.org/toolkit/technology\\_areas/landline\\_phones#Relay](http://www.e-accessibilitytoolkit.org/toolkit/technology_areas/landline_phones#Relay) +- [b-ITU PP Res.175] ITU Plenipotentiary Conference Resolution 175 (Busan, 2014), [\*Telecommunication/information and communication technology accessibility for persons with disabilities and persons with specific needs\*](#). +- [b-ITU WTDC AP] ITU World Telecommunication Development Conference (2014), Action plan. In: *Final Report World Telecommunication Development Conference (WTDC-14)*, Dubai, United Arab Emirates, 2014-03-30/04-10, pp. 43-144. Available [viewed 2019-01-09] at: +- [b-ITU WTDC Res.58] ITU World Telecommunication Development Conference Resolution 58 (Rev Dubai 2014), [\*Telecommunication/information and communication technology accessibility for persons with disabilities, including persons with age-related disabilities\*](#). In: *Final Report World Telecommunication Development Conference (WTDC-14)*, Dubai, United Arab Emirates, 2014-03-30/04-10, pp. 396-410. Available [viewed 2019-01-09] at: +- [b-ITU WTDC Res.68] ITU World Telecommunication Development Conference Resolution 68 (Rev. Dubai, 2014), Assistance to indigenous peoples within the activities of the Telecommunication Development Bureau in its related programmes. In: *Final Report World Telecommunication Development Conference (WTDC-14)*, Dubai, United Arab Emirates, 2014-03-30/04-10, pp. 461-464. Available [viewed 2019-01-09] at: + +- [b-ETSI TS 101-154] ETSI TS 101-154 v.1.11.1 (2012), *Digital video broadcasting (DVB); Specification for the use of video and audio coding in broadcasting applications based on the MPEG-2 transport stream.* +- [b-ISO 9241-210] ISO 9241-210:2010, *Ergonomics of human–system interaction – Part 210: Human-centred design for interactive systems.* +- [b-UNCRPD] United Nations (2006), *Convention on the Rights of Persons with Disabilities*. New York, NY: United Nations. 28 pp. Available [viewed 2019-01-09] at: [http://www.un.org/disabilities/documents/convention/convention\\_accessible\\_pdf.pdf](http://www.un.org/disabilities/documents/convention/convention_accessible_pdf.pdf) +- [b-WFD–IFHHP] WFD, IFHHP (2013). *Co-operation agreement between World Federation of the Deaf and the International Federation of Hard of Hearing People*. Available [viewed 2019-01-09] at: +- [b-WHO ICF] World Health Organization (2017). *International classification of functioning, disability and health (ICF)*. Geneva: WHO. Available [viewed 2018-11-09] at: . +- [b-WHO ICI] World Health Organization (1980). *International classification of impairments, disabilities, and handicaps*. Geneva: WHO. 207 pp. Available [viewed 2019-01-09] at: [http://apps.who.int/iris/bitstream/handle/10665/41003/9241541261\\_eng.pdf;jsessionid=50649198D21FCD7FB7589BC70329A59A?sequence=1](http://apps.who.int/iris/bitstream/handle/10665/41003/9241541261_eng.pdf;jsessionid=50649198D21FCD7FB7589BC70329A59A?sequence=1) + + + +# SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/F/T-REC-F.80-199110-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.80-199110-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..08f40bfe8571afa13f7d07b381e45dc0b0fb00a1 --- /dev/null +++ b/marked/F/T-REC-F.80-199110-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bbc4440e5670f0630ee5e7d613dd733ce580b116e6a439ac87442fc9656cf252 +size 7464 diff --git a/marked/F/T-REC-F.80-199110-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg b/marked/F/T-REC-F.80-199110-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7ae8ba1cf34ff6244d1cb3c2d4e023a34a19df7b --- /dev/null +++ b/marked/F/T-REC-F.80-199110-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:20fc3345c53261222a6a889557ed88c719cc43a240054ea54885dc8f15dc2e3e +size 6030 diff --git a/marked/F/T-REC-F.811-199607-I_PDF-E/raw.md b/marked/F/T-REC-F.811-199607-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..9b1cb10354a683762cd9c6b1a55bf6dc63ab78cf --- /dev/null +++ b/marked/F/T-REC-F.811-199607-I_PDF-E/raw.md @@ -0,0 +1,941 @@ + + +![ITU logo: A globe with a lightning bolt and the letters ITU.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with a lightning bolt and the letters ITU. + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.811** + +(07/96) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +ISDN services + +--- + +**Broadband connection-oriented bearer service** + +ITU-T Recommendation F.811 + +(Previously "CCITT Recommendation") + +--- + +# ITU-T F-SERIES RECOMMENDATIONS NON-TELEPHONE TELECOMMUNICATION SERVICES + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | F.1-F.109 | +| Operating methods for the international public telegram service | F.1-F.19 | +| The gentex network | F.20-F.29 | +| Message switching | F.30-F.39 | +| The international telemessage service | F.40-F.58 | +| The international telex service | F.59-F.89 | +| Statistics and publications on international telegraph services | F.90-F.99 | +| Scheduled and leased communication services | F.100-F.104 | +| Phototelegraph service | F.105-F.109 | +| MOBILE SERVICE | F.110-F.159 | +| Mobile services and multdestination satellite services | F.110-F.159 | +| TELEMATIC SERVICES | F.160-F.399 | +| Public facsimile service | F.160-F.199 | +| Teletex service | F.200-F.299 | +| Videotex service | F.300-F.349 | +| General provisions for telematic services | F.350-F.399 | +| MESSAGE HANDLING SERVICES | F.400-F.499 | +| DIRECTORY SERVICES | F.500-F.549 | +| DOCUMENT COMMUNICATION | F.550-F.599 | +| Document communication | F.550-F.579 | +| Programming communication interfaces | F.580-F.599 | +| DATA TRANSMISSION SERVICES | F.600-F.699 | +| AUDIOVISUAL SERVICES | F.700-F.799 | +| ISDN SERVICES | F.800-F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850-F.899 | +| HUMAN FACTORS | F.900-F.999 | + +*For further details, please refer to ITU-T List of Recommendations.* + +## FOREWORD + +The ITU-T (Telecommunication Standardization Sector) is a permanent organ of the International Telecommunication Union (ITU). The ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Conference (WTSC), which meets every four years, establishes the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics. + +The approval of Recommendations by the Members of the ITU-T is covered by the procedure laid down in WTSC Resolution No. 1 (Helsinki, March 1-12, 1993). + +ITU-T Recommendation F.811 was revised by ITU-T Study Group 1 (1993-1996) and was approved under the WTSC Resolution No. 1 procedure on the 19th of July 1996. + +--- + +## NOTE + +In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +© ITU 1997 + +All rights reserved. No part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and microfilm, without permission in writing from the ITU. + +# CONTENTS + +| | Page | +|-------------------------------------------------------------------------------|-------------| +| 1 Definition ..... | 1 | +| 2 Description ..... | 1 | +| 2.1 General description..... | 1 | +| 2.2 Specific terminology..... | 2 | +| 3 Procedures ..... | 4 | +| 3.1 On demand mode..... | 4 | +| 3.2 Reserved mode..... | 5 | +| 3.3 Permanent mode ..... | 11 | +| 4 Network capabilities for charging ..... | 14 | +| 5 Interworking requirements ..... | 14 | +| 6 Interaction with supplementary services ..... | 14 | +| Annex A – Broadband connection-oriented bearer service ..... | 15 | +| Annex B – Alphabetical list of abbreviations used in this Recommendation..... | 16 | +| Annex C – Narrow-band emulated services ..... | 16 | +| C.1 64 kbit/s bearer communication usable for speech ..... | 16 | +| C.2 64 kbit/s bearer communication usable for 3.1 kHz audio ..... | 17 | +| C.3 64 kbit/s unrestricted bearer service ..... | 17 | +| Annex D – References..... | 18 | + +# **SUMMARY** + +The Broadband Connection-Oriented Bearer Service (BCOBS) provides unrestricted transfer of digital information between $T_B$ reference points, or $S_B$ reference points when $T_B$ and $S_B$ reference points are coincident. This bearer service flexibly supports connection-oriented communications at any bit rate over a virtual channel connection. This bit rate can be approximately constant to support constant bit rate traffic or it can be highly variable to support variable bit-rate traffic. + + + +# BROADBAND CONNECTION-ORIENTED BEARER SERVICE + +*(revised in 1996)* + +# 1 Definition + +The **Broadband Connection-Oriented Bearer Service** (BCOBS) provides unrestricted transfer of digital information between TB reference points, or SB reference points when TB and SB reference points are coincident. This bearer service flexibly supports connection-oriented communications at any bit rate over a virtual channel connection. This bit rate can be approximately constant to support Constant Bit Rate (CBR) traffic or it can be highly variable to support Variable Bit Rate (VBR) traffic. Signalling information transfer will be provided over a separate signalling virtual channel. + +# 2 Description + +## 2.1 General description + +The service provides cell based communications in a bidirectional symmetric, bidirectional asymmetric or unidirectional fashion using virtual channel connections in the Broadband Integrated Services Digital Network (B-ISDN). For each communication of this service, the user and the SP agree on the use of one of the ATM Transfer Capabilities (ATCs) defined in Recommendation I.371[1]. + +This service is provided in Permanent, Reserved and on-Demand mode. + +In permanent mode the user subscribes to one or more permanent communications, specifying all the parameters to characterize them. + +In reserved mode the user gets at subscription the capability to request, to modify and to release reserved communications for the subscription duration. After the subscription, at any time the user can request by management procedures to reserve a communication specifying all the parameters to characterize it. + +In (switched) on-demand mode the user gets at subscription the capability to request, to modify and to release a (switched) on-demand call for the subscription duration. After the subscription, at any time the user can request to set up a call specifying all the parameters to characterize it. + +This Recommendation focuses on the provisioning of the service in a point-to-point communication configuration for the (switched) on-demand mode and in a point-to-point and unidirectional point-to-multipoint configurations for permanent and reserved modes; other communication configurations are for further study. + +The operations (from the user's point of view) necessary for the activation of a communication (from the network point of view) are successively (or simultaneously) as identified in Table 1. + +For these complete operations, there are three possibilities: + +- 1) per management procedures; +- 2) per signalling procedures; +- 3) per fast allocation procedures. + +Table 2 summarises the different options. + +This Recommendation specifies only procedures for the simultaneous request of establishment and activation (i.e. according to the third column). It is compatible with the separation of the request of establishment and activation (i.e. column two and three of Table 2) but procedures are for further study. + +TABLE 1/F.811 + +#### **User request and resulting actions in the network** + +| | User request | Resulting actions in the network | +|---|-----------------------------------------------------------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------| +| 1 | Request of a route including the ATC in order to be connected to (an)other user(s) | Marking the route: it corresponds to the establishment of the associated ATM connection(s) without the resources supporting the communication | +| 2 | Negotiation of the resources, i.e. request of traffic parameter(s), QOS classes and OAM flows per ATM connection supporting the communication | Allocation of the resources: it corresponds to the activation of the communication | + +TABLE 2/F.811 + +#### **Establishment and activation requests** + +| Negotiation Modes | Connection establishment request (without resources): i.e. request of the route including the ATC only | Connection activation request, i.e. negotiation of traffic parameter(s), QOS classes and OAM flows per ATM connection only | Simultaneous request of establishment and activation, i.e. simultaneous request of the route and negotiation of the traffic parameter(s), QOS classes and OAM flows per ATM connection | +|----------------------|--------------------------------------------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| permanent | not applicable | not applicable | at subscription | +| reserved | M | M
compatible S
compatible RM | M | +| (switched) on-demand | S | S
compatible M
compatible RM | S | + +M Made by Management procedures +S Made by Signalling protocols +RM Made by Resource Management cells +NOTES +1 According to Recommendation I.371[1], the use of Resource Management Cell is not specified for DBR and SBR ATCs. +2 In order to allow the user to send RM cells, a minimum level of resources is necessary. + +## **2.2 Specific terminology** + +For the purposes of this Recommendation, the following definitions apply. + +**2.2.1 ATM transfer capability:** See Recommendation I.371[1]. ATM Transfer capabilities are DBR, SBR, ABR, ABT-DT and ABT-IT. The specification of the relevant switching protocol at UNI to negotiate the ATM Transfer capability is contained in Recommendations Q.2931[9] and Q.2961[11]. The management protocols at UNI to negotiate the ATM Transfer capability are for further study. + +**2.2.2 cell delay variation:** See Recommendation I.356[2]. + +- 2.2.3 **cell delay variation tolerance:** See Recommendation I.371[1]. +- 2.2.4 **cell loss ratio:** See Recommendation I.356[2]. +- 2.2.5 **cell transfer delay:** See Recommendation I.356[2]. +- 2.2.6 **initiating manager:** The manager issuing a request to the Service Management Entity (SME). +- 2.2.7 **intrinsic burst tolerance:** See Recommendation I.371[1]. +- 2.2.8 **ISDN number:** A number conforming to the numbering plan and structure specified in Recommendation E.164[4]. +- 2.2.9 **involved manager:** The manager receiving a request issued by another initiating manager. +- 2.2.10 **manager:** In reserved mode, it is the functional entity authorized to send through x reference point (as defined in Recommendation M.3010[3]) the communication registration requests, the communication modification requests and interrogation requests, on behalf of the users to which it is related at the subscription time. +- 2.2.11 **maximum cell rate:** See Recommendation I.371[1]. The Maximum Cell Rate is declared by the user as an integer number of cell per second. +- 2.2.12 **minimum cell rate:** See Recommendation I.371[1]. The Minimum Cell Rate is declared by the user as an integer number of cell per second. +- 2.2.13 **peak cell rate:** See Recommendation I.371[1]. The Peak Cell Rate is declared by the user as an integer number of cell per second. +- 2.2.14 **peak re-negotiation rate:** See Recommendation I.371[1]. The peak re-negotiation rate is declared by the user as an integer number of cell per second. +- 2.2.15 **periodic communication:** A communication whose same scheme of values of source traffic descriptor and QOS class repeats within each period for the communication duration. +- 2.2.16 **quality of service class:** A set of values associated to the following ATM performance parameters: end-to-end Cell Loss Ratio, end-to-end Cell Transfer Delay, end-to-end Cell Delay Variation. +- 2.2.17 **service type:** See Recommendation I.371[1]. +- 2.2.18 **service management entity (SME):** In reserved mode the SME is the functional entity to which manager's communication registration requests, communication modifications requests and interrogation requests are addressed. +- 2.2.19 **source traffic descriptor:** The set of traffic parameters associated to an ATM transfer capability. For DBR it is given by the Peak Cell Rate and the associated Cell Delay Variation Tolerance. For SBR it is given by the Peak Cell Rate and the associated Cell Delay Variation Tolerance, the Sustainable Cell Rate, the Intrinsic Burst Tolerance and the associated Cell Delay Variation Tolerance. For ABR it is given by the Peak Cell Rate and the Minimum Cell Rate. For ABT-DT and ABT-IT it is given by the Maximum Cell Rate and the associated Cell Delay Variation Tolerance, the Peak re-negotiation rate and the associated Cell Delay Variation Tolerance, the Sustainable Cell Rate, the Intrinsic Burst Tolerance and the associated Cell Delay Variation Tolerance. See Recommendation I.371[1]. +- 2.2.20 **sustainable cell rate:** See Recommendation I.371[1]. The Sustainable Cell Rate is declared by the user as an integer number of cell per second. +- 2.2.21 **traffic contract:** See Recommendation I.371[1]. +- 2.2.22 **user:** Each of the subscriber's functional entity that sends or receives ATM cells through TB or TB/SB reference point. It communicates with the manager for the purpose of being controlled and/or monitored. + +# **3 Procedures** + +## **3.1 On demand mode** + +### **3.1.1 Provision/withdrawal** + +By arrangement with the SP. + +The beginning time of provision corresponds to the time of service subscription; the time of withdrawal to the end time of subscription. + +### **3.1.2 Normal procedures** + +##### **3.1.2.1 Activation/deactivation/registration** + +By arrangement with SP. + +##### **3.1.2.2 Invocation and operation** + +User-network signalling is done over a separate signalling channel. + +###### **3.1.2.2.1 Call set-up originating the service** + +Procedures at call set-up are specific for every configuration. However, the following definitions and principles apply to all the configurations. + +The source traffic descriptor consists of a pair of values for every parameter to define the communication in both directions. + +The QOS class defines the QOS required by the user for the communication. If the value is not specified, a default value is assumed. A pair of values is required to define the communication in both directions. + +The call is originated by the user requesting from the network the required bearer service (call set-up request). + +The call request includes the following parameters: + +###### **Communication configuration** + +- ISDN number of the called user. + +###### **ATM traffic contract on communication basis** + +- ATM transfer capability (according to 5.5.1.1/I.371[1], the ATC is the same in both directions if the communication is bidirectional); +- source traffic descriptor in each direction (see Recommendation I.371[1]); +- QOS class in each direction; +- Peak Cell Rate of the OAM F5 user cell flow and associated Cell Delay Variation Tolerance (for DBR, ABT-IT, ABT-DT only) in each direction (the support of the separation of the user OAM F5 flow and the user data flow is an SP option). + +###### **AAL parameters** + +- AAL parameters: required by the network only for interworking purpose (e.g. with N-ISDN). Within B-ISDN the user may specify them for end-to-end matter: the AAL is transparent to the network. + +###### **Service facilities** + +- Service type (optional, i.e. it is a user option to use this parameter). + +For the time being, the identified service types are the emulated narrowband bearer services (see Annex C). If the service type is provided, the consistency between the service type and the other parameters shall be checked within the network (see Annex C); if there is a lack of consistency, the value of the parameters considered within the network are the implicit values as provided in Annex C; afterwards, any notification to the calling user and any indication to the called user(s) will include the implicit values (considered within the network). + +After initiating a call, the calling user will receive an acknowledgement that the network is able to process the call. + +An indication of the incoming call is given to the called user (call set-up indication) including all the parameters of the call request. At called end upon receipt of a call set-up request, if able to support it, the called user will accept the call. + +##### **3.1.2.2.2 Procedures during the communication phase** + +###### **3.1.2.2.2.1 Modification of the traffic characteristics during the call** + +Both user and network may request the change of the traffic characteristics of an already established call. + +Traffic characteristics may be changed after negotiation between the user and the network. Notification of this change shall be sent to both users. + +###### **3.1.2.2.3 Call release** + +The call may be terminated by either or both users by sending a request to the network. If one user terminates a call, an appropriate indication is sent to the other user. + +#### **3.1.3 Exceptional procedures** + +#### **3.1.3.1 Activation/deactivation/registration** + +Not applicable. + +##### **3.1.3.2 Invocation and operation** + +###### **3.1.3.2.1 Failure situation due to user error** + +A user specifying a network-identifiable, improper service request will be given an appropriate failure indication by the network and the call set-up will be ceased. + +A user specifying a non-valid destination address will be given an appropriate failure indication by the network and the call set-up will be ceased. + +When the user's input cell stream violates the traffic contract, the network may take action and discard cells. + +###### **3.1.3.2.2 Failure situation due to resource limitations** + +During the call set-up, if the requested service resources specified by parameters, in particular QOS or source traffic descriptor, are not available in the network, an appropriate failure indication should be sent to the calling user. + +If the network cannot complete the call due to other limitations on network resources, an appropriate failure indication will be given to the requesting user. + +###### **3.1.3.2.3 Failure situation due to the called user state** + +A calling user attempting to establish a call to a user who is not accepting the call will be given an appropriate failure indication by the network. + +A user attempting to establish a call to a user whose terminal equipment fails to respond will be given an appropriate failure indication by the network and the call set-up will be ceased. + +#### **3.1.4 Alternative procedures** + +Not applicable. + +## **3.2 Reserved mode** + +### **3.2.1 Provision and withdrawal** + +Provision and withdrawal are based on schedule 1 (see Figure 1). + +The beginning time of provision corresponds to the time of service subscription; the time of withdrawal to the end time of subscription. + +After the BCOBS is available at time instant $t_1$ as a result of a BCOBS subscription request at time instant $t_0$ , a set of management procedures can be used to request, modify and release reserved communications. The notice time $t_1-t_0$ is greater than or equal to $T_{\min}$ . + +The time $t_3$ of service withdrawal is fixed at subscription or is a result of a subscriber request at time $t_2$ . The duration of the subscription $t_3-t_1$ is greater than a minimum duration $DA_{\min}$ and less than a maximum duration $DA_{\max}$ . + +More than one communication can be available at each UNI at the same time. The subscription stated the maximum number $NB_{\max}$ of communications that can be available at each UNI. + +Either periodic or occasional communications can be reserved. The parameters of each communication are negotiated between the SP and the subscriber in the registration procedure. + +In this subclause, $TS(i)$ identifies the time at which the set-up of the $i$ -th communication and connection shall be completed; the set-up time of the connection can be in advance and the exact time is not the purpose of this Recommendation. + +In this subclause, $TE(i)$ identifies the time from which the release of the $i$ -th communication and connection can be initiated; the exact release time of the $i$ -th connection is not the purpose of this Recommendation. + +The $i$ -th communication can be started at time instant $TA(i)$ explicitly specified at the time instant of communication and connection registration request $RR(i)$ . Communication and connection release occurs at time instant $TD(i)$ explicitly specified also at $RR(i)$ . Communication and connection duration is predetermined: the communication and connection is set up for a specified period of time. As an option, connection release can be initiated from time instant $TE(i)$ following a release request made at time instant $RD(i)$ during the communication and *a priori* undetermined [ $TD(i)-RD(i)$ is as short as possible]. This option corresponds to an unspecified duration of the communication and connection, or to a possibility of unanticipated release. Time $TR(i)$ corresponds to the time instant in which the $i$ -th communication is reserved in response to the request at $RR(i)$ . + +![Timeline diagram showing the sequence of events for the provision of a BCOBS in reserved mode for an occasional communication. The timeline is a horizontal line with arrows at both ends, marked with time instants: t0, t1, RR(i), TR(i), TA(i) = TS(i), RD(i), TD(i) = TE(i), t2, and t3. The label T0106000-96/d01 is at the bottom right.](78ffccd66df9bafd96e3e081110d09dd_img.jpg) + +T0106000-96/d01 + +Timeline diagram showing the sequence of events for the provision of a BCOBS in reserved mode for an occasional communication. The timeline is a horizontal line with arrows at both ends, marked with time instants: t0, t1, RR(i), TR(i), TA(i) = TS(i), RD(i), TD(i) = TE(i), t2, and t3. The label T0106000-96/d01 is at the bottom right. + +| | | +|---------|------------------------------------------------------------------------------------------| +| $t_0$ | Time at which the subscription to the BCOBS is requested | +| $t_1$ | Time at which the BCOBS is available | +| $RR(i)$ | Time at which the reservation request of the $i$ -th communication is made | +| $TR(i)$ | Time at which the $i$ -th communication is notified to the manager to be reserved or not | +| $TS(i)$ | Time at which the $i$ -th communication set-up shall be completed | +| $TA(i)$ | Time at which the $i$ -th communication is activated | +| $RD(i)$ | Time at which the deactivation request for the $i$ -th communication is made | +| $TD(i)$ | Time at which the $i$ -th communication is deactivated | +| $TE(i)$ | Time from which the $i$ -th communication release can be initiated | +| $t_2$ | Time at which the subscription to the BCOBS is requested to be terminated | +| $t_3$ | Time at which the subscription to the BCOBS is terminated | + +NOTE – This figure indicates the set-up and release time of the $i$ -th communication, however more than one communication is possible within a subscription. + +FIGURE 1/F.811 +**Provision of a BCOBS in reserved mode: occasional communication** + +For an occasional communication, the value of source traffic descriptor and of QOS class are constant for the communication duration, except as a result of a registration modification procedure. + +For a periodic communication (see Figure 2), the same scheme of values of source traffic descriptor and QOS repeats within each period for the communication duration. Within a period, a number of time slots are defined (see Figure 3). A time slot is the interval between an activation time $TA_j(i)$ and the related deactivation $TD_j(i)$ . + +![Figure 2/F.811: Provision of a BCOBS in reserved mode: periodic communication. The diagram shows a timeline with several time points: t0, t1, RR(i), TR(i), TS(i), TE(i), t2, and t3. The timeline is divided into periods: Period 1, Period j, Period j + 1, and Last Period. The time points are marked on the timeline as follows: t0 is the start, t1 is the start of Period 1, RR(i) is the reservation request time, TR(i) is the notification time, TS(i) is the set-up completion time, TE(i) is the release time, t2 is the end of the Last Period, and t3 is the termination time. The text 'T0106010-96/d02' is located at the bottom right of the diagram.](4e4be0bd8b235167902f2c03e41da651_img.jpg) + +Figure 2/F.811: Provision of a BCOBS in reserved mode: periodic communication. The diagram shows a timeline with several time points: t0, t1, RR(i), TR(i), TS(i), TE(i), t2, and t3. The timeline is divided into periods: Period 1, Period j, Period j + 1, and Last Period. The time points are marked on the timeline as follows: t0 is the start, t1 is the start of Period 1, RR(i) is the reservation request time, TR(i) is the notification time, TS(i) is the set-up completion time, TE(i) is the release time, t2 is the end of the Last Period, and t3 is the termination time. The text 'T0106010-96/d02' is located at the bottom right of the diagram. + +| | | +|---------|------------------------------------------------------------------------------------------| +| $t_0$ | Time at which the subscription to the BCOBS is requested | +| $t_1$ | Time at which the BCOBS is available | +| $RR(i)$ | Time at which the reservation request of the $i$ -th communication is made | +| $TR(i)$ | Time at which the $i$ -th communication is notified to the manager to be reserved or not | +| $TS(i)$ | Time at which the $i$ -th communication set-up shall be completed | +| $TE(i)$ | Time from which the $i$ -th communication release can be initiated | +| $t_2$ | Time at which the subscription to the BCOBS is requested to be terminated | +| $t_3$ | Time at which the subscription to the BCOBS is terminated | + +NOTE – This figure indicates the set-up and release time of the $i$ -th communication, however more than one communication is possible within a subscription. + +FIGURE 2/F.811 + +##### **Provision of a BCOBS in reserved mode: periodic communication** + +The following parameters shall be specified at subscription as mandatory subscription parameters: + +##### **Communication configuration** + +- Manager identifier; +- ISDN numbers of the users controlled by the manager. + +##### **Time constraints** + +- Subscription beginning time $t_1$ ; +- Subscription end time $t_3$ . + +##### **Access constraints** + +- Maximum number $NB_{max}$ of communications that can be available at each UNI. + +### **3.2.2 Procedures** + +#### **3.2.2.1 Activation, deactivation** + +The activation procedure leads to the effective availability of the resources corresponding to the request source traffic descriptor for the communication. The deactivation procedure leads to deallocate the resources corresponding to the request source traffic descriptor for the communication. + +For an occasional communication $i$ , according to Figure 1, activation occurs at time $TA(i)$ and deactivation at time $TD(i)$ . + +For a periodic communication $i$ -th according to Figure 3, in every period activation occurs at each time $TA_k(i)$ with $k = 1$ to $n$ , and deactivation at each time $TD_k(i)$ with $k = 1$ to $n$ . + +![Diagram showing the definition of time slots within a period 'm'. A horizontal timeline is divided into segments by vertical tick marks. The first segment is labeled with TA_1(i) and TD_1(i). The second segment is labeled with TA_j(i), TD_j(i) = TA_{j+1}(i), and TD_{j+1}(i). The third segment is labeled with TA_{j+2}(i) and TD_{j+2}(i). The final segment is labeled with TA_n(i) and TD_n(i). A double-headed arrow above the timeline spans from the first tick mark to the last, labeled 'Period m'. Below the timeline, a legend defines TA_k(i) as 'Time of the k-th activation of the communication i-th in the period m' and TD_k(i) as 'Time of the k-th deactivation of the communication i-th in the period m'. The code 'T0106020-96/d03' is in the bottom right corner.](4801720824e4b5e2361a5564f91cfb70_img.jpg) + +$TA_k(i)$ Time of the $k^{th}$ activation of the communication $i$ -th in the period $m$ + $TD_k(i)$ Time of the $k^{th}$ deactivation of the communication $i$ -th in the period $m$ + +T0106020-96/d03 + +Diagram showing the definition of time slots within a period 'm'. A horizontal timeline is divided into segments by vertical tick marks. The first segment is labeled with TA\_1(i) and TD\_1(i). The second segment is labeled with TA\_j(i), TD\_j(i) = TA\_{j+1}(i), and TD\_{j+1}(i). The third segment is labeled with TA\_{j+2}(i) and TD\_{j+2}(i). The final segment is labeled with TA\_n(i) and TD\_n(i). A double-headed arrow above the timeline spans from the first tick mark to the last, labeled 'Period m'. Below the timeline, a legend defines TA\_k(i) as 'Time of the k-th activation of the communication i-th in the period m' and TD\_k(i) as 'Time of the k-th deactivation of the communication i-th in the period m'. The code 'T0106020-96/d03' is in the bottom right corner. + +FIGURE 3/F.811 +**Definition of time slots within a period** + +As an SP option for a periodic communication, a deactivation may be simultaneous with an activation. This shall correspond to a source traffic descriptor modification ensuring the continuity of the communication when the QOS class of the previous time slot is equal to the QOS class of the following one. + +#### 3.2.2.2 Registration + +The registration procedure inserts the parameters of a communication in the network after a negotiation between the subscriber(s) and the SP. + +The following parameters are relevant for an occasional communication: + +###### Communication configuration + +- Configuration; +- ISDN numbers of the users. + +###### Time constraints + +- Time $TS(i)$ ; +- Time $TE(i)$ (optional). + +###### ATM traffic contract on communication basis + +- ATM transfer capability (according to 5.5.1.1/I.371[1], the ATC is the same in both directions if the communication is bidirectional); +- Source traffic descriptor, in each direction, (see Recommendation I.371[1]); +- QOS Class, in each direction; +- Peak Cell Rate of the OAM F5 user cell flow and associated Cell Delay Variation Tolerance (for DBR, ABT-IT, ABT-DT only) in each direction (the support of the separation of the user OAM F5 flow and the user data flow is an SP option). + +###### AAL parameters + +- AAL parameters: required by the network only for interworking purpose (e.g. with N-ISDN). Within B-ISDN the user may specify them for end-to-end matter: the AAL is transparent to the network. + +###### Service facilities + +- Service type (optional, i.e. it is a user option to use this parameter). + +Otherwise for a periodic communication the following set of parameters is relevant: + +##### Communication configuration + +- Configuration; +- ISDN numbers of the users. + +###### Time constraints + +- Period: e.g. day, week, etc.; +- Time TS(i); +- Time TE(i) that corresponds to the end of the last period (optional); +- Number N of time slots; +- Beginning time of each time slot in the period: $TA_i$ (with $1 \leq i \leq N$ ); +- End time of each time slot in the period: $TD_i$ (with $1 \leq i \leq N$ ). + +###### ATM traffic contract on communication basis + +- ATM transfer capability (according to 5.5.1.1/I.371[1], the ATC is the same in both directions if the communication is bidirectional); +- Source traffic descriptor for each time slot, in each direction (see Recommendation I.371[1]); +- QOS class for each time slot, in each direction; +- Peak Cell Rate of the OAM F5 user cell flow and associated Cell Delay Variation Tolerance (for DBR, ABT-IT, ABT-DT only) in each direction (the support of the separation of the user OAM F5 flow and the user data flow is an SP option). + +###### AAL parameters + +- AAL parameters: required by the network only for interworking purpose (e.g. with N-ISDN). Within B-ISDN the user may specify them for end-to-end matter: the AAL is transparent to the network. + +###### Service facilities + +- Service type (optional, i.e. it is a user option to use this parameter). + +For the time being, the identified service types are the emulated narrowband bearer services (see Annex C). If the service type is provided, the value of ATM transfer capability, source traffic descriptor, AAL parameters and QOS class shall not be specified by the subscriber and the network shall assign to them the implicit values as provided in Annex C; afterwards, any notification to the calling user and any indication to the called user(s) will include the implicit values (considered within the network). + +If a manager requests the registration of a communication between two or more users all controlled by that manager, the following procedure applies: + +- 1) The manager requests to the SME the registration of a communication providing the relevant parameters. +- 2) The SME shall issue a positive notification (including the global identifier of the communication, the value of VPIs and VCIs assigned to the ATM connection at each UNI, and the values of the relevant network traffic parameters) or negative notification. In the case of positive notification, the communication shall be established as requested. Together with a negative notification, the SME can propose the parameter values that would have made the request viable. + +If a manager (initiating manager) requests the registration of a communication between two or more Users controlled by other managers (involved manager), the following procedure applies: + +- 1) The initiating manager requests to the SME the registration of a communication providing the relevant parameters. +- 2) The SME shall provide either a negative notification to the initiating manager or an indication to each other involved manager reporting the relevant parameters of the communication. Together with the negative notification, the SME can propose the parameter values that would have made the request viable. In the case of negative notification, the procedure stops. +- 3) Each involved manager shall provide the SME either a positive or negative notification. Together with the negative notification, the involved manager can propose the parameter values that would have made the request viable. + +- 4) If each involved manager sends a positive notification, the SME shall provide to all the involved managers and to the initiating manager a positive notification (including the global identifier of the communication, the value of VPIs and VCIs assigned to the ATM connection at each UNI, and the values of the relevant network traffic parameters). Otherwise the SME shall provide a negative notification to the initiating manager together with the parameter values that would have made the request viable for all the involved managers. + +As a service provider option, the manager may request a communication be established as soon as possible [TS(i) = “as soon as possible”]. This manager shall be notified: + +- 1) if the network is not able to establish the communication within a limit time fixed by the SP (Note); +- 2) when the communication is effectively established. + +NOTE – The establishment time foreseen by the network may be notified. + +#### 3.2.2.3 Erasure + +The erasure for the i-th communication applies at the release time TE(i). + +The erasure leads to the deletion of all the parameters in the network of the relevant communication. + +#### 3.2.2.4 Invocation and operation + +After the registration of a communication, the service can be used without invocation during each active phase. Every attempt to send information outside an active phase shall result in cells discarded by the network. In addition, any violation of the traffic contract during the active phases may result in cells discarded by the network; if the traffic contract is violated, the QOS class agreed at registration is not assured. + +#### 3.2.2.5 Registration modification + +The registration modification procedure modifies the parameters in the network of a previously registered communication after negotiation among the SME and all the involved managers. It is not possible to modify an occasional communication in a periodic one or vice versa. + +For an occasional communication the following parameters can be modified in the registration modification procedure: + +- configuration; +- time TE(i); +- source traffic descriptor in each direction. + +Otherwise, for a periodic communication the following parameters can be modified in the registration modification procedure: + +- configuration; +- time TE(i); +- beginning time of one or more time slots in the period: $TA_i$ (with $1 \leq i \leq N$ ); +- end time of one or more time slots in the period: $TD_i$ (with $1 \leq i \leq N$ ); +- source traffic descriptor for one or more time slots, in each direction. + +If a manager requests the modification of a communication between two or more users all controlled by that manager, the following procedure applies: + +- 1) The manager requests the modification of the communication to the SME, providing the relevant parameters. +- 2) The SME shall issue a positive notification (including the values of the relevant network traffic parameters if the source traffic descriptor has changed) or a negative notification. In the case of a positive notification, the communication shall be established as requested. Together with the negative notification, the SME can propose the parameter values that would have made the request viable. + +If a manager (initiating manager) requests the registration of a communication between two or more users controlled by other managers (involved managers), the following procedure applies: + +- 1) The initiating manager requests the modification of the communication to the SME, providing the relevant parameters. +- 2) The SME shall provide either a negative notification to the initiating manager or an indication to each involved manager reporting the parameters to be changed and their values. Together with the negative notification, the SME can propose the parameter values that would have made the request viable. In the case of negative notification, the procedure stops. +- 3) Each involved manager shall provide the SME either a positive or negative notification. Together with the negative notification, the involved manager can propose the parameter values that would have made the request viable. +- 4) If each involved manager sends a positive notification, the SME shall provide to all the involved managers and to the initiating manager a positive notification (including the values of the relevant network traffic parameters if the source traffic descriptor has changed). Otherwise the SME shall provide a negative notification to the initiating manager, together with the parameter values that would have made the request viable for all the involved managers. + +## 3.3 Permanent mode + +### 3.3.1 Provision and withdrawal + +Provision and withdrawal are based on schedule 1 (see Figure 4). + +The beginning time of provision corresponds to the time of service subscription; the time of withdrawal to the end time of subscription. + +The permanent communication *i* of the BCOBS can be started after the connection set-up is completed at time instant $t1(i)$ in response to a request for the subscription of the BCOBS in permanent mode at time instant $t0$ [ $t0 < t1(i)$ ]. + +The time at which the release of the communication is initiated [ $t2(i)$ ] can be specified or unspecified; if the duration is unspecified, the communication (and the connection to support it) is released at the end or after the end of subscription $t3$ . + +More than one communication can be subscribed within the same subscription. The minimum duration of a communication [ $t2(i)-t1(i)$ ] is $DA_{min}$ . The set-up time $t1(i)-t0$ is greater than or equal to $Ta_{min}$ . $DA_{min}$ and $Ta_{min}$ are fixed by the SP. + +![Timeline diagram showing events t0, t1, t1(i), t2(i), and t3 along a horizontal time axis.](29f586959675cafdf81cf934954908eb_img.jpg) + +``` + +graph LR + t0 --- t1 --- t1i[t1\(i\)] --- t2i[t2\(i\)] --- t3 + style t0 stroke-width:0px + style t1 stroke-width:0px + style t1i stroke-width:0px + style t2i stroke-width:0px + style t3 stroke-width:0px + +``` + +$t0$ Time at which the subscription to the BCOBS is requested + $t1$ Time at which the subscription to the BCOBS is effective + $t1(i)$ Time at which the *i*-th communication set-up shall be completed + $t2(i)$ Time from which the *i*-th communication release could be initiated + $t3$ End of subscription + +T0106030-96/d04 + +Timeline diagram showing events t0, t1, t1(i), t2(i), and t3 along a horizontal time axis. + +NOTE – This figure indicates the set-up and release time of the *i*-th communication. However, more than one communication is possible within a subscription. + +FIGURE 4/F.811 +**Provision of a BCOBS in permanent mode** + +Either periodic or occasional communications can be subscribed. The parameters of each communication are negotiated between the SP and the subscriber at the subscription. + +For an occasional communication, the value of the source traffic descriptor and of the QOS are constant for the subscription duration. + +For a periodic communication (see Figure 5), the same scheme of values of the source traffic descriptor and the QOS class is repeated within each period for the subscription duration. + +![Figure 5/F.811: Provision of a BCOBS in permanent mode: periodic communication. The diagram shows a timeline with markers t0, t1, t1(i), t2(i), and t3. Periods are labeled Period 1, Period j, Period j + 1, and Last Period. A legend defines the time markers: t0 (Time at which the subscription to the BCOBS is requested), t1 (Time at which the subscription to the BCOBS is effective), t1(i) (Time at which the i-th communication set-up shall be completed), t2(i) (Time from which the i-th communication release could be initiated), and t3 (End of subscription).](dfe556fea00682b09a59427aaf72051c_img.jpg) + +t0 Time at which the subscription to the BCOBS is requested + t1 Time at which the subscription to the BCOBS is effective + t1(i) Time at which the i-th communication set-up shall be completed + t2(i) Time from which the i-th communication release could be initiated + t3 End of subscription + +T0106040-96/d05 + +Figure 5/F.811: Provision of a BCOBS in permanent mode: periodic communication. The diagram shows a timeline with markers t0, t1, t1(i), t2(i), and t3. Periods are labeled Period 1, Period j, Period j + 1, and Last Period. A legend defines the time markers: t0 (Time at which the subscription to the BCOBS is requested), t1 (Time at which the subscription to the BCOBS is effective), t1(i) (Time at which the i-th communication set-up shall be completed), t2(i) (Time from which the i-th communication release could be initiated), and t3 (End of subscription). + +NOTE – This figure indicates the set-up and release time of the ith communication. However, more than one communication is possible within a subscription. + +FIGURE 5/F.811 +**Provision of a BCOBS in permanent mode: periodic communication** + +#### 3.3.1.1 Occasional communication + +The following subscription parameters shall be agreed by the SP and the subscriber: + +###### Communication configuration + +- Configuration; +- ISDN numbers of the users. + +###### Time constraints + +- Time t1(i); +- Time t2(i) (optional). + +###### ATM traffic contract on communication basis + +- ATM transfer capability (according to 5.5.1.1/I.371[1], the ATC is the same in both directions if the communication is bidirectional); +- Source traffic descriptor in each direction, (see Recommendation I.371[1]); +- QOS class, in each direction; +- Peak Cell Rate of the OAM F5 user cell flow and associated Cell Delay Variation Tolerance (for DBR, ABT-IT, ABT-DT only) in each direction (the support of the separation of the user OAM F5 flow and the user data flow is an SP option). + +###### AAL parameters + +- AAL parameters: required by the network only for interworking purpose (e.g. with N-ISDN). Within B-ISDN the user may specify them for end-to-end matter: the AAL is transparent to the network. + +###### Service facilities + +- Service type (optional, i.e. it is a user option to use this parameter). + +For the time being, the identified service types are the emulated narrowband bearer services (see Annex C). If the service type is provided, the value of ATM transfer capability, source traffic descriptor, AAL parameters and QOS class shall not be specified by the subscriber and the network shall assign to them the implicit values as provided in Annex C. + +#### 3.3.1.2 Periodic communication + +The subscription allows the specification of a period and of a number N of time slots within the period (see Figure 3). The number N of time slots shall be within 1 and $N_{max}$ (fixed by the service Provider). + +For each time slot the values of the source traffic descriptor and of the QOS class can be specified independently from the values they have in other time slots. + +The following subscription parameters shall be agreed by the SP and the subscriber: + +##### Communication configuration + +- ISDN numbers of the users; +- Configuration. + +###### Time constraints + +- Period: e.g. day, week, etc.; +- Time $t1(i)$ ; +- $t2(i)$ that corresponds to the end of the last period (optional); +- Number N of time slots; +- Beginning time of each time slot in the period: $TA_k(i)$ (with $1 \leq k \leq N$ ); +- End time of each time slot in the period: $TD_k(i)$ (with $1 \leq k \leq N$ ). + +###### ATM traffic contract on communication basis + +- ATM transfer capability (according to 5.5.1.1/I.371[1], the ATC is the same in both directions if the communication is bidirectional); +- Source traffic descriptor for each time slot, in each direction (see Recommendation I.371[1]); +- QOS class for each time slot, in each direction; +- Peak Cell Rate of the OAM F5 user cell flow and associated Cell Delay Variation Tolerance (for DBR, ABT-IT, ABT-DT only) in each direction (the support of the separation of the user OAM F5 flow and the user data flow is an SP option). + +###### AAL parameters + +- AAL parameters: required by the network only for interworking purpose (e.g. with N-ISDN). Within B-ISDN the user may specify them for end-to-end matter: the AAL is transparent to the network. + +###### Service facilities + +- Service type (optional, i.e. it is a user option to use this parameter). + +For the time being, the identified service types are the emulated narrowband bearer services (see Annex C). If the service type is provided, the value of ATM transfer capability, source traffic descriptor, AAL parameters and QOS class shall not be specified by the subscriber and the network shall assign to them the implicit values as provided in Annex C. + +#### 3.3.1.3 Subscription modification + +As an SP option, after time $t1$ , it is possible to modify a subscription by: + +- modifying the parameters of a subscribed communication; +- releasing a communication before the time instant fixed at subscription; +- adding a communication not included in the subscription. + +A subscription modification shall be agreed by the SP and the subscriber with a minimum anticipation $T_{a_{min}}$ with respect to the time instant at which the subscription modification is required to be effective. + +#### **3.3.2 Procedures** + +#### **3.3.2.1 Activation, deactivation** + +The activation procedure leads to the effective availability of the resources corresponding to the requested source traffic descriptor for the communication. The deactivation procedure leads to the deallocation of the resources corresponding to the requested source traffic descriptor for the communication. + +For an occasional communication $i$ , according to Figure 4, activation occurs at time $t1(i)$ and deactivation at time $t2(i)$ . + +For a periodic communication $i$ according to Figure 3, in every period activation occurs at each time $TA_k(i)$ with $k = 1$ to $N$ , and deactivation at each time $TD_k(i)$ with $k = 1$ to $N$ . + +As an SP option for a periodic communication, a deactivation may be simultaneous with an activation. This shall correspond to a source traffic descriptor modification ensuring the continuity of the communication when the QOS class of the previous time slot is equal to the QOS class of the following one. + +#### **3.3.2.2 Registration** + +The registration is done at subscription. It is managed by the SP. + +With the registration the parameters of each subscribed communication are stored in the network. A global identifier is assigned to each subscribed communication. + +#### **3.3.2.3 Erasure** + +The erasure is done at the end of the subscription. It is managed by the SP. + +The erasure leads to the deletion of all the parameters of the subscribed communications. + +#### **3.3.2.4 Invocation and operation** + +After the subscription, the service can be used without invocation during each active phase. Every attempt to send information outside an active phase shall result in cells discarded by the network. In addition, any violation of the traffic contract during the active phases may result in cells discarded by the network. + +##### **3.3.2.5 Registration modification** + +The registration modification corresponds to a subscription modification. + +## **4 Network capabilities for charging** + +Charging principles are outside the scope of this Recommendation. + +## **5 Interworking requirements** + +Interworking between the B-ISDN and narrow-band ISDN for the cases included in Annex C is required. A caller should be able to originate a call at the $S_B$ or $T_B$ reference point and terminate at an $S$ or $T$ reference point on the B- or D-channel (and vice versa). + +Interworking between B-ISDN and PSTNs, for data calls which have information transfer characteristics within the capability of the PSTNs, requires the use of an interworking function (including a modem). + +NOTE – The end-to-end delay limits for B-ISDN interactive services are not yet established. It will be necessary to take into consideration the implication of propagation delay on real-time interactive services (e.g. interactive video/audio) introduced by use of satellites when determining these end-to-end delay limits. This is for further study. + +## **6 Interaction with supplementary services** + +Each supplementary service description identifies the applicability to this bearer service. + +## Annex A + +### Broadband connection-oriented bearer service + +(This annex forms an integral part of this Recommendation) + +| Bearer service attributes | Values of attributes | +|------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Information transfer attributes | | +| 1 Information transfer mode | ATM | +| 2 Information transfer capability | Unrestricted | +| 3 Structure | Cell sequence integrity | +| 4 Establishment of communication | (Switched) On-demand, reserved, permanent | +| 5 Symmetry of communication | Bidirectional symmetric, bidirectional asymmetric, unidirectional | +| 6 Communication configuration | Point-to-point, point-to-multipoint (for reserved and permanent mode only). Others for further study (ffs) | +| Access attributes | | +| 7 Access channel and rate | | +| 7.1 For user information | | +| Type of connection | VCC | +| ATM Transfer Capability | DBR, SBR, ABR or ABT-IT, ABT-DT (see Rec. I.371 [1]) | +| QOS class | See Rec. I.356 [2] | +| Information Transfer Rate | Declared by the user on a per connection basis (separately for each time slot for a periodic communication). Values are specified individually for each direction of transmission. Specific parameters apply depending on the ATM transfer capability (see Rec. I.371 [1]) | +| 7.2 For signalling | | +| Type of connection | VCC | +| ATM Transfer Capability | DBR (see Rec. I.371 [1]) | +| QOS class | Class 1 (See Rec. I.356 [2]) | +| Information Transfer Rate | Ffs | +| 7.3 For Management | | +| Type of connection | VCC | +| ATM Transfer Capability | DBR (see Rec. I.371 [1]) | +| QOS class | Class 1 (See Rec. I.356 [2]) | +| Information Transfer Rate | Ffs | +| 7.4 For fast allocation procedures | Ffs | +| 8 Access protocols (Note) | | +| 8.1 Physical layer | Rec. I.432 [5] | +| 8.2 ATM layer | Recs. I.150 [6], I.361 [7] | +| 8.3 Signalling access protocol ATM adaptation layer (AAL) | Recs. Q.2100 [8], Q.2110 [16] and Q.2130 [17] | +| 8.4 Signalling access protocol layer 3 (above AAL) | Recs. of the Q.2900 (DSS 2) Series, Q.2931 [9], Q.2961 [11], Q.2962 [18] and Q.2963.1 [19] | +| 8.5 Information access protocol ATM adaptation layer (AAL) | Type 1 (see Rec I.363.1 [12]), or type 3/4 (see Rec I.363.3 [10]), or type 5 (see Rec. I.363.5 [20]), or empty | +| 8.6 Information access protocol – Layers above AAL up to layer 3 | Specific service dependent | +| General attributes | | +| 9 Supplementary services provided | Ffs | +| 10 Interworking possibilities | See Rec. I.580 [21] | +| 11 Operational and commercial aspects | Out of the scope of this Recommendation | +| NOTE – Management access protocols are for further study. | | + +## Annex B + +## Alphabetical list of abbreviations used in this Recommendation + +(This annex forms an integral part of this Recommendation) + +| | | +|--------|-------------------------------------------------| +| AAL | ATM adaptation layer | +| ABR | Available Bit Rate | +| ABT-DT | ATM Block Transfer – Delayed Transmission | +| ABT-IT | ATM Block Transfer – Immediate Transmission | +| ATC | ATM Transfer Capability | +| ATM | Asynchronous Transfer Mode | +| B-ISDN | Broadband Integrated Services Digital Network | +| BCOBS | Broadband Connection-Oriented Bearer Service | +| CBR | Constant Bit Rate | +| CDV | Cell Delay Variation | +| DBR | Deterministic Bit Rate | +| DSS 2 | Digital Subscriber Signalling System 2 | +| N-ISDN | Narrow-band Integrated Services Digital Network | +| OAM | Operation and Maintenance | +| PSTN | Public Switched Telephone Network | +| QOS | Quality of Service | +| SBR | Sustainable Bit Rate | +| SME | Service Management Entity | +| SP | Service Provider | +| UNI | User-Network Interface | +| VBR | Variable Bit Rate | +| VCC | Virtual Channel Connection | +| VCI | Virtual Channel Identification | +| VPI | Virtual Path Identifier | + +## Annex C + +### Narrow-band emulated services + +(This annex forms an integral part of this Recommendation) + +This annex does not include a complete list of narrow-band emulated services. Additional narrow-band emulated services are for further study. + +### C.1 64 kbit/s bearer communication usable for speech + +The parameters applied to a call requiring this connection-oriented bearer communication are: + +#### ATM traffic contract on communication basis + +- ATM transfer capability: DBR; +- Source traffic descriptor: Peak Cell Rate, 171 Cell/s in both directions (no OAM cell shall be sent); +- QOS class: class 1 (see Recommendation I.356[2]); +- Peak Cell Rate of the OAM F5 user cell flow and associated Cell Delay Variation Tolerance: 0. + +#### **AAL parameters** + +- AAL parameters: AAL1 for voice band signal transport as defined in Recommendation I.363.1[12] in both directions. + +#### **Service facilities** + +- Service type: 64 kbit/s bearer communication usable for speech. + +Inter-operability with “Circuit-mode 64 kbit/s 8 kHz structured bearer service category usable for speech information transfer” [13] is required. In this case the generation and transport of tones and/or announcements should be compliant with Recommendation Q.2931[9]. + +### **C.2 64 kbit/s bearer communication usable for 3.1 kHz audio** + +The parameters applied to a call requiring this connection-oriented bearer communication are: + +#### **ATM traffic contract on communication basis** + +- ATM transfer capability: DBR; +- Source traffic descriptor: Peak Cell Rate, 171 Cell/s in both directions (no OAM cell shall be sent); +- QOS class: class 1 (see Recommendation I.356[2]); +- Peak Cell Rate of the OAM F5 user cell flow and associated Cell Delay Variation Tolerance: 0. + +#### **AAL parameters** + +- AAL parameters: AAL1 for voice band signal transport as defined in Recommendation I.363.1[12] in both directions. + +#### **Service facilities** + +- Service type: 64 kbit/s bearer communication usable for 3.1 kHz audio. + +Inter-operability with “Circuit-mode 64 kbit/s 8 kHz structured bearer service category usable for 3.1 kHz audio information transfer” [14] is required. In this case the generation and transport of tones and/or announcements should be compliant with Recommendation Q.2931[9]. + +### **C.3 64 kbit/s unrestricted bearer service** + +The parameters applied to a call requiring this connection-oriented bearer communication are: + +#### **ATM traffic contract on communication basis** + +- ATM transfer capability: DBR; +- Source traffic descriptor: Peak Cell Rate, 171 Cell/s in both directions (no OAM cell shall be sent); +- QOS class: class 1 (see Recommendation I.356[2]); +- Peak Cell Rate of the OAM F5 user cell flow and associated Cell Delay Variation Tolerance: 0. + +#### **AAL parameters** + +- AAL parameters: AAL1 for circuit transport as defined in Recommendation I.363.1[12]. + +#### **Service facilities** + +- Service type: 64 kbit/s unrestricted bearer service. + +Inter-operability with “Circuit-mode 64 kbit/s unrestricted, 8 kHz structured bearer service category” [15] is required. + +## Annex D + +## References + +(This annex forms an integral part of this Recommendation) + +- [1] ITU-T Recommendation I.371 (1993), *Traffic control and congestion control in B-ISDN.* +- [2] ITU-T Recommendation I.356 (1993), *B-ISDN ATM layer cell transfer performance.* +- [3] ITU-T Recommendation M.3010 (1996), *Principles for a telecommunications management network.* +- [4] CCITT Recommendation E.164 (1991), *Numbering plan for the ISDN era.* +- [5] ITU-T Recommendation I.432 (1993), *B-ISDN user-network interface – Physical layer specification.* +- [6] ITU-T Recommendation I.150 (1995), *B-ISDN asynchronous transfer mode functional characteristics.* +- [7] ITU-T Recommendation I.361 (1995), *B-ISDN ATM layer specification.* +- [8] ITU-T Recommendation Q.2100 (1994), *B-ISDN Signalling ATM Adaptation Layer (SAAL) overview description.* +- [9] ITU-T Recommendation Q.2931 (1995), *Broadband Integrated Services Digital Network (B-ISDN) – Digital Subscriber Signalling System No. 2 (DSS 2) – User-Network Interface (UNI) layer 3 specification for basic call/connection control.* +- [10] ITU-T Recommendation I.363.3 (1996), *B-ISDN ATM Adaptation Layer specification: types 3/4 AAL.* +- [11] ITU-T Recommendation Q.2961 (1995), *Broadband Integrated Services Digital Network (B-ISDN) – Digital Subscriber Signalling System No. 2 (DSS 2) – Additional traffic parameters.* +- [12] ITU-T Recommendation I.363.1 (1996), *B-ISDN ATM Adaptation Layer specification: types 1 and 2 AAL.* +- [13] CCITT Recommendation I.231.2 (1988), *64 kbit/s, 8 kHz structured, usable for speech information transfer.* +- [14] CCITT Recommendation I.231.3 (1988), *64 kbit/s, 8 kHz structured, usable for 3.1 kHz audio information transfer.* +- [15] CCITT Recommendation I.231.1 (1988), *64 kbit/s unrestricted, 8 kHz structured.* +- [16] ITU-T Recommendation Q.2110 (1994), *B-ISDN ATM Adaptation Layer – Service Specific Connection Oriented Protocol (SSCOP).* +- [17] ITU-T Recommendation Q.2130 (1994), *B-ISDN signalling ATM Adaptation Layer – Service specific coordination function for support of signalling at the User-Network Interface (SSCF at UNI).* +- [18] ITU-T Recommendation Q.2962 (1996), *Broadband Integrated Services Digital Network (B-ISDN) Digital Subscriber Signalling System No. 2 (DSS 2) – Connection characteristics negotiation during call/connection establishment phase.* +- [19] ITU-T Recommendation Q.2963.1 (1996), *Broadband Integrated Services Digital Network (B-ISDN) Digital Subscriber Signalling System No. 2 (DSS 2) – Connection modification – Peak cell rate modification by connection owner.* +- [20] ITU-T Recommendation I.363.5 (1996), *B-ISDN ATM Adaptation Layer specification: type 5 AAL.* +- [21] ITU-T Recommendation I.580 (1995), *General arrangements for interworking between B-ISDN and 64 kbit/s based ISDN.* + +# ITU-T RECOMMENDATIONS SERIES + +- Series A Organization of the work of the ITU-T +- Series B Means of expression +- Series C General telecommunication statistics +- Series D General tariff principles +- Series E Telephone network and ISDN +- Series F Non-telephone telecommunication services** +- Series G Transmission systems and media +- Series H Transmission of non-telephone signals +- Series I Integrated services digital network +- Series J Transmission of sound-programme and television signals +- Series K Protection against interference +- Series L Construction, installation and protection of cables and other elements of outside plant +- Series M Maintenance: international transmission systems, telephone circuits, telegraphy, facsimile and leased circuits +- Series N Maintenance: international sound-programme and television transmission circuits +- Series O Specifications of measuring equipment +- Series P Telephone transmission quality +- Series Q Switching and signalling +- Series R Telegraph transmission +- Series S Telegraph services terminal equipment +- Series T Terminal equipments and protocols for telematic services +- Series U Telegraph switching +- Series V Data communication over the telephone network +- Series X Data networks and open system communication +- Series Z Programming languages \ No newline at end of file diff --git a/marked/F/T-REC-F.82-199110-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.82-199110-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..99138d38ec6676f0023e3605f9b4fe5e3951d147 --- /dev/null +++ b/marked/F/T-REC-F.82-199110-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aa65ab234ca7d71e8a87800bda77e2a00132c8ba26ed84ea2f9a86d51172b89d +size 7471 diff --git a/marked/F/T-REC-F.82-199110-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg b/marked/F/T-REC-F.82-199110-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c03fdf9fd74ae13b4d3f7ba7d558496326ba4939 --- /dev/null +++ b/marked/F/T-REC-F.82-199110-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d76546b1d0e88d9d68bb3122fbbe9be8649b0ec3379c023a886f06461b357456 +size 5796 diff --git a/marked/F/T-REC-F.82-199110-I_PDF-E/raw.md b/marked/F/T-REC-F.82-199110-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..dcaef5859bd02fff90546db49f6ed521dd62932c --- /dev/null +++ b/marked/F/T-REC-F.82-199110-I_PDF-E/raw.md @@ -0,0 +1,99 @@ + + +![ITU logo: A globe with the letters ITU and a lightning bolt.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with the letters ITU and a lightning bolt. + +INTERNATIONAL TELECOMMUNICATION UNION + +**CCITT** + +**F.82** + +THE INTERNATIONAL +TELEGRAPH AND TELEPHONE +CONSULTATIVE COMMITTEE + +**TELEGRAPH AND MOBILE SERVICES +OPERATIONS AND QUALITY OF SERVICE** + +--- + +**OPERATIONAL PROVISIONS TO +PERMIT INTERWORKING BETWEEN +THE INTERNATIONAL TELEX SERVICE +AND THE INTEX SERVICE** + +**Recommendation F.82** + +--- + +![CCITT logo: A globe with the letters CCITT.](6ed175c791b5e156d9c98a8dbcc3318c_img.jpg) + +CCITT logo: A globe with the letters CCITT. + +Geneva, 1991 + +## FOREWORD + +The CCITT (the International Telegraph and Telephone Consultative Committee) is a permanent organ of the International Telecommunication Union (ITU). CCITT is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The Plenary Assembly of CCITT which meets every four years, establishes the topics for study and approves Recommendations prepared by its Study Groups. The approval of Recommendations by the members of CCITT between Plenary Assemblies is covered by the procedure laid down in CCITT Resolution No. 2 (Melbourne, 1988). + +Recommendation F.82 was prepared by Study Group I and was approved under the Resolution No. 2 procedure on the 11th of October 1991. + +## --- CCITT NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication Administration and a recognized private operating agency. + +## **Recommendation F.82** + +# **OPERATIONAL PROVISIONS TO PERMIT INTERWORKING BETWEEN THE INTERNATIONAL TELEX SERVICE AND THE INTEX1) SERVICE** + +## **1 Introduction** + +1.1 With the development of the Intex service there may be benefits to both customers and Administrations if an interworking capability with the international telex service is provided. + +The service and operational provisions of the Intex service are described in Recommendation F.150. + +## **2 Scope** + +2.1 This Recommendation sets out service and operational principles to permit interworking between terminals of the international telex service and Intex terminals. + +## **3 Principles of interworking** + +3.1 These principles will provide for, but will not be limited to, real-time communication between terminal equipment. + +3.2 The Quality of Service of the international telex service must not be degraded by any such interworking, nor will there be any burden placed on the international telex service. + +3.3 There will be no requirement for the calling party to be aware of the characteristics of the called party for the purposes of call establishment and information transfer. + +3.4 The format and the content of the message as printed at the transmitting and receiving terminal should be the same. + +3.5 The international destination codes will be in accordance with Recommendation F.69. Further discrimination within national networks will be at the discretion of individual Administrations. + +## **4 Operation** + +### **4.1 *Telex to Intex direction*** + +4.1.1 Calls should be established and cleared using normal telex call set-up and clearing procedures. + +4.1.2 The service and operational provisions of Recommendation F.60 shall apply. + +4.1.3 The answerback returned to the calling telex subscriber at call establishment or in response to a WRU signal during the text transfer stage shall be in accordance with the answerback requirements as specified in Recommendation F.60. + +### **4.2 *Intex to telex direction*** + +4.2.1 Calls should be established and cleared using normal Intex call set-up and clearing procedures. + +--- + +1) The use of the name Intex is conditional on it being freely available in Member countries and hence the name of the service offered may vary from country to country. + +4.2.2 When the called party is a terminal of the international telex service, the service identifier **TLX** will be indicated at the Intex terminal. + +When the call is established in manual mode, the Intex operator should be informed if a previously prepared message contains non-ITA2 characters. The procedures to be applied for automatic call set-up in the case where the Intex operator is not aware of the characteristics of the called party terminal are for further study. + +4.2.3 The received answerback shall be in accordance with Recommendation F.60. There are networks, however, which do not implement answerback composition in accordance with Recommendation F.60 and the interworking provision should take this into account. + +4.2.4 Speed and code conversion may require flow control. Any flow control requirement will be the responsibility of the Intex service. \ No newline at end of file diff --git a/marked/F/T-REC-F.83-199007-I_PDF-E/raw.md b/marked/F/T-REC-F.83-199007-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..572eaddab31e61757b301d25dec118b4bf176abc --- /dev/null +++ b/marked/F/T-REC-F.83-199007-I_PDF-E/raw.md @@ -0,0 +1,254 @@ + + +![ITU logo: A globe with the letters ITU and a lightning bolt.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with the letters ITU and a lightning bolt. + +INTERNATIONAL TELECOMMUNICATION UNION + +**CCITT** + +**F.73** + +THE INTERNATIONAL +TELEGRAPH AND TELEPHONE +CONSULTATIVE COMMITTEE + +**TELEGRAPH AND MOBILE SERVICES +OPERATIONS AND QUALITY OF SERVICE** + +--- + +**OPERATIONAL PRINCIPLES FOR +COMMUNICATION BETWEEN TERMINALS OF +THE INTERNATIONAL TELEX SERVICE AND +DATA TERMINAL EQUIPMENT ON PACKET +SWITCHED PUBLIC DATA NETWORKS** + +**Recommendation F.73** + +--- + +![CCITT logo: A globe with the letters CCITT.](6ed175c791b5e156d9c98a8dbcc3318c_img.jpg) + +CCITT logo: A globe with the letters CCITT. + +Geneva, 1990 + +# **FOREWORD** + +The CCITT (the International Telegraph and Telephone Consultative Committee) is a permanent organ of the International Telecommunication Union (ITU). CCITT is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The Plenary Assembly of CCITT which meets every four years, established the topics for study and approves Recommendations prepared by its Study Groups. The approval of Recommendations by the members of CCITT between Plenary Assemblies is covered by the procedure laid down in CCITT Resolution No. 2 (Melbourne, 1988). + +Recommendation F.73 was prepared by Study Group I and was approved under the Resolution No. 2 procedure on the 2nd July 1990. + +## --- **CCITT NOTE** + +In this Recommendation, the expression "Administration" is used for shortness to indicate both a telecommunication Administration and a recognized private operating agency. + +## **Recommendation F.73** + +# **OPERATIONAL PRINCIPLES FOR COMMUNICATION BETWEEN TERMINALS OF THE INTERNATIONAL TELEX SERVICE AND DATA TERMINAL EQUIPMENT ON PACKET SWITCHED PUBLIC DATA NETWORKS** + +The CCITT, + +*considering* + +(a) the need to allow communication between terminals of the international telex service with terminals on packet switched public data networks; + +(b) that Recommendations F.60, F.69 and other relevant Recommendations define the international telex service; + +(c) that Recommendation X.121 defines the international numbering plan for public data networks; + +(d) that the PAD functions of the telex/packet interworking functions (TPIWF) are in conformance with Recommendations X.3, X.28 and X.29, + +*unanimously declares the view* + +1 that there are benefits in standardizing the operational principles for a terminal of the international telex service to communicate, across international boundaries, with a data terminal equipment (DTE) on a packet switched public data network (PSPDN); + +2 that where provided, the operational principles to achieve communication should be in accordance with this Recommendation. + +## **1 Introduction** + +1.1 The procedures defined in this Recommendation enable telex subscribers to communicate with both packet mode and character mode data terminal equipment (DTE) connected to the PSPDN directly or via other networks. In the reverse direction, the provisions of this Recommendation enable users of packet mode and character mode DTEs, connected to a PSPDN directly or via other networks, to communicate with telex subscribers. + +1.2 This Recommendation applies to user classes 8-13 and 20-23 as defined in Recommendation X.1. Categories of access for DTEs accessing the PSPDN are shown in Recommendation X.10. + +1.3 This Recommendation does not apply to other Telematic services that may be supported by packet switched public data networks and interworking with the telex service. + +For example, interworking between the telex service and the teletex service or the interpersonal messaging service is not within the scope of this Recommendation. Such interworking scenarios are defined in other Recommendations. + +1.4 Call establishment from a telex terminal via a PSPDN to a character-mode DTE connected to the public switched telephone network (PSTN) requires further study. + +## **2 Operational outline** + +2.1 Communication shall be in quasi real-time mode and support interactive operation. Delays may be encountered as defined in § 4.1.2. + +2.2 The interworking shall be established by the provision of a telex/packet interworking function (TPIWF). + +Interworking on an international connection should be via the telex network as shown in Figure 1/F.73. + +![Diagram of the interworking model (Figure 1/F.73) showing connections between Country A and Country B across an international boundary.](9e6062272bbe3ddbb7c0606721d64cf0_img.jpg) + +The diagram illustrates the interworking model across an international boundary between Country A and Country B. In Country A, a 'Telex terminal' is connected to a 'Telex network'. This network connects to another 'Telex network' in Country B across the 'International boundary'. A 'Telex service' is indicated by a double-headed arrow spanning both countries. In Country B, the second 'Telex network' connects to a 'TPIWF' (Telex/Packet Interworking Function) at a 'Network boundary'. The 'TPIWF' is connected to a 'PSPDN' (Packet Switched Data Network). The 'PSPDN' is further connected to an 'Other network', which in turn connects to three types of Data Terminal Equipment (DTE): 'X.25 DTE', 'X.28 DTE', and 'X.25 or X.32 DTE'. A 'Telex terminal' is also shown connected to the second 'Telex network' in Country B. The diagram is labeled 'T0100551-90' at the bottom right. + +Diagram of the interworking model (Figure 1/F.73) showing connections between Country A and Country B across an international boundary. + +FIGURE 1/F.73 +**Interworking model** + +2.3 The point of interworking between the two networks shall be in the same country as the PSPDN. + +2.4 In the telex to PSPDN direction, an Administration may implement either/or both one-stage and two-stage call set-up procedures. + +Where the DTE is assigned a telex number, or its address is represented as part of the national telex numbering plan of the destination country, one-stage selection may be used. + +In all other cases two-stage selection shall be used. + +## **3 Call set-up procedures** + +### *3.1 Telex to PSPDN direction* + +#### *3.1.1 One-stage selection* + +3.1.1.1 The length of the number assigned to the DTE shall be in accordance with the relevant U-Series signalling Recommendations. + +3.1.1.2 The procedures for selection within the PSPDN are a national matter and not covered by this Recommendation. + +3.1.1.3 The call to the TPIWF shall be established using normal telex procedures. The procedures for call establishment within the PSPDN are defined in the relevant X-Series Recommendations. + +3.1.1.4 The number assigned to a user in the TPIWF must appear to be part of the national telex numbering plan. The method of verification is a national matter. If the verification fails, the following procedures should be applied: + +- Where the TPIWF is provided by the Administration which also provides all or part of the telex network, the service signal **NP** may be returned. +- Where the TPIWF is not provided by the Administration which also provides all or part of the telex network, the procedures to be applied shall be in accordance with Recommendation F.74. + +3.1.1.5 The answerback returned by the TPIWF to the calling telex terminal at call establishment and during the text transfer phase shall be in accordance with § 4.3.1.1. The answerback shall be returned in accordance with Recommendation S.6. + +#### 3.1.2 Two-stage selection + +3.1.2.1 In the case of two-stage selection, a national telex number should be assigned to the telex/packet interworking function (TPIWF), and the DTE X.121 address should be input in a second stage of selection. + +3.1.2.2 Connection to the TPIWF shall be established using normal telex procedures. + +3.1.2.3 During the first stage of telex call establishment and until the call connected packet is received, the answerback returned in response to the WRU signal shall be the answerback of the TPIWF. + +3.1.2.4 The format of the TPIWF answerback shall be in accordance with Figure 2/F.73 and Recommendation S.6. + +![](d980a3f9608055996a07f31788baf827_img.jpg) + +| | | | | | | | +|-----------|-----------------------------|-------|---------------|---|---|---| +| ↑   ←   ≡ | TPIWF national telex number | ↓   → | Mnemonic part | → | Ø | ↓ | +|-----------|-----------------------------|-------|---------------|---|---|---| + +T0100561-90 + +| | | | | +|---|-----------------|---|--------------------------------------------------------------------------| +| ↑ | Figure shift | ↓ | Letter shift | +| ← | Carriage return | → | Space | +| ≡ | Line feed | Ø | Telex network identification code in accordance with Recommendation F.69 | + +FIGURE 2/F.73 + +3.1.2.5 After the answerback exchange the telex subscriber shall input the DTE address preferably followed by the character + as an end-of-selection signal. Alternatively, other end-of-selection signals, for example, carriage-return, linefeed, may be used by bilateral agreement. + +3.1.2.6 Having received the address of the requested DTE, the TPIWF shall establish the connection across the PSPDN in accordance with national procedures. On call establishment, the TPIWF receives the call connected PAD service signal. This PAD service signal will, whenever possible, include the identification of the called DTE. + +3.1.2.7 The procedures for call establishment in the PSPDN are a national matter and not covered by this Recommendation. + +3.1.2.8 During the PSPDN call establishment phase, the TPIWF may clear the telex call if the connection through the PSPDN is impossible to establish. Alternatively, the TPIWF may choose not to clear the telex call but to allow the telex user to set up another PSPDN call, possibly with a different DTE. The choice between the two alternatives is a network option. + +3.1.2.9 The provision for telex automatic emitting devices (TAEDs) is for further study. + +### 3.2 *PSPDN to telex direction* + +3.2.1 Selection procedures from the PSPDN DTE to the TPIWF are a national matter. The TPIWF should establish the telex call using normal telex procedures with telex selection information provided by the calling DTE. + +3.2.2 Where an Administration provides one-stage selection in accordance with § 3.1.1 of this Recommendation, only DTEs assigned a telex number are permitted to establish a telex call. The method of verification is a national matter. + +3.2.3 The TPIWF may store the identification of the calling DTE for the duration of the call to generate an answerback if requested by the called telex terminal. The format of the answerback is defined in § 4.3.1.1 or Figure 3/F.73, as appropriate. + +![Diagram of a telex terminal display showing a recall address. The display is divided into two sections. The left section contains three icons: an upward-pointing arrow, a left-pointing arrow, and a horizontal line with three dots. The right section contains the text 'Recall address'.](573c8f9e728e343b6864e1070c14b8cf_img.jpg) + +| | | +|-------|----------------| +| ↑ ← ≡ | Recall address | +|-------|----------------| + +Diagram of a telex terminal display showing a recall address. The display is divided into two sections. The left section contains three icons: an upward-pointing arrow, a left-pointing arrow, and a horizontal line with three dots. The right section contains the text 'Recall address'. + +T0100571-90 + +*Note* – For legends see Figure 2/F.73. + +FIGURE 3/F.73 + +Alternatively, as a network option, the TPIWF may respond to a WRU signal received from the telex network with the answerback of the TPIWF formatted in accordance with Figure 2/F.73. + +3.2.4 After a successful call establishment to the telex terminal, the TPIWF should indicate call connect to the PSPDN. + +3.2.5 If a call is unsuccessful, the TPIWF shall clear the call to the DTE with an appropriate cause code reflecting the received telex service signal. The appropriate cause code is selected from Recommendation X.96. + +3.2.6 The handling of the answerback of the called telex subscriber, automatically returned by the telex network to the TPIWF, is a national matter. + +3.2.7 Upon receipt of the answerback of the called telex terminal, the TPIWF should transmit the answerback/DTE identification of the calling DTE, or TPIWF answerback, as appropriate, to the called telex terminal. + +## 4 **Text transfer phase** + +The provisions for telex automatic emitting devices (TEADs) in the two-stage selection case is for further study. + +### 4.1 *Telex to packet* + +4.1.1 Characters received from the telex network shall be converted in accordance with Recommendation S.18, packetized by the TPIWF, and forwarded to the PSPDN in accordance with Recommendation X.28. + +4.1.2 When flow control prevents the forwarding of further data packets, the TPIWF may attempt to flow-control the calling telex terminal in accordance with Recommendation S.4. If this is unsuccessful, the telex call should be cleared. + +### 4.2 *Packet to telex* + +4.2.1 The user data received from the DTE by the TPIWF shall be transmitted to the telex subscriber. + +4.2.2 The TPIWF shall convert the IA5 characters to ITA2 characters, in accordance with Recommendation S.18, and transmit them to the telex network. The conversion from other character sets to ITA2 is a national matter. The appropriate PAD parameters of the TPIWF shall be set to avoid the occurrence of overprinting at the telex terminal. + +4.2.3 The procedures to be followed by the TPIWF when signals are received on the backward path during forward transmission to the telex terminal shall be in accordance with Recommendation S.4. + +### 4.3 *Answerback formats and WRU processing* + +#### 4.3.1 *Answerback formats* + +4.3.1.1 The DTE answerback format in case of one-stage selection shall be in accordance with Figure 1/F.74. + +4.3.1.2 In the case of two-stage selection, the format of the answerback of the TPIWF should be in accordance with Figure 2/F.73 and the DTE identification, where provided, will be the appropriate recall address. + +In the telex-to-PSPDN direction, it should be noted that, where the DTE identification is provided, the answerback returned by the TPIWF in the data transfer phase differs from the answerback returned during the call establishment phase and is not, therefore, in conformance with Recommendation F.60. + +#### 4.3.2 *WRU processing* + +4.3.2.1 If a WRU signal is received from the telex terminal during the data transfer phase, the TPIWF shall return the answerback as defined in either §§ 3.1.1.5, 3.1.2.3 or 3.2.3, as appropriate to the telex terminal. + +This answerback shall be returned only when all outstanding data has been transmitted to the PSPDN. + +4.3.2.2 The DTE may verify connection to the correct telex terminal by use of the IA5 "ENQ" character as part of a data packet. This should be converted to the ITA2 WRU signal, and transmitted to the telex terminal to trigger the answerback. + +The TPIWF shall forward all outstanding data to the telex terminal before transmission of the WRU signal. The first 20 characters received from the telex subscriber after transmission of the WRU signal should be considered to be the answerback, which should then be returned to the DTE. + +4.3.2.3 The TPIWF should transmit the answerback to the DTE immediately after its reception. + +If no character is received within 2 seconds following transmission of the WRU signal, the TPIWF should continue with text transmission. + +4.3.2.4 The responsibility for the action to be taken where an answerback is not returned in response to the IA5 "ENQ" character from the DTE rests with the DTE. + +4.3.2.5 The DTE on the PSPDN may also cause the TPIWF to send its answerback to the telex network by sending an IA5 "ACK" character. The answerback should not be forwarded until all outstanding data packets have been transmitted to the telex terminal. + +### 4.4 *Call clearing* + +4.4.1 If the telex call is cleared, the TPIWF shall clear the PSPDN call. + +4.4.2 If the PSPDN call is cleared, the TPIWF shall clear the telex call where the DTE was the originator of the call. However, in the telex-to-PSPDN direction, the TPIWF may choose not to clear the telex call where the TPIWF offers a follow-on call facility and it has been requested by the originating telex subscriber in the call set-up. + +*Note* — Follow-on calls are only allowed on telex-to-PSPDN calls. + +## 5 **Abnormal conditions** + +The action to be taken when abnormal conditions occur shall be in accordance with the relevant U-Series Recommendations on the telex side of the TPIWF and with the X-Series Recommendations on the PSPDN side of the TPIWF. \ No newline at end of file diff --git a/marked/F/T-REC-F.87-199103-I_PDF-E/2763901b7a1fd1b5d704cdc450d12ed0_img.jpg b/marked/F/T-REC-F.87-199103-I_PDF-E/2763901b7a1fd1b5d704cdc450d12ed0_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..e6cec042ca9189647dccf857a97fecf31b2b40e3 --- /dev/null +++ b/marked/F/T-REC-F.87-199103-I_PDF-E/2763901b7a1fd1b5d704cdc450d12ed0_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ddbdabf3f835c5f487986597778875f72a6589d0de7768925ad9b1169a13fec0 +size 32165 diff --git a/marked/F/T-REC-F.87-199103-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/F/T-REC-F.87-199103-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9ff35fb80886d11ede195ab5a46c22ce14710d49 --- /dev/null +++ b/marked/F/T-REC-F.87-199103-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:30287180fe1dd821f9408c4cc58b365a422436acfaa9134dcb98f9f1fe8a50fe +size 7439 diff --git a/marked/F/T-REC-F.87-199103-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg b/marked/F/T-REC-F.87-199103-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d86cfe4b80250eb67356d96e2b1f4b36a174e5c1 --- /dev/null +++ b/marked/F/T-REC-F.87-199103-I_PDF-E/6ed175c791b5e156d9c98a8dbcc3318c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9fefac5daf3718de304d029d011812fda0ca9277aa8635888355267a232173f6 +size 5901 diff --git a/marked/F/T-REC-F.87-199103-I_PDF-E/740442c999390734911677f01af0316d_img.jpg b/marked/F/T-REC-F.87-199103-I_PDF-E/740442c999390734911677f01af0316d_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..80e856c073240382c4e3eac2c0dfd7f7abc88578 --- /dev/null +++ b/marked/F/T-REC-F.87-199103-I_PDF-E/740442c999390734911677f01af0316d_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c8de8dc39def108c04e028e9a0bfb3c2c05a0a82d70d58c1100b4caa5187cd4f +size 11650 diff --git a/marked/F/T-REC-F.87-199103-I_PDF-E/d3ca266c298aeb34b019960c6c36f187_img.jpg b/marked/F/T-REC-F.87-199103-I_PDF-E/d3ca266c298aeb34b019960c6c36f187_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..81600047de3b5c0e37e2e93dad91b94e4b708d90 --- /dev/null +++ b/marked/F/T-REC-F.87-199103-I_PDF-E/d3ca266c298aeb34b019960c6c36f187_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:69fad6d64101897c5cfdaca15c7d7358c2c221c798b34f1e6db5fb0fb39858e9 +size 31492 diff --git a/marked/F/T-REC-F.87-199103-I_PDF-E/raw.md b/marked/F/T-REC-F.87-199103-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..657bfec37d60b8809b7b17cc93a48ed08f17c4b5 --- /dev/null +++ b/marked/F/T-REC-F.87-199103-I_PDF-E/raw.md @@ -0,0 +1,411 @@ + + +![ITU logo: A globe with the letters ITU and a lightning bolt.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with the letters ITU and a lightning bolt. + +INTERNATIONAL TELECOMMUNICATION UNION + +**CCITT** + +**F.87** + +THE INTERNATIONAL +TELEGRAPH AND TELEPHONE +CONSULTATIVE COMMITTEE + +**TELEGRAPH AND MOBILE SERVICES +OPERATIONS AND QUALITY OF SERVICE** + +--- + +**OPERATIONAL PRINCIPLES FOR THE +TRANSFER OF MESSAGES FROM +TERMINALS ON THE TELEX NETWORK +TO GROUP 3 FACSIMILE TERMINALS +CONNECTED TO THE PUBLIC SWITCHED +TELEPHONE NETWORK** + +**Recommendation F.87** + +--- + +![CCITT logo: A globe with the letters CCITT.](6ed175c791b5e156d9c98a8dbcc3318c_img.jpg) + +CCITT logo: A globe with the letters CCITT. + +Geneva, 1991 + +# FOREWORD + +The CCITT (the International Telegraph and Telephone Consultative Committee) is a permanent organ of the International Telecommunication Union (ITU). CCITT is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The Plenary Assembly of CCITT which meets every four years, establishes the topics for study and approves Recommendations prepared by its Study Groups. The approval of Recommendations by the members of CCITT between Plenary Assemblies is covered by the procedure laid down in CCITT Resolution No. 2 (Melbourne, 1988). + +Recommendation F.87 was prepared by Study Group I and was approved under the Resolution No. 2 procedure on the 11 of March 1991. + +## --- CCITT NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication Administration and a recognized private operating agency. + +## **Recommendation F.87** + +# **OPERATIONAL PRINCIPLES FOR THE TRANSFER OF MESSAGES FROM TERMINALS ON THE TELEX NETWORK TO GROUP 3 FACSIMILE TERMINALS CONNECTED TO THE PUBLIC SWITCHED TELEPHONE NETWORK** + +The CCITT, + +*considering* + +- (a) the need to allow the transfer of messages from terminals of the international telex service to Group 3 facsimile terminals on the public switched telephone network; +- (b) that Recommendations F.60, F.69 and other relevant Recommendations govern operation of the international telex service; +- (c) that Recommendation E.163 defines the numbering plan for the international telephone service; +- (d) that Recommendation F.72 defines the general principles and operational aspects of international telex store and forward; +- (e) that Recommendations F.160 and F.180 define the general operational provisions for the international public facsimile services; +- (f) that Recommendation T.4 defines the standardization of Group 3 facsimile apparatus; +- (g) that Recommendation T.30 defines the procedure for document facsimile transmission in the public switched telephone network; +- (h) that technical aspects of the international telex service are described in R-, S- and U-Series Recommendations, + +*unanimously declares the view* + +- (1) that there are benefits in standardizing the operational procedures for a terminal of the international telex service to transfer a message, across international boundaries, to a Group 3 facsimile terminal on the public switched telephone network; +- (2) that where this message transfer service is provided, the preferred operational procedures are laid down in this Recommendation. + +## **1. Scope** + +- 1.1 The procedures defined in this Recommendation enable telex subscribers to transfer messages to Group 3 facsimile terminals connected to the PSTN, on a store and forward basis. +- 1.2 The interworking shall be established by provision of a telex/facsimile interworking function (FAXIWF). +- 1.3 International access to the FAXIWF will normally be provided via the telex network (see Figure 1/F.87). Bilateral agreement may be required in the event that other than normal telex service agreements are required. +- 1.4 For telex subscribers accessing a FAXIWF in their own country, national or international destinations may be specified, with international interworking being implemented via the PSTN. See Figure 2/F.87. + +![Diagram of network configuration for interworking between terminals of the international telex service and Group 3 facsimile terminals on the PSTN. The diagram shows a 'Terminal' connected to a 'Telex network' in 'Country A'. This network connects across an 'International boundary' to another 'Telex network', which then connects to a 'FAXIWF' node. The 'FAXIWF' node is connected to a 'PSTN' in 'Country B', which in turn connects to a 'G3 Fax terminal'. A 'Service boundary' is indicated between the second 'Telex network' and the 'FAXIWF' node. An arrow labeled 'Direction of communication' points from left to right. The reference code 'T0101150-91' is in the bottom right corner.](2763901b7a1fd1b5d704cdc450d12ed0_img.jpg) + +Diagram of network configuration for interworking between terminals of the international telex service and Group 3 facsimile terminals on the PSTN. The diagram shows a 'Terminal' connected to a 'Telex network' in 'Country A'. This network connects across an 'International boundary' to another 'Telex network', which then connects to a 'FAXIWF' node. The 'FAXIWF' node is connected to a 'PSTN' in 'Country B', which in turn connects to a 'G3 Fax terminal'. A 'Service boundary' is indicated between the second 'Telex network' and the 'FAXIWF' node. An arrow labeled 'Direction of communication' points from left to right. The reference code 'T0101150-91' is in the bottom right corner. + +FIGURE 1/F.87 + +General network configuration for interworking between terminals of the international telex service and Group 3 facsimile terminals on the PSTN + +![Diagram of network configuration for interworking between terminals of the international telex service and Group 3 facsimile terminals via the international telephone network. The diagram shows a 'Terminal' connected to a 'Telex network' in 'Country A', which connects to a 'FAXIWF' node. The 'FAXIWF' node is connected to a 'PSTN' in 'Country A'. This 'PSTN' connects across an 'International boundary' to another 'PSTN' in 'Country B', which then connects to a 'G3 Fax terminal'. A 'Service boundary' is indicated between the 'Telex network' and the 'FAXIWF' node. An arrow labeled 'Direction of communication' points from left to right. The reference code 'T0101160-91' is in the bottom right corner.](d3ca266c298aeb34b019960c6c36f187_img.jpg) + +Diagram of network configuration for interworking between terminals of the international telex service and Group 3 facsimile terminals via the international telephone network. The diagram shows a 'Terminal' connected to a 'Telex network' in 'Country A', which connects to a 'FAXIWF' node. The 'FAXIWF' node is connected to a 'PSTN' in 'Country A'. This 'PSTN' connects across an 'International boundary' to another 'PSTN' in 'Country B', which then connects to a 'G3 Fax terminal'. A 'Service boundary' is indicated between the 'Telex network' and the 'FAXIWF' node. An arrow labeled 'Direction of communication' points from left to right. The reference code 'T0101160-91' is in the bottom right corner. + +FIGURE 2/F.87 + +Network configuration for interworking between terminals of the international telex service and Group 3 facsimile terminals via the international telephone network + +1.5 The Administration providing the FAXIWF shall be responsible for any necessary barring of international access. + +## **2. Service principles** + +2.1 The basic service offered is: + +- acceptance of messages from the telex network, with accompanying fax (PSTN) addressing information; + +- delivery of these messages to the requested facsimile terminal, with retries if necessary; +- notification message sent to the originator when the message delivery attempt fails and, optionally, when it succeeds. + +2.2 A single-stage selection procedure may be used employing normal telex procedures where the requested facsimile machine is addressed using a telex number. + +2.3 A two-stage selection procedure may be used whereby a call from the telex network gains access to the FAXIWF in the first stage of selection and inputs the called address(es) and message(s) in the second stage of selection, after the return of a call connect signal. To the extent possible, this procedure has been aligned with that for telex store-and-forward facilities defined in Recommendation F.72. + +2.4 Message input from both manual and telex automatic emitting devices should be possible. + +2.5 Messages deposited into the FAXIWF will be sub-divided by the FAXIWF into “pages” which serve as physical “A4” size pages as received at the fax terminal. Otherwise, the input format shall be preserved. + +A page should normally comprise 55 lines of text (of 4.23 mm pitch), or part thereof for pages with fewer than 55 text lines. Upon transmission, it is preferred to have both message reference and pagination information inserted in a header line on each page. + +2.6 In addition to delivering the message as pages of text to the specified fax terminal(s), the FAXIWF may optionally prepare and transmit a “header page” consisting of at least: + +- message reference information (which may include the legal time and date of acceptance); +- calling party identification (telex answerback and recall information, if necessary); +- indication of number of pages in message; +- specified facsimile terminal (PSTN) address; +- optional customer attention information field (see § 3.2.4). + +2.7 Where a header page is not transmitted, the FAXIWF should attach a herald to the beginning of the message indicating the calling party identification (telex answerback and recall information, if necessary). + +2.8 The FAXIWF will attempt to deliver the message as soon as operationally feasible. Deferred delivery may also be provided. + +## 3 Service facilities + +In the following section, the symbol \* indicates that the facility is not normally applicable for one-stage selection. + +### 3.1 General + +In addition to message deposit, the following facilities may be made available: + +- multi-address\*; +- follow-on messages\*; +- positive delivery notification; +- status enquiry request; +- message cancel request\*. + +*Note* – The multi-address facility would not normally be available to incoming international telex calls except by bilateral agreement. + +### 3.2 Message deposit + +3.2.1 Validation of format and/or length should preferably be performed by the FAXIWF on the address(es) of the called fax terminal(s) as specified by the calling party. + +3.2.2 Where the multi-address facility is provided, the FAXIWF should be capable of accepting at least 20 facsimile terminal (PSTN) addresses in a multi-address call. + +3.2.3 An attention information field facility should be provided by the FAXIWF to enable each addressee of a multi-address message to have a relevant attention prefix preceding the message in the header page where such is provided. + +3.2.4 A message reference number shall be available to the calling party for every message accepted by the FAXIWF. For multi-address messages, a separate message reference number may be provided for each address. + +3.2.5 The application of short-code address\* and pre-recorded address lists is a national matter. + +### 3.3 *Follow-on messages* + +Provision may be made to accept follow-on message(s) with their associated address(es) which may be sent immediately after the deposit of a previous message while still retaining the connection to the FAXIWF from the originator. The procedures should be aligned with Recommendation U.43. + +### 3.4 *Status enquiry request* + +3.4.1 Status information on previously-deposited message shall only be returned to the message originator. The procedure used should be generally aligned with Recommendations F.72 and U.80. + +3.4.2 This status information should be available for at least 24 hours after the time of message receipt by the FAXIWF. + +### 3.5 *Message cancel request\** + +If message cancellation is provided, previously deposited messages may only be cancelled by the message originator. + +### 3.6 *Delivery/non-delivery notification message and journals* + +3.6.1 In the case of failure to deliver a message to the requested fax terminal, the FAXIWF shall if possible return a non-delivery notification message to the originator (see § 7.1). + +3.6.2 An Administration may optionally provide a positive delivery notification message to the originator upon successful delivery of the message to the requested fax terminal (see § 7.2). + +3.6.3 The above notifications may also be provided in a summarized form (journals). + +## 4 Access procedures + +### 4.1 *General* + +Two basic procedures for access from a telex terminal should be provided by the FAXIWF: + +#### a) *Interactive operation:* + +- input from manual calling terminals where the TPIWF may return prompt signals; + +#### b) *Non-interactive operation:* + +- input from telex automatic emitting devices where prompt signals may not be required; +- interworking with other types of interworking and conversion facilities. + +### 4.2 *One-stage selection* + +In the one-stage selection procedure, the facsimile terminal is assigned a telex number that is part of the national telex numbering plan. + +#### 4.2.1 *Call establishment* + +4.2.1.1 The originator (or intermediary interworking or conversion facility) shall use normal telex procedures. + +4.2.1.2 The telex number received by the FAXIWF shall be verified as being proper to a registered facsimile terminal. The method of effecting this verification is a national matter. If the verification fails, the following procedures should be applied: + +- where the FAXIWF is provided by the Administration which also provides all or part of the telex network, the service signal NP may be returned; +- where the FAXIWF is not provided by the Administration which also provides all or part of the telex network, the procedures to be applied shall be in accordance with Recommendation F.74. + +4.2.1.3 Having verified the received telex number, the FAXIWF shall return the answerback allocated to the requested facsimile terminal to the originator. + +4.2.1.4 This answerback of the facsimile terminal shall be formatted in accordance with Recommendation F.74. + +4.2.1.5 In accordance with Recommendation S.23, the FAXIWF may transmit a WRU signal to the originating telex subscriber if the backward path remains idle for at least 800 milliseconds. + +4.2.1.6 Having received the answerback allocated to the called facsimile terminal, the originator may input the text of his message. The FAXIWF shall respond to any received WRU signal during this phase of the call by returning the answerback allocated to the facsimile terminal. + +4.2.1.7 The reaction to abnormal conditions during call set-up shall be in accordance with normal telex procedures (see Recommendation U.2071) ). + +### 4.3 *Two-stage selection* + +#### 4.3.1 *General* + +4.3.1.1 Calls to the FAXIWF will be established by means of normal telex procedures. + +4.3.1.2 The received telex answerback shall be processed in accordance with Recommendation U.74 to determine the calling telex address. If the answerback is deemed to be non-processable, the FAXIWF shall request input of the calling address by the telex subscriber by returning the ADD prompt. + +4.3.1.3 Transmission from an automatic terminal is indicated by commencing the procedure with the non-interactive service request (characters CI). The calling terminal is recognized as interactive by the omission of the non-interactive service request (CI). + +4.3.1.4 If the calling telex address is capable of being extracted from the received answerback, the FAXIWF shall return GA, unless a non-interactive service request has been made. + +#### 4.3.2 *Input of address(es)* + +4.3.2.1 For each address required, the originator should transmit the sequence + +FAX
+ () + +where: + +FAX is the service identifier. + +4.3.2.2 The address field
is the only mandatory field of the address line and is the called PSTN fax address formatted in accordance with Recommendation E.163. The address field consists of the international PSTN number of the required fax terminal. The end of the address field should be delimited by the end of address field signal (+), for example, + +P43150145207+ + +--- + +1) Recommendation U.207 has not been officially approved as of the publication date of this Recommendation F.87. + +The attention information field is a 30-character field, delimited by parentheses, for example: + +(ATTN: MR.J.SMITH) + +The inclusion of the field is optional. If specified, the field will be inserted in the header page (see § 3.2.3), otherwise that part of the header page will be left blank. + +An example of a complete address line would be as follows: + +FAX 43150145207+(ATTN: MR.J.SMITH) + +4.3.2.3 The last address line shall be delimited from the message by the end of address sequence/begin text input sequence BT. + +4.3.2.4 It should be possible to mix telex, teletex and facsimile addresses in a multi-address call in accordance with Recommendation F.72. + +4.3.2.5 Once the maximum number of permissible addresses have been input, the FAXIWF shall return the message input prompt (GA) unless a non-interactive service request has been received. + +4.3.2.6 If an end-of-address signal (BT) is received without any valid addresses being received, the call will be cleared by the FAXIWF. + +4.3.2.7 The action to be taken when invalid addresses are received shall be in accordance with § 5.2.3. + +### 4.4 Answerbacks and WRU processing + +4.4.1 In the case of single-stage selection, the FAXIWF shall at all times respond to the reception of a WRU signal from the originator with the "answerback" allocated to the registered terminal, formatted in accordance with Recommendation F.74. + +4.4.2 In the case of two-stage selection, the FAXIWF shall at all times respond to a received WRU signal with its own answerback, consisting of 20 characters formatted in accordance with Figure 3/F.87 and Recommendation F.60 where $\Phi$ is the TNIC of the network to which the FAXIWF is connected. Neither the WRU signal nor the resultant answerback should be stored as part of the message. + +![Diagram of FAXIWF answerback format showing a sequence of symbols and text fields.](740442c999390734911677f01af0316d_img.jpg) + +The diagram illustrates the format of a FAXIWF answerback. It consists of a horizontal sequence of boxes and symbols. From left to right, the elements are: an upward arrow (↑), a leftward arrow (←), a horizontal line with three dots (≡), a box containing the text "FAXIWF national telex number", a downward arrow (↓), a rightward arrow (→), a box containing the text "Mnemonic part", a rightward arrow (→), a circle with a diagonal slash (∅), and a downward arrow (↓). + +Diagram of FAXIWF answerback format showing a sequence of symbols and text fields. + +T0100562-91 + +FIGURE 3/F.87 +Format of FAXIWF answerback + +## 5 Message deposit by the telex subscriber + +### 5.1 General + +5.1.1 All ITA2 spacing characters received during text input shall be stored by the FAXIWF for subsequent transmission to the addressed facsimile terminal. However, neither the WRU signal nor the bell signal, if received as part of the text, will be stored within the message. + +5.1.2 The FAXIWF shall be able to accept messages of at least 24 000 characters. + +5.1.3 If the calling subscriber wishes to cancel a message during the text input phase, the cancel operation signal to be used, where the FAXIWF offers this facility, is FFFF. + +If the FAXIWF detects a cancel operation signal, it should cease message acceptance, issue the message cancel confirmation signal ANUL ANUL and clear the call. + +5.1.4 In case of two-stage selection, if the FAXIWF detects a valid EOM or EOT signal, but no characters have been received as message text, the message will be abandoned and, if not already cleared by the originator, the call shall be cleared by the FAXIWF as in § 5.1.3. + +5.1.5 The reaction to abnormal conditions during message deposit shall be in accordance with Recommendation U.2071). + +### 5.2 *Message security* + +5.2.1 The FAXIWF should only accept messages for delivery to destination addresses accessible from that FAXIWF. + +5.2.2 In case of two-stage selection, the FAXIWF should not accept message input unless acceptable identification of the originator has been received at call set-up. + +5.2.3 The FAXIWF may verify the format of the called address(es). If this verification is unsuccessful for all addresses, the message shall be rejected and the service code ITR should be returned (see Recommendation U.80, § 4.6). However, a positive verification result does not guarantee that the message can be delivered to the given address. + +### 5.3 *End of transaction and end of message signal* + +5.3.1 At the end of each transaction, an end of transmission (EOT) signal is required. This signal is + + + +. However, for follow-on messages, an end of message (EOM) signal is required at the end of each individual message. This can be one of two types, as follows: + +- a) NNNN, which is simply used to separate messages; +- b) NNNNACK, which is used to separate messages and to request the FAXIWF for an input message acknowledgment (IMA) plus reference information of those messages not previously acknowledged. + +5.3.2 Where there is a stop in transmission for 30 seconds, and no EOM or EOT signal is detected, the service code GA will be sent to the originator. + +5.3.3 If the EOM/EOT signal is not received, or transmission does not resume within a further 30 seconds, the FAXIWF shall initiate the clear down procedure. + +5.3.4 In case of two-stage selection, a message received without any EOM or EOT shall nevertheless be delivered with the following text appended: **POSSIBLE INCOMPLETE MESSAGE**. + +### 5.4 *Input transaction accepted* + +The FAXIWF shall send an input transaction accepted code for delivery (ITD) notifying the originator that the message(s) has(have) been accepted and that delivery would be attempted. Delivery should be attempted even if the call is cleared before the ITD is sent. The ITD should be followed by the message reference(s) and, if applicable, number of messages. + +## 6 *Message delivery* + +6.1 Messages deposited in the FAXIWF with appropriate PSTN address(es) shall be delivered as soon as operationally feasible after receipt, unless deferred delivery option is requested. Delivery to the requested fax terminal(s) shall be in accordance with the normal procedures applicable to fax-fax communication (see Recommendation T.30). + +6.2 If the delivery attempt is unsuccessful, i.e. the complete message was not delivered, the FAXIWF may initiate re-attempts in accordance with national rules laid down for re-calling into the PSTN. + +--- + +1) Recommendation U.207 has not been officially approved as of the publication date of this Recommendation F.87. + +6.3 Should all delivery attempts fail, the FAXIWF shall return a non-delivery notification message to the originator in accordance with § 7.1. + +6.4 Should delivery be successfully completed, the FAXIWF may return a positive delivery notification message to the originator where this facility is offered by the Administration operating the FAXIWF. + +6.5 Where a message is only partially delivered, i.e., the connection across the PSTN was cleared for any reason before the EOM or EOT was reached, the FAXIWF shall re-establish the connection and re-send the message commencing after the last successfully delivered page. + +6.6 The reaction to abnormal conditions during message delivery shall be in accordance with the relevant E- and T-Series Recommendations. + +## 7. **Delivery/non-delivery notification messages** + +### 7.1 *Non-delivery notification message* + +7.1.1 The FAXIWF shall return a non-delivery notification message to the originator in respect of failure to deliver an accepted message in its entirety to any requested destination. + +7.1.2 The non-delivery notification message should be returned to the originating telex subscriber as soon as possible after expiry of the retry cycle (see § 6.2 above). + +7.1.3 It is recommended that the message contain at least the following information: + +- message reference; +- time/date of message acceptance by the FAXIWF; +- time/date of final delivery attempt; +- originating telex answerback; +- requested facsimile terminal number; +- called station identity (CSI), if available; +- number of delivery attempts made; +- number of pages transmitted. + +7.1.4 If the FAXIWF is unable to deliver the non-delivery notification message to the originating subscriber after a maximum of 8 attempts to do so, the message should be spilled out to an assistance position for manual processing. + +### 7.2 *Positive delivery notification message* + +7.2.1 An Administration may optionally offer this facility to its subscribers either on a global basis or by request by the calling subscriber at the addressing stage. + +7.2.2 The facility may be requested on a per-address basis by use of the ACK signal in the address line, as follows: + +FAX
+(optional information field),ACK + +for example: + +FAX 43150145207+(ATTN: MR.J.SMITH),ACK + +7.2.3 It is recommended that the positive delivery notification message, where provided, contain at least the following information: + +- message reference; +- time/date of message acceptance by the FAXIWF; +- time/date of final delivery attempt; +- originating telex answerback; +- requested facsimile terminal number; + +- called station identity (CSI), if available; +- number of pages in message; +- number of delivery attempts made; +- number of pages transmitted. + +7.2.4 If the FAXIWF is unable to deliver the positive delivery notification message to the originating subscriber after a maximum of 8 attempts to do so, the message should be spilled out to an assistance position for manual processing. + +## ANNEX 1 (to Recommendation F.87) + +### **List of abbreviations** + +| | | +|--------|--------------------------------------------| +| ACK | Request for positive delivery notification | +| ADD | Address prompt | +| BT | End-of-address/begin text signal | +| CI | Conversation impossible | +| CSI | Called station identity | +| EOM | End-of-message | +| EOT | End-of-transaction | +| FAXIWF | Facsimile interworking function | +| GA | Go ahead | +| IMA | Input message acknowledgment | +| ITD | Input transaction accepted for delivery | +| ITR | Input transaction rejected | +| PSTN | Public switched telephone network | +| TNIC | Telex network identification code | +| WRU | Who-are-you | \ No newline at end of file diff --git a/marked/F/T-REC-F.901-199303-I_PDF-E/raw.md b/marked/F/T-REC-F.901-199303-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..bf67dc69fbaaf419938abb28e39128d181223b6c --- /dev/null +++ b/marked/F/T-REC-F.901-199303-I_PDF-E/raw.md @@ -0,0 +1,242 @@ + + +![ITU logo: A globe with a lightning bolt and the letters ITU.](2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg) + +ITU logo: A globe with a lightning bolt and the letters ITU. + +INTERNATIONAL TELECOMMUNICATION UNION + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.901** + +(03/93) + +**OPERATIONS AND QUALITY OF SERVICE** + +**HUMAN FACTORS – GENERAL +(NOT SPECIFIC TO TELEPHONE SERVICE)** + +--- + +**USABILITY EVALUATION OF +TELECOMMUNICATION SERVICES** + +**ITU-T Recommendation F.901** + +(Previously "CCITT Recommendation") + +--- + +# FOREWORD + +The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of the International Telecommunication Union. The ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Conference (WTSC), which meets every four years, established the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics. + +ITU-T Recommendation F.901 was prepared by the ITU-T Study Group I (1988-1993) and was approved by the WTSC (Helsinki, March 1-12, 1993). + +## --- NOTES + +1 As a consequence of a reform process within the International Telecommunication Union (ITU), the CCITT ceased to exist as of 28 February 1993. In its place, the ITU Telecommunication Standardization Sector (ITU-T) was created as of 1 March 1993. Similarly, in this reform process, the CCIR and the IFRB have been replaced by the Radiocommunication Sector. + +In order not to delay publication of this Recommendation, no change has been made in the text to references containing the acronyms "CCITT, CCIR or IFRB" or their associated entities such as Plenary Assembly, Secretariat, etc. Future editions of this Recommendation will contain the proper terminology related to the new ITU structure. + +2 In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +# **USABILITY EVALUATION OF TELECOMMUNICATION SERVICES** + +*(Helsinki, 1993)* + +## **1 Introduction** + +This Recommendation defines guidelines for measuring the usability of human-to-system interfaces in telecommunications. + +Usability is widely recognized as one of the components of the Quality of Service. It plays a pivotal role in determining the success of new telecommunication services. + +## **2 Definitions** + +The concept of usability and its components has been defined within ISO and ETSI. A specific interpretation of this definition for the human-to-system interface is as follows: + +**usability:** A concept comprising the effectiveness, efficiency and satisfaction with which specified users can achieve specified system tasks in a particular environment; + +**effectiveness:** Measures of the accuracy and completeness of the system tasks achieved; + +**efficiency:** Measures of the accuracy and completeness of system tasks relative to the resources (e.g. time, human effort) used to achieve the specific system tasks; + +**satisfaction:** Measures of the comfort and acceptability of the system to its users and other people affected by its use. + +## **3 Evaluation aspects** + +The current literature and laboratory reports show that there is a world-wide commonality on the experimental procedures for usability evaluation. Both the techniques of the Protocol Analysis1) and controlled experiments have been used to evaluate the human-to-system interface. These techniques include many of the following considerations: + +- 1) decide which usability dimensions are to be measured; +- 2) explore methods by which these can be assessed within the system-specific framework; +- 3) set explicit, quantifiable usability criteria (i.e. reference values); +- 4) design experimental tasks by which to assess the degree to which usability criteria have been met by the system; +- 5) let a sample of experimental subjects complete the set tasks. The subjects can be tested one at a time or together, dependent on experimental design. Subjects should be given instructional material as demanded by the tasks; +- 6) analyse the data statistically; +- 7) modify those areas of the system in which the usability criteria were not met, relying on anecdotal observations of where and when subjects encountered cognitive stumbling blocks; + +--- + +1) Protocol Analysis: Standard evaluation method in cognitive psychology. + +- 8) re-test as under 5) above, using different subjects belonging to the same class but the same tasks as before in the modified system; +- 9) upon completion of the new experiment, analyse data: + - i) by themselves as before, using the same types of analyses, and + - ii) comparing data from both experiments; +- 10) iterate through steps 7) to 9) until usability criteria are clearly met by the system. + +### **3.1 Methods and measures** + +Since usability is a multi-disciplinary concept, no single, universally "correct" usability evaluation method exists. The choice of the most suitable methods for data capture among the many available (such as data log, observation, questionnaires, interviews and so forth) depends on the system under investigation and the usability dimensions deemed important for that system. + +In a public telecommunication terminal, for example, effectiveness would seem to be the most important dimension of usability. People should be able to approach the telecommunication terminal, make use of the service and pay for the service without any fuss. In the case of a public telecommunication terminal, learnability should not be an issue, as one may assume that people will not use a public telecommunication terminal which they cannot readily use. + +The use of both performance and attitude measures is recommended because they focus on different aspects of usability. + +Performance measures, sometimes referred to as objective measures, include performance time, error rate and number of failures to complete the task, whilst attitude (or subjective) measures capture user opinions. + +Checks should be made to ensure that there are no human factors deficiencies (such as poor lighting). + +### **3.2 Criteria** + +The usability criteria are quantified reference values set operationally for each usability component. The data captured is compared against the criteria to assess whether these have been reached. + +### **3.3 Tasks** + +Experimental tasks are to be identified by task analysis and appropriate scenarios made up. The scenarios should cover the most important system functions and be typical of the tasks that potential users will perform in an up-and-running system. + +### **3.4 Users** + +Experimental subjects should be representative of the potential users. A suitable number of subjects should participate in the experiment, so as to guarantee the statistical reliability of the results. + +### **3.5 Analysis of data2)** + +Experimental data is processed to check how well the usability criteria have been met. Problems are identified and solutions proposed. + +## **4 Example** + +An application of the general procedure to a specific instance is provided in the Annex A. + +--- + +2) In certain situations a statistically significant sample is not necessary to bring to light a design problem. + +## **Annex A** (to Recommendation F.901) + +## **Usability evaluation of ISDN videotelephone terminals** + +(This annex forms an integral part of this Recommendation) + +The following is an example of an approach to evaluating the usability of the ISDN videophone service access procedures according to this Recommendation. + +This example contains specific numeric values and specific procedures for illustrative purposes. In an actual evaluation, these values and procedures and other particular details of the test would of course have to be adjusted to suit the situation being tested. + +### **A.1     The user interface** + +The ISDN videophones are here referred to as terminals consisting of a basic ISDN telephone set and a colour video display (typically 9" to 12"). + +The telephone set consists of a numeric keypad, a number of additional keys and a display which provides the user with information (status of the line, dialled number, charging, etc.) and prompts. + +Dialling may be performed according to the "overlap" and "en-bloc" procedure. The "overlap" procedure encompasses the following steps: going off-hook, waiting for dial tone, dialling, changing mode from audio to audiovisual communication. The "en-bloc" procedure consists of dialling, going off-hook and changing mode from audio to audiovisual communication. + +The following functions are provided in addition to dialling: terminal set-up, re-dial, supplementary services, suspend/resume, dialled number editing, mute, hands-free, video pause, self-view, mode change control. They are accessed by means of dedicated keys located near the numeric keypad. + +The self-view facility may be provided through a separate monitor. + +Instructions of use may be provided in written form or verbally by the personnel who installs the equipment. + +### **A.2     Users** + +The videophones have been designed for the general public. + +### **A.3     Environment** + +The videophones for home or office applications are designed for an environment with the following viewing conditions: + +- environmental lighting: $\geq 400$ lux; +- viewing distance: 60 to 120 cm. + +### **A.4     Tasks** + +When using the videophone the user can set up videophone calls with either the "en-bloc" or "overlap" procedure, edit dialled numbers in the "en-bloc" procedure, change mode (from audio to audiovisual and vice versa), display his/her own picture (unless continuously provided on a separate screen), inhibit the transmission of the outgoing picture, release videophone calls. + +### A.5 Usability measures + +To evaluate the videophone usability both user performance and user attitudes are to be measured. The following data are captured: number of failures to complete a task, error rate, task completion time and satisfaction of user procedures. + +### A.6 Test design + +To check whether the usability criteria are met by the videophone user interface, laboratory tests are to be performed. + +In the following an example is provided of how to plan such an experiment for testing the most important aspects of videotelephony. + +#### A.6.1 The subjects + +At least thirty representative users drawn as a random sample (the higher the number, the more statistically reliable the results) take part in the experiment. + +#### A.6.2 The environment + +The videophone terminal is located in premises with 400 lux lighting and is used at 60 cm viewing distance. + +#### A.6.3 The tasks + +In each experimental session a given subject performs the following experimental tasks: + +- Task 1 consists of setting up a videophone call with the “overlap” procedure. The videophone number to be dialled is a 10-digit one. This task has to be repeated five times. +- Task 2 consists of setting up a videophone call with the “en-bloc” procedure. The videophone number to be dialled is a 10-digit one. This task has to be repeated seven times. +- Task 3 consists of editing one digit in the middle of a 10-digit telephone number dialled with the “en-bloc” procedure. This task has to be repeated five times. + +To assess the effect of learnability a subject will typically return for a number of sessions spread over a period of time. + +The experimental conditions are administered in a different random order to each subject. + +#### A.6.4 The usability criteria + +They are as follows: + +##### A.6.4.1 Task 1 + +- 1) 90% of the subjects should be able to set up the five videophone calls within a session using the “overlap” procedure without committing any errors3) or failing to complete a task in the last three attempts; +- 2) the average performance time of the last (correct) attempt should not exceed 60 seconds; +- 3) the MOS4) of satisfaction should be higher than 3 and the distribution should not be significantly bimodal. + +3) Errors are deviations from the correct procedure (excluded misdialling). + +4) MOS (Mean Opinion Score) measured on 5-point rating scales. +Unsatisfied Satisfied + +![A 5-point rating scale diagram. It consists of a horizontal line with five vertical tick marks. Below the tick marks are the numbers 1, 2, 3, 4, and 5. The word 'Unsatisfied' is positioned above the number 1, and 'Satisfied' is positioned above the number 5.](c973a12ab7ab173217244b189dc52577_img.jpg) + +|----------------|----------------|----------------|----------------| +1                2                3                4                5 + +A 5-point rating scale diagram. It consists of a horizontal line with five vertical tick marks. Below the tick marks are the numbers 1, 2, 3, 4, and 5. The word 'Unsatisfied' is positioned above the number 1, and 'Satisfied' is positioned above the number 5. + +##### **A.6.4.2 Task 2** + +- 1) 90% of the subjects should be able to set up the seven videophone calls within a session using the “en-bloc” procedure without committing any errors or failing to complete a task in the last three attempts; +- 2) the average performance time of the last (correct) attempt should not exceed 60 seconds; +- 3) the MOS of satisfaction should be higher than three and the distribution should not be significantly bimodal. + +##### **A.6.4.3 Task 3** + +- 1) 90% of the subjects should be able to complete the five wrong digit correction attempts within a session without committing any errors or failing to complete a task in the last three attempts; +- 2) the average performance time for editing the wrong digit should not exceed 30 seconds; +- 3) the MOS of satisfaction should be higher than three and the distribution should not be significantly bimodal. + +#### **A.6.5 The experimental procedure** + +All the users receive written instructions on how to use the equipment and a description of the tasks to be performed. Training may be provided if the service provider's introduction foresees so. Then the subjects carry out the assigned tasks the specified number of times. + +At each attempt the accuracy (number of errors or failures) and speed (performance time) are recorded. No time-out for task completion is set. In case of failure to complete task, subjects proceed to the next attempt. + +Possible questions on the experiment may be answered by the experimenter, but not ones on the tasks and the use of the terminal. + +At the end of each task the subjects are asked to express their “satisfaction” assessment on a 5-point rating scale. \ No newline at end of file diff --git a/marked/F/T-REC-F.922-202008-I_PDF-E/raw.md b/marked/F/T-REC-F.922-202008-I_PDF-E/raw.md new file mode 100644 index 0000000000000000000000000000000000000000..0f4267d1b6fb1fa98664e9a60083d5a39f732764 --- /dev/null +++ b/marked/F/T-REC-F.922-202008-I_PDF-E/raw.md @@ -0,0 +1,490 @@ + + +**ITU-T** + +TELECOMMUNICATION +STANDARDIZATION SECTOR +OF ITU + +**F.922** + +(08/2020) + +SERIES F: NON-TELEPHONE TELECOMMUNICATION +SERVICES + +Accessibility and human factors + +# --- **Requirements of information service systems for visually impaired persons** + +Recommendation ITU-T F.922 + +## ITU-T F-SERIES RECOMMENDATIONS **NON-TELEPHONE TELECOMMUNICATION SERVICES** + +| | | +|-----------------------------------------------------------------|--------------------| +| TELEGRAPH SERVICE | | +| Operating methods for the international public telegram service | F.1–F.19 | +| The gentex network | F.20–F.29 | +| Message switching | F.30–F.39 | +| The international telemessage service | F.40–F.58 | +| The international telex service | F.59–F.89 | +| Statistics and publications on international telegraph services | F.90–F.99 | +| Scheduled and leased communication services | F.100–F.104 | +| Phototelegraph service | F.105–F.109 | +| MOBILE SERVICE | | +| Mobile services and multdestination satellite services | F.110–F.159 | +| TELEMATIC SERVICES | | +| Public facsimile service | F.160–F.199 | +| Teletex service | F.200–F.299 | +| Videotex service | F.300–F.349 | +| General provisions for telematic services | F.350–F.399 | +| MESSAGE HANDLING SERVICES | F.400–F.499 | +| DIRECTORY SERVICES | F.500–F.549 | +| DOCUMENT COMMUNICATION | | +| Document communication | F.550–F.579 | +| Programming communication interfaces | F.580–F.599 | +| DATA TRANSMISSION SERVICES | F.600–F.699 | +| MULTIMEDIA SERVICES | F.700–F.799 | +| ISDN SERVICES | F.800–F.849 | +| UNIVERSAL PERSONAL TELECOMMUNICATION | F.850–F.899 | +| ACCESSIBILITY AND HUMAN FACTORS | F.900–F.999 | + +*For further details, please refer to the list of ITU-T Recommendations.* + +# Recommendation ITU-T F.922 + +# Requirements of information service systems for visually impaired persons + +## Summary + +Recommendation ITU-T F.922 establishes requirements for an information service system for visually impaired persons (ISS-VIPs), based on the mobile Internet, which connects visually impaired persons (VIPs) and volunteers. With the assistance of volunteers, the system enables VIPs to solve problems not yet handled by automatic assistive technology. The ISS-VIPs is constructed in client/server mode, enabling VIPs and volunteers to access it through their client software and to use its services or to provide its services to another person. The ISS-VIPs provides three kinds of service: image recognition; video assistance; and trip companion. The ISS-VIPs also supports the use of other software tools for assisting VIPs, so as to save valuable volunteer resources for service provision in person. While provision of services to VIP is the top priority, the ISS-VIPs can also be used by non-VIP with special service requirements. + +## History + +| Edition | Recommendation | Approval | Study Group | Unique ID* | +|---------|----------------|------------|-------------|---------------------------------------------------------------------------| +| 1.0 | ITU-T F.922 | 2020-08-13 | 16 | 11.1002/1000/14335 | + +## Keywords + +Accessibility, information service system, Internet, mobile network, visually impaired persons. + +--- + +\* To access the Recommendation, type the URL in the address field of your web browser, followed by the Recommendation's unique ID. For example, . + +## FOREWORD + +The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of telecommunications, information and communication technologies (ICTs). The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. + +The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics. + +The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1. + +In some areas of information technology which fall within ITU-T's purview, the necessary standards are prepared on a collaborative basis with ISO and IEC. + +## NOTE + +In this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. + +Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other obligatory language such as "must" and the negative equivalents are used to express requirements. The use of such words does not suggest that compliance with the Recommendation is required of any party. + +## INTELLECTUAL PROPERTY RIGHTS + +ITU draws attention to the possibility that the practice or implementation of this Recommendation may involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence, validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others outside of the Recommendation development process. + +As of the date of approval of this Recommendation, ITU had not received notice of intellectual property, protected by patents, which may be required to implement this Recommendation. However, implementers are cautioned that this may not represent the latest information and are therefore strongly urged to consult the TSB patent database at . + +© ITU 2020 + +All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior written permission of ITU. + +## Table of Contents + +| | Page | +|------------------------------------------------|------| +| 1 Scope ..... | 1 | +| 2 References..... | 1 | +| 3 Definitions ..... | 1 | +| 3.1 Terms defined elsewhere ..... | 1 | +| 3.2 Terms defined in this Recommendation..... | 1 | +| 4 Abbreviations and acronyms ..... | 1 | +| 5 Conventions ..... | 2 | +| 6 System description..... | 2 | +| 7 Service provision method ..... | 3 | +| 8 Service mode ..... | 3 | +| 9 Server function ..... | 4 | +| 9.1 User interaction ..... | 4 | +| 9.2 System management..... | 4 | +| 9.3 Website management..... | 4 | +| 10 Client function ..... | 4 | +| 10.1 Client for visually impaired persons..... | 4 | +| 10.2 Client for volunteers ..... | 5 | +| 11 Service process procedure ..... | 5 | +| Bibliography..... | 9 | + + + +# Recommendation ITU-T F.922 + +# Requirements of information service systems for visually impaired persons + +## 1 Scope + +This Recommendation establishes requirements for an information service system for visually impaired persons (VIPs), including those for the system composition, service mode, server function, client function and service process procedure. + +## 2 References + +The following ITU-T Recommendations and other references contain provisions which, through reference in this text, constitute provisions of this Recommendation. At the time of publication, the editions indicated were valid. All Recommendations and other references are subject to revision; users of this Recommendation are therefore encouraged to investigate the possibility of applying the most recent edition of the Recommendations and other references listed below. A list of the currently valid ITU-T Recommendations is regularly published. The reference to a document within this Recommendation does not give it, as a stand-alone document, the status of a Recommendation. + +None. + +## 3 Definitions + +### 3.1 Terms defined elsewhere + +This Recommendation uses the following terms defined elsewhere: + +**3.1.1 accessibility** [b-ITU-T F.791]: The degree to which a product, device, service or environment (virtual or real) is available to as many people as possible. + +**3.1.2 assistive technology** [b-ITU-T F.790]: Piece of equipment, product system, hardware, software or service that is used to enable, maintain or improve functional capabilities of individuals with disabilities. + +**3.1.3 disability** [b-ITU-T F.791]: An evolving concept, which refers to the interaction between persons with impairments, and attitudinal and environmental barriers that hinder their full and effective participation in society on an equal basis with others. + +### 3.2 Terms defined in this Recommendation + +This Recommendation defines the following terms: + +**3.2.1 ISS-VIPs user**: A person who accesses the information service system for visually impaired persons (ISS-VIPs) through the mobile Internet, including VIPs and volunteers. + +**3.2.2 information service system for visually impaired persons**: A service system which is established on the mobile Internet that provides information services to visually impaired persons (VIPs), the main purpose of which is to help VIPs to solve problems that cannot yet be solved by assistive technology. This system puts VIPs and volunteers in contact with each other and aids VIPs to get necessary timely visual information with the help of volunteers or to find a trip companion for going out. + +## 4 Abbreviations and acronyms + +This Recommendation uses the following abbreviations and acronyms: + +| | | +|----------|----------------------------------------------------------| +| AI | Artificial Intelligence | +| APP | Application | +| CMS | Content Management System | +| ISS-VIPs | Information Service System for Visually Impaired Persons | +| OCR | Optical Character Recognition | +| PLMN | Public Land Mobile Network | +| VIP | Visually Impaired Person | +| Wi-Fi | Wireless Fidelity | + +## 5 Conventions + +None. + +## 6 System description + +This clause specifies the overall functional architecture of an ISS-VIPs that is established over the Internet and mobile Internet, and composed of server and client units. It also specifies the operation methods of the ISS-VIPs. + +Figure 1 shows how VIPs and volunteers connect to the ISS-VIPs. They first connect to a public land mobile network (PLMN) or wireless fidelity (Wi-Fi) by using their mobile phone. Through the PLMN or Wi-Fi, they connect to the Internet. Through the Internet, they connect to the ISS-VIPs. + +![Schematic diagram of an information service system for visually impaired persons. On the left, four icons represent 'Volunteer' and 'Visually impaired persons' (two of each). Each icon has a mobile phone and a signal wave pointing to a cloud labeled 'PLMN' or 'Wi-Fi'. Both clouds are connected to a central cloud labeled 'Internet'. The 'Internet' cloud is connected to a dashed box labeled 'Service system'. Inside the 'Service system' box, from top to bottom, are: 'Database' (two cylinders), 'Servers' (three mobile phones), 'Operation and maintenance' (two people at computers), and 'Customer service' (two people at computers). The label 'F.922(20)_F01' is at the bottom right of the service system box.](5860ad6bd2a2dd8d1ab12864b8f90f37_img.jpg) + +Schematic diagram of an information service system for visually impaired persons. On the left, four icons represent 'Volunteer' and 'Visually impaired persons' (two of each). Each icon has a mobile phone and a signal wave pointing to a cloud labeled 'PLMN' or 'Wi-Fi'. Both clouds are connected to a central cloud labeled 'Internet'. The 'Internet' cloud is connected to a dashed box labeled 'Service system'. Inside the 'Service system' box, from top to bottom, are: 'Database' (two cylinders), 'Servers' (three mobile phones), 'Operation and maintenance' (two people at computers), and 'Customer service' (two people at computers). The label 'F.922(20)\_F01' is at the bottom right of the service system box. + +**Figure 1 – Schematic diagram of an information service system for visually impaired persons** + +In the central part of Figure 1, one cloud stands for the Internet, with others to its left labelled Wi-Fi and PLMN have their connections to the Internet shown. + +In the left part of Figure 1, VIPs and volunteers connect to the Wi-Fi and the PLMN cloud by using their mobile phones. + +In the right part of Figure 1, the ISS-VIPs connects to the Internet, which is composed of database, servers, operation and maintenance staff and customer service staff. + +The operation methods of the ISS-VIPs are as follows: + +- the ISS-VIPs accepts and manages user information, assists users to communicate with each other, and provides technical and service support for users when they access the service system; +- VIPs and volunteers can access the ISS-VIPs via their mobile phones, request or accept service requirements after a communication connection is established and receive or provide voluntary service through the network. + +## 7 Service provision method + +When users use the ISS-VIPs application (APP), it provides two choices: local technology and remote in-person service. Users can choose to use local processing technology (such as optical character recognition (OCR)) to solve their problems, or use the in-person service of the ISS-VIPs to find a volunteer to help them. + +![Flowchart of the service provision method. A Visually impaired person (VIP) sends a request (1) to an APP. The APP uses an AI engine (2) to decide between 'Local technology' (Yes) or 'ISS-VIP' (No). If 'Yes', local processing (3) occurs, and results (4) are returned to the APP and then to the VIP. If 'No', the APP sends a request (5) to the ISS-VIP. The ISS-VIP offers three services: Image recognition, Video assistance, and Trip companion. The ISS-VIP then forwards the request (6) to Service staff/volunteers. Finally, the ISS-VIP helps the VIP establish contact (7) with the staff or volunteer.](b3baf3a29b67c7425d2562ddbc52f0cc_img.jpg) + +``` + +graph LR + VIP[Visually impaired person] -- (1) --> APP((APP)) + APP -- (2) --> LT{Local technology} + LT -- Yes -- (3) --> APP + LT -- No -- (5) --> ISS_VIP[ISS-VIP] + ISS_VIP -- (6) --> Staff[Service staff/volunteers] + Staff -- (7) --> VIP + APP -- (4) --> VIP + subgraph ISS_VIP [ISS-VIP] + IR[Image recognition] + VA[Video assistance] + TC[Trip companion] + end + +``` + +F.922(20)\_F02 + +Flowchart of the service provision method. A Visually impaired person (VIP) sends a request (1) to an APP. The APP uses an AI engine (2) to decide between 'Local technology' (Yes) or 'ISS-VIP' (No). If 'Yes', local processing (3) occurs, and results (4) are returned to the APP and then to the VIP. If 'No', the APP sends a request (5) to the ISS-VIP. The ISS-VIP offers three services: Image recognition, Video assistance, and Trip companion. The ISS-VIP then forwards the request (6) to Service staff/volunteers. Finally, the ISS-VIP helps the VIP establish contact (7) with the staff or volunteer. + +**Figure 2 – Service provision method** + +Figure 2 shows how the ISS-VIPs provides services to VIPs, as follows: + +- (1) a VIP requests services; +- (2) by using an artificial intelligence (AI) engine, the APP decides how to help a VIP select local technology or remote in-person service; +- (3) if the APP chooses to use local technology, then local processing begins; +- (4) the APP returns the results processed by local technology to the VIP; +- (5) if the APP chooses to use the remote in-person service, the APP sends the VIP's service request to the ISS-VIPs; +- (6) the ISS-VIPs forwards the service request to the staff or volunteer group and waits for the response of an available staff member or volunteer; +- (7) the ISS-VIPs helps the VIP to establish contact with the staff member or volunteer, so that the latter can provide service for the former. + +## 8 Service mode + +This clause specifies the following service modes of the ISS-VIPs, which help volunteers provide service to VIPs: + +- image recognition service: answer the questions of VIPs about images they upload; +- video assistance service: guide VIPs to do one thing according to their requests through a video connection; +- trip companion service: help a VIP to find an available volunteer to go to a destination together. + +## **9 Server function** + +This clause specifies the functional requirements of servers for the ISS-VIPs, including user interaction, system management and web management. + +### **9.1 User interaction** + +The user interaction function includes the following sub-functions: + +- personal information interaction: process user's personal account information and service mode configuration information; +- service data interaction: receive users' requests in real-time, process service data, support the communication between user-to-system and user-to-user until an integrated service procedure is completed; +- system information interaction: issue system information in real-time, accept and process user feedback opinions. + +### **9.2 System management** + +The system management function includes the following sub-functions: + +- operation and maintenance management: perform the functions of system configuration, performance monitoring, failure detection, etc. +- data management: maintain and manage user account information, service information, user feedback opinions, etc. +- service management: perform the functions of service policy configuration, service status tracking, abnormal conditions processing, etc. +- security management: perform the functions of access control, permission control, data backup, disaster recovery, etc. +- statistical analysis: count and analyse user data, service data, operation data, etc. + +### **9.3 Website management** + +The website of the ISS-VIPs is managed by its content management system (CMS), which performs the functions of information collection, information editing and input, information retrieval, access control, supervisory control, etc. The webmaster can access the CMS in the web manner, and manipulate the website through the CMS. + +## **10 Client function** + +This clause specifies the functional requirements of clients of the ISS-VIPs, including clients for VIPs and volunteers. + +### **10.1 Client for visually impaired persons** + +The client for VIPs performs the following functions: + +- identity registration: provides personal information according to system requirements, so as to get the qualification to use the ISS-VIPs service; +- login authentication: provides personal account information according to ISS-VIPs requirements when accessing the ISS-VIPs; +- personal information maintenance: registers and manages personal information according to users' own needs; +- AI engine selection: selects the most suitable manner to help the VIP automatically, including selecting local technology or remote in-person service; + +- image recognition: sends an image recognition service requirement to the ISS-VIPs, and accepts the reply from a volunteer; +- video assistance: sends a video assistance service requirement to the ISS-VIPs, and accepts the service of a volunteer through a video connection; +- trip companion: sends a trip companion service requirement to the ISS-VIPs, and gets in touch with a volunteer through the ISS-VIPs; +- service record: saves information related to services that a VIP has used; +- social network sharing: sends information related to the ISS-VIPs service to social networks. + +### 10.2 Client for volunteers + +The client for volunteers performs the following functions: + +- identity registration: provides personal information according to system requirements, so as to get the qualification to use system service; +- login authentication: provides personal account information according to system requirements when accessing the ISS-VIPs; +- personal information maintenance: registers and manages personal information according to users' own needs; +- service mode set: sets service type and time, etc. according to personal situations; +- image recognition: accepts an image recognition service requirement from the ISS-VIPs, and provides a reply to the VIP; +- video assistance: accepts a video assistance service requirement from the ISS-VIPs, and provides service to a VIP through a video connection; +- trip companion: accepts a video assistance service requirement from the ISS-VIPs, gets in touch with a VIP through the ISS-VIPs and provides service. +- service record: saves information related to the services that a volunteer has provided; +- social network sharing: sends information related with the ISS-VIPs service to a social network. + +## 11 Service process procedure + +This clause specifies the complete information transmission and processing procedures of the image recognition service, video assistance service and trip companion service. + +The image recognition service process procedure is as illustrated in Figure 3. + +![A sequence diagram illustrating the image recognition service process procedure. The diagram is divided into three vertical lanes: 'Visually impaired person', 'Information service system', and 'Volunteer'. The process starts with 'Start' in the Information service system lane, followed by '1) Issue service item.' which points to the Visually impaired person lane. The Visually impaired person then performs '2) Take a picture and send a request.' which points to the Information service system lane. The system then performs '3) Forward the request.' which points to the Volunteer lane. The Volunteer performs '4) Send a reply about the picture.' which points to the Information service system lane. The system then performs '5) Forward the reply, and record the service status.' which points to the Visually impaired person lane. Finally, the Visually impaired person performs '6) Receive the reply.' which leads to the 'End' node in the Information service system lane.](d4af765160d04ecef538e5066006dc77_img.jpg) + +``` + +sequenceDiagram + participant VIP as Visually impaired person + participant ISS as Information service system + participant V as Volunteer + Note right of ISS: Start + ISS->>VIP: 1) Issue service item. + VIP->>ISS: 2) Take a picture and send a request. + ISS->>V: 3) Forward the request. + V->>ISS: 4) Send a reply about the picture. + ISS->>VIP: 5) Forward the reply, and record the service status. + VIP->>ISS: 6) Receive the reply. + Note right of ISS: End + +``` + +F.922(20)\_F03 + +A sequence diagram illustrating the image recognition service process procedure. The diagram is divided into three vertical lanes: 'Visually impaired person', 'Information service system', and 'Volunteer'. The process starts with 'Start' in the Information service system lane, followed by '1) Issue service item.' which points to the Visually impaired person lane. The Visually impaired person then performs '2) Take a picture and send a request.' which points to the Information service system lane. The system then performs '3) Forward the request.' which points to the Volunteer lane. The Volunteer performs '4) Send a reply about the picture.' which points to the Information service system lane. The system then performs '5) Forward the reply, and record the service status.' which points to the Visually impaired person lane. Finally, the Visually impaired person performs '6) Receive the reply.' which leads to the 'End' node in the Information service system lane. + +**Figure 3 – Image recognition service process procedure** + +As shown in Figure 3, the image recognition service process procedure is as follows: + +- 1) the ISS-VIPs issues service items that include image recognition; +- 2) after VIPs get the service information, they can take a picture and send a related question to the ISS-VIPs; +- 3) on receipt of a request from a VIP, the ISS-VIPs forwards it to several online volunteers; +- 4) the volunteers reply to the ISS-VIPs, to explain the content of the picture in response to the question from the VIP; +- 5) the system forwards the replies to the VIP; depending on the setting of the system or VIP user preference, the system sends the first reply, or the first three replies or all replies to the VIP; +- 6) the VIP receives the replies. + +The video assistance service process procedure is as illustrated in Figure 4. + +![A sequence diagram illustrating the video assistance service process procedure. The diagram is divided into three vertical lanes: 'Visually impaired person', 'Information service system', and 'Volunteer'. The process starts with 'Start' in the Information service system lane, followed by '1) Issue service item.' (represented by a document shape). The Visually impaired person then sends '2) Send a video assist service request.' to the Information service system. The Information service system forwards it with '3) Forward the request.' to the Volunteer. The Volunteer replies with '4) Accept the request, send a reply.' to the Information service system. The Information service system then '5) Accept the reply, assist both persons to set up a video connection.' This leads to a parallel section where '6a) Shoot a video and ask related questions.' occurs in the Visually impaired person lane and '6b) Answer the question according to the video.' occurs in the Volunteer lane. Both parallel actions lead to '7) Record the service status after the video communication stops.' in the Information service system lane, which finally leads to 'End'.](4e4be0bd8b235167902f2c03e41da651_img.jpg) + +``` + +sequenceDiagram + participant VIP as Visually impaired person + participant ISS as Information service system + participant V as Volunteer + Note right of ISS: Start + ISS->>VIP: 1) Issue service item. + VIP->>ISS: 2) Send a video assist service request. + ISS->>V: 3) Forward the request. + V->>ISS: 4) Accept the request, send a reply. + ISS->>VIP: 5) Accept the reply, assist both persons to set up a video connection. + Note left of ISS: Parallel section + VIP->>ISS: 6a) Shoot a video and ask related questions. + V->>ISS: 6b) Answer the question according to the video. + Note right of ISS: After parallel section + ISS->>VIP: 7) Record the service status after the video communication stops. + Note right of ISS: End + +``` + +A sequence diagram illustrating the video assistance service process procedure. The diagram is divided into three vertical lanes: 'Visually impaired person', 'Information service system', and 'Volunteer'. The process starts with 'Start' in the Information service system lane, followed by '1) Issue service item.' (represented by a document shape). The Visually impaired person then sends '2) Send a video assist service request.' to the Information service system. The Information service system forwards it with '3) Forward the request.' to the Volunteer. The Volunteer replies with '4) Accept the request, send a reply.' to the Information service system. The Information service system then '5) Accept the reply, assist both persons to set up a video connection.' This leads to a parallel section where '6a) Shoot a video and ask related questions.' occurs in the Visually impaired person lane and '6b) Answer the question according to the video.' occurs in the Volunteer lane. Both parallel actions lead to '7) Record the service status after the video communication stops.' in the Information service system lane, which finally leads to 'End'. + +F.922(20)\_F04 + +**Figure 4 – Video assistance service process procedure** + +As shown in Figure 4, the video assistance service process procedure is described as follows: + +- 1) the ISS-VIP issues service items that include video assistance; +- 2) after VIPs get the service information, they can send a video assistance service request to the ISS-VIPs; +- 3) on receipt of a request, the ISS-VIPs forwards it to several online volunteers; +- 4) the volunteers wishing to provide service reply to the ISS-VIPs; +- 5) when the ISS-VIPs receives the first reply, it establishes a video connection between the VIP and the volunteer; +- 6a) the VIP shoots a video and asks questions through the video connection; +- 6b) the volunteer answers the questions according to the video, until the VIP's problem is solved. +- 7) after video communication stops, the ISS-VIPs records the service status. + +The trip companion service process procedure is as illustrated in Figure 5. + +![Figure 5 – Trip companion service process procedure flowchart. The diagram is a swimlane flowchart with three lanes: Visually impaired person, Information service system, and Volunteer. The process starts with 'Start' in the Information service system lane, followed by '1) Issue service item.' (oval). The flow then goes to '2) Send a trip companion service request.' (rectangle) in the Visually impaired person lane. From there, it goes to '3) Forward the request and issue it.' (rectangle) in the Information service system lane. Next is '4) Send a reply for providing service.' (rectangle) in the Volunteer lane. Then '5) Accept the reply, assist both persons to get up in touch with each other.' (rectangle) in the Information service system lane. A horizontal line separates the pre-service phase from the post-service phase. In the post-service phase, '6a) Accept trip companion service.' (rectangle) in the Visually impaired person lane and '6b) Provide trip companion service.' (rectangle) in the Volunteer lane both lead to '7a) Feedback opinions to the system.' and '7b) Feedback opinions to the system.' respectively. Both '7a)' and '7b)' lead to '8) Record the service status.' (oval) in the Information service system lane, which finally leads to 'End' (oval).](b9ecbc3baefab13719e000faa6e0c7eb_img.jpg) + +``` + +graph TD + subgraph Visually impaired person + V2[2) Send a trip companion service request.] + V6a[6a) Accept trip companion service.] + V7a[7a) Feedback opinions to the system.] + end + subgraph Information service system + Start([Start]) + I1[1) Issue service item.] + I3[3) Forward the request and issue it.] + I5[5) Accept the reply, assist both persons to get up in touch with each other.] + I8[8) Record the service status.] + End([End]) + end + subgraph Volunteer + Vol4[4) Send a reply for providing service.] + Vol6b[6b) Provide trip companion service.] + Vol7b[7b) Feedback opinions to the system.] + end + + Start --> I1 + I1 --> V2 + V2 --> I3 + I3 --> Vol4 + Vol4 --> I5 + I5 --> V6a + I5 --> Vol6b + V6a --> V7a + Vol6b --> Vol7b + V7a --> I8 + Vol7b --> I8 + I8 --> End + +``` + +Figure 5 – Trip companion service process procedure flowchart. The diagram is a swimlane flowchart with three lanes: Visually impaired person, Information service system, and Volunteer. The process starts with 'Start' in the Information service system lane, followed by '1) Issue service item.' (oval). The flow then goes to '2) Send a trip companion service request.' (rectangle) in the Visually impaired person lane. From there, it goes to '3) Forward the request and issue it.' (rectangle) in the Information service system lane. Next is '4) Send a reply for providing service.' (rectangle) in the Volunteer lane. Then '5) Accept the reply, assist both persons to get up in touch with each other.' (rectangle) in the Information service system lane. A horizontal line separates the pre-service phase from the post-service phase. In the post-service phase, '6a) Accept trip companion service.' (rectangle) in the Visually impaired person lane and '6b) Provide trip companion service.' (rectangle) in the Volunteer lane both lead to '7a) Feedback opinions to the system.' and '7b) Feedback opinions to the system.' respectively. Both '7a)' and '7b)' lead to '8) Record the service status.' (oval) in the Information service system lane, which finally leads to 'End' (oval). + +F.922(20)\_F05 + +**Figure 5 – Trip companion service process procedure** + +As shown in Figure 5, the trip companion service process procedure is described as follows: + +- 1) the ISS-VIPs issues service items that include trip companions; +- 2) after VIPs get the service information, they can send a trip companion service request to the ISS-VIPs; +- 3) on receipt of a request, the ISS-VIPs forwards it to several online volunteers; +- 4) volunteers wishing to provide service reply to the ISS-VIPs; +- 5) when the ISS-VIPs receives the first reply, it puts the VIP and the volunteer in touch with each other; +- 6a) the VIP accepts the trip service and goes to the destination with the volunteer; +- 6b) after the trip, the VIP feeds opinions back to the ISS-VIPs; +- 7a) the volunteer provides service and goes to the destination with the VIP; +- 7b) after the trip, the volunteer feeds opinions back to the ISS-VIPs; +- 8) after the service is over, the ISS-VIPs records the service status. + +## Bibliography + +- [b-ITU-T F.790] Recommendation ITU-T F.790 (2007), *Telecommunications accessibility guidelines for older persons and persons with disabilities.* +- [b-ITU-T F.791] Recommendation ITU-T F.791 (2007), *Accessibility terms and definitions.* +- [b-ITU-T F.921] Recommendation ITU-T F.921 (V2) (2018), *Audio-based indoor and outdoor network navigation system for persons with vision impairment.* + + + + + +## SERIES OF ITU-T RECOMMENDATIONS + +| | | +|-----------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| +| Series A | Organization of the work of ITU-T | +| Series D | Tariff and accounting principles and international telecommunication/ICT economic and policy issues | +| Series E | Overall network operation, telephone service, service operation and human factors | +| Series F | Non-telephone telecommunication services | +| Series G | Transmission systems and media, digital systems and networks | +| Series H | Audiovisual and multimedia systems | +| Series I | Integrated services digital network | +| Series J | Cable networks and transmission of television, sound programme and other multimedia signals | +| Series K | Protection against interference | +| Series L | Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation and protection of cables and other elements of outside plant | +| Series M | Telecommunication management, including TMN and network maintenance | +| Series N | Maintenance: international sound programme and television transmission circuits | +| Series O | Specifications of measuring equipment | +| Series P | Telephone transmission quality, telephone installations, local line networks | +| Series Q | Switching and signalling, and associated measurements and tests | +| Series R | Telegraph transmission | +| Series S | Telegraph services terminal equipment | +| Series T | Terminals for telematic services | +| Series U | Telegraph switching | +| Series V | Data communication over the telephone network | +| Series X | Data networks, open system communications and security | +| Series Y | Global information infrastructure, Internet protocol aspects, next-generation networks, Internet of Things and smart cities | +| Series Z | Languages and general software aspects for telecommunication systems | \ No newline at end of file diff --git a/marked/H/T-REC-H.110-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/H/T-REC-H.110-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..64672153e39c851d9e16fdbc702e526c7ba7e8a7 --- /dev/null +++ b/marked/H/T-REC-H.110-198811-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:454a9958ffe168868cb7d38a0eb24418dafe31a7a4245c992089b2316ac37d3e +size 7392 diff --git a/marked/H/T-REC-H.110-198811-I_PDF-E/67b83c4adaaa4de02d367168308deb2a_img.jpg b/marked/H/T-REC-H.110-198811-I_PDF-E/67b83c4adaaa4de02d367168308deb2a_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ddfffa910e28ff3e98c36de085fa70ab9b6144b7 --- /dev/null +++ b/marked/H/T-REC-H.110-198811-I_PDF-E/67b83c4adaaa4de02d367168308deb2a_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c0d518ccf105d7c3a8f7a8a2f048b260c53ac2d71145f7145f5636409fbc1299 +size 1526 diff --git a/marked/H/T-REC-H.110-198811-I_PDF-E/78d5774278a3f4a614f8c0ae485ce8d9_img.jpg b/marked/H/T-REC-H.110-198811-I_PDF-E/78d5774278a3f4a614f8c0ae485ce8d9_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..87bb962aa3987f91a58d36c01bcc7fa2b0ffe827 --- /dev/null +++ b/marked/H/T-REC-H.110-198811-I_PDF-E/78d5774278a3f4a614f8c0ae485ce8d9_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:875958d956f951dd8dfe008c7c25297d903c24af336b998a18cde8f9bf264904 +size 47275 diff --git a/marked/H/T-REC-H.110-198811-I_PDF-E/8e065eabc40a3645a569db7e9cd0da32_img.jpg b/marked/H/T-REC-H.110-198811-I_PDF-E/8e065eabc40a3645a569db7e9cd0da32_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..6660aa4fcc3cc66339f7b4e0e701c53c70af528c --- /dev/null +++ b/marked/H/T-REC-H.110-198811-I_PDF-E/8e065eabc40a3645a569db7e9cd0da32_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f409bfea9221e9e61ba94d6674b12a6267c551bb56b1fa5c7c158a2116771d46 +size 2364 diff --git a/marked/H/T-REC-H.110-198811-I_PDF-E/9ebd85380ef496499e5f23ec0d9cd744_img.jpg b/marked/H/T-REC-H.110-198811-I_PDF-E/9ebd85380ef496499e5f23ec0d9cd744_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d39a162c67016f6ff970bff6ff70a21a2ea91fec --- /dev/null +++ b/marked/H/T-REC-H.110-198811-I_PDF-E/9ebd85380ef496499e5f23ec0d9cd744_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dd320dda6af85d171a7f77da95182a5be9910606d0a145cc18d493102f8d402e +size 1337 diff --git a/marked/H/T-REC-H.110-198811-I_PDF-E/a612b25163221507295cc3d2fb679c10_img.jpg b/marked/H/T-REC-H.110-198811-I_PDF-E/a612b25163221507295cc3d2fb679c10_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..49e8a31d25d17c9a2ffb9efe55748a2368479a70 --- /dev/null +++ b/marked/H/T-REC-H.110-198811-I_PDF-E/a612b25163221507295cc3d2fb679c10_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1071f82fce05f1bb655be464e025eca735632bc679f56fe4968ee6f26b52f2f7 +size 2111 diff --git a/marked/H/T-REC-H.110-198811-I_PDF-E/a7d78d22e465dea388b31d0739f9d0cd_img.jpg b/marked/H/T-REC-H.110-198811-I_PDF-E/a7d78d22e465dea388b31d0739f9d0cd_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..61a5d687767e17a644052276b7bcfe7e6f636f75 --- /dev/null +++ b/marked/H/T-REC-H.110-198811-I_PDF-E/a7d78d22e465dea388b31d0739f9d0cd_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:27fa1805887ab8f2f54db91cbc2f37e642df27d7c798fd4621b6dbce5f5babea +size 89328 diff --git a/marked/H/T-REC-H.110-198811-I_PDF-E/c1c7af7ea36be0323047962df57d75b0_img.jpg b/marked/H/T-REC-H.110-198811-I_PDF-E/c1c7af7ea36be0323047962df57d75b0_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..51618f83881272d85c98834519f1423584e4e6cd --- /dev/null +++ b/marked/H/T-REC-H.110-198811-I_PDF-E/c1c7af7ea36be0323047962df57d75b0_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c1202dfad294dbfc5796bff5ff29e5d4fa76ec58f62920b2475dd60288ee50fc +size 1431 diff --git a/marked/H/T-REC-H.110-198811-I_PDF-E/d0abac95583b52a3b35f74a215567334_img.jpg b/marked/H/T-REC-H.110-198811-I_PDF-E/d0abac95583b52a3b35f74a215567334_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..910fe5d6924b0ee7224c29403605a07be50dfc0d --- /dev/null +++ b/marked/H/T-REC-H.110-198811-I_PDF-E/d0abac95583b52a3b35f74a215567334_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:64c2ab6a747c8c10626f6a2af6d0bd4e55efa1434004c82bbc2ff261b2dde6f0 +size 60157 diff --git a/marked/H/T-REC-H.110-198811-I_PDF-E/d980a3f9608055996a07f31788baf827_img.jpg b/marked/H/T-REC-H.110-198811-I_PDF-E/d980a3f9608055996a07f31788baf827_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c0e8c872008c1a9aebcf624668a7e77ece8577e9 --- /dev/null +++ b/marked/H/T-REC-H.110-198811-I_PDF-E/d980a3f9608055996a07f31788baf827_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:84e54bfb5209401b5aff5cb5abafa26a29b769232ac56e40734e8f0d2a3cec68 +size 60849 diff --git a/marked/H/T-REC-H.110-198811-I_PDF-E/e40ba44ba095240702ed2099e01d23ce_img.jpg b/marked/H/T-REC-H.110-198811-I_PDF-E/e40ba44ba095240702ed2099e01d23ce_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ecbf789c9643a0fc6e4290e292530b62482f7b3b --- /dev/null +++ b/marked/H/T-REC-H.110-198811-I_PDF-E/e40ba44ba095240702ed2099e01d23ce_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:689335369cae74942964ad2ab7aa4b62c0c37251c256b2bd348643afe74cf23d +size 2682 diff --git a/marked/H/T-REC-H.110-198811-I_PDF-E/f3e03accc76df483950e65a9fb19c20e_img.jpg b/marked/H/T-REC-H.110-198811-I_PDF-E/f3e03accc76df483950e65a9fb19c20e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..26179edb67ce62201124325ebeb5b6ac78768543 --- /dev/null +++ b/marked/H/T-REC-H.110-198811-I_PDF-E/f3e03accc76df483950e65a9fb19c20e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9c108c43a822f4ccea965d7ba5ff2f279ca510631f78c71ffbc214eb7c44691c +size 1594 diff --git a/marked/H/T-REC-H.110-198811-I_PDF-E/f9a14fbfecbd7d059226cc93677d721b_img.jpg b/marked/H/T-REC-H.110-198811-I_PDF-E/f9a14fbfecbd7d059226cc93677d721b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..5a0ecb2434ac63f9f66a5cbb0143448a07d4370e --- /dev/null +++ b/marked/H/T-REC-H.110-198811-I_PDF-E/f9a14fbfecbd7d059226cc93677d721b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a81b17db9f0747eb2b289f364e2ab2d0dd9b903783b34c5bf703ccc195fc6f5a +size 44880 diff --git a/marked/H/T-REC-H.222.1-199603-I_PDF-E/1c953f32bd34345dfd68fddf8a3736d6_img.jpg b/marked/H/T-REC-H.222.1-199603-I_PDF-E/1c953f32bd34345dfd68fddf8a3736d6_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..77d124cc05987f0b663a1ae5414ee0ca27ebed97 --- /dev/null +++ b/marked/H/T-REC-H.222.1-199603-I_PDF-E/1c953f32bd34345dfd68fddf8a3736d6_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8e8f315ad00bc9f8482c52abe8bf27c4c93c27ccb75c6e87000ef275a0ad0573 +size 26710 diff --git a/marked/H/T-REC-H.222.1-199603-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/H/T-REC-H.222.1-199603-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f6651533b19e71f53022f9d3bc37df5449eecdf3 --- /dev/null +++ b/marked/H/T-REC-H.222.1-199603-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:679479fbab17a4509d20946d7f2eabe41f64ce19c19b3e02835d4aa352bb1647 +size 8239 diff --git a/marked/H/T-REC-H.222.1-199603-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg b/marked/H/T-REC-H.222.1-199603-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..2baf923515c715b616191ab15e5aefdb0b0f10ff --- /dev/null +++ b/marked/H/T-REC-H.222.1-199603-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3977f9e0a4f620929b2fd294aaf0662db55d585d86bdd6ca04e2d69a6464d0db +size 40538 diff --git a/marked/H/T-REC-H.235.1-200509-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg b/marked/H/T-REC-H.235.1-200509-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..baf79732cdabbe6f1e0d68ad7f499566f23f72fb --- /dev/null +++ b/marked/H/T-REC-H.235.1-200509-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1c97e052632f2400398c0ea914036d0509b4f4054808aaf8eb5f79576f491710 +size 28335 diff --git a/marked/H/T-REC-H.235.1-200509-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/H/T-REC-H.235.1-200509-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f79471c5659742863ea900cf3f090a359b2ca164 --- /dev/null +++ b/marked/H/T-REC-H.235.1-200509-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:596fc75d81fdc29178f84e408ac1f1fb684372d34b0b14f490046ec0665bcaae +size 8185 diff --git a/marked/H/T-REC-H.235.1-200509-I_PDF-E/90ee16ccc0ad16aeca48087797d7b07f_img.jpg b/marked/H/T-REC-H.235.1-200509-I_PDF-E/90ee16ccc0ad16aeca48087797d7b07f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9f94ef2e0b1f205abc6c601294bd6222451a886e --- /dev/null +++ b/marked/H/T-REC-H.235.1-200509-I_PDF-E/90ee16ccc0ad16aeca48087797d7b07f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e8b9aa6cb0b13702308cf69bf3e801c0411175c22a7341ceab45fd243a80bd18 +size 27696 diff --git a/marked/H/T-REC-H.235.1-200509-I_PDF-E/9b9d2abd741ed4bafe7f78f89961c663_img.jpg b/marked/H/T-REC-H.235.1-200509-I_PDF-E/9b9d2abd741ed4bafe7f78f89961c663_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d2ad300183a4f8adea22c3c84206391c2adb36ba --- /dev/null +++ b/marked/H/T-REC-H.235.1-200509-I_PDF-E/9b9d2abd741ed4bafe7f78f89961c663_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:738c4cda6e7d79e7b8f976ca4c897fc93ece89bbaa3407ffa93f148772fcd70c +size 27099 diff --git a/marked/H/T-REC-H.235.10-202203-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg b/marked/H/T-REC-H.235.10-202203-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..2585edab110a1df8ff9f636903e1b13280ccbfda --- /dev/null +++ b/marked/H/T-REC-H.235.10-202203-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5c17359bdc002ff65ca8303c0bbee4fc5e9e3b5a78247d0bd146ea55b66047d8 +size 6147 diff --git a/marked/H/T-REC-H.235.10-202203-I_PDF-E/dbe553cf16dd14073b89a8263a428664_img.jpg b/marked/H/T-REC-H.235.10-202203-I_PDF-E/dbe553cf16dd14073b89a8263a428664_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9e8ffdc9cc3486a8064e70835f5f603f34ff01d6 --- /dev/null +++ b/marked/H/T-REC-H.235.10-202203-I_PDF-E/dbe553cf16dd14073b89a8263a428664_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:35d37e79108bb2f02eff9f4127c4eaf181442afb2f1f17af314dba26b1141cbb +size 41789 diff --git a/marked/H/T-REC-H.248.23-201303-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg b/marked/H/T-REC-H.248.23-201303-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7cb4d5d11511a2e361553824d508fde008fab400 --- /dev/null +++ b/marked/H/T-REC-H.248.23-201303-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:abbadb5a4216778c87e818254a2116b3161c9b23f839b90107c90e8f05b9e8c3 +size 3538 diff --git a/marked/H/T-REC-H.248.34-201202-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/H/T-REC-H.248.34-201202-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..650401cbff29ec75514e8e21d999ea0060c24983 --- /dev/null +++ b/marked/H/T-REC-H.248.34-201202-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2cc0ee724f22fbb2b9f797c79d7733c99b4b66fe8c758557f2cdf18ede34e6b4 +size 6733 diff --git a/marked/H/T-REC-H.248.36-201303-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg b/marked/H/T-REC-H.248.36-201303-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..dad02182b30e94f721c7f50726f60641c96df521 --- /dev/null +++ b/marked/H/T-REC-H.248.36-201303-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:45bf920a7798c0111eed0c87f3f277105f8935167814eaedd42f96a3f983bba0 +size 4024 diff --git a/marked/H/T-REC-H.248.40-201303-I_PDF-E/410562339ce067fdc6fa41940c118658_img.jpg b/marked/H/T-REC-H.248.40-201303-I_PDF-E/410562339ce067fdc6fa41940c118658_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..db66e0aaea4bcf473a0261c3b8e4e86343c10403 --- /dev/null +++ b/marked/H/T-REC-H.248.40-201303-I_PDF-E/410562339ce067fdc6fa41940c118658_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:070264218d96e5c3467a60333c2fc0aa52138aef36f1995ac9527e7dc17fd310 +size 56137 diff --git a/marked/H/T-REC-H.248.40-201303-I_PDF-E/5a9282ac54ca7bc50f1d2ab6cfb376ba_img.jpg b/marked/H/T-REC-H.248.40-201303-I_PDF-E/5a9282ac54ca7bc50f1d2ab6cfb376ba_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..da9606429f1c9abaa0a1d8eb2ba007b664a2631c --- /dev/null +++ b/marked/H/T-REC-H.248.40-201303-I_PDF-E/5a9282ac54ca7bc50f1d2ab6cfb376ba_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3c1c116102ad2c28fccc4e2fae05793dd8713b860db1f004dbfa3456e1c7b4fe +size 100048 diff --git a/marked/H/T-REC-H.248.40-201303-I_PDF-E/9b6b5924b48bf2fd5f347f88f06f45b3_img.jpg b/marked/H/T-REC-H.248.40-201303-I_PDF-E/9b6b5924b48bf2fd5f347f88f06f45b3_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d3289e25ba9c47d81430dae589ac61e5d858902f --- /dev/null +++ b/marked/H/T-REC-H.248.40-201303-I_PDF-E/9b6b5924b48bf2fd5f347f88f06f45b3_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:755ec3671c440ce5a6f9780d54e545e6a220390481d6f192a8889a5571d5ee86 +size 34587 diff --git a/marked/H/T-REC-H.248.40-201303-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg b/marked/H/T-REC-H.248.40-201303-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7cb4d5d11511a2e361553824d508fde008fab400 --- /dev/null +++ b/marked/H/T-REC-H.248.40-201303-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:abbadb5a4216778c87e818254a2116b3161c9b23f839b90107c90e8f05b9e8c3 +size 3538 diff --git a/marked/H/T-REC-H.248.40-201303-I_PDF-E/d4af765160d04ecef538e5066006dc77_img.jpg b/marked/H/T-REC-H.248.40-201303-I_PDF-E/d4af765160d04ecef538e5066006dc77_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ae0577ba49f1220cea8b57891f734900b8c0d7a5 --- /dev/null +++ b/marked/H/T-REC-H.248.40-201303-I_PDF-E/d4af765160d04ecef538e5066006dc77_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:95b519ed939c17a4ae80b0d3a79fb7191edea9e9f6d0298c73dd87e633ecba53 +size 78691 diff --git a/marked/H/T-REC-H.248.58-200806-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg b/marked/H/T-REC-H.248.58-200806-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d5a35041124e73f88d6638cd17f42a2eb4f6781a --- /dev/null +++ b/marked/H/T-REC-H.248.58-200806-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8f24b214263bab4142643453a604fb5196f9397e2500c675cde79d53ce5e5d8e +size 3645 diff --git a/marked/H/T-REC-H.248.58-200806-I_PDF-E/352c5fab6f936356e9570761a02ab71e_img.jpg b/marked/H/T-REC-H.248.58-200806-I_PDF-E/352c5fab6f936356e9570761a02ab71e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..b791431f04af573f9b4413e86745c9dbe45caa6a --- /dev/null +++ b/marked/H/T-REC-H.248.58-200806-I_PDF-E/352c5fab6f936356e9570761a02ab71e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b93eb13b38eef583f23fbb10b4549a89731b4b6acba7d045cc373fd28f2ee590 +size 56465 diff --git a/marked/H/T-REC-H.248.58-200806-I_PDF-E/91be14371a97fb5ce9eeb29ae18d07c3_img.jpg b/marked/H/T-REC-H.248.58-200806-I_PDF-E/91be14371a97fb5ce9eeb29ae18d07c3_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..2dfd9a0b863d75ae3fd013ad362956b61b27b495 --- /dev/null +++ b/marked/H/T-REC-H.248.58-200806-I_PDF-E/91be14371a97fb5ce9eeb29ae18d07c3_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dd355466244c9a245068fd395a732c97fa702ed5e24b3cdada8ffbad6c77978d +size 68648 diff --git a/marked/H/T-REC-H.248.60-200903-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg b/marked/H/T-REC-H.248.60-200903-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d5a35041124e73f88d6638cd17f42a2eb4f6781a --- /dev/null +++ b/marked/H/T-REC-H.248.60-200903-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8f24b214263bab4142643453a604fb5196f9397e2500c675cde79d53ce5e5d8e +size 3645 diff --git a/marked/H/T-REC-H.248.76-201009-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg b/marked/H/T-REC-H.248.76-201009-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f0b9ad00e787802057e514c23d50e5a006468505 --- /dev/null +++ b/marked/H/T-REC-H.248.76-201009-I_PDF-E/84a1d09fb489061482111515543b60dc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f5710b4101c2ac05321b5fe04d0b292318b8649c0ac50f587a3b3c0d3e4dde04 +size 7496 diff --git a/marked/H/T-REC-H.248.8-201303-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg b/marked/H/T-REC-H.248.8-201303-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..0ad98c6e45b6c52f074f0054407ca88a640d8f1d --- /dev/null +++ b/marked/H/T-REC-H.248.8-201303-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:58019f38be268f19eaa80c6d56d0f6b39d5741569a8f3ccc04117fa95d9c6465 +size 4016 diff --git a/marked/H/T-REC-H.248.98-201602-I_PDF-E/08dce7ad4c512fdf0c0cde60415fade6_img.jpg b/marked/H/T-REC-H.248.98-201602-I_PDF-E/08dce7ad4c512fdf0c0cde60415fade6_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..865c9203684ad08cb533b088360bb3e4d1df49a0 --- /dev/null +++ b/marked/H/T-REC-H.248.98-201602-I_PDF-E/08dce7ad4c512fdf0c0cde60415fade6_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e37145458a5f5b7754d0be7bfb1ab5ce8298a84576a7c3a4e910e530b48c9909 +size 76994 diff --git a/marked/H/T-REC-H.248.98-201602-I_PDF-E/0b3d9fe35da3ee0c88f1420bb9ed7a03_img.jpg b/marked/H/T-REC-H.248.98-201602-I_PDF-E/0b3d9fe35da3ee0c88f1420bb9ed7a03_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..21b92efa6f53a34eff18fe4405dadb410fc8ad78 --- /dev/null +++ b/marked/H/T-REC-H.248.98-201602-I_PDF-E/0b3d9fe35da3ee0c88f1420bb9ed7a03_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5c6cb0c41b088bf69e3776ef53109cdd51e472303fe8388c5ec7bab18d525b56 +size 83222 diff --git a/marked/H/T-REC-H.248.98-201602-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/H/T-REC-H.248.98-201602-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..22dc7ff0fb30f976cc1e576c45d2b7e65fa5ddc7 --- /dev/null +++ b/marked/H/T-REC-H.248.98-201602-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aa14933e776f7d0be8b2d60b81cbb6c1ee49deb5080e0281da05296f48189055 +size 6859 diff --git a/marked/H/T-REC-H.248.98-201602-I_PDF-E/5478f70a6cef3e5672b2b22d28830cfb_img.jpg b/marked/H/T-REC-H.248.98-201602-I_PDF-E/5478f70a6cef3e5672b2b22d28830cfb_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..e4f943f25c2f9cfa1754a3c008eae51cc2ada6e9 --- /dev/null +++ b/marked/H/T-REC-H.248.98-201602-I_PDF-E/5478f70a6cef3e5672b2b22d28830cfb_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:60c850e6c31e3cbfe0a36a2b23fb4736af31e5105ca6ef5c83f301cf4d502574 +size 37874 diff --git a/marked/H/T-REC-H.248.98-201602-I_PDF-E/692541e65db4dc852988ce77ebb60ce5_img.jpg b/marked/H/T-REC-H.248.98-201602-I_PDF-E/692541e65db4dc852988ce77ebb60ce5_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a821a74b670473f151fe9bd467c7a811e9e032f5 --- /dev/null +++ b/marked/H/T-REC-H.248.98-201602-I_PDF-E/692541e65db4dc852988ce77ebb60ce5_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b9343fda8489e66e42199815cd4e1ac7e61149890bb9663db7b5d1919a0fc8ec +size 108283 diff --git a/marked/H/T-REC-H.248.98-201602-I_PDF-E/77959075c823bb5169480d7b8ff82a63_img.jpg b/marked/H/T-REC-H.248.98-201602-I_PDF-E/77959075c823bb5169480d7b8ff82a63_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..5c7f99d01bae9114ca694623add7a9973de2d5b2 --- /dev/null +++ b/marked/H/T-REC-H.248.98-201602-I_PDF-E/77959075c823bb5169480d7b8ff82a63_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3fc79d323069cf88637b7fdd7fd145346fb08ae9a57300a70148710ade93b2d6 +size 122803 diff --git a/marked/H/T-REC-H.248.98-201602-I_PDF-E/9870bf462aa0d916a16d14b5a100c60a_img.jpg b/marked/H/T-REC-H.248.98-201602-I_PDF-E/9870bf462aa0d916a16d14b5a100c60a_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3bb7de73143ad2934bd53825dc0d05290277d645 --- /dev/null +++ b/marked/H/T-REC-H.248.98-201602-I_PDF-E/9870bf462aa0d916a16d14b5a100c60a_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ea2bfc1bf0f258b69b60bff1027dc84f798e871717c77cfd6a8cb2c5f9d52a50 +size 45804 diff --git a/marked/H/T-REC-H.248.98-201602-I_PDF-E/9a5927586a691c4908aa2cf98bd47ebb_img.jpg b/marked/H/T-REC-H.248.98-201602-I_PDF-E/9a5927586a691c4908aa2cf98bd47ebb_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..507254ce78e9cccd92874728beb44651ecdcdb03 --- /dev/null +++ b/marked/H/T-REC-H.248.98-201602-I_PDF-E/9a5927586a691c4908aa2cf98bd47ebb_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:66e6ace353e3e970d05c61f8f8dec17cc1b466704331cbd9ac1d438047d65167 +size 106519 diff --git a/marked/H/T-REC-H.248.98-201602-I_PDF-E/a4b963a07cc368283154762c4b156fe7_img.jpg b/marked/H/T-REC-H.248.98-201602-I_PDF-E/a4b963a07cc368283154762c4b156fe7_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..e8795a9bf51680d31608881c83fad030c28457b8 --- /dev/null +++ b/marked/H/T-REC-H.248.98-201602-I_PDF-E/a4b963a07cc368283154762c4b156fe7_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0e1b458d1cddd5580f00d5f2c7f9bb3abb3e8b807dff0a99f1ffaed6e11d3321 +size 61438 diff --git a/marked/H/T-REC-H.248.98-201602-I_PDF-E/c2fc2621e8206d24427b56bcb2398fc0_img.jpg b/marked/H/T-REC-H.248.98-201602-I_PDF-E/c2fc2621e8206d24427b56bcb2398fc0_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ecf83cdbd613764c10b21df85ce920f208346211 --- /dev/null +++ b/marked/H/T-REC-H.248.98-201602-I_PDF-E/c2fc2621e8206d24427b56bcb2398fc0_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d2fff15e9fc88ef9e9ad62598ec81df824d31f01d639861dc7b87e21a95534e2 +size 51288 diff --git a/marked/H/T-REC-H.248.98-201602-I_PDF-E/e451401f8fa77b466f401d5fce15b26c_img.jpg b/marked/H/T-REC-H.248.98-201602-I_PDF-E/e451401f8fa77b466f401d5fce15b26c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..90dfa34b710be81d0d4408283f2199cbaf291da8 --- /dev/null +++ b/marked/H/T-REC-H.248.98-201602-I_PDF-E/e451401f8fa77b466f401d5fce15b26c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ca8bdc68ee2a91d59a01f849bfef9f3714d7719a55691d4cf1f46d47fcab8c36 +size 48112 diff --git a/marked/H/T-REC-H.248.98-201602-I_PDF-E/e64c7b989e5bdb2708cd7aefd18b06e1_img.jpg b/marked/H/T-REC-H.248.98-201602-I_PDF-E/e64c7b989e5bdb2708cd7aefd18b06e1_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ac4ece1a3604dc6c8bd130aaf336db93bb42c790 --- /dev/null +++ b/marked/H/T-REC-H.248.98-201602-I_PDF-E/e64c7b989e5bdb2708cd7aefd18b06e1_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d7fac9900fade756128356771f75cc57070568562d935c0b5efbd2fd1b490602 +size 226786 diff --git a/marked/H/T-REC-H.248.98-201602-I_PDF-E/f9c64800d9bace9b4315646d1057be3c_img.jpg b/marked/H/T-REC-H.248.98-201602-I_PDF-E/f9c64800d9bace9b4315646d1057be3c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..dd0b1a83add72e51f8e77c91ea0153271456840f --- /dev/null +++ b/marked/H/T-REC-H.248.98-201602-I_PDF-E/f9c64800d9bace9b4315646d1057be3c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c6eb388d8e028b6714d06752baa434d2869d82803774a791999dc712ed357caf +size 44793 diff --git a/marked/H/T-REC-H.430.3-202309-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/H/T-REC-H.430.3-202309-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ac295b1b0c7f14b1c5a33ef6682b1b017c812477 --- /dev/null +++ b/marked/H/T-REC-H.430.3-202309-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37aed4fc114a39843438175326e0b7e1d09a9d6075723d82fea32040fd560ee4 +size 7192 diff --git a/marked/H/T-REC-H.430.3-202309-I_PDF-E/082ba09313df59d76a7bfbdde8ec877d_img.jpg b/marked/H/T-REC-H.430.3-202309-I_PDF-E/082ba09313df59d76a7bfbdde8ec877d_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..49e1283b068ee33c8509dd788a4f034fe40dc3d0 --- /dev/null +++ b/marked/H/T-REC-H.430.3-202309-I_PDF-E/082ba09313df59d76a7bfbdde8ec877d_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f29bef802b200c25aa92dceb112f3ddba10b40fb72186d29273794ffc8eaa778 +size 47443 diff --git a/marked/H/T-REC-H.430.3-202309-I_PDF-E/09955ff8214ffb6947951fc0f60eb6ab_img.jpg b/marked/H/T-REC-H.430.3-202309-I_PDF-E/09955ff8214ffb6947951fc0f60eb6ab_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..325c226daa5ecb5ce19e31ff3b3fb836c121d6bb --- /dev/null +++ b/marked/H/T-REC-H.430.3-202309-I_PDF-E/09955ff8214ffb6947951fc0f60eb6ab_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4ddea2f0505660b5d32a7ad47d374c77e2fc3e7ff74eb6ba56dc740b69891f5b +size 84801 diff --git a/marked/H/T-REC-H.430.3-202309-I_PDF-E/3267a096e9ca525744d8cd820f12eb59_img.jpg b/marked/H/T-REC-H.430.3-202309-I_PDF-E/3267a096e9ca525744d8cd820f12eb59_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..29e77c3e2bd54bd8acdfd6e4637c83ddd580218e --- /dev/null +++ b/marked/H/T-REC-H.430.3-202309-I_PDF-E/3267a096e9ca525744d8cd820f12eb59_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9c07c385917e913094fae49e82378a2b00626b3ea24c24c1c1352c0cfa537759 +size 41241 diff --git a/marked/H/T-REC-H.430.3-202309-I_PDF-E/3376f33dd811eb3e39ff47117c958ed9_img.jpg b/marked/H/T-REC-H.430.3-202309-I_PDF-E/3376f33dd811eb3e39ff47117c958ed9_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c14d2c659e2e5844c995bc677347d055e3525aee --- /dev/null +++ b/marked/H/T-REC-H.430.3-202309-I_PDF-E/3376f33dd811eb3e39ff47117c958ed9_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e4b96a0b3f748fac9773259692e2f657e1619eaec5b34eedb2e4bea1d3149dd7 +size 15903 diff --git a/marked/H/T-REC-H.430.3-202309-I_PDF-E/40aa08ca83c29a9281c0096fb9335f55_img.jpg b/marked/H/T-REC-H.430.3-202309-I_PDF-E/40aa08ca83c29a9281c0096fb9335f55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..b06918b58787a83cc786934cf0df5d6185378f0c --- /dev/null +++ b/marked/H/T-REC-H.430.3-202309-I_PDF-E/40aa08ca83c29a9281c0096fb9335f55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d7e9efa508659b0b59fe2dbbe0f7ebc4092701e54dfa31452e07cf67707e54bb +size 23378 diff --git a/marked/H/T-REC-H.430.3-202309-I_PDF-E/8c348bf9c2c81b018017ae1d19506a9a_img.jpg b/marked/H/T-REC-H.430.3-202309-I_PDF-E/8c348bf9c2c81b018017ae1d19506a9a_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d5b093dc5232923593b5c970310bac6c19a915c8 --- /dev/null +++ b/marked/H/T-REC-H.430.3-202309-I_PDF-E/8c348bf9c2c81b018017ae1d19506a9a_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8d3cdbe3ec06235d55be453f8748751216460470901fa0891e21547080ec4597 +size 60388 diff --git a/marked/H/T-REC-H.430.3-202309-I_PDF-E/9c1d3678db4a12d5864cb2a4def1135d_img.jpg b/marked/H/T-REC-H.430.3-202309-I_PDF-E/9c1d3678db4a12d5864cb2a4def1135d_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a4499f974997cd17391860b202d0ecf21beae203 --- /dev/null +++ b/marked/H/T-REC-H.430.3-202309-I_PDF-E/9c1d3678db4a12d5864cb2a4def1135d_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:942c25d838383da417c453b7cf08195d810b66e5007f44496253e6553b45e03c +size 23694 diff --git a/marked/H/T-REC-H.430.3-202309-I_PDF-E/a0739aaf13fa5a632d4faa830f6b2708_img.jpg b/marked/H/T-REC-H.430.3-202309-I_PDF-E/a0739aaf13fa5a632d4faa830f6b2708_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..0bfae14e41d729f542f527e18db17051425cc411 --- /dev/null +++ b/marked/H/T-REC-H.430.3-202309-I_PDF-E/a0739aaf13fa5a632d4faa830f6b2708_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2590e2e3fc0d13661b663aff887a87fe535e79d934b29758df047030420a31ab +size 71625 diff --git a/marked/H/T-REC-H.430.3-202309-I_PDF-E/a251a846ffb8c1750262be87b489eeec_img.jpg b/marked/H/T-REC-H.430.3-202309-I_PDF-E/a251a846ffb8c1750262be87b489eeec_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..038da0a503b8bb121b8c9bfac9ec09011615630c --- /dev/null +++ b/marked/H/T-REC-H.430.3-202309-I_PDF-E/a251a846ffb8c1750262be87b489eeec_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4159e974c2ddc07acbc397c352f1b8b29bfaa51009888235f00554aa74c466a6 +size 26384 diff --git a/marked/H/T-REC-H.430.3-202309-I_PDF-E/aaf3e6e44cdeabd6d1df869c5f392ea1_img.jpg b/marked/H/T-REC-H.430.3-202309-I_PDF-E/aaf3e6e44cdeabd6d1df869c5f392ea1_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..cbbd19f3c74f30bec9573dd855c35620e9f87c00 --- /dev/null +++ b/marked/H/T-REC-H.430.3-202309-I_PDF-E/aaf3e6e44cdeabd6d1df869c5f392ea1_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c5551d48625d89f19cc0445b24ba48b4bcb43fa6771cf29dbe6e446aefb3ad51 +size 102640 diff --git a/marked/H/T-REC-H.430.3-202309-I_PDF-E/d9c0a780cd22626253dab4aa41699e2f_img.jpg b/marked/H/T-REC-H.430.3-202309-I_PDF-E/d9c0a780cd22626253dab4aa41699e2f_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..eca66794a877c961b6d3488b5b46e9e9f4d78a8b --- /dev/null +++ b/marked/H/T-REC-H.430.3-202309-I_PDF-E/d9c0a780cd22626253dab4aa41699e2f_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e61177e9d645ec980637d1b96f79a1d687a6837782041cee531c905156b7da03 +size 58701 diff --git a/marked/H/T-REC-H.430.3-202309-I_PDF-E/e9a1dfde7b55d69a0a56f5bbe35328d4_img.jpg b/marked/H/T-REC-H.430.3-202309-I_PDF-E/e9a1dfde7b55d69a0a56f5bbe35328d4_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ba824ea299f4211c088ef98cb615cdf229d95fbe --- /dev/null +++ b/marked/H/T-REC-H.430.3-202309-I_PDF-E/e9a1dfde7b55d69a0a56f5bbe35328d4_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:09cd67ca8ba2d0cdbadcd62f924cceca6c25991d0528aa56dc53dac04a2c2ce1 +size 41190 diff --git a/marked/H/T-REC-H.430.3-202309-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/H/T-REC-H.430.3-202309-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..b267271888888e9fdd8581b8aca6b9284fae5762 --- /dev/null +++ b/marked/H/T-REC-H.430.3-202309-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cf797318e117ce235884e71b9b574cc33aaf516a072f2422bebb000c5c70d928 +size 109453 diff --git a/marked/H/T-REC-H.460.10-200403-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/H/T-REC-H.460.10-200403-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..939eb6a57fac47a375526475ba6c31bf67e08b30 --- /dev/null +++ b/marked/H/T-REC-H.460.10-200403-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:52a289bbb49d106b16f646b5050f62e5c7cbe1bb201f8c767547c3039abbf359 +size 8245 diff --git a/marked/H/T-REC-H.460.16-200501-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg b/marked/H/T-REC-H.460.16-200501-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..36cf51f8ba07e44aa7a4b009f3ba08eb2b570b1c --- /dev/null +++ b/marked/H/T-REC-H.460.16-200501-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:11780faeee75ba39ab4ed9de67a9e67abec0d9ba3068ebec661dea70c16e29d3 +size 4338 diff --git a/marked/H/T-REC-H.460.16-200501-I_PDF-E/1b5a812c8aa20fd5cba28e97001d32de_img.jpg b/marked/H/T-REC-H.460.16-200501-I_PDF-E/1b5a812c8aa20fd5cba28e97001d32de_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..599be445a86a510c7f608ad239223c1cf75a7cd7 --- /dev/null +++ b/marked/H/T-REC-H.460.16-200501-I_PDF-E/1b5a812c8aa20fd5cba28e97001d32de_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ceeee029341c1b3906fc3c87a0c34e83c480b9e58df2dc46acf91d242bd72080 +size 39125 diff --git a/marked/H/T-REC-H.460.16-200501-I_PDF-E/4ee27dbf5ef12e7b58b0ef0937bc5a5e_img.jpg b/marked/H/T-REC-H.460.16-200501-I_PDF-E/4ee27dbf5ef12e7b58b0ef0937bc5a5e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..192421e16cf7104e47183e9b1c858406e6682f97 --- /dev/null +++ b/marked/H/T-REC-H.460.16-200501-I_PDF-E/4ee27dbf5ef12e7b58b0ef0937bc5a5e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7bc63ff66b122112ead0c4cf05eb17df48ee55f9f44d8dfafd8bc7e9b93da4a6 +size 41168 diff --git a/marked/H/T-REC-H.460.16-200501-I_PDF-E/724c7777b608e53be38b12b6fb3c43bc_img.jpg b/marked/H/T-REC-H.460.16-200501-I_PDF-E/724c7777b608e53be38b12b6fb3c43bc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..bdb07c86956f03df168c550bea0eba6bc1dda10a --- /dev/null +++ b/marked/H/T-REC-H.460.16-200501-I_PDF-E/724c7777b608e53be38b12b6fb3c43bc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9934c5b52ba9623ab63c4f1e0feb62fb1de3619d07bcf30e722226ab32e2ea8f +size 34965 diff --git a/marked/H/T-REC-H.460.16-200501-I_PDF-E/78ffccd66df9bafd96e3e081110d09dd_img.jpg b/marked/H/T-REC-H.460.16-200501-I_PDF-E/78ffccd66df9bafd96e3e081110d09dd_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..345484895d59af6a53ea9d4392e10f66e28d9cef --- /dev/null +++ b/marked/H/T-REC-H.460.16-200501-I_PDF-E/78ffccd66df9bafd96e3e081110d09dd_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b5271fc0d414a118fac217003778fd7bdc80a0499ec007b3626cd5473a9b46ea +size 31780 diff --git a/marked/H/T-REC-H.460.16-200501-I_PDF-E/ddc7460821484f1ae2835c67955c554c_img.jpg b/marked/H/T-REC-H.460.16-200501-I_PDF-E/ddc7460821484f1ae2835c67955c554c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..55b073ccdab597c6a50f127bc4516909ffac56c6 --- /dev/null +++ b/marked/H/T-REC-H.460.16-200501-I_PDF-E/ddc7460821484f1ae2835c67955c554c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aed29399f07202a581a376299836e36969771711795668ba72e323c910e59c2b +size 24067 diff --git a/marked/H/T-REC-H.460.16-200501-I_PDF-E/eefe19c5e14dc4d6c316b7f7fbb7d7d7_img.jpg b/marked/H/T-REC-H.460.16-200501-I_PDF-E/eefe19c5e14dc4d6c316b7f7fbb7d7d7_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..840fc20102ec5ab55ea5346ae3e6c5f91e8b4167 --- /dev/null +++ b/marked/H/T-REC-H.460.16-200501-I_PDF-E/eefe19c5e14dc4d6c316b7f7fbb7d7d7_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a99d8255c102ee6d7123518bcf30f94ad558f31221fd16d654287041299b2092 +size 58635 diff --git a/marked/H/T-REC-H.460.17-200509-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg b/marked/H/T-REC-H.460.17-200509-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..36cf51f8ba07e44aa7a4b009f3ba08eb2b570b1c --- /dev/null +++ b/marked/H/T-REC-H.460.17-200509-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:11780faeee75ba39ab4ed9de67a9e67abec0d9ba3068ebec661dea70c16e29d3 +size 4338 diff --git a/marked/H/T-REC-H.460.5-200211-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/H/T-REC-H.460.5-200211-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..939eb6a57fac47a375526475ba6c31bf67e08b30 --- /dev/null +++ b/marked/H/T-REC-H.460.5-200211-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:52a289bbb49d106b16f646b5050f62e5c7cbe1bb201f8c767547c3039abbf359 +size 8245 diff --git a/marked/H/T-REC-H.530-200203-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg b/marked/H/T-REC-H.530-200203-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..939eb6a57fac47a375526475ba6c31bf67e08b30 --- /dev/null +++ b/marked/H/T-REC-H.530-200203-I_PDF-E/2dfa6ac3edfe874f68aa0cbccaa42322_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:52a289bbb49d106b16f646b5050f62e5c7cbe1bb201f8c767547c3039abbf359 +size 8245 diff --git a/marked/H/T-REC-H.530-200203-I_PDF-E/410562339ce067fdc6fa41940c118658_img.jpg b/marked/H/T-REC-H.530-200203-I_PDF-E/410562339ce067fdc6fa41940c118658_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..6f14b9e92948fb0e08f2793b3eab5ddd02562116 --- /dev/null +++ b/marked/H/T-REC-H.530-200203-I_PDF-E/410562339ce067fdc6fa41940c118658_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:678a02197653f107232e372d49cc325670932cf599effb519bfb1d5eb6e5be7a +size 81476 diff --git a/marked/H/T-REC-H.530-200203-I_PDF-E/4cde160bcc69b7b6c81b648dd0e4252e_img.jpg b/marked/H/T-REC-H.530-200203-I_PDF-E/4cde160bcc69b7b6c81b648dd0e4252e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..5c8c5a92ad87dde6e2216412c80208bca2969a7d --- /dev/null +++ b/marked/H/T-REC-H.530-200203-I_PDF-E/4cde160bcc69b7b6c81b648dd0e4252e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0f402877428c76ae15657164a3652a4bda96ab07cfc84e6fbe5042739ea64a0d +size 91814 diff --git a/marked/H/T-REC-H.530-200203-I_PDF-E/5b8a756d9a71c35f17db8bcb90b438a3_img.jpg b/marked/H/T-REC-H.530-200203-I_PDF-E/5b8a756d9a71c35f17db8bcb90b438a3_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..13e487e4c0c9be8987a7410a58794af1b5c33269 --- /dev/null +++ b/marked/H/T-REC-H.530-200203-I_PDF-E/5b8a756d9a71c35f17db8bcb90b438a3_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:28bca76e3e931f236686de94be893aebaa7b67f2675ed27b21467cb2da385216 +size 63780 diff --git a/marked/H/T-REC-H.530-200203-I_PDF-E/724c7777b608e53be38b12b6fb3c43bc_img.jpg b/marked/H/T-REC-H.530-200203-I_PDF-E/724c7777b608e53be38b12b6fb3c43bc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..b415348aa38bcb302d6b2b8fd245dcef39a80d80 --- /dev/null +++ b/marked/H/T-REC-H.530-200203-I_PDF-E/724c7777b608e53be38b12b6fb3c43bc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1b567a089a4a35d87ec2e0ac4890678527529589e90782ed2ac6684e55599ab3 +size 102087 diff --git a/marked/H/T-REC-H.530-200203-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg b/marked/H/T-REC-H.530-200203-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..58b5926805664f73fa0f61030016d5a4bc652f9a --- /dev/null +++ b/marked/H/T-REC-H.530-200203-I_PDF-E/76b0cd79baaedd942af4dc42f2e764b8_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8d84d489a9e3022fb86d1e3e02034133da14d98b24c84cd6933a30f183cc2c89 +size 81475 diff --git a/marked/H/T-REC-H.530-200203-I_PDF-E/86b4670fc1a5a694821ee92b99c1209a_img.jpg b/marked/H/T-REC-H.530-200203-I_PDF-E/86b4670fc1a5a694821ee92b99c1209a_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ac1ce4470047a85e6df8df8e538f4b3dda784b79 --- /dev/null +++ b/marked/H/T-REC-H.530-200203-I_PDF-E/86b4670fc1a5a694821ee92b99c1209a_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:afd3ef5e5bc1ae150a6cd14705add498e713bf93067b7baf2bed7e1aec66d5ed +size 47148 diff --git a/marked/H/T-REC-H.530-200203-I_PDF-E/90ddb84c323b956e2d50a54d3f870566_img.jpg b/marked/H/T-REC-H.530-200203-I_PDF-E/90ddb84c323b956e2d50a54d3f870566_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ea4c0d25a8f35596f544bc356ab28ee43e481b82 --- /dev/null +++ b/marked/H/T-REC-H.530-200203-I_PDF-E/90ddb84c323b956e2d50a54d3f870566_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:865fd6f377ad0fb3b228d0ae1780c80cfd27142d8bc5238437a120219c78483e +size 65567 diff --git a/marked/H/T-REC-H.530-200203-I_PDF-E/b2ea162a0f53d5e0504b7d28346e0754_img.jpg b/marked/H/T-REC-H.530-200203-I_PDF-E/b2ea162a0f53d5e0504b7d28346e0754_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..63c91eeb35e0b02961faa5f64a9101bc64043ab5 --- /dev/null +++ b/marked/H/T-REC-H.530-200203-I_PDF-E/b2ea162a0f53d5e0504b7d28346e0754_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:647db43146e5bd6c491c391cdb89b0a499e2228774dc8bac5eef26f680d8b773 +size 44079 diff --git a/marked/H/T-REC-H.530-200203-I_PDF-E/c5452f95f3b28f1bfe29e84fbc2e1267_img.jpg b/marked/H/T-REC-H.530-200203-I_PDF-E/c5452f95f3b28f1bfe29e84fbc2e1267_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..bcd0f975ac9f3d88fa38d55ac16446034a350676 --- /dev/null +++ b/marked/H/T-REC-H.530-200203-I_PDF-E/c5452f95f3b28f1bfe29e84fbc2e1267_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:03d5ba8cd7d2381ff3fb12cf516e6375fa0cedc49010b9dcea8dee697117d0b8 +size 44404 diff --git a/marked/H/T-REC-H.530-200203-I_PDF-E/e69b9188aa2c14ec6b21c83f711fef65_img.jpg b/marked/H/T-REC-H.530-200203-I_PDF-E/e69b9188aa2c14ec6b21c83f711fef65_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..fb74241b279a9d67036843f894184669e50ba606 --- /dev/null +++ b/marked/H/T-REC-H.530-200203-I_PDF-E/e69b9188aa2c14ec6b21c83f711fef65_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c796b532a2572177802ad8c17baa5a9e553c30d1ca5896f9ec138130e73ea696 +size 45533 diff --git a/marked/H/T-REC-H.530-200203-I_PDF-E/fa859e4e468bfb2710a94527f2c504af_img.jpg b/marked/H/T-REC-H.530-200203-I_PDF-E/fa859e4e468bfb2710a94527f2c504af_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..0b38852f9b6af14e59e7aed36e74455b2fc8fe31 --- /dev/null +++ b/marked/H/T-REC-H.530-200203-I_PDF-E/fa859e4e468bfb2710a94527f2c504af_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:192f9e8f7143803d8fe07719779732b99daccf71114e48ad1c2dc45c30346e22 +size 101450 diff --git a/marked/H/T-REC-H.553-202512-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/H/T-REC-H.553-202512-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c1629394c02f697f60ae954ad36bf274fa8082cb --- /dev/null +++ b/marked/H/T-REC-H.553-202512-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3763acd7a0ed56aa30f3f90468c1fb0cb54573eb9045bca3eddf8976b4f0f41d +size 7219 diff --git a/marked/H/T-REC-H.553-202512-I_PDF-E/acfc53eca625d62b38aa2563efa95c3e_img.jpg b/marked/H/T-REC-H.553-202512-I_PDF-E/acfc53eca625d62b38aa2563efa95c3e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..6645cbe9846d7483360df2dfd62b44743b2b1f13 --- /dev/null +++ b/marked/H/T-REC-H.553-202512-I_PDF-E/acfc53eca625d62b38aa2563efa95c3e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ef7e265302728c43a0e19fdc136b672697c8183cc69757ae16ce19cbed3478c0 +size 40296 diff --git a/marked/H/T-REC-H.553-202512-I_PDF-E/f4fdd410cdb84df81274da55721e56fb_img.jpg b/marked/H/T-REC-H.553-202512-I_PDF-E/f4fdd410cdb84df81274da55721e56fb_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..fa4fb0ce6cec5f8501b4e9a0bb07681c16d6c4ea --- /dev/null +++ b/marked/H/T-REC-H.553-202512-I_PDF-E/f4fdd410cdb84df81274da55721e56fb_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dac7f810d94e625dcb5057d0987660d63071e4aabfa9f8c15b974e537f584ad6 +size 184053 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9002ee68e2b9ebf63d27d39ba8c73bdafb0e6be6 --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/1d7527f4316cfe2d342b08d1653d1592_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8df701536c25ab8246d2868de3c20c8a4e25168bb52c5eb9078278ab1b5b4fd8 +size 6859 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/20cb2d2cc18d663dc1b82453c67250bb_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/20cb2d2cc18d663dc1b82453c67250bb_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..41b8dde11bc13f268fead1806ca2f8a4991d8056 --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/20cb2d2cc18d663dc1b82453c67250bb_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ce866750df1db845d6aa612071da43ecb338ff047fc6a2800d41521d27db29c6 +size 21874 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/2119293733c74fc64de88aefd597f4bb_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/2119293733c74fc64de88aefd597f4bb_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..2a69f661c3ddc78180f93d27ec52ca95592dbc41 --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/2119293733c74fc64de88aefd597f4bb_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f98d833a7a80befd03f8c015a358fe194cd86f650712cababf3865d95995cf50 +size 18670 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/2eb73eee0ca246cd09f43c2044f3fb49_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/2eb73eee0ca246cd09f43c2044f3fb49_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..8c3e4082c944ad1a500bb515a3349b7e73b5134c --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/2eb73eee0ca246cd09f43c2044f3fb49_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a45d1f98ff1a40f7e6b11a84f9e35789c1512e6cf23ea681b6f34fbeb4314625 +size 18846 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/3d1817e8551f9c226a5f561108d3b3f9_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/3d1817e8551f9c226a5f561108d3b3f9_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..aaf0ecf3d8100f0ac70a8ef7a6181d1349bd0c4a --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/3d1817e8551f9c226a5f561108d3b3f9_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b62814e6d876319fb4b5e92176f4b07a67fa872bce4fd8da89bc55f6721ec51d +size 108332 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/46357bd4781f8f8f776667fb096f54a9_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/46357bd4781f8f8f776667fb096f54a9_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..8f162692e2505e5b95c6eed70f6190c220b70003 --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/46357bd4781f8f8f776667fb096f54a9_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6ef2949c2663e387605342d4b69fa16a8e738d17da563fcf469f19ef99396625 +size 18607 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/474a819357587e34949a3e110ff19b30_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/474a819357587e34949a3e110ff19b30_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7467b8d44f93b5bcc6a8dc15091a2af591f0492c --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/474a819357587e34949a3e110ff19b30_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cb712b0076482c4f8c688515456d34155960c078b70d7aa311e60e4780fb7336 +size 98549 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/5e92d9e8e9ce204e405bff2367f88176_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/5e92d9e8e9ce204e405bff2367f88176_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9724beb17e24b7fe316c6b7db936569d447b8a52 --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/5e92d9e8e9ce204e405bff2367f88176_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0137e3a1e1e2bb202b214a1f8b4c8d82b9633708285eb6e7b61a3a62b93cddc5 +size 41506 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/701bc79e78b382bcfd3ba85597dbb9c3_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/701bc79e78b382bcfd3ba85597dbb9c3_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..5edee0c13b7f785ce4aaa81d3495d8523db97e64 --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/701bc79e78b382bcfd3ba85597dbb9c3_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e5f970238776fb27ea775e58ff882c9be165ea9f7d02817332b92a80a215dfb9 +size 53103 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/75e4b78ee25f885d73120e3066a5253e_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/75e4b78ee25f885d73120e3066a5253e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..8258da428f03de35003732b16ded051a407815ee --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/75e4b78ee25f885d73120e3066a5253e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e69a1bf104db9b9d76c5866af8fa1cdaf8f44385176e0c2af9235fb9fca6a4fb +size 29961 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/7e1c9b51e067a48cd0fcc9748d8bd8d8_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/7e1c9b51e067a48cd0fcc9748d8bd8d8_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a291ce9e19656bf91e7723b3350436823847e8cf --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/7e1c9b51e067a48cd0fcc9748d8bd8d8_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3e7fc629f535fa418059274bab7ad63ac3631117ca1e5e3cc49a8f15a7327bf1 +size 21864 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/9252ccfbbe9e34cb108f0060f2b563f1_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/9252ccfbbe9e34cb108f0060f2b563f1_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ec73fc7af980162f28370b4fd3a094de1ac7a43a --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/9252ccfbbe9e34cb108f0060f2b563f1_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7b035b8b7c6cdbfe8082c61bc2de9bdab4f5a7abe9dbdb2038be6240810c54a1 +size 56253 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/92e26d86c5a582766cb9afdbd2c1aee7_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/92e26d86c5a582766cb9afdbd2c1aee7_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..dc4c90d925fef28f727f917b60d30c461de558a7 --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/92e26d86c5a582766cb9afdbd2c1aee7_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e96a4e44c51ae2185b9b4f4b9e4c19c1367e3d51dbdb5fc39cd047fc2460e115 +size 16003 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/a83ba9e3e2c1e21dd69953a7b09e45b4_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/a83ba9e3e2c1e21dd69953a7b09e45b4_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..25f8d04e3d9b346cf3ee0c95119be571daefd42e --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/a83ba9e3e2c1e21dd69953a7b09e45b4_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:91aa99158a137a5bce3633a1a699d269f412d97dfbe2556ce7047399009279ce +size 11679 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/a97518a839da75f8379c578562b01bc2_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/a97518a839da75f8379c578562b01bc2_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..2f98ba14945504527bdc87c5b240739f69332b3c --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/a97518a839da75f8379c578562b01bc2_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:77b63ceacac0657008653f461bde9b56cd57b05ba0a1a23b2a764303a8fbd041 +size 55979 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/b54b769d67164910c7c5b2a0ea74257b_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/b54b769d67164910c7c5b2a0ea74257b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c8d2e9cfac2140aebe7faa48101666bdf6774b42 --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/b54b769d67164910c7c5b2a0ea74257b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b0702fea5767c8d671f3a520c99cffc82ce6f1f7e7cb96690da34d93b466c074 +size 117666 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/caeb1637753f79f4cd005eda3768d1d5_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/caeb1637753f79f4cd005eda3768d1d5_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..bf0c5a46cb0f4295432b06eeb3d539f0aea52d79 --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/caeb1637753f79f4cd005eda3768d1d5_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0d196ffc4b1d90db77c6a682e4dc361db57a3ba365c6256ccaec52cb853a3fe7 +size 127556 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/ccb9a8830d90ffdcecfc4ef45ce72dbe_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/ccb9a8830d90ffdcecfc4ef45ce72dbe_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..315d0e6f2949cf7a3893bcfd6bd3d01cd70a6115 --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/ccb9a8830d90ffdcecfc4ef45ce72dbe_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e453566aa9a793cf51c7a06ece1ca69cd9c24f469556ae4822323603479d228d +size 17047 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..7451ab220936c5f14096e9e475c39c6527b60103 --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:66c84a782afe30a24267f8e94801d44e71ab812432b8ae33d67d25458df3c961 +size 33095 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/d04c50badc78d5ba47bf4e352af4a754_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/d04c50badc78d5ba47bf4e352af4a754_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9c4dc428038f872d103d7a8bd97a864cfe5b5aa3 --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/d04c50badc78d5ba47bf4e352af4a754_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f7853e7a83ea08e18df3d4dd660d71675854b811d9bcd5b60c9d7b4b32361cac +size 116437 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/d53cd0fd1cf896a9353fd63de1505ba2_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/d53cd0fd1cf896a9353fd63de1505ba2_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..b9a499a0bcc270830d94f228d25acda2f6b11c3d --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/d53cd0fd1cf896a9353fd63de1505ba2_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7eba823afe1ea7fb941c9fb3d59b40229f1a6779b04d1757138ed075a4a7280d +size 30679 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/e404de9a69c05473fbd6af28ee32311b_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/e404de9a69c05473fbd6af28ee32311b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f29d92fa4201a8a0e83cf998660930d7dc70e5af --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/e404de9a69c05473fbd6af28ee32311b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bb27873c4b64fe899d13234de1cfaaa2b2a22e602f53768eb603eb0eef78b6ff +size 17754 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/e4aab374983ba0fc24496fe7cd11c06c_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/e4aab374983ba0fc24496fe7cd11c06c_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d0a15ce60ce816f754031dd13c17153697af25f2 --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/e4aab374983ba0fc24496fe7cd11c06c_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cb97dba26f72d099a28e564fa0f8b9a10c932d2e6c4cc372cb17108ac99a5df6 +size 61528 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/e8f5117fb031f413a836525061addd61_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/e8f5117fb031f413a836525061addd61_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..68c661beae5a36aabd1e5dcf57ca7b5db148d020 --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/e8f5117fb031f413a836525061addd61_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:46f74595b8dae5b2092d84e6f73a9bcafa0b27e8017b2b6099c3160a9d94c434 +size 114182 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/e90987faabad6a6665cd8ed1151dc474_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/e90987faabad6a6665cd8ed1151dc474_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..180ea3243582cf84b426da546ab7bc32bb25590d --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/e90987faabad6a6665cd8ed1151dc474_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9559e8884eeb85c2b8f62db84e97656e62284a4501f25b3b8b24deae95eac6dd +size 94670 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/f59dafee4c3bbc563111e5b1b6048c3d_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/f59dafee4c3bbc563111e5b1b6048c3d_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1f7a75329ddfea7dc43a1747f0681764b66578d3 --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/f59dafee4c3bbc563111e5b1b6048c3d_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6887a3a18fd10bd25150a4b5f503bba1f57b4f4c1df932fb592e14223cf8f83b +size 17997 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/f943b07ab747bfd85e302e5a31c20ba8_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/f943b07ab747bfd85e302e5a31c20ba8_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ca7806727ae96b8a30e5b60370f47ce1c229fe1f --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/f943b07ab747bfd85e302e5a31c20ba8_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:09a9da5383ea3a13cdad146307caf38c0f7d972e5251d42c0d941d868db50219 +size 30758 diff --git a/marked/H/T-REC-H.627-202008-I_PDF-E/fd76efce549d3713543bb5ed9b023c2e_img.jpg b/marked/H/T-REC-H.627-202008-I_PDF-E/fd76efce549d3713543bb5ed9b023c2e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..6086285db57a4e853c102574d960d51625a466ee --- /dev/null +++ b/marked/H/T-REC-H.627-202008-I_PDF-E/fd76efce549d3713543bb5ed9b023c2e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:da418369db2a23fafc46104e720575709d12d12347b716cbeec1bbb4a8bdd404 +size 54194 diff --git a/marked/H/T-REC-H.705.2-202309-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/H/T-REC-H.705.2-202309-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ac295b1b0c7f14b1c5a33ef6682b1b017c812477 --- /dev/null +++ b/marked/H/T-REC-H.705.2-202309-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37aed4fc114a39843438175326e0b7e1d09a9d6075723d82fea32040fd560ee4 +size 7192 diff --git a/marked/H/T-REC-H.705.2-202309-I_PDF-E/0f985b39edc1d52ba3600c438bc8f0a5_img.jpg b/marked/H/T-REC-H.705.2-202309-I_PDF-E/0f985b39edc1d52ba3600c438bc8f0a5_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3f1173ed1e4889b310c741b2ae3c4b7de1e0d5eb --- /dev/null +++ b/marked/H/T-REC-H.705.2-202309-I_PDF-E/0f985b39edc1d52ba3600c438bc8f0a5_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1ead26c1b0490590a9a4c9b85817bad2279f6d59255dbbed9bb59ca5021931b7 +size 54656 diff --git a/marked/H/T-REC-H.705.2-202309-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg b/marked/H/T-REC-H.705.2-202309-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..54bd9f3cf4456cdc6232a471fa605a03e3669561 --- /dev/null +++ b/marked/H/T-REC-H.705.2-202309-I_PDF-E/33ed1f9b27c7c21c797aa928b0f06851_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c3561ecc37a2a2c4d6a516c4a6a5477d9c711e2e7416c261b6b00b9012aa3312 +size 47579 diff --git a/marked/H/T-REC-H.705.2-202309-I_PDF-E/36117e9cb27c58484cd8d3e5f9dc7ac3_img.jpg b/marked/H/T-REC-H.705.2-202309-I_PDF-E/36117e9cb27c58484cd8d3e5f9dc7ac3_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9ed9dc08752cd676daa5a9e4231f1766bea70198 --- /dev/null +++ b/marked/H/T-REC-H.705.2-202309-I_PDF-E/36117e9cb27c58484cd8d3e5f9dc7ac3_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ae1dd64f8a52492371fbd1e5ad1422ebc8b5f323300929d032d72f46a22588ad +size 17076 diff --git a/marked/H/T-REC-H.705.2-202309-I_PDF-E/3da1a07cb87051bf616c9876db958cf0_img.jpg b/marked/H/T-REC-H.705.2-202309-I_PDF-E/3da1a07cb87051bf616c9876db958cf0_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..50df0fcffcc6142d29ea9e10a5ece178c86af50a --- /dev/null +++ b/marked/H/T-REC-H.705.2-202309-I_PDF-E/3da1a07cb87051bf616c9876db958cf0_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:28a4acb9f3d69e81b2c2097ba37ee2794f635020cbe2852a73fe9ed74507f47e +size 71250 diff --git a/marked/H/T-REC-H.705.2-202309-I_PDF-E/4cc7cdce3d498d8b0ba033a9be24ade5_img.jpg b/marked/H/T-REC-H.705.2-202309-I_PDF-E/4cc7cdce3d498d8b0ba033a9be24ade5_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..2866218962877d9e36a74c020e885e2d06cf18b6 --- /dev/null +++ b/marked/H/T-REC-H.705.2-202309-I_PDF-E/4cc7cdce3d498d8b0ba033a9be24ade5_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:166a6c740a999ae58004d335daa7935c6b0dba7ab7875bdd4bd8c2249ae797b3 +size 45270 diff --git a/marked/H/T-REC-H.705.2-202309-I_PDF-E/4ee27dbf5ef12e7b58b0ef0937bc5a5e_img.jpg b/marked/H/T-REC-H.705.2-202309-I_PDF-E/4ee27dbf5ef12e7b58b0ef0937bc5a5e_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..716a7873962d94425e1db04cafcff0bd4a61fd9e --- /dev/null +++ b/marked/H/T-REC-H.705.2-202309-I_PDF-E/4ee27dbf5ef12e7b58b0ef0937bc5a5e_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aad61015dc3af0f8ab4d308342efeba92e63d30fd5355da37197e7b02cea0ce8 +size 54192 diff --git a/marked/H/T-REC-H.705.2-202309-I_PDF-E/5b8a756d9a71c35f17db8bcb90b438a3_img.jpg b/marked/H/T-REC-H.705.2-202309-I_PDF-E/5b8a756d9a71c35f17db8bcb90b438a3_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..dd4a0f68b2ecd10641dc95872f340ee65285d742 --- /dev/null +++ b/marked/H/T-REC-H.705.2-202309-I_PDF-E/5b8a756d9a71c35f17db8bcb90b438a3_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9580ecf8b3b3c52423724798d94bc99f7a6cfc867f5af4eb568e73fcb9ac40b9 +size 66063 diff --git a/marked/H/T-REC-H.705.2-202309-I_PDF-E/5dfc130b129ace4df375839020a5700d_img.jpg b/marked/H/T-REC-H.705.2-202309-I_PDF-E/5dfc130b129ace4df375839020a5700d_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c161f7fac3f1dc19085ba3860ed5511f008066fb --- /dev/null +++ b/marked/H/T-REC-H.705.2-202309-I_PDF-E/5dfc130b129ace4df375839020a5700d_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:06af5e5f4843a16f3e0204a92d6b335c3b09b06e13b270f3abd3832659357cad +size 50774 diff --git a/marked/H/T-REC-H.705.2-202309-I_PDF-E/9c6461e1e94afae4dec455e69a2ce152_img.jpg b/marked/H/T-REC-H.705.2-202309-I_PDF-E/9c6461e1e94afae4dec455e69a2ce152_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..6af7da16a7cb22659c3745c17bfc3fb4c069ed84 --- /dev/null +++ b/marked/H/T-REC-H.705.2-202309-I_PDF-E/9c6461e1e94afae4dec455e69a2ce152_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6bc44426f5a5b921b6f74e5ae8b2d79620e29f641a65641c1d0e3403766d8d6c +size 20731 diff --git a/marked/H/T-REC-H.705.2-202309-I_PDF-E/9cd90f495b95ad2116ff780248c26d95_img.jpg b/marked/H/T-REC-H.705.2-202309-I_PDF-E/9cd90f495b95ad2116ff780248c26d95_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..6857332501a6cbd752f621b0ea45e6940a633a82 --- /dev/null +++ b/marked/H/T-REC-H.705.2-202309-I_PDF-E/9cd90f495b95ad2116ff780248c26d95_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dfd1fe4ee1b4e8d7cf1efbf11c02913b7fea889f410ff8430dbb71c34b39d793 +size 18805 diff --git a/marked/H/T-REC-H.705.2-202309-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg b/marked/H/T-REC-H.705.2-202309-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..8c20fe3e0e33ae86ed6a3a5a6865fd2913ae3e92 --- /dev/null +++ b/marked/H/T-REC-H.705.2-202309-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:883a72d97867fff4d3300ee87cb4fedab219d722a5533e2f493bb2a50079d8db +size 121273 diff --git a/marked/H/T-REC-H.705.2-202309-I_PDF-E/f4fdd410cdb84df81274da55721e56fb_img.jpg b/marked/H/T-REC-H.705.2-202309-I_PDF-E/f4fdd410cdb84df81274da55721e56fb_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..12a73bda508e3e104a53d798093919e53a43b377 --- /dev/null +++ b/marked/H/T-REC-H.705.2-202309-I_PDF-E/f4fdd410cdb84df81274da55721e56fb_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7afdfa9c25ebc9c92ee8e420e14b3693128568b9125e7b1fb0acb83537320c73 +size 63791 diff --git a/marked/H/T-REC-H.741.2-201206-I_PDF-E/18442e4e239480f0c3c95b547aa8fde2_img.jpg b/marked/H/T-REC-H.741.2-201206-I_PDF-E/18442e4e239480f0c3c95b547aa8fde2_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c6409056fbea9ce04c5fae81653ad7816cf0b77f --- /dev/null +++ b/marked/H/T-REC-H.741.2-201206-I_PDF-E/18442e4e239480f0c3c95b547aa8fde2_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cafdabf6f686e55e539459636df22882edd2413fb0a242101acd897df3457e0a +size 4294 diff --git a/marked/H/T-REC-H.741.2-201206-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg b/marked/H/T-REC-H.741.2-201206-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..b310ac210c3a7914165adf06403f98576a85c994 --- /dev/null +++ b/marked/H/T-REC-H.741.2-201206-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0953d8a48897d2f76d40d07a4f72e9fd2630b3ba8f1e39b9ba5bc493583058ef +size 3635 diff --git a/marked/H/T-REC-H.741.2-201206-I_PDF-E/c78c2eefd86269d1740ab85a916f24f2_img.jpg b/marked/H/T-REC-H.741.2-201206-I_PDF-E/c78c2eefd86269d1740ab85a916f24f2_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..dff60ef7008131bae51e278e1f54704d5c44844b --- /dev/null +++ b/marked/H/T-REC-H.741.2-201206-I_PDF-E/c78c2eefd86269d1740ab85a916f24f2_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3d4bab8977d9da048bd52bf2bbff02df0e2426d4ee0b594490f5e04bd320ce15 +size 3265 diff --git a/marked/H/T-REC-H.751-201303-I_PDF-E/08dce7ad4c512fdf0c0cde60415fade6_img.jpg b/marked/H/T-REC-H.751-201303-I_PDF-E/08dce7ad4c512fdf0c0cde60415fade6_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..1f1d2d3b26dced57341c8967acd452c5ffa4dcf5 --- /dev/null +++ b/marked/H/T-REC-H.751-201303-I_PDF-E/08dce7ad4c512fdf0c0cde60415fade6_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:292afb9035e8c976e9cb360458ffe2fe7e15ebdec6e53b599242ecf4eb3a63b0 +size 76952 diff --git a/marked/H/T-REC-H.751-201303-I_PDF-E/0eb742ed939b1846d05da644664fa9b7_img.jpg b/marked/H/T-REC-H.751-201303-I_PDF-E/0eb742ed939b1846d05da644664fa9b7_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..029c1a15ca6c8c34393354dc2f76a439e2913494 --- /dev/null +++ b/marked/H/T-REC-H.751-201303-I_PDF-E/0eb742ed939b1846d05da644664fa9b7_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6013925da3e0b6f19d4ab09b2835436111e7210d22240fbce8ea6b07472fb400 +size 755 diff --git a/marked/H/T-REC-H.751-201303-I_PDF-E/11edb7fcedf09ac6a817f8d7b8c61eec_img.jpg b/marked/H/T-REC-H.751-201303-I_PDF-E/11edb7fcedf09ac6a817f8d7b8c61eec_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..3779626c7379b946662db53418663a45e674ce43 --- /dev/null +++ b/marked/H/T-REC-H.751-201303-I_PDF-E/11edb7fcedf09ac6a817f8d7b8c61eec_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6e77f8c8cb4ef5ba9b7595bab15521a4d4f09ac959e9e59b9e7a2cbffdaa2e1c +size 149510 diff --git a/marked/H/T-REC-H.751-201303-I_PDF-E/5b8a756d9a71c35f17db8bcb90b438a3_img.jpg b/marked/H/T-REC-H.751-201303-I_PDF-E/5b8a756d9a71c35f17db8bcb90b438a3_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d5a87f188a3c438789998f93055227df1b378b56 --- /dev/null +++ b/marked/H/T-REC-H.751-201303-I_PDF-E/5b8a756d9a71c35f17db8bcb90b438a3_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0bb03fa230f782939567a399c955f46c428d61ecd7e5b8c1b35c9dc02f68e254 +size 49230 diff --git a/marked/H/T-REC-H.751-201303-I_PDF-E/84e5b251aa38db522f76f5cc3afcb853_img.jpg b/marked/H/T-REC-H.751-201303-I_PDF-E/84e5b251aa38db522f76f5cc3afcb853_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ac44a04e876b582cb9525221facd0c66b3398b76 --- /dev/null +++ b/marked/H/T-REC-H.751-201303-I_PDF-E/84e5b251aa38db522f76f5cc3afcb853_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9a92cd77115d0bf506f9d61a27d03c3ac7a23c246c971513d7b501239e989d59 +size 48820 diff --git a/marked/H/T-REC-H.751-201303-I_PDF-E/a05e675f8651ae7ccea1d0d68691d1a9_img.jpg b/marked/H/T-REC-H.751-201303-I_PDF-E/a05e675f8651ae7ccea1d0d68691d1a9_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..b568b24ab48e345d194acc19285263b9a393e501 --- /dev/null +++ b/marked/H/T-REC-H.751-201303-I_PDF-E/a05e675f8651ae7ccea1d0d68691d1a9_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f7157dc267d284aad7be10aaac5ff88c05e5b460439dc99e34a0be301aad7ae3 +size 33533 diff --git a/marked/H/T-REC-H.751-201303-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg b/marked/H/T-REC-H.751-201303-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..b310ac210c3a7914165adf06403f98576a85c994 --- /dev/null +++ b/marked/H/T-REC-H.751-201303-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0953d8a48897d2f76d40d07a4f72e9fd2630b3ba8f1e39b9ba5bc493583058ef +size 3635 diff --git a/marked/H/T-REC-H.751-201303-I_PDF-E/aaf3e6e44cdeabd6d1df869c5f392ea1_img.jpg b/marked/H/T-REC-H.751-201303-I_PDF-E/aaf3e6e44cdeabd6d1df869c5f392ea1_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..5737b2a3d3deec6041daf3c598e0a626a4fe9f08 --- /dev/null +++ b/marked/H/T-REC-H.751-201303-I_PDF-E/aaf3e6e44cdeabd6d1df869c5f392ea1_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2b0b33fcc3b25c046bd4f923c001cdb108c389b5f512f2c9bef557f09f9a6e3a +size 102347 diff --git a/marked/H/T-REC-H.752-201511-I_PDF-E/40f30e4d577a17052f8b1e6dc802a0d8_img.jpg b/marked/H/T-REC-H.752-201511-I_PDF-E/40f30e4d577a17052f8b1e6dc802a0d8_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..451c14fa03f76313ffa73fdf0cc03e04b85fbd24 --- /dev/null +++ b/marked/H/T-REC-H.752-201511-I_PDF-E/40f30e4d577a17052f8b1e6dc802a0d8_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:246a061d9944d1c84a644ff642fbdee7c1f514a2f93d01773b1a091be3b9ebb0 +size 40690 diff --git a/marked/H/T-REC-H.752-201511-I_PDF-E/81a4cbf0b3c4cbc065efdf8f800dadde_img.jpg b/marked/H/T-REC-H.752-201511-I_PDF-E/81a4cbf0b3c4cbc065efdf8f800dadde_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..db7444654a5d5cf2537e1a918cc5a17d8d4d78d7 --- /dev/null +++ b/marked/H/T-REC-H.752-201511-I_PDF-E/81a4cbf0b3c4cbc065efdf8f800dadde_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d63db10095175420e2311401c4b75ba2c7233cc599d9ccda656c4e164e943b28 +size 41945 diff --git a/marked/H/T-REC-H.752-201511-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg b/marked/H/T-REC-H.752-201511-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..d029dd4fbf2b92152823b8298396c96c7282e426 --- /dev/null +++ b/marked/H/T-REC-H.752-201511-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:476aa07e8e41a13b741a431c04ad0f9bd3a7bd0a3822f14a669b12f2abaf0e0d +size 3828 diff --git a/marked/H/T-REC-H.752-201511-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg b/marked/H/T-REC-H.752-201511-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..ab033dd53f3754372f977b645b0413f5976d1874 --- /dev/null +++ b/marked/H/T-REC-H.752-201511-I_PDF-E/cfef993dcc8fb513de79eb1f93cf26ae_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7ad77e064e91e6dae3a7f5c5e1c668c3e9bf8c3e4f3fde84599e4da0eef4d387 +size 90208 diff --git a/marked/H/T-REC-H.830.1-201704-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg b/marked/H/T-REC-H.830.1-201704-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..bd7c24e83d75a6b1baa7fb4210f76345166390c9 --- /dev/null +++ b/marked/H/T-REC-H.830.1-201704-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5866ab521163bd4f911ae6e27d0a1cf1f491c665e9b87b7d7076098a143f4bff +size 3944 diff --git a/marked/H/T-REC-H.830.9-201704-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg b/marked/H/T-REC-H.830.9-201704-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..71ed3c0020d190a2d540306cfa27958ba9631e0e --- /dev/null +++ b/marked/H/T-REC-H.830.9-201704-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aa0958fa01dbc1a49b82df2d67f8f7fe3d8c908be26a9aa58bbd633a65d02c76 +size 3952 diff --git a/marked/H/T-REC-H.842-201911-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg b/marked/H/T-REC-H.842-201911-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..9470bdd196ceb2c96f43350ca01e20d42262056e --- /dev/null +++ b/marked/H/T-REC-H.842-201911-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b3679e5ea0a64fee7e562a2e09065769675e8693575948e6e7703a6ba88c575e +size 4013 diff --git a/marked/H/T-REC-H.845.13-201704-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg b/marked/H/T-REC-H.845.13-201704-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..a713c9d3a38a0b0ded423795a4e6fb9889f1cf7e --- /dev/null +++ b/marked/H/T-REC-H.845.13-201704-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5ac2e2cf21c51a63df761924b63730e9562d1cb9210d5b37a96e62976d508279 +size 4125 diff --git a/marked/H/T-REC-H.845.2-201808-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg b/marked/H/T-REC-H.845.2-201808-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..cd85c5f0cf29767d5fce2cdb865aece52387c01d --- /dev/null +++ b/marked/H/T-REC-H.845.2-201808-I_PDF-E/14a22f23ced8ba1d63ece69861dbaacc_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:09082581f21d9548df66739aba3c0950a4cd19b319365b8f1c69a3129a386b3c +size 4070 diff --git a/marked/H/T-REC-H.845.7-201704-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg b/marked/H/T-REC-H.845.7-201704-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..529e4f72f14cdf7f41f56cfbfb26655bf8181c59 --- /dev/null +++ b/marked/H/T-REC-H.845.7-201704-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:abb9e3ec6dfc8364cedc39a68cf796e9d3ff5529c79a1811eedeb12c6d6e7583 +size 3787 diff --git a/marked/H/T-REC-H.850-201911-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg b/marked/H/T-REC-H.850-201911-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..eecea66e6f8b7943bd3a02f8d6352624927c66cd --- /dev/null +++ b/marked/H/T-REC-H.850-201911-I_PDF-E/a3dc41dc3df86ea68d266af2bf95cf5b_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d3429ad8405a074d669fdc452abd60c4606b6ea5176105a922b2228634392d7d +size 3700 diff --git a/marked/H/T-REC-H.850.1-202008-I_PDF-E/e8ba5d4a3a22e24e44f7935ea26afcb0_img.jpg b/marked/H/T-REC-H.850.1-202008-I_PDF-E/e8ba5d4a3a22e24e44f7935ea26afcb0_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..2ae5038e42e5b0b3a2c7a9a699311e757edc4ee1 --- /dev/null +++ b/marked/H/T-REC-H.850.1-202008-I_PDF-E/e8ba5d4a3a22e24e44f7935ea26afcb0_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b67fcb836f55d09b337bf8854477a6bbfd8bed388699b21ceee6dc06427ab22 +size 3728 diff --git a/marked/H/T-REC-H.850.6-202008-I_PDF-E/e8ba5d4a3a22e24e44f7935ea26afcb0_img.jpg b/marked/H/T-REC-H.850.6-202008-I_PDF-E/e8ba5d4a3a22e24e44f7935ea26afcb0_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..f8cebc2c809c6cea575a387f92a98cab7db28208 --- /dev/null +++ b/marked/H/T-REC-H.850.6-202008-I_PDF-E/e8ba5d4a3a22e24e44f7935ea26afcb0_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:23ed41e6f076554ee4096c5faa5c47d494277049862427daf8ffe1976f506a4f +size 4182 diff --git a/marked/H/T-REC-H.862.6-202309-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg b/marked/H/T-REC-H.862.6-202309-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c1629394c02f697f60ae954ad36bf274fa8082cb --- /dev/null +++ b/marked/H/T-REC-H.862.6-202309-I_PDF-E/0538daaa5583c23e17db3a12f2281a55_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3763acd7a0ed56aa30f3f90468c1fb0cb54573eb9045bca3eddf8976b4f0f41d +size 7219 diff --git a/marked/H/T-REC-H.862.6-202309-I_PDF-E/7f17c430b9598e4d748a8041457810b3_img.jpg b/marked/H/T-REC-H.862.6-202309-I_PDF-E/7f17c430b9598e4d748a8041457810b3_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..c074673f4732370c1de58e1422081d83a48b6ccb --- /dev/null +++ b/marked/H/T-REC-H.862.6-202309-I_PDF-E/7f17c430b9598e4d748a8041457810b3_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cd13f04d957abcc3b14852229c5c5418374f06216990a23972d7946875142d26 +size 102405 diff --git a/marked/H/T-REC-H.862.6-202309-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg b/marked/H/T-REC-H.862.6-202309-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg new file mode 100644 index 0000000000000000000000000000000000000000..43cb95a230ef68d17e4a5bba6bc6d302111e6f52 --- /dev/null +++ b/marked/H/T-REC-H.862.6-202309-I_PDF-E/ebff22fb5dd6f50a90e44dca0f82f285_img.jpg @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:305f3563b00f77089482b95eaa72add77b08a66077b4df3d0f8b1968abe35e89 +size 60443